Different NES2 (top-loader) revisions

Different NES2 (top-loader) revisions
by Jeroen on 2011-01-10 (#72563)

---
MOD EDIT: This discussion was split from Retro duo Nes top loader vertical line issue
---

The problem in the nes2 was iirc interference from other parts of the board. This happend near the ppu's composite output and thus causes lines. Later nes2 boards (VERY VERY rare) fixed it by moving the ppu away from the interference.

by Xious on 2011-01-11 (#72590)

The 'VERY VERY rare' NES2 boards also have the same multi-out port as the SFC/SNES/N64. I'm pretty sure I've narrowed down the specific source of interference and exactly where its path brings it to cause the lines. As I said, when I'm positive, I'll post about it in detail.

The lines shouldn't exist on a RetroDuo; but they have to come from the same source, else they wouldn't look the same.

by game-tech.us on 2011-05-27 (#78823)

Jeroen wrote:
Xious wrote:
The 'VERY VERY rare' NES2 boards also have the same multi-out port as the SFC/SNES/N64.

actually they found one that used rf. (altough that might be just uncommon and hence not documented. Vs very very rare.)

Couple comments to add. I don't think the 'later' production run NES2 motherboards were any different than the first units. As far as I know they only upgraded units that were sent in for repair and changed the motherboard to either the RF or AV interference free output. The RF upgraded board I have is the only one i've heard of, but that most likely because not many people take them apart if there aren't any annoying lines in the video output.
I've had units with really low and really high serial numbers and the upgraded RF mobo I have was not a high serial number unit. In fact the only difference i've seen is the 800 number sticker on the back, but most of the repaired units i've seen had a new sticker put over the old serial number sticker, usually a NN900...
I wrote all this up on my webpage.

by game-tech.us on 2011-05-27 (#78824)

Xious wrote:
I'm pretty sure I've narrowed down the specific source of interference and exactly where its path brings it to cause the lines. As I said, when I'm positive, I'll post about it in detail.

Check this threadout and you'll find the solution was found a few years ago.

by koitsu on 2011-05-28 (#78833)

akaviolence wrote:
Xious wrote:
I'm pretty sure I've narrowed down the specific source of interference and exactly where its path brings it to cause the lines. As I said, when I'm positive, I'll post about it in detail.

Check this threadout and you'll find the solution was found a few years ago.

There is nothing absolutely conclusive in that thread from what I can tell. The supposed "fix" for the problem was confirmed to not fix the issue entirely by an individual who posted in the thread. There was a later comment from someone who did a write-up, but the parts list required parts from a front-loading NES, which is absolutely retarded (sorry, it's true).

As such, I'm still waiting for someone to write up a well-documented (with photos preferably) step-through guide for the modification, including a list of parts needed, where to buy them, etc... And trust me, I know the pain of having to create such a list; I get literally 2-3 mails a week from Internet users complaining that I haven't completed my Logitech mouse fix walk-through document.

But simply put: I, like most people, aren't willing to tear my top-loader apart and "take educated guesses" using a nesdev forum post solely as a source of information (when the information within the thread conflicts itself, and again, nothing concrete in it). So until someone actually writes something up, few-to-none will risk destroying their consoles. Really.

by koitsu on 2011-05-28 (#78834)

akaviolence wrote:
Jeroen wrote:
Xious wrote:
The 'VERY VERY rare' NES2 boards also have the same multi-out port as the SFC/SNES/N64.

actually they found one that used rf. (altough that might be just uncommon and hence not documented. Vs very very rare.)

Couple comments to add. I don't think the 'later' production run NES2 motherboards were any different than the first units. As far as I know they only upgraded units that were sent in for repair and changed the motherboard to either the RF or AV interference free output. The RF upgraded board I have is the only one i've heard of, but that most likely because not many people take them apart if there aren't any annoying lines in the video output.
I've had units with really low and really high serial numbers and the upgraded RF mobo I have was not a high serial number unit. In fact the only difference i've seen is the 800 number sticker on the back, but most of the repaired units i've seen had a new sticker put over the old serial number sticker, usually a NN900...
I wrote all this up on my webpage.

