Split from
here:
NES flash cart is more complex than anything Tototek has done because of all the mappers.
I wish I could do preorders on my NES flash cart (
www.nespowerpak.com) but I have learned that is a bad idea when dealing with China fabs. Main part left is making new cart plastics. I have the solid model now so I am starting to get production quotes. I also have a PowerPak Lite ram board that supports 8 mappers and is writable through CopyNES. All these and more will be on a website when I get those new case plastics...
o_O Where does one get THAT?
Nice. If that goes into mass production...
Quote:
"I gotta gets me one a deez."
bunnyboy wrote:
I wish I could do preorders on my NES flash cart (
www.nespowerpak.com)
Me, too! You better post it here.. first!
NC
Looks very nice. This could be the NES counterpart to the GBA SuperCard.
Will programs other than the menu be able to read and write from the FAT file system on the CF card?
I see the CIC slot in your prototype board is unpopulated. Do you plan on using CIClone, -5 V, or just an instruction leaflet to cut lock pin 4?
And how much do you intend to charge?
tepples wrote:
Will programs other than the menu be able to read and write from the FAT file system on the CF card?
Getting other apps to use the CF card would be an interesting process. The game would have to load some code into the internal sram, that code would have to enable the boot rom and call the card routines. Would get pretty messy with all the variable space that could overlap but it is possible.
tepples wrote:
I see the CIC slot in your prototype board is unpopulated. Do you plan on using CIClone, -5 V, or just an instruction leaflet to cut lock pin 4?
I will be populating it with a multi region Ciclone, but there is also space for a Nintendo CIC if someone wants to add one for some reason. Theres really no reason to use 5V tricks now that the Ciclone is done. Maybe someone should refurbish Color Dreams etc carts to use it
tepples wrote:
And how much do you intend to charge?
Somewhere in the range of $120-140, will depend a lot on the injection molding price and if I want to make boxes too. That looks unlikely now because places have a 1000 box minimum for printed paperboard thats die cut. Price would have been at the low end if [michael] didn't disappear from #nesdev
nice work! Me and my friend (and probably others I know too) will order one as soon as we can
Will it be possible to develop other custom mappers on it too? that would be awesome
Still hopeful this project can see the light of day... keep it up!
(And try to avoid getting Dugg when you start selling...)
bunnyboy wrote:
Getting other apps to use the CF card would be an interesting process. The game would have to load some code into the internal sram, that code would have to enable the boot rom and call the card routines.
A few 8-bit operating systems loaded much of their code into a part of RAM that was bankswitched with the boot ROM. For instance, ProDOS on Apple II left behind a 256-byte stub for the "MLI" syscall dispatcher, which bankswitched to the OS, looked up the syscall in the jump table, and jumped to it. It's not too messy if the standard ld65 link script for homebrew designed to use this cartridge specifies that, say, $7F00-$7FFF is reserved for use by the operating system.
Quote:
tepples wrote:
And how much do you intend to charge?
Somewhere in the range of $120-140
That sounds acceptable, especially for early adopters like us. But if this enters larger-scale production, can the price come down?
bunnyboy,
What percentage of the NES/Famicom game library will work perfectly on your cart? Can you put up a video on youtube, showing your cart in action?
I certainly hope that cart goes into production!
It would make my collection of hacked carts less painful to use! (burning EPROMs and digging through piles of all the different cart types).
I'd like a spot in line to pick one up, thanks.
Its not a matter of IF it will go into production, its WHEN I will have all the parts to sell it. The first 100 boards are already made. Waiting on new cart plastics which I hope will be 3-4 weeks but that is China schedule so it could be 6 weeks... When the first batch is sold I will be able to lower the price but I have no idea how long that will take.
Right now I have all USA mappers except mmc5 working (some may be buggy), which is all but ~10 games. Only USA game that will never work is Action 52 because it has too much prg. Also have about 30 famicom/pirate mappers done but I dont know how many games that is. In time the only carts that will never work are FDS and anything bigger than 512KB prg or 512KB chr. Custom user designed mappers are also possible. All the mappers are stored on the CF card so they are very easily upgraded.
I can get a video together next week, must sleep before weekend of bike races!
If there is going to be a long delay would you consider selling the board without the plastic? Would it be particularly hard to put it into another plastic cartridge case (after some suitable modification with a dremel)?
Are you going to release the documention so we could make our own mapper files?
Will your board work with the GameGenie and ProActionReplay?
Thanks for the great work so far.
In an earlier thread, it was stated that the Power PAK has built-in GameGenie support. I wouldn't use a real GG because you don't want to change the boot ROM by accident (even 8-letter codes aren't guaranteed to be safe here). The Pro Action Replay, as I recall, modifies RAM (through a special NMI handler that temporarily overrides the cart's NMI routine), which can cause problems while the boot ROM is running if it modifies a critical section of RAM. The PAR also maps in registers in the $5xxx range (if I remember right), so if the boot ROM uses that register space, there may be further conflicts.
Long story short, I would not attach any cheat device to this cart.
perhaps you may consider selling DIY kits at a lower cost?
also if the plastic case is a bulk of the cost on your end I think meny would be move than happy to skip it
I'd like to reserve a slot for one of the first batch if that's possible. I'm not concerned if it has a case or not; I need to adapt it to FC anyway.
Possible to hack up a grey Nintendo plastic cart to work?
I didn't see a center hole screw.
drk421 wrote:
Possible to hack up a grey Nintendo plastic cart to work?
I didn't see a center hole screw.
I hope the final plastic shells are clear like in the picture of the prototype, as it looks very much like a Game Action Replay
loopy wrote:
Seconded. I'd pick up a bare board right away if it's documented (or at least has some sample mapper code).
Epicenter wrote:
I'd like to reserve a slot for one of the first batch if that's possible. I'm not concerned if it has a case or not; I need to adapt it to FC anyway.
Yeah, if it is possible to reserve them, let me/us know. I'll put down a down payment, or whatever I need to do
NC
No Carrier wrote:
loopy wrote:
Seconded. I'd pick up a bare board right away if it's documented (or at least has some sample mapper code).
Epicenter wrote:
I'd like to reserve a slot for one of the first batch if that's possible. I'm not concerned if it has a case or not; I need to adapt it to FC anyway.
Yeah, if it is possible to reserve them, let me/us know. I'll put down a down payment, or whatever I need to do
NC
same here
Same here. I'm good for one.
-Rob
I want that device the very day it come out !!
Plus, I may be buying a series of plastic cases this summer or this fall when it'll be time to complete and sell hard copies of Dragon Skill (however, that's another topic).
Excellent! Original hardware + lots of ROMs = BIG WIN
If the price isn't crazy, I would love one of these.
flojomojo wrote:
Excellent! Original hardware + lots of ROMs = BIG WIN
If the price isn't crazy, I would love one of these.
Well although I am interested in this its still emulates the memory mappers using the FPGA and there is bound to be a few bugs/inaccuracies its not quite Original hardware, but the next possible best thing in one normal sized cart
I am definitely in for one of these things! Awesome work bunnyboy!
If it is possible to reserve, count me in for one
CF
Bunnyboy, just in case you are keeping tabs on how much interest there is, I am also interested. Would purchase it at any price - it's the device I've wanted for 20 years, and have been (unsuccessfully) designing for at least 3 years.
Your Compact Flash design, which uses the 6502 to access the memory, is a design I avoided because of the difficulty of doing FAT32 in 6502 assembly. But it certainly simplifies a whole bunch of other parts of the design, and reduces cost as well. Plus, I absolutely love that you've used the Ciclone and that you are doing your own housings. I know it's easy enough to scrap an old cart to get a donor CIC or housing - but I hate the idea of people destroying their carts, they are practically antiques at this point.
Great job, you've got my business.
So who's doing the smashmypowerpak.com video?
No seriously, I'll buy one..
Long time luker forced to register just to say:
About sodding time! ^_^
You can count me in for buying one, with or without a case. (got a dead cart I can use)
I'll definitely want one.
Al
I also want one for sure. This is the coolest NES project I have seen in a very long time. Great work Bunnyboy. Do we have any kind of time frame for this right now, or is it still in the early stages?
~~NGD
Count me in as interested too.
If memory serves me right, the boot ROM in the cart was designed to be upgraded through CopyNES (something I don't have). Is CopyNES required to do a boot ROM upgrade? If so, I will definitely not be buying until enough time has passed to ensure there are no issues with the boot ROM code. (I probably won't be getting one of the first 100 anyway, mainly due to price.)
The only issue that still bothers me (though it only affects a small number of games) is INES header ambiguity. How (if at all) does the Power PAK handle SOROM and SXROM boards (MMC1 variants with 16K/32K of WRAM)? Also, how is the MMC6 distinguished from the MMC3 (are StarTropics 1 and 2 playable)? If NES 2.0 ever got anywhere, this would be easier to deal with (whatever happened to NES 2.0 anyway?).
Looks great. I would be interested in one, finished product, in shell.
Does the boot ROM configure the FPGA? If so, it would definitely be a good idea to allow for the boot ROM to be upgrade via a special program placed on the compact flash card, so that users can update the boot ROM without a CopyNES. Otherwise, I'd suggest only selling the first batch to people who already own a CopyNES. People that own a CopyNES will be more technically inclined, and capable of upgrading the boot ROM in order to fix incompatibilities. Then the later batches can be sold to more casual gamers.
Can you still buy CopyNES's? I sent kevtris an email a few weeks ago and never heard back.
~~NGD
It would seem to me it must be software configurable through the CF card and PC with a card reader. Once something like this requires another third party device, you start peeling off potential users and splinter/dilute demand.
If possible, it's best to reduce the likelihood of V.1, v 1.5 etc hardware revisions as much as possible. Unless the goal here is to truly only serve the needs of the most hardcore of NES users (a subset of a subset of a subset of gamers!)
Better a delay, than a crippled, completely commercially un-viable first version, at least IMO. The community can very quickly turn negative on a project, just from perceptions. For bunnyboy's and his projects sake, this hopefully won't occur.
Also, I think bunnyboy should be permitted to charge a price that allows him to make a few bucks, if not now, hopefully in the future. Hard work has value. No one here has really complained, and hopefully they don't. It is obvious many many hours have been poured into this.
Bravo!
dvdmth wrote:
The only issue that still bothers me (though it only affects a small number of games) is INES header ambiguity. How (if at all) does the Power PAK handle SOROM and SXROM boards (MMC1 variants with 16K/32K of WRAM)? Also, how is the MMC6 distinguished from the MMC3 (are StarTropics 1 and 2 playable)? If NES 2.0 ever got anywhere, this would be easier to deal with (whatever happened to NES 2.0 anyway?).
If the flash carts read iNES 2.0 ROMs, then this definitely isn't a problem as long as the boot ROM has this implemented and can programm the FPGA to emulate the MMC1 with different pin functions.
its dose seem very interesting, be sure to let us know when your getting close to being ready to sell
I've been waiting a *very* long time for something exactly like this. Count me in for one as soon as they are ready.
Wow, this thread is bringing all of the lurkers out of the woodwork!
NC
If even the lurkers are showing up, I guess I should say I'm very interested in getting one of these too!
My only fear comes from the fact that I don't own a copyNES... and have no interest in getting one. Will I really be unable to update the ROM? I guess I wouldn't buy it if that was the case.
as long as the .bin is supplyed and the boot rom can reprogrammed with a willem its not a problem for me, it being a serface mount chip cant stop me. Although if that is the case pads to connect progam wires to would be handy
peppers wrote:
as long as the .bin is supplyed and the boot rom can reprogrammed with a willem its not a problem for me, it being a serface mount chip cant stop me. Although if that is the case pads to connect progam wires to would be handy
That's not really going to work, since the program lines on EPROMs/Flash Roms are held at ground in read mode, and need Vcc to program. Only way to reprogram would be to pull.
P.S. Add me to the list please.
I am sick so hopefully this is coherent!
Quote:
Will your board work with the GameGenie and ProActionReplay?
GameGenie devices would probably break the boot rom, which is why I added support for 5 codes. There is space on the fpga for more Game Genie hardware so that 5 could be upgraded later.
Quote:
perhaps you may consider selling DIY kits at a lower cost?
I will not be doing bare boards, the chips are not hand solderable. I will also not be skipping the plastics because this will be the jumpstart to get them for homebrew and other projects.
Quote:
if it's documented (or at least has some sample mapper code).
I should be releasing good docs so people can build their own mappers, but that may take a while. I have not yet decided how much of my source and schematics I will have available.
Quote:
still emulates the memory mappers using the FPGA and there is bound to be a few bugs/inaccuracies
There are bugs now, there will be bugs later, yes I will be very open about known bugs
That is the whole reason its upgradable! Stuff like video colors and audio pitch should be fine because that is still using the real NES hardware. The board does NOT work in any clone I have tested (yobo, nex, fc twin) and I doubt I will try to figure out why.
Quote:
Do we have any kind of time frame for this right now, or is it still in the early stages?
*Hopefully* 2-4 weeks. Boards are done, need to get cart plastics, print labels, build website, etc.
Quote:
Is CopyNES required to do a boot ROM upgrade?
CopyNES IS required to update the boot rom, unless you unsolder it from the board. CopyNES is NOT required to update mappers or add new mappers. Those files are stored on the CF card.
The boot rom is not nearly all the code. The only thing it does is reset the cf card, check for file system etc, then load up a code module from the card. Modules can use the boot rom to do things like file io and graphics but they could also have their own instead. It is much like the CopyNES where small modules are loaded into ram. All of the mapper fpga files are on the CF card. The only boot rom bugs that I can think of would be if the cf card has a broken file system, or the scrolling in pal systems (may get garbage on top and bottom lines).
For those that dont have a CopyNES I will likely have a free upgrade service for critical stuff. Hopefully it will never be needed!
Quote:
INES header ambiguity
I will be adding iNES 2.0 so right now the wram saving for games that need more than 8KB is buggy. That is a good example of the code modules, the game loading code for each mapper is just a file on the CF card so it can be updated anytime.
See
www.nespowerpak.com/mappers.png for the current mapper support level. Not all bugs are marked yet (like big wram saving) but gives a general idea where its at now.
bunnyboy wrote:
CopyNES is NOT required to update mappers or add new mappers. Those files are stored on the CF card.
Ah, perfect! Count me in then.
bunnyboy wrote:
See
www.nespowerpak.com/mappers.png for the current mapper support level. Not all bugs are marked yet (like big wram saving) but gives a general idea where its at now.
What is "bad mapper"? I noticed that the mappers with the Officer Ugg symbol in
Kevin Horton's FPGA NES mapper grid are marked as "bad" in your grid, but there are others. Are pirate multis generally marked as "bad mapper", or only when all known dumps are bigger than 512 KiB PRG?
9 (P*ROM) buggy, 10 (F*ROM) good. Aren't they the same thing except for the difference between 8 KiB and 16 KiB banking?
99 (Vs. Unisystem, similar to CNROM): How are you handling the DIP switches? And does the loader support ROM patches to change the palette data, or must ROMs be pre-patched on the PC first?
As long as NES 2.0 is supported, and as long as the boot ROM is thoroughly tested (as it sounds like it has been), I no longer have any concerns. Hoping for the best!
BTW, can you offer a text version of the mapper grid? (Not a biggie.)
Quote:
What is "bad mapper"?
