SNES/Famicon PCB Revisions and RGB Video

The place for all discussion on gaming hardware
User avatar
James-F
Posts: 87
Joined: Fri Mar 23, 2018 11:01 am

Re: SNES/Famicon PCB Revisions and RGB Video

Post by James-F »

Why not continue this thread and fix the 2chip blur?

At the output of the PPU2 there are Q3-Q8 which buffer the RGB signal from the PPU2 to the S-ENC encoder and AV Multiout.
Did someone try to completely remove these Q4-Q8 transistors and rebuild the output?
If the PPU (DAC) itself is any similar to the 1CHip or the N64 why not try to use Current Sink resistors at the DAC output instead of these Transistors?


http://www.chrismcovell.com/gotRGB/snesblur.html
Oscilloscope images shows that there is stupidly slow rise and fall times in the 2chip SNES, which I assume comes from theses transistors after the PPU.
I do not have a spare 2chip or an oscilloscope to tinker with, so I need some help from the community to try make the 2chip great again,
especially when the 1Chip is so damn rare and expensive this age, plus, nothing beats the 2chip for compatibility and they are plentiful and cheap.
borti4938

Re: SNES/Famicon PCB Revisions and RGB Video

Post by borti4938 »

These slow rise times come out of the S-PPU2. I’ve lifted the RGB pins long time ago and measured it on the oscilloscope.
Also the outputs were not build to drive load resistors to GND.
User avatar
James-F
Posts: 87
Joined: Fri Mar 23, 2018 11:01 am

Re: SNES/Famicon PCB Revisions and RGB Video

Post by James-F »

Ah bummer.
I assume the only solution would be removing the PPU2 and replacing it with a FPGA like the NESRGB.
Last edited by James-F on Sat Mar 31, 2018 10:28 am, edited 1 time in total.
User avatar
FBX
Posts: 2347
Joined: Wed Feb 18, 2015 10:18 am
Location: DFW area, Texas
Contact:

Re: SNES/Famicon PCB Revisions and RGB Video

Post by FBX »

James-F wrote:Ah bummer.
I assume the only solution would be removing the PPU2 and replacing it with a FPGA like the NESRGB.
Which you might as well get a Super Nt. It's got the compatibility of the 2-chip with perfect digital upscaled output.
User avatar
James-F
Posts: 87
Joined: Fri Mar 23, 2018 11:01 am

Re: SNES/Famicon PCB Revisions and RGB Video

Post by James-F »

I take back what I said about not liking the 1Chip because of on-screen glitches, they are really minor and few.
After modding, tuning, and testing many favorite games, I find myself very pleased with the 1Chip.
syboxez
Posts: 94
Joined: Sun Jul 24, 2016 4:33 am

Re: SNES/Famicon PCB Revisions and RGB Video

Post by syboxez »

James-F wrote:Ah bummer.
I assume the only solution would be removing the PPU2 and replacing it with a FPGA like the NESRGB.
Is that even a possibility? Something like what the NESRGB or Hi-Def NES does would be amazing on the SNES, but I'm not sure if that's possible. Maybe someone who knows more than me could chime in. SMD isn't a problem for me, but you could manage by lifting pins and having a QSB for it.
PixelDharma
Posts: 108
Joined: Tue Jan 18, 2011 6:47 am

Re: SNES/Famicon PCB Revisions and RGB Video

Post by PixelDharma »

Another batch of Voultar’s RGB bypass board came and went. I waited weeks to buy one, and then it turns out he doesn’t ship outside the US, not even to here in Canada. Now it’s sold out again.

Does anyone have any advice for obtaining his board outside the US?
User avatar
parodius
Posts: 720
Joined: Wed Jan 26, 2005 5:54 am
Location: Singapore

Re: SNES/Famicon PCB Revisions and RGB Video

Post by parodius »

My sales thread : 2020/07/20..MASTER.VER.
User avatar
James-F
Posts: 87
Joined: Fri Mar 23, 2018 11:01 am

Re: SNES/Famicon PCB Revisions and RGB Video

Post by James-F »

PixelDharma wrote: Does anyone have any advice for obtaining his board outside the US?
If you can solder just buy everything on ebay for like 3$ and DIY.
borti4938

Re: SNES/Famicon PCB Revisions and RGB Video

Post by borti4938 »

Or just use open source designs or look at videogameperfection.com for modding kits. There are soooo many options ;)
PixelDharma
Posts: 108
Joined: Tue Jan 18, 2011 6:47 am

Re: SNES/Famicon PCB Revisions and RGB Video

Post by PixelDharma »

^ thanks all. Good to know it isn’t his exclusive part ;) I wasn’t totally sure.
SamIAm
Posts: 475
Joined: Thu Mar 03, 2005 1:09 am

Re: SNES/Famicon PCB Revisions and RGB Video

Post by SamIAm »

I've got a 1CHIP-02 Super Famicom that I play pretty much exclusively on old broadcast monitors. Since I haven't done any brightness fixes, the image is ... well, too bright. However, simply turning down the monitors' contrast knobs until white no longer blooms, which is extremely fast and easy, seems to do a pretty decent job of bringing things into balance.