I would love to know what the Q1 and Q2 replacements are exactly (zoomed in pictures of the silkscreening on both parts); possibly replacing those is all people need to do? My (old-style) PCB has silkscreening on the PCB itself that matches the silkscreening on both Q1 and Q2:

Old-style Q1 = 2SC2021
Old-style Q2 = 2SA937

The redesigned new-style PCB seems to indicate Nintendo moved a lot of the existing parts around, and re-did some of the PCB circuitry, while also saving/minimising parts at the same time (for example the 74LS139 decoder/demultiplexer is completely gone and probably replaced with something much more simple). Chances are they did this to solve the horrible video output problem in combination with reducing costs. It's really too bad they still don't offer replacement PCBs. Man what I would give for someone to design and manufacture a new one (even if it meant de-soldering existing the CPU/PPU/etc. on my top-loader and installing them on a different board).

by game-tech.us on 2011-05-28 (#78842)

koitsu wrote:
There is nothing absolutely conclusive in that thread from what I can tell.

If I posted that the dartcloud circuit along with lifting pin 21 was a conclusive fix would you believe me and take it as conclusive evidence? I'm just one guy with little clout around here, but i've modded hundreds of nes2 units, I used to do the gamesx mod, but then found leonk's thread and tried the various fixes and found the two fixes mentioned above work perfectly. Go to my site and check out the comparison pic of video capture of a front loader composite vs modded nes2 composite and stock nes2 rf. I didn't do a write up since there already is one, I never read through it since I know its only one guys version and not the only way to do it. I also tried the various cut this trace at this spot and found nothing but lifting the pin actually gets rid of the interference.
koitsu wrote:
The supposed "fix" for the problem was confirmed to not fix the issue entirely by an individual who posted in the thread.

Again, one guys trial, i've done well over 70 of them using the above methods, I know because I bought 100 transistors and have about 30 left.
koitsu wrote:
There was a later comment from someone who did a write-up, but the parts list required parts from a front-loading NES, which is absolutely retarded (sorry, it's true).

I agree to a point, but its also retarded to think thats the only way it can be done, in fact I think its also mentioned in the thread that the major parts are already in the nes2, but substitutes are easily implemented.
koitsu wrote:
As such, I'm still waiting for someone to write up a well-documented (with photos preferably) step-through guide for the modification, including a list of parts needed, where to buy them, etc... And trust me, I know the pain of having to create such a list;
But simply put: I, like most people, aren't willing to tear my top-loader apart and "take educated guesses" using a nesdev forum post solely as a source of information (when the information within the thread conflicts itself, and again, nothing concrete in it). So until someone actually writes something up, few-to-none will risk destroying their consoles. Really.

Thats fine, send it to me and i'll mod it!

by game-tech.us on 2011-05-28 (#78844)

koitsu wrote:
I would love to know what the Q1 and Q2 replacements are exactly (zoomed in pictures of the silkscreening on both parts); possibly replacing those is all people need to do? My (old-style) PCB has silkscreening on the PCB itself that matches the silkscreening on both Q1 and Q2:

If you're talking about the pics of the repairs i've found, that unit still had the lines, its not the transistors causing the problem, its the trace leading away from pin 21, as mentioned in LeonK's thread.
koitsu wrote:
The redesigned new-style PCB seems to indicate Nintendo moved a lot of the existing parts around, and re-did some of the PCB circuitry, while also saving/minimising parts at the same time (for example the 74LS139 decoder/demultiplexer is completely gone and probably replaced with something much more simple). Chances are they did this to solve the horrible video output problem in combination with reducing costs. It's really too bad they still don't offer replacement PCBs. Man what I would give for someone to design and manufacture a new one (even if it meant de-soldering existing the CPU/PPU/etc. on my top-loader and installing them on a different board).

