SNES/Famicon PCB Revisions and RGB Video

[RGB0b]

Borti, what do you think? This is the new board we've been messing with, not the ones being sold. Isn't there an extra filter feature of the THS7374 that can be toggled?

[borti4938]

I wanted to redesign it anyway. Hence I can add a solderpad for disabling the filters anyway.
At the moment I'm thinking about leaving away the brightness adaptation from and keep it outside e.g. by soldering additional resistors on top of R6, R7 and R8 or replacing them or by soldering a resistor in front of AVCC.
What do you think?

[RGB0b]

Edit:

I wanted to redesign it anyway. Hence I can add a solderpad for disabling the filters anyway.
At the moment I'm thinking about leaving away the brightness adaptation from and keep it outside e.g. by soldering additional resistors on top of R6, R7 and R8 or replacing them or by soldering a resistor in front of AVCC.
What do you think?

Also, to answer your question, I think it would be best to keep the brightness resistors on the board to make things as simple as possible for end users. My favorite feature of the newer board was having amplified csync, so people can get RGBs easily from the via's on the bottom of the board and not have to solder on top. That might seem silly to those of us who do this stuff all the time, but little details like that make it easier for beginners to add RGB to their consoles.

[xga]

Hey Borti,

I recently picked up a SNES RGBAmp V2.1 board for my Super Famicom Jr. I noticed on your Osh Park profile that the board is now up to V3.3b. I wasn't able to find the list of changes between V2.1 and V3.3b (perhaps they're in this thread and I should go back and read it from the beginning?). Is it worth me skipping this revision and getting the latest one or are the changes mainly for compatibility or something?

[RGB0b]

I enlisted Bob's help on the PQ issue I posted the pic of concerning the SNES mini's "borti" board mod. We're pretty certain now that it's an issue with the board and not the SNES mini. Bob sent me a screencap under the same zoom settings on a 1CHIP-03 modded with the same board, and it looked to have the exact same PQ issues as the SNES mini pic I posted.

Okay, I turned the filter off by connecting pin 9 to 5v rather then ground (page 7, thanks to db Elec for the tip): http://www.ti.com/lit/ds/symlink/ths7374.pdf

That's better!:

[FBX]

Much better! Now it looks consistent in quality with the unmodded 1CHIP consoles.

[nissling]

So finally it looks like the package has arrived to Sweden.

And I got it today. The difference between the CPU-01 and 1CHIP01 is so huge I won't even be making any comparisons. Just at the startup menu of the 240p test suite I was nearly floored by its quality. This is simply fantastic.

[mvsfan]

they really dont compare well.

[RGB32E]

Okay, I turned the filter off by connecting pin 9 to 5v rather then ground (page 7, thanks to db Elec for the tip): http://www.ti.com/lit/ds/symlink/ths7374.pdf

Very cool! I recall seeing the LPF pin when I installed a THS7374 in a mini 3+ years ago, but never tried bypassing the LPF!

Looks like Micomsoft left provisions for bypassing the LPF in their application of the THS7374 in the XSYNC-1, but decided to enable it. I had wondered if the switch on the underside of the XSYNC-1 was a toggle for this, but have confirmed it isn't wired for that function.


R40 connects THS7374 pin 9 to ground via a 4.7k ohm resistor. R42 is left unpopulated - LHS is 5VDC, RHS is pin 9. Remove R40 and jumper (or use a resistor) R42 to bypass the LPF!

This brings up an interesting point though - The THS7314 used in many projects has an undefeatable LPF. Perhaps people can get a slight improvement in sharpness from your THS7314 circuits if you updated them for use with the THS7374!

[RGB0b]

Very cool! I recall seeing the LPF pin when I installed a THS7374 in a mini 3+ years ago, but never tried bypassing the LPF!

Oh yeah, I remember that post!

Looks like Micomsoft left provisions for bypassing the LPF in their application of the THS7374 in the XSYNC-1, but decided to enable it. I had wondered if the switch on the underside of the XSYNC-1 was a toggle for this, but have confirmed it isn't wired for that function.

I've never used the XSYNC-1 before. Isn't that just a SCART to DSUB box that strips sync and works with the XCAPTURE-1?

This brings up an interesting point though - The THS7314 used in many projects has an undefeatable LPF. Perhaps people can get a slight improvement in sharpness from your THS7314 circuits if you updated them for use with the THS7374!

I believe the filter is slightly different in the THS7314. I want to see for sure, so I plan on using FBX's capture methods to compare a SNES Mini with the "Regular" RGB mod, THS7314, THS7374...and each with and without the brightness attenuating resistors. Should be done in about a week.

[RGB32E]

I've never used the XSYNC-1 before. Isn't that just a SCART to DSUB box that strips sync and works with the XCAPTURE-1?

It's a Japanese RGB21 splitter that outputs RGBS on the 8DIN (MSX pinout), RGBHV on the HD15 (VGA pinout), breaks out audio to RCAs and 3.5mm TRS. It was released in the same window as the XRGB-mini and the SC-500N1/DVI capture card.