Would doing the 750 ohm resistor fix outlined on RetroRGB's page do anything to improve my image quality compared to this contrast adjustment? If it only makes things more convenient, then I don't really need it. I suppose attenuated signal levels would give me less AC coupling? Is there anything else?
User avatar
Syntax
Posts: 1774
Joined: Wed Aug 09, 2017 12:10 am
Location: Australia

Re: SNES/Famicon PCB Revisions and RGB Video

Post by Syntax »

Look at it like this.

You tv set accepts .7vpp video signals.
Without the termination resistors your Famicom is outputting closer to .8vpp.

A great way to test this is the IRE option in 240p suit.
Currently you will not see a dramatic difference between 100% IRE and 90% IRE.
Adding the resistors will fix this.

As soon as the signal goes beyond. 7vpp you lose that information forever. Lowering brightness will not bring it back. The tv cannot read beyond .7vpp.

So while you can lower contrast you are still missing a good percentage of video signal.

I think those resistors can be fit to the scart head for ease.
SamIAm
Posts: 475
Joined: Thu Mar 03, 2005 1:09 am

Re: SNES/Famicon PCB Revisions and RGB Video

Post by SamIAm »

It seems to me that if the monitor's input is really clipping the signal at 0.7vpp, then 100% white from my Super Famicom should be the same as 100% white from any other system, but it's not - it's brighter. I could understand if the range above 0.7vpp is getting distorted somehow without the brightness fix, but is that really what's happening?
User avatar
Syntax
Posts: 1774
Joined: Wed Aug 09, 2017 12:10 am
Location: Australia

Re: SNES/Famicon PCB Revisions and RGB Video

Post by Syntax »

The range above clips it's not distorted.
Imagine a bar, anything above is gone.


Now think of in game brightness as non linear, most brighness settings will not be 100% they will sit lower but by running the .8vpp you have pushed the majority closer to the top, so it looks brighter than your other systems.


Test with 240p suit IRE to see how bad your white crush is.
SamIAm
Posts: 475
Joined: Thu Mar 03, 2005 1:09 am

Re: SNES/Famicon PCB Revisions and RGB Video

Post by SamIAm »

Syntax wrote:Test with 240p suit IRE
That's exactly what I'm talking about. 240p test suite 100 IRE (or any other 100% white) is brighter on the Super Famicom than it is on other systems. How can that be if the signal is being clipped?
User avatar
Syntax
Posts: 1774
Joined: Wed Aug 09, 2017 12:10 am
Location: Australia

Re: SNES/Famicon PCB Revisions and RGB Video

Post by Syntax »

SamIAm wrote:
Syntax wrote:Test with 240p suit IRE
That's exactly what I'm talking about. 240p test suite 100 IRE (or any other 100% white) is brighter on the Super Famicom than it is on other systems. How can that be if the signal is being clipped?
It's brighter because your sending info which was sitting lower to the top.

The bar was probably a poor analogy. It's more like a ceiling. Forgive me I'm a builder not an EE.

What does 100% ire look like compared to 90% ire? Can you see a difference?
SamIAm
Posts: 475
Joined: Thu Mar 03, 2005 1:09 am

Re: SNES/Famicon PCB Revisions and RGB Video

Post by SamIAm »

Syntax wrote:What does 100% ire look like compared to 90% ire? Can you see a difference?
I won't be able to test this for at least several hours, unfortunately.
It's brighter because your sending info which was sitting lower to the top.
I'm afraid I don't understand what you mean by this.

This is basic clipping, right?