Bad mapper is any of the FFE hacked ones, anything with more than 512KB prg or chr, and the FDS. There shouldnt be any mappers that have bank switching too small to work.
Quote:
9 (P*ROM) buggy, 10 (F*ROM) good. Aren't they the same thing except for the difference between 8 KiB and 16 KiB banking?
One bg row in punch out goes weird so there is a bankswitching problem, 10 likely has the same bug but it didnt appear in the games I tried.
Quote:
99 (Vs. Unisystem, similar to CNROM): How are you handling the DIP switches?
DIP switches will not work, can't physically set them without something attached to expansion port. Will have to buy a kevtris console for that one!
Quote:
And does the loader support ROM patches to change the palette data, or must ROMs be pre-patched on the PC first?
Right now it copies the rom file exactly, so it would have to be patched before going on the CF card. The loading module could patch after loading but it would have to be able to identify the rom first. CRC would slow down the loading, dont know if it would be too slow?
Thanks for pointing things like the DIP switches out, its exactly what I need to put on the website.
It will be a good idea to build a game compatibility list. Maybe a handful of early adopters can split the total NES game list into evenly divided chunks and play test each game? Maybe a phased release would be a good idea. First sell to a small number of early adopters that have a CopyNES, let them help test it on all official Nintendo-made systems, test every game, and test it on PAL systems, etc.
After the initial release, a second release can be done for more casual gamers.
New member here, long time AA forum member... Spent the last 30 minutes mopping up the drool I made when I first heard about this.
I got the news from atariage.com forum and I would love to have this one as well. Add me to the buyer list, I'd need to scrape up some money first.
I didn't see this mentioned:
#1 Would the cart work in NES deck whose lockout chip was disabled by sniping pin 4? This mod only affected one cart the NWC cart but that is a rare one and most will never own it. If your CIC clone behaves like standard Nintendo CIC, it can probably ignore disabled lockout chip and work fine.
#2: Would thicker Type II/III CF and Microdrive work as well? Actually I shouldn't be worried abou tthat as the 16MB CF card I have will hold more than a dozen NES games.
Wait, 2-4 weeks, really?? OMG I am so hitting my bond fund.
cd_vision wrote:
OMG I am so hitting my bond fund.
007 was an N64 game, not an NES game. Or did you mean James Bond Jr?
I'm interested too, so if you do have a list of buyers, count me in
uzumaki wrote:
#1 Would the cart work in NES deck whose lockout chip was disabled by sniping pin 4? This mod only affected one cart the NWC cart but that is a rare one and most will never own it. If your CIC clone behaves like standard Nintendo CIC, it can probably ignore disabled lockout chip and work fine.
My NWC cart works fine in my pin-4-snipped NES. Never had a problem.
I know the NWC cart uses the Reset line for something critical that other carts don't use...were you confusing it with that?
Oh, and will this cart play the NWC rom (dip switches an issue there too)? I know MANY people have been dying to try this cart on real hardware. This may be the first real opportunity people have to do so. Without dip switch settings, I think the game hovers around 5:00 something instead of the standard 6:21 competition time.
-Rob
DIP switch support can be done if the VS System ROMs are patched to read DIP settings in a different manner (i.e. by accessing special registers mapped in by the Power PAK to hold DIP settings). The user would have to select DIP settings prior to game startup. The ROMs will have to be patched anyway to overcome the various security mechanisms Nintendo had in place, as well as to offer a way to insert coins through a controller button press (it's hard to play a game without money, unless its DIP settings are set to Free Play of course).
The only game whose DIP settings will work without modifications is the NWC.
It's probably too late or not a good idea for other reasons but it would have been nice to have a DIP bootrom that we could just pull out and re-program ourselves.
rbudrick wrote:
uzumaki wrote:
#1 Would the cart work in NES deck whose lockout chip was disabled by sniping pin 4? This mod only affected one cart the NWC cart but that is a rare one and most will never own it. If your CIC clone behaves like standard Nintendo CIC, it can probably ignore disabled lockout chip and work fine.
My NWC cart works fine in my pin-4-snipped NES. Never had a problem.
I know the NWC cart uses the Reset line for something critical that other carts don't use...were you confusing it with that?
Possible. I know that at 1990 tourney several NES decks were all connected together. Maybe that was how it's used and not needed for single player mode?
Either way if it works, great. I haven't played that one since I was at Detroit tourney March 1990.
If the bootrom is upgradeable via copyNES (which I have at least
), does it do a serial write to upgrade? (Just wondering if it programs over 1 pin or mutliple) Shouldn't be *too* bad to wire a cart slot to a pc and make a small rom upgrade ap.
uzumaki wrote:
I know that at 1990 tourney several NES decks were all connected together. Maybe that was how it's used and not needed for single player mode?
If only so that all three players' player 2 start buttons could be wired together.
gannon wrote:
If the bootrom is upgradeable via copyNES (which I have at least
), does it do a serial write to upgrade? (Just wondering if it programs over 1 pin or mutliple) Shouldn't be *too* bad to wire a cart slot to a pc and make a small rom upgrade ap.
why would it be nessaserry to wire to the cart slot, is there a reason simpley soldering wires to the pins and useing a programmer cant work?
it would not be overley difficult for me to pull the chip or maybe cut a few of the traces and repair them when done but I cannot see why that would be nessasery, I know bunnyboy said boot rom reprogramming would probubley be unnessasery but I still want to be able to. If its "held in read mode" like this "leonk" said then I do not see how the GG would be able to screw it up the onley thing the way I see it witch could be held in ground to disable wrights would be WE but that would be unnessery and I do not see why this would be done
Sure you could solder straight to the card edge or PCB, but those options would require opening of the case and excess solder on the card edge could inhibit insertion to a slot (more likely to inhibit non-official/non-zif connector types)
As far as the GG, I'd assume register mapping/bootrom conflicts would be the reason that it wouldn't work, but I've never looked at the tech behind the GG much beyond what it does.
I ment the pins of the "Boot ROM" not the cart connector it would be easyer to devise a programming methade if I knew what it was
You do realize that my post covered that method, correct?
Anyways, guess I can wait the few weeks for a complete item, I doubt it'd kill me
um no you didint I mentions this earleyer and someone said
leonk wrote:
peppers wrote:
as long as the .bin is supplyed and the boot rom can reprogrammed with a willem its not a problem for me, it being a serface mount chip cant stop me. Although if that is the case pads to connect progam wires to would be handy
That's not really going to work, since the program lines on EPROMs/Flash Roms are held at ground in read mode, and need Vcc to program. Only way to reprogram would be to pull.
P.S. Add me to the list please.
then bunnyboy said
bunnyboy wrote:
CopyNES IS required to update the boot rom, unless you unsolder it from the board.
then I desided leonk's comments dont make sence you can power the chip from the programmer and...wont go into my full logic. but bunnyboys comment seems to conferm it
tepples wrote:
cd_vision wrote:
OMG I am so hitting my bond fund.
007 was an N64 game, not an NES game. Or did you mean James Bond Jr?
I'm pretty sure you're just being cute, but no, I'm talking about an ETF in bonds I hold with my stockbroker. BTW not to get way off topic here but if anyone wants more details, just PM me. You get very respectable returns without volatility and risk. If you have been socking away funds in CDs this is a good alternative to that. You should be able to buy shares with any broker.
Getting back on topic...I have been waiting quite a while for one of these, you can put me on the list for one. It's a definite purchase for me.
Count me in on purchasing one!
Ill take one as well, better set up a PayPal account the money will be coming in fast for one of these.
ill take one too.! One question. Do the Roms come already in it? Or do we upgrade them to it? If thats the case will software be provided and will we need anything else to connect it to a computer?
parpunk wrote:
Do the Roms come already in it?
One has to be crazy in order to sell something with ROMs in it. Unless you mean public domain ROMs. This should not be a problem for anyone though... ROMs are some of the easiest things to obtain on the internet. Along with porn. Just don't ask for them in here.
Quote:
Or do we upgrade them to it? If thats the case will software be provided and will we need anything else to connect it to a computer?
You probably just copy the files to the CF card. If you don't have one, a device that makes them accessible on the PC shouldn't be expensive at all. I'm sure there are some cheap card readers that read many types of cards and plugs into your USB port.
tokumaru wrote:
parpunk wrote:
Do the Roms come already in it?
One has to be crazy in order to sell something with ROMs in it. Unless you mean public domain ROMs.
What about copyrighted ROMs whose authors have authorized redistribution? That's what (PD) actually means in practice.
Quote:
This should not be a problem for anyone though... ROMs are some of the easiest things to obtain on the internet. Along with porn. Just don't ask for them in here.
What about (PD) porn ROMs?
Quote:
Quote:
Or do we upgrade them to it? If thats the case will software be provided and will we need anything else to connect it to a computer?
You probably just copy the files to the CF card. If you don't have one, a device that makes them accessible on the PC shouldn't be expensive at all.
CF to NES adapter: Below $150.
CF card: $10 on Froogle.
CF writer: $5 on Froogle.
Seeing homebrew come to life: Pr*c*l*ss.
I've been looking forward to this since you first posted about it a long while ago. Just let me know where, when, and how and my wallet is ready.
I want one! Who do I have to kill to get one?
Pricing?
its seems pretty fair to say that everybody here wants one
We are lucky everyboady is in differint parts of the world or there would be fights
Count me in too. I've been looking forward to this for a long long time.
bunnyboy do you have a waiting list ?
If so please add me to it.
I want one to!
Just in case you do keep a list.
Another lurker from the old PC Jr. board. I don't have a copyNES yet, just plenty of time for testing. I should also mention that I've got more than a few friends that would purchase this as well.
I dont mean to be a pest but I would very much like to get an update as to how things are comeing along
Manual printer arrived this morning, label printer arrives monday, Ciclones are hopefully on their way, cart plastics are shipping from HK soon, tons of boring stuff coming together
Got one more try to find a box... Haven't had any time to do mappers so no updates there.
As always it is 2-4 weeks away from being sold. No accurate estimate is the reason there is no list, no preorders, etc.
Yay label printer just arrived early!
Fixed punctuation on the title since I didnt actually start this thread, correct name is PowerPak (not Power Pak, Power Pack, Power Flash, Flash Pak, or any other incorrect versions people have sent me
)
Good to hear some news. I should have asked earlier, how well do you expect the PowerPak to perform on a PAL machine?
Hey, I just trought something... This cannot be a hoax, right ? Because I'm going to be so depressed if it is... I just cannot wait to get one.
To kill time, I'd be wishing to solve something : Are the 512kb CHR and 512kb PRG SRAMs, EEPROMs or Flash ROMs ? I guess the 512kb CHR is SRAM, because there is no other SRAM for games having CHR-RAM, so this definitely must be SRAM. But since there is additional 32kb of PRG RAM, the main 512kb of PRG can be either type of memory.
bunnyboy wrote:
...Ciclones are hopefully on their way...
I don't like the sound of that. Have you heard from Kevtris? I've gotten the impression (based on people trying to get a CopyNES) that he might be AWOL (I hope I'm wrong, though).
not sure if I have already posted but put me on the list =)
Bregalad wrote:
This cannot be a hoax, right ?
If it was a hoax I would say something like: I cant take pictures because my NES just broke and my brother crashed his car which had my camera in it but the board now plays Hellraiser perfectly!
Bregalad wrote:
Are the 512kb CHR and 512kb PRG SRAMs, EEPROMs or Flash ROMs ?
512KB chr = sram, 512KB prg = sram, 32KB wram = sram, 4 screen ram = sram inside fpga, boot rom = flash. sram is everywhere because there is no write delay or limit.
dvdmth wrote:
Have you heard from Kevtris?
The trick is not to use email
I don't think he is answering CopyNES questions because he only has a few boards (no parts) and nobody is working on the Windows host software anymore. For the Ciclones he just had a busted programmer so its taking a bit longer than expected.
If the NES PowerPak is a big success, I hope to see "PowerPak" carts for other systems. Other than the GBA and DS, flash carts for other systems leave allot to be desired.
FDS will be the next (cheaper!) board, almost all the design will directly transfer over so it should go fairly fast. SNES is also on the list because the FAT code will port quickly. I mostly skipped the SNES generation consoles so it is less interesting
The overall architecture might be massively overkill for Atari etc, don't know what types of universal carts are already available there. I wouldnt want to do the file list display code on 2600....
Questions for BunnyBoy:
* How will PowerPak deal with games that uses batterybackup? Will this be saved on the compactflash aswell?
* What will the estimated price be for this cool unit?
* Will it work on both PAL/NTSC NES?
You're mentioning an FDS card aswell. Will this require a real FDS or would it be "emulating" the FDS?
oRBIT2002 wrote:
* How will PowerPak deal with games that uses batterybackup? Will this be saved on the compactflash aswell?
When you hit reset on the console you have an option to save battery ram back to the cf card if it was enabled. You can only overwrite existing save files, not create new ones.
oRBIT2002 wrote:
* What will the estimated price be for this cool unit?
Pretty set at $135 now, add your own CF card and reader for ~$150 total.
oRBIT2002 wrote:
* Will it work on both PAL/NTSC NES?
Should be no problem on PAL, except there may be garbage at the top and bottom of the screen when scrolling through the file listings.
oRBIT2002 wrote:
Will this require a real FDS or would it be "emulating" the FDS?
This is loading .fds files off a CF card on a normal NES. No extra hardware needed, never replace a drive belt
Hardware is easier/cheaper so the cart will be cheaper. But my list of new projects is huge so don't expect anything for a couple months...
So, pressing RESET will actually not reset the game but will bring back your menu?
How will we be able to purchase this cool piece of hardware? Will you perhaps have an online shop?
Shippingcost to Sweden?
The FDS cart seems really cool aswell, can't wait. How will you deal with FDS games that writes to disk?
oRBIT2002 wrote:
So, pressing RESET will actually not reset the game but will bring back your menu?
It does reset the game (no pausing/resuming) but starts up with my boot code instead of the game. Power is never removed so the wram is still good.
oRBIT2002 wrote:
How will we be able to purchase this cool piece of hardware? Will you perhaps have an online shop? Shippingcost to Sweden?
There will be website to buy, international rates just went up so it will be ~$11USD.
oRBIT2002 wrote:
How will you deal with FDS games that writes to disk?
Same as the wram on PowerPak, when you hit reset you will have the option to overwrite a previous .fds file with the new data. I might have to do something like padding at the end of the original file in case the new data is longer than original.
I can't wait for this to be released! Please tell us that you have enough in the initial batch to cover at least everyone who has posted in this thread!
No Carrier wrote:
I can't wait for this to be released! Please tell us that you have enough in the initial batch to cover at least everyone who has posted in this thread! :)
By the time they're ready, all 100 of them will have been spoken for, and then some!
It could be worse, though (picturing the store about to sell PS3's but only have four in stock and a huge mob of people trying to get them...).
Well if this has become the official "waiting list", sign me up too! ... assuming Samus Parker doesn't chew on any of the parts
Just for the record, I'm ordering one the day it becomes avaliable! =) I want to be at the top of that waiting list also!
@bunnyboy:
How will you deal with disk-side swapping with the FDS-cart? Also, what about loading-times, would it be alot faster? And if so, is there a risk of breaking compability?
Yeah, will it have two modes of operation?
-one for fast loading (increment pointer every poll)
-increment every 8*96KHz or so
Quote:
How will you deal with disk-side swapping with the FDS-cart?