I believe the filter is slightly different in the THS7314.

The slope is more gradual (5th order) and the corner frequency is lower on the 7314 (removes more info at a lower frequency).

From the datasheets:

THS7374:
– 6th-Order 9.5-MHz (–3 dB) Butterworth
– –1 dB Passband Bandwidth at 8.2-MHz
– 54-dB Attenuation at 27-MHz

THS7314:
– 5th-Order 8.5-MHz (–3dB) Butterworth
– –1dB Passband Bandwidth at 7-MHz
– 47dB Attenuation at 27-MHz

The lower the corner frequency, the blurrier things can get. The OSSC provides a good example, as it shows you the corner frequency being used by the LPF IC!

Should be done in about a week.

Looking forward to your results. I anticipate that the overall cleanest results will be with the bypassed THS7374.

Also, Tim's N64RGB uses the THS7374. I just checked a v1.1 and yes, the LPF is BYPASSED! This could explain why I noticed that the N64RGB output is a little clearer than the basic THS7314 mod! (Compare the MM shots)

[ZellSF]

Got a non-1CHIP SNES. Getting a dedicated SNES for Super Turrican before getting an Amiga... my prioritizations are a bit messed up.

First impression was oddly enough "this actually looks better than my 1CHIP". For a pretty simple reason: my TV has forced noise reduction that looks awful, the sharper image of the 1CHIP triggers it more. Run through a DVDO VP50 though and the 1CHIP looks noticeably better than the non-1CHIP, but not really to the point where I would bother buying one if I already had a good non-1CHIP. It's not a that huge difference.

Then again, I really don't care much about image quality (more than most average people, but not more than most people here).

A bit more on topic, Final Fantasy Mystic Quest is on one of those list of 1CHIP problems. That game has garbage in the overscan area even on non-1CHIP systems.

[Einzelherz]

Which chipset is it, out if curiosity?

[ZellSF]

Which chipset is it, out if curiosity?

No idea, no tools or motivation to open it.

[borti4938]

Also, Tim's N64RGB uses the THS7374. I just checked a v1.1 and yes, the LPF is BYPASSED! This could explain why I noticed that the N64RGB output is a little clearer than the basic THS7314 mod! (Compare the MM shots)

I don't know what you have seen, but the internal LP filters are NOT bypassed on v1.1 and later!!! The filters are only bypassed if and only if one closes J4.

[Pasky]

How is the LPF bypassed with the 7314 & 7374?

[borti4938]

With the THS7314 you cannot bypass the internal filters. With the THS7374 you can bypass them by setting pin 9 to high (bypassed) or low (not bypassed).

[Pasky]

I see, thank you.

[Seraphic]

So if you were using a SNES for example with a THS7374 that had its LPF bypassed, but plugged into an XSync-1 using JP21, it would have a LPF reapplied?

And on a side note, years ago I purchased a Nintendo's SHVC-010 as I was under the mistaken impression that it used C-Sync and in fact uses Composite-Sync. Sony's SCPH-1050 also uses Composite-Sync as C-Sync is not enabled on PlayStation. But after looking over RetroRGB's website, the PlayStation (think he is talking about doing this to a PS1 not PS2 here) can have its C-Sync enabled, but requires direct solder to the Video Chip so he just recommends sync-on-luma. But RetroRGB website is PAL focused so not sure if sync-on-luma carries over to the NTSC units. Would it be worth the trouble to enable C-Sync on a PS2 vs using Luma-Sync?

I have a SNES SHVC-010 going into one XSync-1 and a PS2 SCPH-1050 going into another XSync-1. But I guess I need a properly enabled C-Sync cable for the SNES and a Sync-On-Luma for the PlayStation 2. Not much luck finding them though as retrogamingcables only has PAL/SCART for sale with no ETA on JP/21

Oh and RGB32E, were you able to deduce what that switch on the bottom of the XSync-1 does?

[Einzelherz]

Are you having problems with the composite as sync or the luma sync? From what I understand if the lesser outputs work fine, it's best to leave them target than chasing that unicorn.

Even composite sync on PS1/2 works well with my monitors, but I know some processors are finicky.

[RGB0b]

Seraphic, I don't mean any disrespect, but your post doesn't really make sense. What exactly are you trying to do and how does it apply to this thread?

My site focuses on all revision consoles, not "PAL-focused". Are you confusing the terms "PAL" and "SCART"?

[Seraphic]

Are you having problems with the composite as sync or the luma sync? From what I understand if the lesser outputs work fine, it's best to leave them target than chasing that unicorn.

No, not really. Just recall that C-Sync recommended whenever possible.

Even composite sync on PS1/2 works well with my monitors, but I know some processors are finicky.

There would be no benefit using a cable wired for sync over luma instead of composite sync?

Seraphic, I don't mean any disrespect, but your post doesn't really make sense. What exactly are you trying to do and how does it apply to this thread?

Hey now, I started this thread. I am heading out to work, so didn't have much time to write it out as clearly as I should. But My post was primary about the SNES asking about the Xsync-1, SNES SHVC-010, THS7374 and LPF.