Image

If the wave on the left is my Genesis and the wave on the right is my Super Famicom, shouldn't their 100% white be the same, i.e. the level of the solid red line?
User avatar
Syntax
Posts: 1774
Joined: Wed Aug 09, 2017 12:10 am
Location: Australia

Re: SNES/Famicon PCB Revisions and RGB Video

Post by Syntax »

See how the wave on the leftt only touched the top line but the wave on the right slammed into it and is in contact for a longer period?
SamIAm
Posts: 475
Joined: Thu Mar 03, 2005 1:09 am

Re: SNES/Famicon PCB Revisions and RGB Video

Post by SamIAm »

Syntax wrote:See how the wave on the leftt only touched the top line but the wave on the right slammed into it and has a larger amount touching for a longer period?
That would explain a longer period of white, but not greater amplitude.
Last edited by SamIAm on Sat Apr 07, 2018 3:37 am, edited 2 times in total.
User avatar
Syntax
Posts: 1774
Joined: Wed Aug 09, 2017 12:10 am
Location: Australia

Re: SNES/Famicon PCB Revisions and RGB Video

Post by Syntax »

So your saying a completely topped out/clipped signal should look the same on your tv as a nice .7vpp signal?
(All white image)
Last edited by Syntax on Sat Apr 07, 2018 3:48 am, edited 1 time in total.
SamIAm
Posts: 475
Joined: Thu Mar 03, 2005 1:09 am

Re: SNES/Famicon PCB Revisions and RGB Video

Post by SamIAm »

Syntax wrote:So your saying a completely topped out/clipped signal should look the same on your tv as a nice .7vpp signal?
For 100% white, yes. To go back to that image I posted above, both the 0.7vpp Genesis and the 0.8vpp Super Famicom should be sitting at the same place: right on the ceiling, i.e. the red line.

Now that you mention it, I do remember playing this Super Famicom on a consumer Sony and not seeing noticeably brighter whites, so that TV was probably clipping the signal. The broadcast monitors I use now, on the other hand, display the Super Famicom's 100% white as far brighter than any other system's.
User avatar
FBX
Posts: 2347
Joined: Wed Feb 18, 2015 10:18 am
Location: DFW area, Texas
Contact:

Re: SNES/Famicon PCB Revisions and RGB Video

Post by FBX »

SamIAm wrote: The broadcast monitors I use now, on the other hand, display the Super Famicom's 100% white as far brighter than any other system's.
If they are CRT, perhaps they can handle the extra voltage and truly are overblowing the white levels. I imagine if you were to switch to a digital display, the RGB conversion would cause clipping and crush your SNES color range. I did the 750 ohm attenutation on my lines and works great for me.
SamIAm
Posts: 475
Joined: Thu Mar 03, 2005 1:09 am

Re: SNES/Famicon PCB Revisions and RGB Video

Post by SamIAm »

Yep, they're CRT. With luck, they'll last me a long while yet, but I'll be sure to check for clipping in any digital converter I use.

By the way, how does your white level compare to other systems with the 750 ohm resistors in place? I've seen a couple of people suggest going with something more like 800 to keep it from being too dark. Ideally, I'd like the SFC to be right in line with other systems.
Last edited by SamIAm on Sun Apr 08, 2018 12:07 am, edited 1 time in total.
nissling
Posts: 454
Joined: Sun May 10, 2015 8:12 am
Location: Stockholm, Sweden

Re: SNES/Famicon PCB Revisions and RGB Video

Post by nissling »

SamIAm wrote:If the wave on the left is my Genesis and the wave on the right is my Super Famicom, shouldn't their 100% white be the same, i.e. the level of the solid red line?
No, because 100% white isn't the peak for analoge video. It ranges from -7.5% to 109% amplitudes. For SMPTE, legal levels are 0-109% and for digital this was made into video levels, which are usually considered to be 16...235 but in reality, 235 is just the reference point for white. On a properly calibrated display you should have levels all the way up to 255, though far from all systems allow this.
User avatar
TheShadowRunner
Posts: 273
Joined: Sun Feb 24, 2013 7:41 pm

Re: SNES/Famicon PCB Revisions and RGB Video

Post by TheShadowRunner »

nissling wrote:
SamIAm wrote:If the wave on the left is my Genesis and the wave on the right is my Super Famicom, shouldn't their 100% white be the same, i.e. the level of the solid red line?
No, because 100% white isn't the peak for analoge video. It ranges from -7.5% to 109% amplitudes. For SMPTE, legal levels are 0-109% and for digital this was made into video levels, which are usually considered to be 16...235 but in reality, 235 is just the reference point for white. On a properly calibrated display you should have levels all the way up to 255, though far from all systems allow this.
That's why Sam said their 100%, he was talking relative to the maximum white level they can output, I think.
Anyway, on 1CHIP (connected to a Xrgb mini via RGB) I see a difference between 90% and 100% IRE, it's not obvious but it's there so I haven't pushed further.
What I wonder though is why mod the system itself, isn't it possible to rectify the RGB signal in the cable? Much more graceful imho.
User avatar
Syntax
Posts: 1774
Joined: Wed Aug 09, 2017 12:10 am
Location: Australia