I have thought of 4 possible ways so far:
1. physical buttons, no space, adds cost
2. watch for controller button combo to swap to next disk, game has to be pollling the controller to work
3. patch bios to do auto swap, it would just try each side thats loaded from the file until it hits the right one
4. patch bios to jump back to boot code for file listing
I don't know how many games do not use the bios to ask for disk swap, those wouldnt work with patched bios options.
Quote:
Also, what about loading-times, would it be alot faster?
The original or fast loading can easily be an option in the config screen. Don't know what other config options there would be yet. Are Game Genie codes used on the FDS?
bunnyboy wrote:
Are Game Genie codes used on the FDS?
AFAIK the Game Genie wasn't released in Japan.
lynxsolaris wrote:
Well if this has become the official "waiting list", sign me up too! ... assuming Samus Parker doesn't chew on any of the parts
AFAIK, there is no official waiting list yet. It will probably be a good idea for bunnyboy to require payment before a person is added to the waiting list. That way he doesn't have a bunch of dead beats on his waiting list.
Quote:
Should be no problem on PAL, except there may be garbage at the top and bottom of the screen when scrolling through the file listings.
Actually, I would suggest you check this up. Both Memblers's EPROM emulator and the one I built myself have had weird electrical problems giving garbage sprites on an NTSC system. (mine still does) So even if the chance of anything screwing up is insignificant, it's better to be safe than sorry.
And needless to say, you can sign me up on the waiting list as well. :)
Game Genie was not released there. Late in the FC's life (death?) the Pro Action Rocky did come out, though. No word on whether FDS works w/ it, though.
The Game Master Boy's save state works, though.
-Rob
Jagasian wrote:
AFAIK, there is no official waiting list yet. It will probably be a good idea for bunnyboy to require payment before a person is added to the waiting list.
No problem there. I already have the money set aside
Bananmos wrote:
Both Memblers's EPROM emulator and the one I built myself have had weird electrical problems giving garbage sprites on an NTSC system.
Might this be timing related? The NTSC PPU runs slightly faster than the PAL one (5.37 MHz vs. 5.32 MHz).
That said, I do agree that PAL testing should be done before selling internationally, if possible. There are a number of things that can potentially go wrong due to hardware differences (make sure the flash card is still read properly, make sure PPU rendering remains accurate and that IRQ's are triggered as they should, make sure the Ciclones work as advertised, and so forth).
Quote:
If it was a hoax I would say something like: I cant take pictures because my NES just broke and my brother crashed his car which had my camera in it but the board now plays Hellraiser perfectly! Smile
Yes, but nothing prevents you to do better hoax than what Evildragon did.
Quote:
(make sure the flash card is still read properly, make sure PPU rendering remains accurate and that IRQ's are triggered as they should, make sure the Ciclones work as advertised, and so forth).
If there would be minor IRQ and/or Ciclones problems on a PAL console, I'd definitely buy anyway. But if there would be significant problems, that would sucks very badly.
About the battery backup, this means that you have to save normally, then leave your sit, press reset, then the menu will be able to backup the WRAM to the flash card ? That's decent I think, but how do you detect resets ? Does the FPGA check for M2 being tree-states, and enable the BIOS back if it is ?
Bregalad wrote:
About the battery backup, this means that you have to save normally, then leave your sit, press reset, then the menu will be able to backup the WRAM to the flash card ? That's decent I think, but how do you detect resets ? Does the FPGA check for M2 being tree-states, and enable the BIOS back if it is ?
That's my understanding of how the cart works. You save within the game, press Reset, then tell the PowerPak to save WRAM to the flash card. The WRAM isn't battery backed, so you have to save to the flash card or else lose the saved data.
There are a couple of different ways to detect resets. You can monitor Phi-2, or you can monitor PRG A0 (I remember hearing about a cart that used this technique - it works because during CPU execution A0 never stays constant for more than a few cycles). I might point out that a number of mappers require certain registers to be re-initialized upon reset, so even if the BIOS weren't swapped back in, the FPGA would still need to know when a reset occurs for those mappers.
throw my name on the list as well.
While this would work in most cases, this would be improper, because a game doing sofware reset using something like jmp ($fffc) or just doing dummy read to $fffc or $fffd would reset the cart.
Quote:
I might point out that a number of mappers require certain registers to be re-initialized upon reset, so even if the BIOS weren't swapped back in, the FPGA would still need to know when a reset occurs for those mappers.
Nintendo mappers initialize their internal registers when they detect the supply voltage is rising on power-up, but my undestanding is that they are not affected by hardware resets at all I'm talking about the Nintendo's MMC chips. I don't know about the other ASICs.
As noted before there is no list, no preorders, no reservations
I won't do orders without taking money, I won't take money until I have a set date, and there will be no set date until the day everything is actually ready. Yes I will post to this forum first. May have to release in smaller batches so it isn't super backed up.
Reset is detected by watching M2 compared to an on board clock. Can't watch for reset vector because starting the game needs that too, and some games will read that location when its bank swapped to something else.
Just to keep the hoax going, here is an obviously photoshopped photo of the final plastics:
http://www.nespowerpak.com/redpowerpak.jpg (yes it is real) The CF slot is cnc milled, doesn't look like hand held dremel. Don't have final labels yet but the test ones are very water resistant, acid free, etc. Rubbing while under faucet did nothing
Don't have final manuals yet either, still testing color laser paper. Both the high quality labels and manuals will be on all my future products so contact me if you want your homebrew game produced.
OMFG
I think we should agree not to "digg" anything for a wile after bunny starts accepting orders just to give the fourm members here a chance
Bregalad wrote:
Nintendo mappers initialize their internal registers when they detect the supply voltage is rising on power-up, but my undestanding is that they are not affected by hardware resets at all I'm talking about the Nintendo's MMC chips. I don't know about the other ASICs.
I'm fairly sure the MMC5 detects resets. Its upper ROM bank, according to the info I have, is mapped to the last bank of ROM on power-up AND on reset, even though the bank is switchable. It's also my understanding that the MMC5 (and possibly the MMC3/MMC6) will disable WRAM upon reset. The Nintendo World Championship cart also detects reset, but I believe it gets its reset signal from the lockout chip.
There are also some pirate mappers that use a reset trigger to select which game to play (i.e. you get a different game each time you hit reset). I do not know if these games can be supported on the PowerPak, nor do I care since they are pirated games in the first place.
Quote:
It's also my understanding that the MMC5 (and possibly the MMC3/MMC6) will disable WRAM upon reset.
Actually they disable it on power-up, because I see no reason for the WRAM to be corrupted when the system is reseting itself, but I do for when the power goes up and down.
For the MMC5, mabe you're right, I have no idea about it.
Red cart...beautiful. Perfect choice for the Powerpak
-Rob
peppers wrote:
I think we should agree not to "digg" anything for a wile after bunny starts accepting orders just to give the fourm members here a chance
The forum members should be the early adopters, as we will be more tolerant of issues with the early versions of the firmware. But if it does turn out to be as good as it looks, we should all spread the word of its greatness.
I love the look of the clear red plastic cart shell. By the way, what is the status of the Nintendo patents on the shape of their NES cart plastic shells?
Jagasian wrote:
I love the look of the clear red plastic cart shell.
I agree. The red is very appealing! Perhaps, at some point, we will see a gold Zelda style shell for the pak.
Jagasian wrote:
peppers wrote:
I think we should agree not to "digg" anything for a wile after bunny starts accepting orders just to give the fourm members here a chance
The forum members should be the early adopters, as we will be more tolerant of issues with the early versions of the firmware. But if it does turn out to be as good as it looks, we should all spread the word of its greatness.
I love the look of the clear red plastic cart shell. By the way, what is the status of the Nintendo patents on the shape of their NES cart plastic shells?
sure but in due time alter the regulars here get there chance its onley fair
Ok wait, it is photoshopped? The cartridge in the other pictures has the label space on the opposite side. I was just thinking how sweet it would be to replace some of the shells on my games that have munched labels.
So the final thing, are you putting the label on the righthand side like official games, or no..?
dvdmth wrote:
There are also some pirate mappers that use a reset trigger to select which game to play (i.e. you get a different game each time you hit reset). I do not know if these games can be supported on the PowerPak, nor do I care since they are pirated games in the first place.
Unless you get into stuff like
X-Men for Sega Genesis, which asks the player to get up and press the console's reset button after defeating a boss. This does not work on a Nomad.
Jagasian wrote:
By the way, what is the status of the Nintendo patents on the shape of their NES cart plastic shells?
Patents expire 20 years after filing. Design patents expire sooner.
The last pict is correct, first picts are just mirrored. The cart plastic is close to a copy of the NES ones.
It took many mods but I verified PowerPak works fine on PAL systems, tested using Mattel Version. The top or bottom row is corrupted while scrolling through file listings so I may fix that. Didnt notice any irq problems, SMB3 works good.
To get the AUS system working I had to:
1. solder new power plug
2. disable lockout (no ciclones yet)
3. add pullups to controller
4. get pal->ntsc converter
5. use slingbox because no other tvs would find the ntsc signal
6. play in 1950s mode because only the b/w part was picked up
7. cope with ~5 sec delay for slingbox to compress
parhaps you may just want to get one of them cheap video capture cards they would work well enoth for testing sake and can display a pal signel just fine probubley would have been cheaper than the converter (around $25-15)
The PAL->NTSC converter was $6, which is probably why it doesnt really work! I have a tv card but it takes really bad screenshots.
Just posted a low quality video at
http://www.youtube.com/watch?v=pU1m_-SX7ro where unfortunately the cursor doesn't show up. I will have to do a better one that shows game genie stuffs too.
bunnyboy wrote:
Just posted a low quality video at
http://www.youtube.com/watch?v=pU1m_-SX7ro where unfortunately the cursor doesn't show up. I will have to do a better one that shows game genie stuffs too.
Awesome!
No Carrier wrote:
bunnyboy wrote:
Just posted a low quality video at
http://www.youtube.com/watch?v=pU1m_-SX7ro where unfortunately the cursor doesn't show up. I will have to do a better one that shows game genie stuffs too.
Awesome!
I agree
bunnyboy wrote:
Just posted a low quality video at
http://www.youtube.com/watch?v=pU1m_-SX7ro where unfortunately the cursor doesn't show up. I will have to do a better one that shows game genie stuffs too.
That looks like directories and long file names. What, exactly, is the file system support? FAT16 with standard long file name support, or does the PowerPak support some non-standard long file name extension?
Card is FAT32 with long file names and full directory tree, ~28 chars displayed. Only directories and .NES files are displayed when choosing a game (.NSF will be added later). Only dirs and .SAV files are shown when choosing and saving the battery ram. I was planning FAT16 too but it is even more of a mess than FAT32
No special software is needed to load the games onto the card, it is just a plain disk. Microdrives would also likely work assuming they didn't drain too much power or are too big physically.
Only the first logical partition on the CF card is used. The root dir of that partition is shown first. The folder holding all the mappers must be in the root dir (POWERPAK dir in movie), everything else you can arrange how you want.
bunnyboy wrote:
Card is FAT32 with long file names and full directory tree, ~28 chars displayed. Only directories and .NES files are displayed when choosing a game (.NSF will be added later). Only dirs and .SAV files are shown when choosing and saving the battery ram. I was planning FAT16 too but it is even more of a mess than FAT32
Darn, that shoots to pieces my plans to use the same CF card with a GBA Movie Player (PocketNES) and the PowerPak (TheRealThingNES). But anyway, how is FAT16 more of a mess than FAT32?
FAT16 has variable cluster sizes for volumes in the cf card range, and root dir is "special" and can hold only a small number of files. There are more variations of the FAT16 headers depending on which OS formats the disk, Win95 and XP do two completely different things. Also wastes more space when your drive is 512MB+, tho that isn't really a concern.
Random GBA Movie Player website (
http://www.gbamovie.com/ztt.htm) says it does FAT32, is that wrong or a different product?
The supercard may be what he is referring to it dose not support 32 but is a better device and has pocket nes bulit in they are all considered a gba movie player
I am also a supercard user and it would be nice to use the same card paticularley to play nes games off of both devices but oww well I do not mind buying a new one
If a game has battery backup and there is no .sav file present on the cf card for that game, can one be created? Something you posted earlier seemed to suggest that the answer is no, that the PowerPak can only save over existing ones. If that's the case, this can be annoying, particularly since Nestopia (the emulator I use) compresses the .sav files it generates (and there's no way to turn that off, at least not in the Mac version).
VirtuaNES is the only emulator I have which can put the .sav file in the same folder as the original file (and it will do it by default). Other emulators typically allow for placing the .sav files in a sinlge folder you chose where you want, but hardly have the option to just place them in the same folder as the ROM. So I always use VirtuaNES for RPGs, and Nestopia for action games.
bunnyboy wrote:
FAT16 has variable cluster sizes for volumes in the cf card range
In theory, FAT32 has variable cluster sizes too.
Quote:
and root dir is "special" and can hold only a small number of files.
True, a lot of file systems limit the number of items in / at formatting time. But at least I keep roms in the /nes folder.
Quote:
There are more variations of the FAT16 headers depending on which OS formats the disk, Win95 and XP do two completely different things.
Citation? I'd like to read about the differences.
Quote:
Also wastes more space when your drive is 512MB+, tho that isn't really a concern.
But even 1 GB cards come preformatted for FAT16 because a lot of digital cameras and other devices can't understand FAT32. Having to reformat cards before using them will generate more support e-mails.
Quote:
Random GBA Movie Player website (
http://www.gbamovie.com/ztt.htm) says it does FAT32, is that wrong or a different product?
Current version of GBA Movie Player works fine with FAT32, but I seem to remember that older versions of some GBA CF adapter's firmware (GBAMP? SuperCard?) had trouble with FAT32.
tepples wrote:
Current version of GBA Movie Player works fine with FAT32, but I seem to remember that older versions of some GBA CF adapter's firmware (GBAMP? SuperCard?) had trouble with FAT32.
I can tell you for a fact that the latest version of the SC fw still cannot support FAT32.
Quote:
If a game has battery backup and there is no .sav file present on the cf card for that game, can one be created?
No, it will not create any new files. You can only overwrite existing ones. Someone could add new file creation later but doing all the FAT searching and cluster allocation was too much at the time.
Quote:
Nestopia (the emulator I use) compresses the .sav files it generates
This is a known problem with Nestopia and its actually pretty stupid. It takes an 8KB file, uses zip to compress to ~200 bytes, renames it to .sav instead of .zip, then the file takes up a 4-32KB cluster of disk so no space is saved. I don't plan on coding zip for nes so the decompression will have to be done manually when copying over the files. Maybe something like Automator can be set up to do it automatically.
Quote:
In theory, FAT32 has variable cluster sizes too.
FAT32 cluster sizes don't change until the volume size is over 8GB. When I started this the CF docs said 4GB was the physical limit so I don't know how the 8GB cards exist yet
bunnyboy wrote:
Maybe something like Automator can be set up to do it automatically.
I can easily set up a script to decompress the files. My main concern is that the average end user (which you hopefully will get the chance to sell to) won't know to do this. One remedy would be to test for compressed files and tell the user if he tries to load one that is compressed.
You might also want to provide a blank 8KB or 64KB .sav file for end users who don't have emulators to use as a starting point, giving instructions in the manual to duplicate, rename, and load that file to the flash card.
bunnyboy wrote:
Quote:
In theory, FAT32 has variable cluster sizes too.
