I think it might be more of an accepted idea if you were able to put together a kind of pro/con list for Bunnyboy.
Give him the list of people would commit to purchasing an AVS if it was implemented (you mentioned you knew several) and answer questions like:
Do you know if the speed running community will even accept it regardless if implemented?
What number of additional sales would it lead to?

Ultimately what you're asking for is a limited feature which has really no usefulness to the average home gamer and could cause compatibility issues when enabled (which leads to support requests which take time/money to reply to).

The issue here is that the screen telling them to push a button only if they see the screen is displayed after the change. One of the first responses people do when the screen goes blank is to start hitting buttons and stuff to try to jog the system out. This has the effect of getting the AVS stuck in an unsupported display mode.

Perhaps it would be wise to update the firmware to always display 60Hz during the menu and only switch to 50Hz when a game is booted, except when performing a 50Hz display test from the menu. This way NTSC users won't get stuck with unusable signal if they can return to a 60Hz option menu to change it back.

 

We are not discussing the difference between 50 & 60 Hz.

It is a discussion to meet NES accuracy which is a setting for 60.0988 Hz.

Make your own NES clone hardware if you care about it so much. If you lack the ability to, then perhaps this is not the hill you should die on.

Please keep the discussion civil and about relevant factual information.

It is up to BunnyBoy as to whether he would like to implement the feature or not.

Before you get too defensive of the criticism, let me explain the logic train the several users missed when I brought up pal/NTSC. You jumped all over me when I mentioned it saying its a separate issue not on the table for discussion. I used this already existing option to illustrate why timing cycles will probably not be changed. Not because the option itself is involved in your OP.

Here we go.

You give a user a button and tell them "don't touch that button because you don't need it and won't notice the difference and your tv probably can't accept it anyways." Some users will listen. Some may never go in to find the button. But a few will ignore what is said and will always press the button, because it's there and they are curious. And then flash! The screen goes away while 10 seconds count down waiting for input to confirm the change. The user freaks out. They start pressing buttons. Confirmed. Now their system is in an unusable mode. And they can't fix it. They turn it off and on, still can't see anything, and can't change it back. So they go to support and ask for help. Brian has to walk them through troubleshooting and help fix the problem, while they may or may not be truthful or explain what happened very well. This takes time and money for him to do.

This already is happening with users who pressed the "PAL" button.

The kicker though? Many modern tvs accept 50hz because it is a used/accepted international rate. Mine does in the USA. Yet this is still happening.

But what about an out of spec refresh rate that a far smaller percentage of tvs can handle properly? How many more issues will be created? How much time will be spent fixing it because 5 or 10 people cried out, not because they knew it would work on their system, but because they wanted to test and see what would happen because they didn't know yet? That is a very big risk for almost no return.

You can't guarantee it will work. He tested already and it works so little of the time that he made a decision to not make it an option. He already addressed it. I would not anticipate that it will be changed in the future.

I read through his entire thread and am having a hard time tying to point to exactly what the issue is.

1) Casual players don't care about the insignificant "delay." It has no bearing on their experiance as the end user.

2) While perhaps an issue for competitive players/speed runners, most of the folks in this lot would never play on anything but an original console anyway, so this point is moot.

Perhaps someone could clarify exactly what, and more
importantly why, it needs to be fixed (or the user given an option)?

Yeah, I think this is the bigger issue. The OP said he talked to several Twitch users who would be interested, but that's still a small part of the potential AVS market. Not to make them insignificant, but I would imagine most speedrunners want a common starting point, which would be an original NES on a CRT. And even if you could sort out the issues with the AVS in this regard, I think many still won't adopt it.

 

It's 0.2% people. Play for an entire hour and your system a handful of seconds behind an original NES. Really who the hell cares? For world record attempts, you need to count actual "frames", not hh:mm:ss format. Even when recording off real hardware, some capture cards will sync to the NES but record the video frames at 59.94Hz. This will often result in the audio track drifting slightly out of sync after a long time. Other capture cards may lock the refresh rate to 59.94 using a frame buffer, and drop a couple NES frames every minute or so, since the NES refresh rate is slightly higher than the capture rate.

So by using any digital capture medium other than analog VHS, the frame count will be locked to 59.94Hz and may or may not be cycle accurate to the game. Even the most accurate emulators use Vsync to prevent sprite tearing, locking on to your Monitor refresh at 60Hz.

So even recording original hardware using various capture devices can result in variances depending on the hardware used. For casual play as well as Twitch streaming, the AVS if perfectly acceptable. Not sure why the twitch guys got up in arms over it. Sounds like someone's just butthurt because HDMI consoles don't output an authentic analog composite signal. And if they demand authentic analog outputs, then they can wait for the NT2 and pay 2.5x more for the privilege. And I'm sure the NT2 will bend the HDMI output off the original composite spec exactly the way the AVS and the HiDef mods currently do.  

So true, but the ultimate HD mods and systems would win over even the elite players.  CRTs are big, heavy, and use a lot power, replacing them is always nice. 

Never give up on shaving off that lag.  So impressive they've come this far.

The solution for this is to make the required button combination complex enough so that people won't hit it accidentally.

 

I hate to say it but I also have a hard time believing the difference is as small as what is being mentioned here when you can see it in less than 4 minutes of recording.

Here is a link to the exported comparison video on youtube in case anyone missed it:




I also think it is more complicated than just saying the Hz are different since when I play or record my NESRGB NES through a FrameMeister's RGB in & HDMI out side by side with an unmodified NES it takes 8 hours to notice any difference at all.

The NESRGB mod does not interact with the CPU on an NES which leads me to believe the problem may lie in the CPU clock itself and not the PPU but BunnyBoy would know more about those details than any of us.

The issue here is that the AVS must be locked to exact HDMI framerate in order to be compatible with all existing HDTVs. Keeping native NES timings and compatibility would mean the NES PPU and display will need "desync", with the result that a framebuffer is used. This adds a minimum of 16ms of lag as well a larger FPGA is necessary to store the buffer. So for the AVS to have perfect timing and perfect compatibility, it would need a more expensive FPGA and result in a laggy gameplay experience. By that point, you might as well just be playing an emulator box.

AVS has zero buffer because it literally sends the scanlines to the HDMI port as they are rendered by the PPU. This results in an effective signal lag of one scanline or a few microseconds. This is a far cry from the type of lag experienced, often dozens of milliseconds lag added by expensive upscale devices like Framemeister or xRGB Mini. The AVS not only is cheaper than these aftermarket scaler devices, but there is no analog-to-digital conversion because the signal never leaves the digital domain.

So essentially you have a nearly flawless high definition, lag-free gameplay experience, at the expense of running .2% slower than the native NES clock. so you purists can keep your RGB mods and fancy upscalers, and lose every time at games like Punchout and Megaman because of the added lag. I will be gaming on the AVS and experiencing pure unmolested digital bliss.

That video fits the math exactly.  At 9:50.25 on the on screen clock the NES Mario starts moving, at 9:51.22 the AVS starts going.  1 sec difference in 590 secs = 0.17%. = 0.03ms/frame.

  Framemeister uses frame buffers so it might drop/duplicate frames to get the HDMI to 60hz.  Someone needs to buy me one for xmas to test stuff like this   

  CPU and PPU speed are directly tied together from the same common clock.  Adjusting one without the other will break some games.