Re: SNES/Famicon PCB Revisions and RGB Video

Post by Syntax »

Better to attenuate signals before they travel down long lines, although in this case the difference would be bugger all, and it makes the mod work with stock scart cables.
User avatar
James-F
Posts: 87
Joined: Fri Mar 23, 2018 11:01 am

Re: SNES/Famicon PCB Revisions and RGB Video

Post by James-F »

Analog RGB inputs in CRTs have some headroom before clipping; it's not like if amplitude goes above 713mV it is certainly clipped, no.
With stock SFC 1Chip-02 I can clearly see all IRE steps with equal intensity between them on my Sony KV-29FX20K.
0.8vAC peak will most definitely NOT clip any RGB inputs on a CRT, but it will be slightly brighter than other 'on spec' consoles.

When I modded my N64 I had almost 1.5V amplitude on the RGB lines,; when I turned contrast down to almost zero on the CRT the image went back to normal without any clipping.
So there is definitely quite a lot of voltage headroom in the analog RGB inputs in CRTs.

The problem starts when the analog voltage is digitized to exact specification,, where in the digital domain there is no headroom outside of spec.
This can bee seen in Framemeister (don't own) and the OSSC (own).
The OSSC TVP7002 chip has pre-digitization Gain control so levels of the 1Chip can be fixed without any modification to the vintage hardware, I do not know about the Framemeister.

@borti4938
Will it be possible to add a Coarse Gain/Offset (page 12, 38) which scales 0.5Vpp-2.0Vpp to the 10bit ADC of the OSSC?
Currently the OSSC has the Fine Gain/Offset controls which are post ADC, while all Coarse parameters are PRE ADC and missing from the OSSC.
I think the Pre (coarse) ADC controls are much more important, than the Post (fine) controls that can be done on the LCD anyway.
It will be really nice not having to mod a stock 1Chip or any other slightly off-spec console with the OSSC once the Pre ADC control are implemented. :wink:
http://www.ti.com/lit/ds/symlink/tvp7002.pdf

edit:
I'll post the request for borti in the OSSC thread also.
User avatar
James-F
Posts: 87
Joined: Fri Mar 23, 2018 11:01 am

Re: SNES/Famicon PCB Revisions and RGB Video

Post by James-F »

As said, the clipping occurs in the digitization stage of the OSSC or Framemeister, NOT in the 1Chip itself.

paul_nl posted a test firmware for the OSSC which has Pre-ADC-Gain control that can accept up to 2.0vPP without clipping.
https://www.videogameperfection.com/for ... -controls/
Furthermore, paul_nl mentioned that the Framemeister also has this option called "ADC Level" so there is no need to modify your vintage fat 1chip console at all in the first place to correct the levels.

If you have a fat 1chip that already has RGB but you've yet to install the 750/1.2k resistors, please try this option.
User avatar
TheShadowRunner
Posts: 273
Joined: Sun Feb 24, 2013 7:41 pm

Re: SNES/Famicon PCB Revisions and RGB Video

Post by TheShadowRunner »

James-F wrote:As said, the clipping occurs in the digitization stage of the OSSC or Framemeister, NOT in the 1Chip itself.

paul_nl posted a test firmware for the OSSC which has Pre-ADC-Gain control that can accept up to 2.0vPP without clipping.
https://www.videogameperfection.com/for ... -controls/
Furthermore, paul_nl mentioned that the Framemeister also has this option called "ADC Level" so there is no need to modify your vintage fat 1chip console at all in the first place to correct the levels.

If you have a fat 1chip that already has RGB but you've yet to install the 750/1.2k resistors, please try this option.
hmm there is no such thing as an "ADC level" setting on the Mini.
http://www.micomsoft.co.jp/XRGB-mini_Ver203_OSD_E.pdf
Is paulb_nl refering to "A/D LEVEL", same thing?
But on my unmodified fat 1CHIP-01 + Mini I do not have the shading becoming solid starting from column C (as on the MLiG RGB309 video @ 8:22), I see distinct columns up to F.
I use default A/D LEVEL setting on the mini and a SHVC-010 RGB cable.
Edit: so now I'm starting to wonder if certain 1CHIPs, maybe the early revisions, do not have this white crush issue at all?
Last edited by TheShadowRunner on Fri Apr 13, 2018 8:14 pm, edited 1 time in total.
Post Reply