FAT32 cluster sizes don't change until the volume size is over 8GB. When I started this the CF docs said 4GB was the physical limit so I don't know how the 8GB cards exist yet
The only limit of the CF interface is the limit of the ATA version it's based on, and that's well over 130 GB (2^28 sectors). As for the save file, you could bundle a Windows based app that searches for all .nes files in a folder, looks at the battery bit, and creates a blank 8 KiB .sav file as needed.
I am curious the cluster size is what windows calls the"Allocation size unit" correct?
this can be manually set to whatever you like if you format the card under disk management so if the device expects a paticular size and it is differint by defalt such as with a 8gb it would be nessasery to set it manually?
also when I format a card I usually set this to as low as it can go doing so lowers access speeds because it has to read through more sections but if you have a large amount of small files you can fit a surpriseingley large amount of extra stuff on it because you dont have a waisted large section for the end of each file,it would also probubley be nessasery to stop that for this?
Think you might make it auto-recognize rom & sav files based on folder name or something? (so you could just point to the folder and it'd recognize that there's a rom in the folder matching the folder name, etc)
like a dir structure of
/Root/Rom_Name/Rom_Name.nes
/Root/Rom_Name/Rom_Name.sav
FAT32 support and long file names sounds great! With my gigantic clunky SWCDX2, I use a Kingbyte based parallel port external drive enclosure along with a CF-IDE adapter. It lets my SNES load ROMs and saves off of a 2GB compact flash card, but I am stuck with FAT16 and no long file name support, so it isn't possible to load every SNES/SFC ROM into one CF card
Regarding your NES file browser interface, I'd suggest that by default it boots into the game ROM listing and if possible the ROM browser also remembers the last selected game.
Yay got my video camera back so I have a much better PowerPak demo video at
http://www.youtube.com/watch?v=rR5ea3S_XaQ This one actually shows the cursor.
Also have a cool video of my cnc mill cutting the PowerPak cases at
http://www.youtube.com/watch?v=xTfeOkCmvFM
I am looking at FAT16 again because it has to be done as a boot rom upgrade, not just files on card. Hopefully everything will be selling sometime next week! I will have to edit my posts so the release date estimates are accurate...
Watching your miller was both the most boring and exciting thing I've done all week.
Goddamn I can't wait to own one of these.
I just watched the demo... it's pretty cool, although I think the file browsing could use some better aesthetics (not major).
1. There's no point in showing the "/." directory, so I'd filter it out. The "/.." directory can stick around, although you might want to rename it to something like "(go back 1 directory)" so it's less confusing to those who don't know what "/.." means.
2. If a file name is too long, I personally think it's better to show the beginning and end of the file name, with an ellipsis (...) in between. (That's how OSX's Finder shows long names, and it's nice particularly since often the only difference between consecutive entries comes at the end of the name.)
3. On the options screen (after selecting a ROM), it would help if the name of the ROM (in its entirety if possible) were displayed on the screen. You may also want an option to go back to the ROM listing (I assume you can go back by pressing B, but it's still helpful for a novice user if an explicit on-screen choice is present).
4. I counted 29 lines of visible text on the screen. How much space do you allow for overscan on each edge of the screen? My TV cuts off nearly two full rows of tiles from the top of the screen and almost nothing from the bottom. Thus, if the first file listed is within the top 16 pixels of the screen, my TV will cut it off. Likewise, any graphical artifacting on the bottom of the screen (which I can see in a number of games BTW) will be visible.
I am very impressed with what I'm seeing. You have obviously put a huge amount of effort into this project, and I surely hope it pays dividends for you. The NES has waited too long for something like this to come around, but late is much better than never at all. Keep it up!
My PayPal account is primed and ready.
What kind of warranty will these come with?
Those are good UI tips, some may make it into the final version
Filtering and renaming might be tricky but that would have to be a boot rom update. Won't be doing any auto matching of game to save because some people (aka me) have multiple saves for each rom. Displaying the name of the game and save is on my list and is just a module update. Remembering the info between resets might happen too, but that gets dangerous if the files have moved.
If you want to go back to rom listing you just hit the reset button and start over
It is assumed one top and one bottom rows are cut off, which is where the bits of old file names show up when scrolling. Nothing important is in the border ~3 rows.
I haven't worked out the warranty yet. Obviously "will not break for a long time" but guaranteeing games will run bug free cannot be a part of it. Also can't make any promises of upgrade or fix schedule.
bunnyboy wrote:
It is assumed one top and one bottom rows are cut off, which is where the bits of old file names show up when scrolling. Nothing important is in the border ~3 rows.
There are guidelines. For UI, you will want to use the
title safe area. The BBC recommends leaving 5 percent of the top and 5 percent of the bottom unoccupied by text. On the NTSC NES, 5 percent of the screen is 12 scanlines. So don't put anything important in the top two rows or the bottom two rows.
bunnyboy wrote:
It is assumed one top and one bottom rows are cut off, which is where the bits of old file names show up when scrolling. Nothing important is in the border ~3 rows.
This won't work on my TV. The name of the first file on the screen will be almost completely invisible. I'd suggest, if it's workable, to allow for 12-pixel margins at the top and bottom of the screen (this will allow for 27 entries to be visible at once). This should be enough to allow the top row to be readable on any set (mine might cut off a scanline or two, but most of the text graphics will still be visible).
In order to avoid glitches while scrolling, you will probably want to have the screen scroll a full tile at a time, rather than scrolling in smaller increments. This way, you can erase the filename scrolling out of range during the saame V-Blank in which you draw the new file name, then have the scroll occur at the same time these updates are done. If you prefer to keep the smoother scrolling shown in the demo, you'll need to do a split-screen effect in order to clip the BG display at the top and bottom, but that's probably more effort than it's worth (I'd rather keep the graphical glitches than to undergo that much effort to remove them, especially since they're confined to the edges where artifacting is common in commercial games).
Quote:
This won't work on my TV. The name of the first file on the screen will be almost completely invisible.
And that is the beauty of having the redundant /. directory displayed, that is the one that will be chopped
Only place that wont happen is the root directory, so you can just copy the powerpak directory to the card first so it will be cut.
Title safe area is good idea but it doesnt always work on crts (my tv cuts 1 row on top and
3 full rows on the bottom, and no its not the alignment or hold) and my lcd tv usually shows all the screen including stuff I want hidden. Never seen a tv that cuts the sides, that would be much harder to deal with here...
bunnyboy wrote:
And that is the beauty of having the redundant /. directory displayed, that is the one that will be chopped
TouchƩ.
The NTSC resolution is wider than the NES's output, so I don't think you can expect overscan issues horizontally (except at the corners with a curved screen edge). It's the vertical direction that's critical, and yes, CRTs do vary in terms of which lines get cut off, as well as how many (most cut around 16 lines, but there's no rule here).
You know, in light of this discussion, I finally realize why my cable box has a "screen position" configuration which changes where on screen the menus and program guide are displayed (as if it mattered).
Me (and probably many others) will be happy if upgrades are solved by software and not by having to use a CopyNES or similar.
I still do not belive it would be all that difficult to bulid a simpler programming device if it becomes nessasery after all no matter what he has on there its still just a eeprom it could even be replaced with a diferint one if its realy something strange (unless its real strange)
but as he said the mapper stuff is on the CF card and if the device is working the boot rom probubley wont need an update
peppers wrote:
I still do not belive it would be all that difficult to bulid a simpler programming device if it becomes nessasery after all no matter what he has on there its still just a eeprom it could even be replaced with a diferint one if its realy something strange (usless its real strange)
but as he said the mapper stuff is on the CF card and if the device is working the boot rom probubley wont need an update
One solution would be for the boot ROM to contain just enough code to read a ROM image off of a fixed location on the CF card. The loaded ROM would contain the file browser, game ROM loader, game save loader, Game Genie code support, etc.
That way it is extremely easy to upgrade the multi-game menu, as everything except for basic booting and FAT32, could be loaded from CF. Basic booting and FAT32 most likely won't need to be changed, while people will definitely want to change and customize the multi-game menu.
Jagasian wrote:
One solution would be for the boot ROM to contain just enough code to read a ROM image off of a fixed location on the CF card. The loaded ROM would contain the file browser, game ROM loader, game save loader, Game Genie code support, etc.
This is an excellent idea! Would it be possible to use some of the RAM in the cart to hold the code for all this stuff, loaded from the CF card? I really hope it is, 'cause I loved the idea! I just want to get the best out of the PowerPak, in spite of not having a CopyNES...
Looks like I got FAT16 working, have to do more extensive tests tho. Nobody better ask for FAT12 of they are off the (non existent) list!
I was actually really close at my first attempt but missed a few more parts of the root dir funkyness.
peppers wrote:
I still do not belive it would be all that difficult to bulid a simpler programming device
The hard part about an external programmer is the cart connector. The rest of the hardware could be done as simply as a parallel port connection and ~2x 8bit registers, or a $8 pic usb chip.
Jagasian wrote:
One solution would be for the boot ROM to contain just enough code to read a ROM image off of a fixed location on the CF card
Apparently I didnt explain the code modules very well. The boot rom basically does just load up a 1KB module from the POWERPAK dir into internal NES ram. The logo screen is one module, .nes display is another module, options screen is another, loading the game is a module, etc. Modules could load more code from the card into the PowerPak wram or prgram. Boot rom holds all the code for game genie decoding, file system processing, copying strings to the screen, etc but your module doesnt have to use those. You could even replace the first intro screen with your own code and never use the boot rom again except for nmi (if you feel like writing your own fat handling).
CopyNES should really never be needed, just if there is some major bug like card handling corrupts data. No that should never happen because no fat table writing is ever done... You should get a CopyNES anyways to use my other ram cart!
Ahh, so that's what you were referring to when talking about modules. Makes perfect sense now. As long as the boot ROM always works to the point of the first module beginning execution, there's no way we'll ever have to worry about a boot ROM update, as it can be completely removed from the system (so to speak) afterwards if it ever comes to that.
And no, I'm not going to ask for FAT12.
bunnyboy wrote:
Looks like I got FAT16 working, have to do more extensive tests tho. Nobody better ask for FAT12 of they are off the (non existent) list!
You're in luck. In practice, FAT12 isn't used for volumes bigger than 16 MB. The smallest CF card you can buy nowadays is 256 MB.
bunnyboy,
Will it be possible for you to document the NES PowerPak and provide a free dev kit, so that mapper support and custom multi-game menus can be developed by other people, in a way that doesn't involve you giving away your intellectual property?
Jagasian wrote:
Will it be possible for you to document the NES PowerPak and provide a free dev kit, so that mapper support and custom multi-game menus can be developed by other people
Original plan was to have sample mapper files (code module asm and xilinx ise files) so people could develop their own. I wasn't expecting for people to want to change the game menus so I will have to figure out how to document that well. Both are in the "later" time frame, definitely not when everything is first available next week. Any part of the documentation pretty much has to show how the whole thing works, so there is no way to protect against someone copying parts or everything. It can all be reverse engineered anyways but that would not be fun...
When I get this baby in my mailbox, I'll stay home from work for a month, coding and playing.
bunnyboy wrote:
Any part of the documentation pretty much has to show how the whole thing works, so there is no way to protect against someone copying parts or everything. It can all be reverse engineered anyways but that would not be fun...
Are there any lawyers here? What kind of patent protection is possible on a cartridge such as this? (Not sure it would help though.)
Probably very little patent protection possible, given that the SuperCard for GBA is prior art for everything except the reconfigurable mapper.
bunnyboy wrote:
Both are in the "later" time frame, definitely not when everything is first available next week.
Is your Paypal account going to be ready next week?
Ive been checking up regularly, both here and at
http://www.nespowerpak.com/ and I have just noticed that there are pictures of empty blue and clear cart shells. Do we get to pick the colour we want or is it 'get what your given'? (not that I mind, but I personally would like a clear shell so when Im not using it I can put is on display and gawp at the marvelous circuit design.
BTW, you
will be shipping to the UK, won't you? Please?
You better had do, for the sake of humanity!
The blue, clear and red cases looks like rendered images, most likely, the only real one is the other red one that is on a table (you see the light reflecting from it).
Bregalad wrote:
The blue, clear and red cases looks like rendered images, most likely, the only real one is the other red one that is on a table (you see the light reflecting from it).
I, on the other hand, think they are all real
tepples wrote:
Probably very little patent protection possible, given that the SuperCard for GBA is prior art for everything except the reconfigurable mapper.
It should be possible for him to copyright the software related such as the boot ROM not that is really matters since he would not have the means to go after any copier
It appears none of the components have any hardware protection to prevent dumping of the code I would not be surprised to see a cloned "commercial" version made by some group most likely based in Hong Kong like neoflash for example who made a genesis device last year if I remember correctly. Witch may not be all that bad of a thing depending on the motive of the developer sence it would mean more people both the team and people who bought one working on perfecting the mapper emulation
Quote:
The blue, clear and red cases looks like rendered images
I wish I was that good at photoshop!
Cases are all real, photos taken outside on top of glossy printer paper. Camera did something weird with the colors on the clear one... All will be for sale on my site but only red is being used for PowerPak and PowerPak Lite, and blue is being used for NWC. All colors can be used for homebrew production if anyone has a finished game.
peppers wrote:
I would not be surprised to see a cloned "commercial" version made by some group most likely based in Hong Kong
I doubt this would happen due to the cost. Maybe single mapper variants, but no large multimapper devices.
This design is a single-mapper device, it just has a reconfigurable portion. Hopefully Neoflash or others won't stoop so low as to copy a unprotected device.
bunnyboy wrote:
...and blue is being used for NWC.
OMG!
You are releasing NWC standalone reproduction soon too? Any idea on cost on this? Hopefully I'll have enough cash for this and the PowerPak
I just can't wait to get one...
bunnyboy, what do you think about implementing simple nsf player into the bios ? or maybe have it emulated thru some drivers on the mem card, same as with the mapper drivers.
Quote:
what do you think about implementing simple nsf player into the bios ?
NSF will be added just like any other mapper through a mappers update. Doesn't have to be in the boot rom. Choosing a .nsf instead of .nes file would likely bypass the options screen and go directly to the player, which will have play/pause/prev/next controls.
bunnyboy,
The anticipation is killing me. Have you thought of using a wiki to let the community help maintain a detailed console/game compatibility list? Maybe start it out with a row for every GoodNES V3.10 ROM, as that is what many people will be using. Then a column for compatibility, and a column for details. Since it is GoodNES, many compatibility problems will be due to bad headers, bad dumps, etc, which such a table can address in a single location.
What's this?
http://www.nespowerpak.com/redboard.jpg
The suspense is terrible. I hope it'll last.
MMC1 cart for copynes maybe? Something along those lines I think
It kinda looks like Memblers UNROM-like board
cd_vision wrote:
What's this?
http://www.nespowerpak.com/redboard.jpgThe suspense is terrible. I hope it'll last.
bunnyboy is going to release 3 carts:
1. NES PowerPak: a multi-game flash cart
2. Homebrew Game Kit: probably what is in the picture you linked
3. NWC Cart
I finished writing and designing the manual for the NWC cart last night and sent it to BunnyBoy. I didn't want to mention the project until he did. There is some information in the book that has never before been revealed, basically a good amount of exclusive info. I hope you folks enjoy it when you get your paws on it.