In my opinion not worth the trouble when a simple circuit and pin lift can give you front loader quality video output.

by koitsu on 2011-05-30 (#78917)

Until someone provides a replacement circuit (amplification board) that doesn't require pieces from a NES front-loader, like I said, most people aren't going to go through with this mod. I also cease to understand why one has to "harvest parts from a front-loader"; I have a very hard time believing that replacement parts can't be purchased from numerous vendors. If people know what the parts are, what they do, their specs, etc. then getting replacements shouldn't be a problem -- unless, of course, someone doesn't know what the specs are of the parts, which makes that amplification board questionable at best. Make sense?

Additionally, consider the fact that top-loading NES units are being sold on eBay for between US$200-500. Ask yourself if you'd bother to mod something that expensive (at this point in time) when there's absolutely nothing concrete in said 11-page thread. Yes, I see some sort of horrible-god-awful "document" that requires me to "open autorun.exe" -- good luck getting me to ever download or run some random executable like that.

Furthermore, there are other posts -- example 1, example 2, example 3 (be sure to see my rebuttal points to the design in example 3) -- which are newer/more recent and reference what appears to be an alternate amplification circuit. But since I can't read circuit diagrams I can't determine if they require parts from a front-loading NES or not.

So again: until something concrete/sane is written up, and well-documented in a commonly-used format (read: HTML or Wiki page with images/photos and diagrams, NOT some random executable), most people are not going to risk destroying their top-loader.

In my opinion, what needs to happen is a merge of the two modifications: example 3 provides a circuit that apparently uses standard parts (read: you do not need to "harvest" anything from a front-loader), while examples 1/2 provide a wiring diagram that goes to composite (RCA) out. Combine these two things and you end up with a circuit that uses standard parts which goes to a composite (RCA) out. That is what people are looking for.

Effectively what we have right now is a "very informative thread" (not really) that resembles this.

by kyuusaku on 2011-05-30 (#78919)

The transistor doesn't matter at all, it's a common collector amplifier (voltage buffer, current amplifier). 2N3906 are fine.

I got those values off a modded Famicom but 200R and 120R would be better since they make exactly 75 ohm Zout.

by game-tech.us on 2011-05-30 (#78924)

koitsu wrote:
I also cease to understand why one has to "harvest parts from a front-loader";

As I said in my previous post, they are in the top loader as well as other substitutes.
koitsu wrote:
Additionally, consider the fact that top-loading NES units are being sold on eBay for between US$200-500.

I've bought 3 on ebay under $70 in the last couple months.
koitsu wrote:
Yes, I see some sort of horrible-god-awful "document" that requires me to "open autorun.exe" -- good luck getting me to ever download or run some random executable like that.

I'm with ya there...
koitsu wrote:
Furthermore, there are other posts -- example 1, example 2, example 3 (be sure to see my rebuttal points to the design in example 3) -- which are newer/more recent and reference what appears to be an alternate amplification circuit.

I haven't tried his circuit, but TuEE is a good guy, DonMega of Sega modding.
koitsu wrote:
But since I can't read circuit diagrams

Oh...

by Bregalad on 2011-05-31 (#78949)

I agree this thread is so messy. First there is so much conflicting info. Many people show how they did complicated things like adding an external amplification board or did a stereo mod or wathever. This is completely unecessarly - the video is already amplified just not impedence matched, and the NES was designed to be in mono.

Removing the RF was a major pain for me, but I had to do it because it wasn't compatible with european RF standard so I had to AV mod it to get any sound.
The impedance was much higher than 75 ohm if you just remove the box so you also have to change the reistors so that the impedance matches.
I don't think there is ANY capacity/highpass-filter involved in the video. I guess the PPU already outputs the correct voltage amplitude, at an offset of -0.7 V to compensate the BJT transistor. The only reason the transistor is needed is for impedance matching. I guess it wasn't possible to make the DAC which is inside the PPU directly with an output as low as 75 ohm, this is just a guess though. Maybe connecting the PPU directly could even work hehe.

by TmEE on 2011-06-01 (#78965)

The PPU output is high impedance and unable to drive any TV directly, the transistor is just a simple buffer to beef up the signal. The weak signal from PPU will pick up noise form address lines etc. particularly notable if you just press you finger against the video out and some digital I/O line... the signal is very weak.