My site focuses on all revision consoles, not "PAL-focused". Are you confusing the terms "PAL" and "SCART"?

My mistake then, I misinterpreted. No, I understand the SCART connection. There is the PAL layout and the JPN layout.

[RGB0b]

There would be no benefit using a cable wired for sync over luma instead of composite sync?

In most scenarios, there would not be a benefit. The most common issue I've run into isn't related to video quality, it's compatibility: My NEC XM29, USB3HDCAP and a few switches I've tried will only accept csync and not luma or composite video.

Hey now, I started this thread. I am heading out to work, so didn't have much time to write it out as clearly as I should. But My post was primary about the SNES asking about the Xsync-1, SNES SHVC-010, THS7374 and LPF.

LOL, sorry. The (more expensive) shielded cables from both of the main cable sellers are excellent quality, so I'd personally use those with csync instead of an original OEM cable. Other people may disagree though.

I'm still working on the differences between using each THS chip vs the 3-wire mod...as well as getting new comparison shots of as many different SNES revisions as I can. Should be another week (have to wait for parts to arrive), but so far it's looking like all the Mini mods are pretty similar, provided the LFP is off on the 7374. Detailed pics coming very soon!

My mistake then, I misinterpreted. No, I understand the SCART connection. There is the PAL layout and the JPN layout.

From that perspective, you are correct: I focus on SCART and not JP21, since SCART is much more common.

[Seraphic]

In most scenarios, there would not be a benefit. The most common issue I've run into isn't related to video quality, it's compatibility: My NEC XM29, USB3HDCAP and a few switches I've tried will only accept csync and not luma or composite video.

Would there also be no benefit in re-enabling C-Sync on the PlayStation with the direct solder to the Video Chip you talked about in that paragraph on your website? And is that possible on the PS2 or just PS1? I mean, instead of using Composite or Luma sync.

The (more expensive) shielded cables from both of the main cable sellers are excellent quality, so I'd personally use those with csync instead of an original OEM cable. Other people may disagree though.

Who are the two main cable sellers? retrogamingcables said the do not know when they will be selling JP21 cables again.

[FBX]

Who are the two main cable sellers? retrogamingcables said the do not know when they will be selling JP21 cables again.

The other I know of is:

http://www.ebay.com/usr/retro_console_accessories

However, there's an on-and-off thing going with her store, where it will be basically closed for a couple days, then open again, and so on.

[RGB32E]

I don't know what you have seen, but the internal LP filters are NOT bypassed on v1.1 and later!!! The filters are only bypassed if and only if one closes J4.

I don't know either, as I double checked my v1.1, and yes, you have to close J4 to bypass the THS7374 LPF, as pin 9 is always tied to ground via a 4k7 resistor (same as the XSYNC-1). Perhaps the differences between the two LPFs account for the slight differences in picture clarity?

Oh and RGB32E, were you able to deduce what that switch on the bottom of the XSync-1 does?

I dug out the manual and translated it:

Code: Select all

(7) Sync Mode Select Switch
There are times when the image signal (sync signal) and device will not properly display the image from the SD-500N1/DVI, and also will not properly capture. In such situation, please turn the switch OFF. Please use OFF as the standard method.

I'd have to spend more time tracing the circuits in the XSYNC-1 in an attempt to figure out what specifically the switch changes with sync output, and if it only applies to the HD15/VGA output, or both outputs.

[mvsfan]

Seraphic: The other one would be Retro console accessories in the us. they sell Jp21 wired cables and you can get them sheilded .

[RGB0b]

Would there also be no benefit in re-enabling C-Sync on the PlayStation with the direct solder to the Video Chip you talked about in that paragraph on your website? And is that possible on the PS2 or just PS1? I mean, instead of using Composite or Luma sync.

It honestly depends on your setup. I'm sorry, I know that sounds like a cop-out answer, but it's the truth. On my Sony BVM monitor, I see zero difference between using luma and csync on my PS1. PS2 is more complicated though: You need to have an older revision with the right chip. I've never even been able to test this myself, I've only seen websites where other people did it.

Who are the two main cable sellers? retrogamingcables said the do not know when they will be selling JP21 cables again.

http://www.retrogamingcables.co.uk/
http://stores.ebay.com/Retro-Accessories

[FBX]

But RetroRGB website is PAL focused so not sure if sync-on-luma carries over to the NTSC units. Would it be worth the trouble to enable C-Sync on a PS2 vs using Luma-Sync?

A friend made me a Sync-on-Luma cable for my NTSC PS2 and it works beautifully. I wanted to buy one, but the usual suppliers were out of stock on the parts needed to mass produce them at the time.

[the_crayon_king]

If you guys want a cheap solution to the LPF issue you could use one of these (haven't tested them yet):

NS1 N64s:
https://easyeda.com/hotdog6394/N64_THS_7374-h30K6sOa8

SNES (designed for minis):
https://easyeda.com/hotdog6394/SNES_THS7374-Bb5ngyQwq

The snes one had room for a little toggle switch to be installed as well.