-Rob
Nintendo World Championships 1990.
-Rob
Jagasian wrote:
bunnyboy is going to release 3 carts:
1. NES PowerPak: a multi-game flash cart
2. Homebrew Game Kit: probably what is in the picture you linked
3. NWC Cart
What about the PowerPak Lite?
this seems to be the motherboard in that case
http://www.nespowerpak.com/repropak.jpg
I've probably missed something here, but what's the point of doing an NWC-cart, can someone explain perhaps?
oRBIT2002 wrote:
I've probably missed something here, but what's the point of doing an NWC-cart, can someone explain perhaps?
They are extremely rare and expensive to own. I'd imagine that they are fun to play with friends, and probably hard to reproduce on your own due to dip switches.
I'd buy one.
NC
AHHHHHHHHHHHHHH!!!1111
EVERYBODY PANIC!!!
http://www.retrousb.com/
Damn you all, I'll be first in line!
Seriously though, I never realized that bunnyboy worked for Retrozone. There are other pictures here:
http://www.retrousb.com/products_pictures/
...such as a picture of the PowerPak Lite, which I guess is the FDS-only cart...?
Just wondering, isn't it unfair to produce carts with a seal of quality on them (even if it doesn't seem to be exactly like the real ones, the resemblance is still high).
And... I cannot wait to get those stuff.
PS : It looks like you forgot to put the pins 1, 35, 36 and 72 to a width of 3mm instead of 2mm like the others (I should absolutely find something to criticate to hide that I'm scared by all the stuff bunnyboy does).
EDIT 2 : Why not produce NWC carts like the original ones to get money from stupid collectors are Bananmos already suggested ? This money could be re-invesed to homebrew game developement, paying stuff like high costs for plastic moulding, PCB fab, etc... and would give a good start to hardware homebrew.
oRBIT2002 wrote:
I've probably missed something here, but what's the point of doing an NWC-cart, can someone explain perhaps?
To hold tournaments with friends, family and gaming conventions, or just to get a high score, like any scored game. Many folks never got a chance to compete or experience this "game." It was a much dreamt-after idea to many to be able to have a repro cart, but it was considered extraordinarily difficult to produce until now. The game is not sufficiently playable in an emulator due to the switches, and it seems emu authors don't want to add a menu to compensate for them. Before now, people had to own a real cart just to play it in true tournament time, and we all know almost no one can afford these.
Bregalad wrote:
Just wondering, isn't it unfair to produce carts with a seal of quality on them (even if it doesn't seem to be exactly like the real ones, the resemblance is still high).
That seal is just a seal, that is, the animal. It's sort of a gold circle with a gold seal inside. Cute huh? Get it? Sealie Computing makes it, so they have a "Seal" of Quality. Hah! Genius.
Quote:
EDIT 2 : Why not produce NWC carts like the original ones to get money from stupid collectors are Bananmos already suggested ? This money could be re-invesed to homebrew game developement, paying stuff like high costs for plastic moulding, PCB fab, etc... and would give a good start to hardware homebrew.
REQUEST DENIED.
-Rob
Thanks for all answers about NWC, seems like a cool idea.
Noticed that RetroZone are actually selling some interesting nesdev-stuff now, check it out!
Jagasian wrote:
AHHHHHHHHHHHHHH!!!1111
EVERYBODY PANIC!!!
This isn't the War of the Worlds, is it?
Jagasian wrote:
Seriously though, I never realized that bunnyboy worked for Retrozone.
Doesn't surprise me, though I didn't know it either. Nice to know he's not entirely alone on this venture.
Jagasian wrote:
...such as a picture of the PowerPak Lite, which I guess is the FDS-only cart...?
He talked about the PowerPak Lite before. As I recall, it's an SRAM-based cartridge programmed through CopyNES that supports (I believe) 8 mappers. It's essentially a stripped-down version of the PowerPak for those who have CopyNES and don't want to shell out the money for the high-end version.
I like the idea of reproductions, as long as they aren't counterfeits. The distinction between a reproduction and a counterfeit is a reproduction presents itself as not the real thing, while a counterfeit tries to fool consumers into thinking that it is the real thing. Considering that the reproduction of the NWC uses a plastic cart shell and PCB that both look very different from the real thing, it is definitely not a counterfeit.
But how can they sell copies of the NWC software, which is surely copyrighted by Nintendo?
dvdmth wrote:
Jagasian wrote:
Seriously though, I never realized that bunnyboy worked for Retrozone.
Doesn't surprise me, though I didn't know it either. Nice to know he's not entirely alone on this venture.
I don't work for RetroZone, I am RetroZone
Other than my wife who occasionally does packaging and two friends who do art there are no other employees here. Surprisingly the cat and guinea pig don't help often...
Buy RetroPorts and fund NES development! Don't buy RetroPads, they take more time to build.
Jagasian wrote:
But how can they sell copies of the NWC software, which is surely copyrighted by Nintendo?
This reproduction (and ALL repros) is illegal because of the copyrighted code. Even translations are illegal, they still contain the majority of the original copyrighted program. If Nintendo complained I would have to shut it down immediately. The damages would be debateable because it was never for sale publically but the copyright still stands.
bunnyboy wrote:
dvdmth wrote:
Jagasian wrote:
Seriously though, I never realized that bunnyboy worked for Retrozone.
Doesn't surprise me, though I didn't know it either. Nice to know he's not entirely alone on this venture.
I don't work for RetroZone, I am RetroZone
Other than my wife who occasionally does packaging and two friends who do art there are no other employees here. Surprisingly the cat and guinea pig don't help often...
Buy RetroPorts and fund NES development! Don't buy RetroPads, they take more time to build.
I've already bought a SNES Retropad when you first started Retrozone. Back then you were making them with 3rd party SNES controller clones. Earlier this year I bought two NES RetroPorts, which I like much better than the Retropads, as they don't involve destroying classic hardware. The emulation crowd would love RetroPorts that supported direct access to the raw peripherial data, so that support could be added for any console peripherial, via emulation software. Again, just an idea for another product that you could mark the price up on and sell it as a "pro" version, while all you have to do is change the firmware.
bunnyboy wrote:
This reproduction (and ALL repros) is illegal because of the copyrighted code. Even translations are illegal, they still contain the majority of the original copyrighted program. If Nintendo complained I would have to shut it down immediately. The damages would be debateable because it was never for sale publically but the copyright still stands.
You might want to re-think how you approach the reproduction business, so that you can sell reproductions without violating copyright. If your reproduction business functioned as a Kinkos, where you merely let customers "print" their ROMs to carts, then you can have the customer email you the ROM they want printed... placing the copyright burden on them.
Since you are a USB expert, for a future product, consider making full fledged console copiers, like the old SNES copiers. A RetroZone copier could be shaped like a Game Genie with a USB port, which supports USB drives as well as direct linking with a PC.
http://www.nespowerpak.com/powerpak2.jpg
Words cannot discribe my awe!
Not much longer now I hope, got my credit card statement today and everything is in the black and ready for the purchase! ^_^
Sounds dangerous to sell a copyrighted ROM on a new cartridge. Aren't you afraid of being sued or something else?
oRBIT2002 wrote:
Sounds dangerous to sell a copyrighted ROM on a new cartridge. Aren't you afraid of being sued or something else?
It is a risk bunnyboy is willing to take. Unless Nintendo plans on releasing the NWC on the Wii Virtual Console, the risk is arguably very small since Nintendo isn't likely to suffer much in terms of damages. If Nintendo does complain, bunnyboy could theoretically ask for permission to sell the cart and share the profits therefrom with Nintendo (not that Nintendo would go for such a deal, but there's no reason he can't ask). I wouldn't ask Nintendo first, though, as it will grab their attention which otherwise may never come about.
The Ciclone chip is also a potential for legal action, but again I doubt Nintendo would bother because they wouldn't have much to gain.
nwc carts are up on retrousb, but the first batch is sold out
PsychedelicShaman wrote:
nwc carts are up on retrousb, but the first batch is sold out
Did they really sell out that fast? Or are they simply not available yet? Here is the link in question:
http://www.retrousb.com/index.php?productID=146
The Ciclone in itself doesn't copy the sofware internal to the CIC, it just runs a equivalent algorithm.
Quote:
Not much longer now I hope, got my credit card statement today and everything is in the black and ready for the purchase! ^_^
I think I also would have to send a request for a credit card, and I don't think how it would do, but I hope this will be okay to get a powerpack soon.
"install.php exists in the Shop-Script root folder. Please delete this file."
Happened to notice that text at the top of the RetroZone site, in hidden letters. Don't know if it's anything important or not, but I decided to post just in case.
Blah! Why did you put the NWC up for sale at 3am when everyone is sleeping!?!? Most people wouldnt have a chance to buy one =(
BunnyBoy, I hope you got my final revision to the NWC manual. I hope you didn't print up a bunch of the second to last revision. My apologies if you did. I sent it late last night. Hope I wasn't too late, but there were a few corrections that needed to be made and a few artifacts from converting to .pdf that needed to be fixed along with a few other things.
-Rob
since we are now discussing retrozone products in general I have been thinking about the classic to gamecube line, wouldnāt it be possible to make a different adapter witch can plug into the wiimote I know it would drain the batteries faster but it would be pretty nifty
bunnyboy,
Please give us a heads up before you start selling the PowerPak. I'll cry if I miss the first batch
Also, another idea for a NES related product would be a new motherboard and case for toaster NES / Famicom ICs. It could be what the NEX should have been, but built using official Nintendo made ICs, ripped from old toaster NES and RF Famicom systems. The two biggest things that the NES scene has been lacking for all of these years are a good means for playing ROMs on a real system and a console that has all of the benefits of the toaster NES and AV Famicom combined.
dvdmth wrote:
Unless Nintendo plans on releasing the NWC on the Wii Virtual Console, the risk is arguably very small since Nintendo isn't likely to suffer much in terms of damages.
Nintendo has released one component of NWC (Super Mario Bros.) on Virtual Console and another component of NWC (Tetris) on Nintendo DS. Besides, even in the case where actual damages are zero, statutory damages can be up to $150,000.
I wish you wouldnt go on a boot this copyright shit, the last thing we all want is for bunnyboy to get scarred off and never finnish this.
And if bunnyboy gets sued for selling repros, he will never finish this either.
Feeling like a kid night the before christmas, I keep visiting bunnyboys site several times/day to see if the PowerPak is available..
tepples wrote:
And if bunnyboy gets sued for selling repros, he will never finish this either.
I'm not sure he will get sued. I've been selling reproduction for over 3 years now, and except for Yahoo shutting down my original web site, nothing really happened.
Then again, I'm in Canada.. so the laws are different here for the states. (We're allowed to copy/rip DVD's, rip CD's to MP3's, and use bit-torrent!)
In any case, one should always check the laws of where they operate before starting any venture.
leonk wrote:
Then again, I'm in Canada.. so the laws are different here for the states. (We're allowed to copy/rip DVD's, rip CD's to MP3's, and use bit-torrent!)
Here in Brazil they are starting to bitch about those things, but we do it anyway. But they are mostly going after the people who sell pirate stuff (games and movies, mainly) on the streets.
Then again, doing those things you mentioned is allowed in the US too AFAIK, as long as you rip your own CD's and DVD's, and do not use bittorrent to download coprighted material you do not own.
I don't think bunnyboy has any reason to worry... Nothing that's NES-related has been profitable for Nintendo for a while. Plus, if someone really wants to save by not buying a game released for the virtual console, they will most likely download the ROM, not buy a repro. Tepples is going to say something anyway, but I don't care. =)
oRBIT2002 wrote:
Feeling like a kid night the before christmas, I keep visiting bunnyboys site several times/day to see if the PowerPak is available.. :)
Aren't we all?
leonk wrote:
I've been selling reproduction for over 3 years now
^^He makes darn nice ones too.
I'm 100% for emulation, roms, download of commerical roms, and even possibly home made repros. However, I'm much more septic about SELLING repros, wich is selling something you didn't actually make nor buy. I think that's not a very good thing, reagardless of any law. I wouldn't buy any NWC cart, as I already have all 3 of SMB & Tetris and Rad Racer. I don't understand what's the point of playing with a time limit, it should be incredibly frustrating to have the gameplay halting (exept for Tetris that I would never stand playing it more than 1 minute, let alone 6 minutes).
Bregalad wrote:
I'm 100% for emulation, roms, download of commerical roms, and even possibly home made repros. However, I'm much more septic about SELLING repros, wich is selling something you didn't actually make nor buy.
buying all the components and assembling them together so a rom image can be written to it sounds like making a product to me. i don't think he plans on selling the image, but the hardware he put together.
Bregalad wrote:
I don't understand what's the point of playing with a time limit, it should be incredibly frustrating to have the gameplay halting (exept for Tetris that I would never stand playing it more than 1 minute, let alone 6 minutes).
it's all about the competative aspect of the game.
Curious, but would it be impossible to sell development boards with the pinout differences between ROM/EPROM corrected, so that no rewiring would be needed?
you mean this?
http://www.retrousb.com/index.php?productID=137
onley flaw is I dont seem to see a spot for some sram but thats not too hard to currect
bunnyboy made a remark last week that everything would be available this week. But what exactly does that mean?
never-obsolete wrote:
buying all the components and assembling them together so a rom image can be written to it sounds like making a product to me. i don't think he plans on selling the image, but the hardware he put together.
He's selling both the hardware AND the software in a single package. If what you said were true, the cart would have a blank flash ROM (or something of the sort) which you'd have to program yourself before using. If he were selling only the hardware, there would be no legal issues whatsoever (frankly, I'm surprised that isn't what he's doing).
peppers wrote:
onley flaw is I dont seem to see a spot for some sram but thats not too hard to currect
If you look at the other items in the NES Hardware section of RetroZone, you'll see some references to an MMC1-based ReproPak with WRAM support, such as:
"62256 - 32KB 28 pin DIP SRAM for ReproPak CHRRAM, and ReproPak MMC1 CHRRAM and WRAM"
That said, it would've been nice to see WRAM support on the ReproPak already on the site. Just because NROM, U*ROM, CNROM, and A*ROM boards didn't have WRAM doesn't mean it can't be supported on new boards that mimic these mappers.
dvdmth wrote:
He's selling both the hardware AND the software in a single package. If what you said were true, the cart would have a blank flash ROM (or something of the sort) which you'd have to program yourself before using. If he were selling only the hardware, there would be no legal issues whatsoever (frankly, I'm surprised that isn't what he's doing).
if i send him a dump of my smb cart, would he be selling me the software? or just the hardware with my back-up on it. thats not the case with NWC because of the very limited production. i guess it depends on how you wanna see it/the courts decide.
I wonder if he has gotten his cicclones yet
peppers wrote:
you mean this?
http://www.retrousb.com/index.php?productID=137onley flaw is I dont seem to see a spot for some sram but thats not too hard to currect
Heh. Guess I coulda read the item description eh?
peppers wrote:
I wonder if he has gotten his cicclones yet
Well, there is a picture of the Ciclone on the site, so I'm assuming he has at least one (unless the pic came from Kevin Horton).
For those who haven't seen it, the PowerPak Lite is now listed on RetroZone (unavailable, of course).
tokumaru wrote:
hen again, doing those things you mentioned is allowed in the US too AFAIK, as long as you rip your own CD's and DVD's, and do not use bittorrent to download coprighted material you do not own.
Nope. not any more.
The US a few year back passed the digital millennium act. It makes decrypting any copyright material illegal. DVD's are encrypted, and hence making backups of DVD's is illegal.
This is why all DVD2One style programs are all sold outside the USA.
Different countries have different rules. E.g. In Finland, it's legal to copy encrypted format that has been proven to be easily defeated to the point where anyone can do it, etc. etc.
Canada for the longest time has been on the USA's top list of places where copyright laws are broken, and the government isn't doing anything about it.
But the US government decided not to do anything about it because Canada is the USA's #1 trade partner in the world. Go figure.
In any case, back on topic. Please let us know when this wonderful product does come out!
Powerpak Lites are in stock now.
dvdmth wrote:
If he were selling only the hardware, there would be no legal issues whatsoever (frankly, I'm surprised that isn't what he's doing).
Actually, the ciclone chip which I believe all this hardware has goes against the digital millennium act (since it's reverse engineering of copyright material) and hence selling it in the USA is technically illegal.
To make it legal in the states, the boards will need to have a spot for an individual to solder their own chip lockout IC which they can desolder from a real cart.
The digital millennium act does not effect other countries.
For more info, subscribe to great podcasts like twit or dl.tv. There was lots of talk about this the last few months (listen to Windows Weekly epside from 2 weeks ago when they talked about it in detail).
I just noticed that PowerPak Lite is on RetroZone!
I have a couple of questions about it.
1) does the copynes require the flash programming mode added? (see:
http://www.tripoint.org/kevtris/Project ... flash.html)
2) Will it work with Quietust's Windows XP/2K host software??
3) I noticed the lockout chip is not included. Can we add it if we want to?
4) Is it safe to assume that the moment the power is lost, all data on the RAM chips is lost? (i.e. the game can only be played while inside the CopyNES, and the only safe way to do it is by clicking the play cart button?)
Thanks for any info you may provide.
leonk - website clarified and questions
answered in new thread
leonk wrote:
Actually, the ciclone chip which I believe all this hardware has goes against the digital millennium act (since it's reverse engineering of copyright material) and hence selling it in the USA is technically illegal.
O RLY? See
Chamberlain v. Skylink and
Lexmark v. Static Control, which have limited the scope of the DMCA with respect to access control methods used as a generic lock-in mechanism rather than around specific copyrighted works.
Apparently Retrozone's NWC reproductions did go on sale, and did sell out within minutes. Imagine if the reproduction ends up being more rare than the real thing!
Jagasian wrote:
Apparently Retrozone's NWC reproductions did go on sale, and did sell out within minutes. Imagine if the reproduction ends up being more rare than the real thing!
What's your source for the info?
dvdmth wrote:
Jagasian wrote:
Apparently Retrozone's NWC reproductions did go on sale, and did sell out within minutes. Imagine if the reproduction ends up being more rare than the real thing!
What's your source for the info?
NES Forums
http://www.nesforums.com/showthread.php?t=2545
Ahh, OK. Restock within a week, eh?
Unless Nintendo intervenes, these carts shouldn't be rarer than the originals.
*waiting impatiently for the PowerPak release*
Cosign. I went out and bought my CF card last night. Thought I might start getting my rom collection moved over. I noticed in the video you have them sorted by mapper, but can I just arrange them any way I want?
Guys, do you think the "NES-scene" will change when the PowerPak is released? It will for sure affect me anyway, alot more NES-coding.
oRBIT2002 wrote:
Guys, do you think the "NES-scene" will change when the PowerPak is released? It will for sure affect me anyway, alot more NES-coding.
I think that a wide spread means for people to be able to play ROMs on a real NES or Famicom will push emulators to become even more accurate, as people realize there are still differences when playing a ROM on a real NES versus an emulator.
oRBIT2002 wrote:
Guys, do you think the "NES-scene" will change when the PowerPak is released? It will for sure affect me anyway, alot more NES-coding.
That got me thinking too. Is not a coder (yet, in the asm sense) but what about a custom mapper with access to the CF card, a CF mapper perhaps if the PowerPak's design will allow it.
oRBIT2002 wrote:
*waiting impatiently for the PowerPak release*
I hope the PowerPak will be released in a long time.....
....so that I will have a cash card when it will come out. I don't want to be unable to order it when it's out, and then have it sold out when I'm finally able to order it.
oRBIT2002 wrote:
Guys, do you think the "NES-scene" will change when the PowerPak is released?
Hell yeah. It would more closely resemble
the GBA scene and the DS scene, where people actually expect to be able to run homebrew on authentic Nintendo hardware rather than on a PC based emulator.
Hojo_Norem wrote:
but what about a custom mapper with access to the CF card, a CF mapper perhaps if the PowerPak's design will allow it.
bunnyboy has stated before that programs should be able to read and write sectors on the CF card. A program would put game code in a main bank and file system code in an auxiliary bank, with a jump table up in the $Fxxx range. This would give an execution environment reminiscent of the FDS or 8-bit micros. The only thing missing is the keyboard.
Hojo_Norem wrote:
That got me thinking too. Is not a coder (yet, in the asm sense) but what about a custom mapper with access to the CF card, a CF mapper perhaps if the PowerPak's design will allow it.
Interesting idea..
tepples wrote:
The only thing missing is the keyboard.
Another project for bunnyboy
Bregalad wrote:
oRBIT2002 wrote:
*waiting impatiently for the PowerPak release*
I hope the PowerPak will be released in a long time.....
....so that I will have a cash card when it will come out. I don't want to be unable to order it when it's out, and then have it sold out when I'm finally able to order it.
I am unsure how it works in your conutry but here when you apply for a creadit card you will have it in 3 days or sooner
peppers wrote:
Bregalad wrote:
oRBIT2002 wrote:
*waiting impatiently for the PowerPak release*
I hope the PowerPak will be released in a long time.....
....so that I will have a cash card when it will come out. I don't want to be unable to order it when it's out, and then have it sold out when I'm finally able to order it.
I am unsure how it works in your conutry but here when you apply for a creadit card you will have it in 3 days or sooner
AFAIK, many Europeans don't have credit cards. Americans, on the other hand, all have multiple credit cards and tons of debt.
you can still get therm over there and its not a difficult prosess to do so
If you have a checking account, you can get a debit card that draws money from the checking account, usually one with the VISA or MasterCard logo on it. PayPal and other merchants take these debit cards. Are checking accounts also uncommon in Europe?
No, pretty much everyone has debit cards around here (Sweden). It's basically a matter of going in to your bank and saying "I want one" (as long as you're over 18 ), and they'll send it to you in a couple of days.
Questions for bunnyboy concerning PowerPak:
1) Willl RAM at $6000 be enabled for all ROMs? (Or perhaps there was a new flag in the official iNES header telling if additional RAM was present or not?)
2) Is CHRRAM available for mapper 0 ROMs aswell?
3) When powering up a "real" NES cartridge, RAM will contain garbage (as I've heard), how will this work with PowerPak? Do you clear the RAM before the user starts a ROM? Perhaps not that important but I think it's interesting.
4) When will I be able to order it?
Since I'm just 18 since 2 months, I the bank send me a quote and I replied, saying I'd like to get my hand on a MasterCard or something similar, wich is now available. I doubt the whole process will be termined until a couple of weeks, knowing how slow the administrations are in my country (and proabaly most other countries too).
Jagasian wrote:
I think that a wide spread means for people to be able to play ROMs on a real NES or Famicom will push emulators to become even more accurate, as people realize there are still differences when playing a ROM on a real NES versus an emulator.
Actually, the biggest differences are in the mapper emulation, as far as I can see. Even the MMC3 hasn't been fully 100% accurately emulated yet. Since the PowerPak isn't using a real MMC3 for it's mapper 4 emulation, I can imagine that it too will not be 100% accurate.
The good thing about the PowerPak is that it doesn't need to emulate things outside the game cart. But there are and will be many bugs to fix inside the PowerPak in the foreseeable future.
But it still is a great development tool nevertheless.
leonk wrote:
Actually, the biggest differences are in the mapper emulation, as far as I can see. Even the MMC3 hasn't been fully 100% accurately emulated yet.
What? Audio and video are definitely the furthest off in most emulators. I would say that any emulator which doesn't hack the IRQ counter (puts all CPU/PPU accesses through a cartridge handler) is 100% accurate. My WIP emulator's MMC3 model is derived from logic I've synthesized and run on a real console, I think it's pretty sound...
leonk wrote:
Since the PowerPak isn't using a real MMC3 for it's mapper 4 emulation, I can imagine that it too will not be 100% accurate.
The good thing about the PowerPak is that it doesn't need to emulate things outside the game cart. But there are and will be many bugs to fix inside the PowerPak in the foreseeable future.
Why would you assume there are logic "bugs", especially in MMC3? If he didn't want to, bunnyboy wouldn't even have to design his MMC3, he could just use FunkyFlash's which clearly is 100% working despite the implementation being a little strange.
kyuusaku wrote:
What? Audio and video are definitely the furthest off in most emulators.
The operative word here is "most," for there are emulators (such as Nestopia) with very good video and audio (maybe not 100% perfect, but really close).
kyuusaku wrote:
I would say that any emulator which doesn't hack the IRQ counter (puts all CPU/PPU accesses through a cartridge handler) is 100% accurate.
And "most" emulators don't implement A12-based IRQ's.
kyuusaku wrote:
Why would you assume there are logic "bugs", especially in MMC3? If he didn't want to, bunnyboy wouldn't even have to design his MMC3, he could just use FunkyFlash's which clearly is 100% working despite the implementation being a little strange.
I can think of several sources for MMC3 problems.
* Support for both MMC3 and MMC6 without CRC detection or NES 2.0 headers will require ignoring $A001 writes and always having WRAM available at $6000-7FFF.
* At least one game out there (name escapes me) will not work if WRAM is present.
* On some carts, writing $00 to $C000 will cause only one IRQ to occur, whereas other carts will generate IRQs on every A12 clock. I think Nestopia has to address this difference to get one or two games to work right.
* Don't know if any games are affected, but on some carts if $C001 is written after the IRQ counter is reloaded but before the next A12 clock, the counter will be OR'ed with $80 on the next A12 clock. (Kevin Horton found this behavior a while back IIRC. Might be related to the $C000 thing listed above.)
I'll be darned if the first set of drivers for the PowerPak can handle these issues, especially when NES 2.0 hasn't matured (and I doubt it ever will, unless Kev actually gets around to publicizing the mapper and submapper assignments).
dvdmth wrote:
kyuusaku wrote:
What? Audio and video are definitely the furthest off in most emulators.
The operative word here is "most," for there are emulators (such as Nestopia) with very good video and audio (maybe not 100% perfect, but really close).
And since "most" don't have very good A/V, yet have a working even if hacky implementation of the MMC3, I would argue that A/V is more of an issue than MMC3 emulation for "most" emulators.
dvdmth wrote:
kyuusaku wrote:
I would say that any emulator which doesn't hack the IRQ counter (puts all CPU/PPU accesses through a cartridge handler) is 100% accurate.
And "most" emulators don't implement A12-based IRQ's.
But some do which disproves "MMC3 hasn't been fully 100% accurately emulated yet."
dvdmth wrote:
I can think of several sources for MMC3 problems.
* Support for both MMC3 and MMC6 without CRC detection or NES 2.0 headers will require ignoring $A001 writes and always having WRAM available at $6000-7FFF.
He could allow the user to select MMC3 or MMC6 fusemaps...
dvdmth wrote:
* At least one game out there (name escapes me) will not work if WRAM is present.
* On some carts, writing $00 to $C000 will cause only one IRQ to occur, whereas other carts will generate IRQs on every A12 clock. I think Nestopia has to address this difference to get one or two games to work right.
* Don't know if any games are affected, but on some carts if $C001 is written after the IRQ counter is reloaded but before the next A12 clock, the counter will be OR'ed with $80 on the next A12 clock. (Kevin Horton found this behavior a while back IIRC. Might be related to the $C000 thing listed above.)
Fusemap selector again for different revisions?
dvdmth wrote:
I'll be darned if the first set of drivers for the PowerPak can handle these issues, especially when NES 2.0 hasn't matured (and I doubt it ever will, unless Kev actually gets around to publicizing the mapper and submapper assignments).
All of these problems would be from shortcomings of iNES, not bunnyboy's flawed MMC3 logic... You know everything doesn't have to revolve around iNES and iNES 2.0, PowerPak could let the user decide what to do in an ambiguous situation. PowerPak is already far more difficult to use than an original game, I don't think it would be unreasonable to leave a few options to the user.
Quote:
* Support for both MMC3 and MMC6 without CRC detection or NES 2.0 headers will require ignoring $A001 writes and always having WRAM available at $6000-7FFF.
Just my 2 cents 'bout it, I think relying on this kind of stuff when coding is VERY bad coding. This hardware WRAM locks only exists as an attempt to prevent the RAM data to be corrupted at power up and down, and has no purpose when actualy emulating the game.
Quote:
* At least one game out there (name escapes me) will not work if WRAM is present.
The name is Low G Man, and this is even worse programming as above. The game could in theory crash anytime on a real hardware as well, because the contend on reads from $6000-$7ffff is undefined if no WRAM is present. The only reason it doesn't is due to the data bus capacity. I don't know why this game had the seal of quality with such bad programming, but anyway the WRAM size flag could emulate it "correctly" I guess.
Bregalad wrote:
Just my 2 cents 'bout it, I think relying on this kind of stuff when coding is VERY bad coding. This hardware WRAM locks only exists as an attempt to prevent the RAM data to be corrupted at power up and down, and has no purpose when actualy emulating the game.
Why does the MMC3 have a read-only mode then? When the feature was designed, there was a clear intent to try to avoid RAM corruption due to a software bug. (That said, most if not all MMC3 games leave WRAM fully enabled at all times because the WRAM is used for more than just saved game data. That same cannot be said for the MMC6, however, but that's only two games and I don't know if either game has any bugs that could corrupt saved data if WRAM were left on.)
I will also point out that the PowerPak is being marketed as a development tool. A developer working on an MMC3 game would like to know whether or not his work will function on real MMC3 hardware. You can't know 100% without testing on the real hardware, but part of the drive towards accuracy is to mimic the real hardware as clase as possible (which is what sparked this discussion to begin with).
If I were designing the MMC3 implementation of the PowerPak, I would ignore $A001 writes. But then I wouldn't claim 100% accuracy of the MMC3 either.
Quote:
The name is Low G Man, and this is even worse programming as above. The game could in theory crash anytime on a real hardware as well, because the contend on reads from $6000-$7ffff is undefined if no WRAM is present. The only reason it doesn't is due to the data bus capacity. I don't know why this game had the seal of quality with such bad programming, but anyway the WRAM size flag could emulate it "correctly" I guess.
Thanks for the game title. I could be wrong, but I think there's also an SNES game that relies on open bus at a certain address to function (I remember a BSNES update needing to address this issue a while back). It's a rather primitive form of copy protection, not a particularly good one for the reason you cited, but copy protection nonetheless. Nintendo did require developers to submit any special copy protect schemes their software used when requesting approval for a game. I guess Nintendo had no problem with using open bus as a means to do this (either that or the developer managed somehow to slip it past Nintendo).
Bunyboy, you my friend are going to make yourself a big chunk of change if you manage this right. Good work, and good luck! If I had that kind of money to just throw around I would certainly buy one for the price you are offering.
Edit: OOO, just saw your website, very cool. I will be checking this often to see when you release the full blown PowerPak.
dvdmth wrote:
A developer working on an MMC3 game would like to know whether or not his work will function on real MMC3 hardware.
Developing an MMC3 (real MMC3) game sounds like a highly impractical move to me, anyways. I would really advise against it. The only source of MMC3 chips is existing carts, and getting to those would involve a whole lot of desoldering and/or rewiring. What's more, is that there's several revisions of the chip.
I never understood why people want to write new programs using the MMC chips (besides the convenience). Sounds like the road to madness.
Nowadays, I prefer discrete logic + CHR-RAM myself, but in the past I thought the MMC3 was the way to go... The only thing I miss is scanline IRQ's/cycle counters.
Problem is that when developing using emulators, we must rely on the existent mappers. I have some ideas for a perfect mapper for me, but I have no means to produce a cart with my specifications, and wouldn't know how to modify emulators to support it either.
My ideal cart would have 32KB of CHR-RAM that could be mapped to the 8KB of pattern tables in small chunks (1KB probably) so that I'd have the benefits of CHR-RAM and be able to do CHR-ROM-like tricks (large background animations), up to 1MB of PRG-ROM, divided in banks of 8KB (the last bank could be hardwired), 32KB of WRAM, and a cycle counter that triggers IRQ's. Oh, and mirroring selection. I would be able to do anything with that! =)
tokumaru wrote:
My ideal cart would have 32KB of CHR-RAM that could be mapped to the 8KB of pattern tables in small chunks (1KB probably) so that I'd have the benefits of CHR-RAM and be able to do CHR-ROM-like tricks (large background animations), up to 1MB of PRG-ROM, divided in banks of 8KB (the last bank could be hardwired), 32KB of WRAM, and a cycle counter that triggers IRQ's. Oh, and mirroring selection. I would be able to do anything with that! =)
What you want is something like an FME-7.
How good is the FME-7 support in PowerPak?
Memblers wrote:
I never understood why people want to write new programs using the MMC chips (besides the convenience). Sounds like the road to madness.
I can think of one reason, game hacks. There are countless number of web sites dedicated to game hacks. A lot of those developers are currently using emulators to test their patches. I'm sure many would love to see their creations on real carts.
The ideal PowerPak will give them the ability to test their patches during the development cycle.
With the PowerPak, issues like the now infamous "Mario Adventure" not working on a real MMC3 cart shouldn't happen again; as long as the MMC3 emulation on the PowerPak is 100% accurate.
That's what I was hoping for when I first wrote that comment.
tokumaru wrote:
Problem is that when developing using emulators, we must rely on the existent mappers. I have some ideas for a perfect mapper for me, but I have no means to produce a cart with my specifications, and wouldn't know how to modify emulators to support it either.
Historically, it's been simply easier and cheaper to use existing PCB boards that were mass produced to develop and distribute homebrew. (According to Wikipedia, 33-45 million NES games were sold in the US alone!)
Even if the PowerPak supported custom uploadable mappers, mass producing PCB's to sell these games would be a daunting task for most developers.
leonk wrote:
With the PowerPak, issues like the now infamous "Mario Adventure" not working on a real MMC3 cart shouldn't happen again
If I undersand what I read on another site about Mario Adventure, it won't work on real hardware because it relies on a emulator bug to work. Could a tweaked version of the MMC3 code be made so it could work on a real NES via the PowerPak?
I think Mario Adventure used sprites from both pattern tables, which is a big no-no with MMC3 IRQ's (all sprites MUST come from $1000-1FFF for the MMC3 to work properly).
From an emulation standpoint, the less accurate "H-Blank trigger" for MMC3 IRQ's is easier to implement, because an emulator can easily predict in advance when the IRQ will fire in that model. However, in the PowerPak an H-Blank trigger will be harder because you only have access to the PPU address and data buses, and you'll have to do a more complex analysis of the PPU's activity to determine when H-Blank occurs. Such a mechanism will only be reliable if the PPU isn't turned on/off mid-frame.
Hojo_Norem wrote:
Could a tweaked version of the MMC3 code be made so it could work on a real NES via the PowerPak?
I hardly think that the possibility of playing
one single game (not even an original program, but a hack) justifies the creation of a buggy MMC3 implementation.
You could make a custom mapper for it by editing the header and then modifying the mmc3 mapper file if you really wanted to with the Powerpak's setup I suppose
gannon wrote:
You could make a custom mapper for it by editing the header and then modifying the mmc3 mapper file if you really wanted to with the Powerpak's setup I suppose
The INES header is screwed up enough as it is. Let's not make it worse.
dvdmth wrote:
From an emulation standpoint, the less accurate "H-Blank trigger" for MMC3 IRQ's is easier to implement, because an emulator can easily predict in advance when the IRQ will fire in that model. However, in the PowerPak an H-Blank trigger will be harder because you only have access to the PPU address and data buses, and you'll have to do a more complex analysis of the PPU's activity to determine when H-Blank occurs. Such a mechanism will only be reliable if the PPU isn't turned on/off mid-frame.
Unless the mapper is set up to count one scanline every 113.667 or 106.563 CPU cycles, and then wait for the last hblank in the traditional MMC3 manner.
tokumaru wrote:
I hardly think that the possibility of playing one single game (not even an original program, but a hack) justifies the creation of a buggy MMC3 implementation.
I don't see too much of a reason against a Mario Adventure submapper, given that quite a few other mappers are used on fewer than 5 games: MMC2 in PNROM (Punch-Out!!), CPROM (Videomation), MMC3 in TQROM (two Rare pinball games), MMC3 in TLSROM (a couple football games), MMC6 in HKROM (StarTropics 1 and 2), FME-7 (Batman ROTJ and Ufouria), Nichibutsu's bass-ackward version of UNROM (Crazy Climber), etc.
tepples wrote:
quite a few other mappers are used on fewer than 5 games
But those were actual commercial games, which the iNES header tries to accommodate. If people start assigning submapper numbers for each badly programmed hack/game left and right, we'll run out of numbers because of stuff that's not worth it.
The author of
Mario Adventure should fix his hack to comply with the MMC3 specs, just as the original game where the hack was based on did.
Now, if someone is to program a game that can't possibly use the existing mappers, then I'm all for the creation of a new mapper with the needed features. If a hardware counterpart exists, even better.
I just don't think that we need more mappers, specially submappers to accommodate bad code. One of the reasons is that emulators will have a hard time to keep up with all the changes (and old emulators may never support the new ROMs). This is very bad, because if they don't support new homebrew software, they will be seen more like piracy inducers than tools to aid development, something we're all trying to avoid.
tokumaru wrote:
tepples wrote:
quite a few other mappers are used on fewer than 5 games
But those were actual commercial games, which the iNES header tries to accommodate.
This is a notability criterion. Some notability criteria can be interpreted as "inducement" under copyright law.
Quote:
If people start assigning submapper numbers for each badly programmed hack/game left and right, we'll run out of numbers because of stuff that's not worth it.
Which is why eventual support for something other than iNES is a good idea. I think it was Bananmos who tried to drill in everyone that NES homebrew is more than only software development; it is also hardware development.
Quote:
The author of Mario Adventure should fix his hack to comply with the MMC3 specs, just as the original game where the hack was based on did.
Is that necessarily possible given the amount of different stuff that can potentially be in a scene? It might have to use MMC5 instead.
Quote:
Now, if someone is to program a game that can't possibly use the existing mappers, then I'm all for the creation of a new mapper with the needed features.
Would somebody have to go through an exhaustive list of all mappers used in commercial games in one's region and do one's best to hack the game to use each before deciding on extending a mapper?
Quote:
Quote:
If people start assigning submapper numbers for each badly programmed hack/game left and right, we'll run out of numbers because of stuff that's not worth it.
[...]
The author of Mario Adventure should fix his hack to comply with the MMC3 specs, just as the original game where the hack was based on did.
Is that necessarily possible given the amount of different stuff that can potentially be in a scene? It might have to use MMC5 instead.
for Tecmo Super Bowl i need 0x340 more bytes of sram then the mmc3 provides to update it to 32 teams. it would be much easier to create a submapper that has multiple sram banks like the mmc5 (possibly with the register in the $5xxx range), then trying to hack it to the mmc5 board (as i'm trying to do). i'm not sure if it's even gonna be possible because the new irq's are causing very noticeable shearing and the 8x16 sprites are a nightmare in themselves.
i think it depends on your goals. i would like to put this on the extra TSB cart i have, but a new submapper would make that impossible. though emulators could be easily extended.
I don't like creating submappers for the sole purpose of allowing game hacks to work. If you were to ask an emulator author to add support for a hack, s/he wouldn't be too thrilled about it, even if it is as simple as providing for more WRAM on the MMC3.
Porting an MMC3 to MMC5 is definitely doable. If you're having IRQ troubles, chances are it's because of two things:
1. Differences in the IRQ counter. This will particularly show up if multiple IRQ's are desired in a single frame.
2. Different timing. The MMC3's IRQ fires at PPU cycle 260, whereas the MMC5 triggers the IRQ at PPU cycle 0 (a difference of ~87 CPU cycles on NTSC). You'll need to add an extra delay to the IRQ handler to compensate.
What's the big deal with 8x16 sprites? Being an MMC3 game, all sprites should come from $1000-1FFF, so you'd only need to update $5124-5127 for sprite banks. Background banks are set through $5128-512B in 8x16 mode. Even if there were sprites from $0000-0FFF, all you'd have to do is be sure $5120-5123 mirrors $5128-512B and you're set. (I'm not missing something, am I?)
Boy, have we gone off-topic here.
I guess Nintendo would have allowed anyone to use the MMC3's higher CHR adress lines to adress various WRAM banks if you were a commercial game developper back then, in the same fashion than SOROM SXROM does with the MMC1. This could have leaded new board to be created for you. Since no one ever used this feature, there is no "official" way to get more than 8kb of WRAM with MMC3, but this is technically possible, and probably would have been possible.
this should probably be split a few post back...
dvdmth wrote:
I don't like creating submappers for the sole purpose of allowing game hacks to work. If you were to ask an emulator author to add support for a hack, s/he wouldn't be too thrilled about it, even if it is as simple as providing for more WRAM on the MMC3.
i agree, unless the author actually created a hardware compatible board. since i'm not an EE person this probably isn't gonna happen.
dvdmth wrote:
Porting an MMC3 to MMC5 is definitely doable. If you're having IRQ troubles, chances are it's because of two things:
1. Differences in the IRQ counter. This will particularly show up if multiple IRQ's are desired in a single frame.
2. Different timing. The MMC3's IRQ fires at PPU cycle 260, whereas the MMC5 triggers the IRQ at PPU cycle 0 (a difference of ~87 CPU cycles on NTSC). You'll need to add an extra delay to the IRQ handler to compensate.
What's the big deal with 8x16 sprites?
after looking at it again, all the graphic errors/crashes stem from irqs not firing/firing at the wrong time. the shearing is being caused by the mmc3 reg8001 "interpreter" taking more cycles then a usual write to $8001.
Bregalad wrote:
I guess Nintendo would have allowed anyone to use the MMC3's higher CHR adress lines to adress various WRAM banks if you were a commercial game developper back then, in the same fashion than SOROM SXROM does with the MMC1. This could have leaded new board to be created for you. Since no one ever used this feature, there is no "official" way to get more than 8kb of WRAM with MMC3, but this is technically possible, and probably would have been possible.
along those lines i was thinking about using the TQROM board by reading the new save data from sram to chr-ram at start up and writing it back at power down. a good portion of sram is used as game logic ram and needs not be saved at power down. two problems with that is the 64k limitation for chr-rom and the rareity of the board.
So, I guess the Power Paks went on sale and all sold out in 15 minutes. Bummer. I was really hoping to get one of these. Such a drag.
See, I gotcha for a second, didn't I?! Yes, I'm full of shit. Are we back on topic yet?
-Rob
tepples wrote:
This is a notability criterion. Some notability criteria can be interpreted as "inducement" under copyright law.
What?! Sorry, I'm not a native or very experienced english speaker, so I'm having a bit of a hard time figuring what you're saying there...! =)
Quote:
Which is why eventual support for something other than iNES is a good idea. I think it was Bananmos who tried to drill in everyone that NES homebrew is more than only software development; it is also hardware development.
I agree with you here. In a perfect world, mappers would be created with mapper definitions, specifying all the wiring, what is connected where on the board, what components are used, and the IC's would all be emulated. This would allow for a shit load of new "mappers" without the need to code anything. If all the standard IC's were emulated, new code would only have to be written for new chips, instead of for different board combinations. Then emulators would not support "mappers", but mapper chips.
But in the current situation, if new mapper numbers are assigned to each small customization every John Doe makes, it'll be chaos.
Quote:
It might have to use MMC5 instead.
Don't all programmers have to deal with hardware limitations? Why not game hackers? If the current hardware is not enough to do what you want, you either cope with that and make changes so that the project can be done anyway, or use new hardware.
Quote:
Would somebody have to go through an exhaustive list of all mappers used in commercial games
And the list will only grow bigger if people start creating "new" mappers left and right because they are too lazy to look a suitable one up.
rbudrick wrote:
So, I guess the Power Paks went on sale and all sold out in 15 minutes. Bummer. I was really hoping to get one of these. Such a drag.
See, I gotcha for a second, didn't I?! Yes, I'm full of shit. Are we back on topic yet?
-Rob
LOL, is my calendar wrong? I thought we were in June, but I guess it's still the beginning of April. (Dumb computer's automatic time setup must have failed again....)
I haven't heard anything from bunnyboy since last week. Hopefully everything is going OK. I wonder what is holding up the release of the CF based PowerPak. Maybe he is building a game compatibility chart? Maybe there was a design bug?
Jagasian wrote:
I haven't heard anything from bunnyboy since last week. Hopefully everything is going OK. I wonder what is holding up the release of the CF based PowerPak. Maybe he is building a game compatibility chart? Maybe there was a design bug?
Bunnyboy said he is waiting on the manuals (the artist is sick). That's the hold up right now. Hopefully, we'll see the CF PowerPak Monday or Tuesday.
Manuals? Does anyone read them anyway?
Quote:
along those lines i was thinking about using the TQROM board by reading the new save data from sram to chr-ram at start up and writing it back at power down. a good portion of sram is used as game logic ram and needs not be saved at power down. two problems with that is the 64k limitation for chr-rom and the rareity of the board.
Not anything to do with TQROM. Actually, I think there should be a "standard" assignation with mappers wich nonstandard configuration, as long as the iNES 2.0 header exists.
iNES 2.0 can technically have a game with MMC3 + 8k CHRRAM + 32k batteried RPGRAM or something like that, but since no commercial games ever had that, there is no official way to emulate it either.
I think there should be a standard sytem that should improve this situation, based on what the few Famicom and NES games using standard mappers with nonstandard configuration does. This system should in no way break the compatibitily with any existing game (at least any commercial one).
The system would tell wich adress lines would be used for any combination of more RPGROM (a.ka. SUROM, SXROM), more PRG RAM (a.k.a. SOROM, SXROM), more CHR RAM (no known board), and/or CHR ROM + CHR RAM (TQROM).
I think when more CHR RAM is present, it should be banked like normal (i.e. CHRROM), exept it's RAM instead. When more PRG ROM or PRG RAM are present, they should be banked with some of the higher adress bits. When a 74HC32 (or similar logic) is used to select a CHR ROM or a CHR RAM, another higher CHR adress bit should be used. Wich bits are used when is a hot subject, however I think "hypotetical" boards shouldn't be rejected as opposed to "real" boards, as long as credible backward compatibility with "real" boards is fully assured.
oRBIT2002 wrote:
Manuals? Does anyone read them anyway?
For the NWC cart, if you don't read the manual, you're missing out, says I!
See, that was a shameless plug for my mad skillz.
-Rob
rbudrick wrote:
For the NWC cart, if you don't read the manual, you're missing out, says I!
See, that was a shameless plug for my mad skillz.
-Rob
Actually, I'm looking forward to seeing what you put in the manual. Didn't I read somewhere that you went to the NWC??
Yeah, you probably did. I'm currently trying to track down as many staff members and finalists as possible for interviews, along with pics of the carts, and whatever other items I can find from the event....trying to make a book for NWC fans.
If you are interested, here's a really cool interview I did that kind of kickstarted this whole process:
http://www.planetnintendo.com/thewarpzone/champ.html
The NWC has always been an incredible moment of videogame history for me. It's been a fun project thus far, but slow moving.
-Rob
the orginal NWC cart seems to have switches witch I assume were for setting the play time.
so how do you do whatever them switches did?
edit: I see now the switches are for setting the play time and the reprobution has it set to 6:21 onley
The manuals are must haves for collectors. That's the reason for waiting for the manuals.
peppers wrote:
I see now the switches are for setting the play time and the reprobution has it set to 6:21 onley
I'd like to ask for clarification here. The info I have on the NWC doesn't show 6:21 as a possible DIP setting. If the switches are set to 0100, the time is 6:15, and if they're set to 0101 the time is 6:34.
Quote:
The info I have on the NWC doesn't show 6:21 as a possible DIP setting
I was confused about that too, but just guessed they measure from a different starting point. The official contest time is marked as 6:15 on the gray cart, typically quoted at 6:21, and takes 6:22 stopwatch time from hitting the start button to the time over screen appearing.
Might it not count the time spent in the UI before Super Mario Bros. starts?
bunnyboy >
When do you think the powerpak will be ready to order? Got some cash here I want to get rid of
Also when will the NWC carts be up for sale again?
tepples wrote:
Might it not count the time spent in the UI before Super Mario Bros. starts?
That may be it. I (finally) found the ROM today (don't nobody ask aboit it!!), and I have yet to get around to testing or looking at it. If the timer doesn't begin ticking until ~6 seconds after the start button is pressed, then that would explain everything.
bunnyboy wrote:
Quote:
The info I have on the NWC doesn't show 6:21 as a possible DIP setting
I was confused about that too, but just guessed they measure from a different starting point. The official contest time is marked as 6:15 on the gray cart, typically quoted at 6:21, and takes 6:22 stopwatch time from hitting the start button to the time over screen appearing.
The times on the cartridge are only approximate. I'm guessing that the programmers made the timer based on clock cycles of the system...maybe the formula they used was off by a bit over a span of minutes...like about 1 second per minute, a small amount of time. The label time is simply incorrect. In fact, the timer at the real competition started at 6:21 and counted down.
-Rob
Hey Bunnyboy! You've been silent for awhile now. I hope you're busy manufacturing PowerPaks and nothing else.
Would be nice to know when they can be ordered..
rbudrick wrote:
The times on the cartridge are only approximate. I'm guessing that the programmers made the timer based on clock cycles of the system...maybe the formula they used was off by a bit over a span of minutes...like about 1 second per minute, a small amount of time. The label time is simply incorrect. In fact, the timer at the real competition started at 6:21 and counted down.
-Rob
The IRQ timer is based on the CPU clock signal. The time required, in CPU cycles, for the IRQ to fire is
Code:
2^25 * (N + 16)
where N is the 4-bit value represented by the DIP switches. If N=4, this works out to 671,088,640 cycles, or almost exactly 6:15 (374.96 seconds).
they are up on retrousb, but unnavailable ?!? did they sell out already?
Edit: NWC repros are back up!
PsychedelicShaman wrote:
they are up on retrousb, but unnavailable ?!? did they sell out already?
No. Check
http://www.nintendoage.com/
I have already ordered my powerpack get um wile the gettins good if you feel the need to do so
Yep they are up. BUY NOW!
Ordered one! But where's the https:// encryption during entering creditcard numbers?
Anyone knows if there's a way to check the paypal transaction id online? Can't seem to find anything... Haven't got the confirmation e-mail yet..
Quote:
But where's the https:// encryption during entering creditcard numbers?
All I can say is do not buy Shop-Script for shopping cart handling. Endless problems including SSL and PayPal integration, tech support from Russia that changes files without telling you, etc etc etc. I will have to start looking for better software soon...
Just got mine ordered. For those who want to know, PayPal's currency conversion makes the PowerPak cost almost exactly Ā£76.12 including P&P.
A bargain if you ask me.
I got a 512MB CF card months ago in anticipation, soon it will be time to put it to use!
w00t! finally! Already got one ordered!
bunnyboy wrote:
Quote:
But where's the https:// encryption during entering creditcard numbers?
All I can say is do not buy Shop-Script for shopping cart handling. Endless problems including SSL and PayPal integration, tech support from Russia that changes files without telling you, etc etc etc. I will have to start looking for better software soon...
people can use the paypal checkout and this is a non issue
Got my order in! What's the largest size CF card I can use?
-Rob
Quote:
What's the largest size CF card I can use?
1GB works fine, havent tested 2-8GB but I dont expect any problems with them. 256MB is about the size of every USA game.
dvdmth wrote:
rbudrick wrote:
The times on the cartridge are only approximate. I'm guessing that the programmers made the timer based on clock cycles of the system...maybe the formula they used was off by a bit over a span of minutes...like about 1 second per minute, a small amount of time. The label time is simply incorrect. In fact, the timer at the real competition started at 6:21 and counted down.
-Rob
The IRQ timer is based on the CPU clock signal. The time required, in CPU cycles, for the IRQ to fire is
Code:
2^25 * (N + 16)
where N is the 4-bit value represented by the DIP switches. If N=4, this works out to 671,088,640 cycles, or almost exactly 6:15 (374.96 seconds).
2^25=33,554,432. That's fine, but does N care whether 1, 2, 3, or 4 switches are up? Does it add the extra time together or only take the highest or lowest switch into consideration?
-Rob
EDIT: Never mind...the 4 bit value, gotcha. Well anyway, with that forumla in mind, why is the REAL time still around 6:21?
just ordered mine (+ one for a friend) now!
can't wait to have it here.
I have a question I have not thought of before
I know they are not all planned and it will not ever emulate them all or at least I assume so but dose the powerpack have the hardware potential to emulate all current mappers?
btw I am very glad you took the time to add FAT16 support this will be helpfull to me
oRBIT2002 wrote:
Ordered one! But where's the https:// encryption during entering creditcard numbers?
You have a better shot at winning the powerball than someone actually stealing your CC # while using a shopping cart. Ill take that chance.
or just use a paypal account and have an encrypted payment...
Quote:
Occasionally graphics (background or sprites) appear "sparkly" or "flickering". Reset usually fixes the problem.
How often does that actually happen? That is so weird!
Some kind of confirmation e-mail would be a nice thing after you've placed your order. Didn't get any from paypal either (I used the creditcard option on your website and did get a paypal transaction id).
I got emails both about the payment and when the package had been shipped, but I used paypal so maybe it just works in that case.
mic_ wrote:
I got emails both about the payment and when the package had been shipped, but I used paypal so maybe it just works in that case.
package has been shipped? From where did you get this e-mail?
Actually it said "Your purchase will be shipped by PayPal Shipping with U.S. Postal ServiceĀ®", not that it _had_ been shipped.. But it was sent out by
service@intl.paypal.com
Bunnyboy, you might be interested in
this NWC thread which I started in the Nesemdev section. I recently found out how to run an IRQ test in the NWC cartridge, and I'm hoping that someone can run the test on a real cart and post the results.
Got mine today!
HOLY SHIT does this thing ROCK!
bunnyboy you've done an
AMAZING job with this!! It's so easy to use I had it up and running in minutes.
Only slight issue I've had so far is that the CF ejector tab needs to be tucked in when pushing the cart slot down into the NES, otherwise it hangs up.
tokumaru wrote:
Quote:
Occasionally graphics (background or sprites) appear "sparkly" or "flickering". Reset usually fixes the problem.
How often does that actually happen? That is so weird!
I have experenced this it dose not start happening wile your playing just every once in a wile a game will load doing this
if it dose not do it when it first loaded it dose not seem to happen so it is probubley a issue with the mapper or rom loading
Ah, I see... thanks! I'm still waiting for mine, I live quite far from the US! =)
EDIT: Hey, I just noticed I reached 1000 posts with this one! =) I had never written so much anywhere on the web before! =D
Most everything I've tried so far works like a charm! But it seems there are a few games that have consistent partial graphics corruption (like a piece of a tile). Some of the top of my head - RC Pro Am (car antennas and place indicator has garbage), Donkey Kong 3 (DK has garbled fists on the title, second level has minor vine corruption), Excite Bike (bike, rider, start gates and temp gage all seem to have the same minor corruption)
Don't think its the ROMS as they seem fine in Nestopia.
remowilliams wrote:
Don't think its the ROMS as they seem fine in Nestopia.
Well, Nestopia does use CRC's to identify ROMs, so bad headers do not affect it most of the time...
I have a question I should have asked earlier.
The PowerPak uses an FPGA chip to store mapper settings. These FPGA chips are rated for a certain number of writes, like Flash and EEPROM. At some point, however, they will fail a rewrite. How often is the FPGA in the PowerPak written to? Every time a game is loaded or every time the mapper files are updated?
If the FPGA is overwritten with the mapper information needed for the game every time a new game is loaded, that means that the lifespan of the device could fail within a certain number of years.
Great Hierophant wrote:
These FPGA chips are rated for a certain number of writes, like Flash and EEPROM.
I didn't know that. That is a concern.
Quote:
How often is the FPGA in the PowerPak written to? Every time a game is loaded or every time the mapper files are updated?
I'm guessing every time a game is loaded...
The Spartan 2 FPGAs are SRAM based, so they do not have a stated write limit like Flash and EEPROMs. The chip is reprogrammed twice (once on boot, once before game starts) so if I was using a CPLD that would be a problem. I have seen write limits as low as 1000 cycles on those, but even the typical 10000 could fail after years of use. That would have to be extremely frequent resetting!
Quote:
The Spartan 2 FPGAs are SRAM based, so they do not have a stated write limit like Flash and EEPROMs. The chip is reprogrammed twice (once on boot, once before game starts) so if I was using a CPLD that would be a problem. I have seen write limits as low as 1000 cycles on those, but even the typical 10000 could fail after years of use. That would have to be extremely frequent resetting!
That is very refreshing to read. If its SRAM based, it will probably still work after you and I are dead.
tokumaru wrote:
Well, Nestopia does use CRC's to identify ROMs, so bad headers do not affect it most of the time...
Yes, I thought about that as well. Is there a NES emulator that does use all the info (headers, etc) so I can narrow this down? The list of games with minor graphical defects is growing here.
remowilliams wrote:
tokumaru wrote:
Well, Nestopia does use CRC's to identify ROMs, so bad headers do not affect it most of the time...
Yes, I thought about that as well. Is there a NES emulator that does use all the info (headers, etc) so I can narrow this down? The list of games with minor graphical defects is growing here.
I think Nestopia allows you to turn off the ROM database (the Mac version definitely does). You can also try Nintendulator if you're using Windows.
If you're seeing minor graphical glitches in many games, however, I'm not so sure that can be explained by corrupted headers. I'm wondering if maybe the CHR SRAM chip is bad, causing certain addresses to fail (which in turn would corrupt any graphics at those locations). Hopefully I am wrong here.
If the menus work before the game is loaded, I don't see how the SRAM can be bad. A higher address line could be though.
kyuusaku wrote:
If the menus work before the game is loaded, I don't see how the SRAM can be bad. A higher address line could be though.
Yeah, I've seen no graphical corruption in the menus. As I try more games, I'm seeing more (of the same minor) problems. I've checked the header info (for Excitebike for example) which seems correct, and I've tried roms other than my own, and had the same issues. Also have tried different CFs. While I can never for instance get Excitebike to run without corruption, if I use my actual cart it works fine.
Also - I've noticed that some games will occasionally start without corruption (DK3 as an example), and play OK. And sometimes they'll start corrupting ingame.
If you can, try it on a different system. If that doesn't help then you may be looking at a bad powerpak.
As I posted in the "problem games" thread, different consoles seem to handle this cart differently. What causes it, though? Slight voltage or timing variations?
remo, what's the serial number on the bottom of your NES? I have a toaster that exhibits very similar glitches, and I wonder if they're the same revision.
I ask myself if all that graphics corruption things people are talking about isn't just due to their connectors being dirty.
Then I feel the need to repeat my caution: Do now blow into PowerPak. Use a cotton swab soaked in rubbing alcohol to clean the edge connector.
I can assure you, the glitches I've experienced are not due to dirty connectors. I cleaned each console and reseated the PowerPak several times, with identical results.
If your NES is a front-loader, a bigger concern is whether or not the 72-pin connector inside the console is worn down. If a game fails to run on a front-loader, it's more likely because of a worn connector, not because of dirt (although dirt can also play a role).
If you haven't replaced the connector inside the NES, you might consider doing that and seeing if that helps (both with the PowerPak and with other cartridges). My brother replaced the connector in my NES a couple of years ago, and it's amazing how much better the system works now.
BMF54123 wrote:
remo, what's the serial number on the bottom of your NES? I have a toaster that exhibits very similar glitches, and I wonder if they're the same revision.
N26617413
I replaced the connector with a new one last year, so I know its not a dirt/fatigue issue. That and the fact that I have no problem with all my other carts.
Holy shit this looks radical!
remowilliams wrote:
I replaced the connector with a new one last year, so I know its not a dirt/fatigue issue. That and the fact that I have no problem with all my other carts.
Same here. I don't believe it's a connector issue, something physically different inside certain NES consoles isn't playing nice with the PowerPak.
This reminds me of a very similar issue I have with one of my pirate Famicom carts, Adventure Island 3. When inserted into a Famicom, or an NES (toaster OR top-loader) with a Gyromite adapter, the game works fine; however, switching to another, somewhat longer (though identically wired) adapter, or adding a Game Genie, will cause some sprites to flicker and show the wrong graphics. My guess is that the pirate MMC3 is timed so precisely that adding something between the cart and console can interfere with the signal just enough to break it. Perhaps something similar is happening here?
dvdmth wrote:
If your NES is a front-loader, a bigger concern is whether or not the 72-pin connector inside the console is worn down. If a game fails to run on a front-loader, it's more likely because of a worn connector, not because of dirt (although dirt can also play a role).
If you haven't replaced the connector inside the NES, you might consider doing that and seeing if that helps (both with the PowerPak and with other cartridges). My brother replaced the connector in my NES a couple of years ago, and it's amazing how much better the system works now.
If it were the connector, I'd think you'd never even get the menu to work, much less individual roms on the Powerpak.
YAY Earth Bound for NES for less than $1,000!
I WILL BUY ONE OVER THE SUMMER!!!!!
Please continue in
another topic.