Super Famicom Jr RGB mod: Amp or Resistors or straight out?

[niall]

G'day, I've been diagnosing a strange RGB issue with my Super Famicom Jr. I followed this guide to RGB mod it (http://gamesx.com/wiki/doku.php?id=av:snes2rgb) which makes no mention of an RGB amp, but I did neglect to add the 75ohm resistors.

Other guides like this one (http://retrorgb.com/snesminirgb.html) say you also need an RGB amp, and still other guides have just said to connect RGB straight to the multi-out with no resistors or amp.

Which one is correct?

EDIT: I'm using an NTSC cable with 220uF capacitors, and this is the Super Famicom Jr (Japan) not the US SNES2 (sounds like they're identical but just to clarify)

[muckyfingers]

G'day, I've been diagnosing a strange RGB issue with my Super Famicom Jr. I followed this guide to RGB mod it (http://gamesx.com/wiki/doku.php?id=av:snes2rgb) which makes no mention of an RGB amp, but I did neglect to add the 75ohm resistors.

Other guides like this one (http://retrorgb.com/snesminirgb.html) say you also need an RGB amp, and still other guides have just said to connect RGB straight to the multi-out with no resistors or amp.

Which one is correct?

EDIT: I'm using an NTSC cable with 220uF capacitors, and this is the Super Famicom Jr (Japan) not the US SNES2 (sounds like they're identical but just to clarify)

Use the resistors, 75-100 ohms. The RGB signal is very bright and can cause colors to look washed out, if not using the resistors. You will only need an RGB amp if you plan on using it on an Arcade RGB monitor.

[niall]

Ok cheers. Any idea if not using resistors could have resulted in my Trinitron vs. Loewe issue with the UMSA? (http://shmups.system11.org/viewtopic.ph ... 9#p1006369) Signal works fine direct to the Trinitron, but if fed through the UMSA to Trinitron it doesn't display anything (sync has locked though) but it's fine on the Loewe through UMSA.

I'm feeding everything to an Extron Crosspoint over BNC, so the UMSA is connected up there to feed my SCART displays whereas I'm using BNC for the rest.

EDIT: and should I be taking RGB from the SRGB-A chip, or from the holes like the latter retrorgb.com guide? So many different guides out there!

[muckyfingers]

Ok cheers. Any idea if not using resistors could have resulted in my Trinitron vs. Loewe issue with the UMSA? (http://shmups.system11.org/viewtopic.ph ... 9#p1006369) Signal works fine direct to the Trinitron, but if fed through the UMSA to Trinitron it doesn't display anything (sync has locked though) but it's fine on the Loewe through UMSA.

I'm feeding everything to an Extron Crosspoint over BNC, so the UMSA is connected up there to feed my SCART displays whereas I'm using BNC for the rest.

EDIT: and should I be taking RGB from the SRGB-A chip, or from the holes like the latter retrorgb.com guide? So many different guides out there!

I soldered mine directly to the chip, but the holes should be the same thing. Your issue could have something to do with the resistors, the generic RGB/Scart to YUV converters won't work correctly with the Sega Genesis unless the RGB lines have 75 ohm resistors in place.

[niall]

Anyone know why some have amped RGB? Been reading and it sounds like 105Ohm will get closest to 0.7vpp - I can't understand why you would amp it though..?

[niall]

Anyone know why some have amped RGB? Been reading and it sounds like 105Ohm will get closest to 0.7vpp - I can't understand why you would amp it though..?

Or we're they amping another source of RGB and bypassing?

[ApolloBoy]

Anyone know why some have amped RGB? Been reading and it sounds like 105Ohm will get closest to 0.7vpp - I can't understand why you would amp it though..?

You will only need an RGB amp if you plan on using it on an Arcade RGB monitor.

[niall]

I can't understand why you would amp it though..?

You will only need an RGB amp if you plan on using it on an Arcade RGB monitor.

Sorry I should have clarified - I was comparing RGB off the CPU and amping it vs. just taking it from the RGB chip. I did see the reference to Arcade monitors above but I glossed over it as it's out of the ballpark for me - I'm pretty focused on consumer 32" and 29" CRTs for the moment.

I still haven't solved the Trinitron mystery though - no change there, but I'm still feeding a pair of 32" Loewes via SCART fine. 1 with RGBS BNC->UMSA->SCART, and the other with a direct BNC -> SCART which I slapped together in a hurry. Neither UMSA or BNC->SCART cable gives any video on the Trinitron, whereas a RetroGamingCable SCART works fine on the Trintron. All I can think of now is that I have dry joints on the female SCART->BNC, but the Loewes are getting enough signal and amplifying it, but the Trinitron can't cope, or .... ??? I'll re-solder the female SCART in the morning and if that fails I guess I'll pick up someone else's female SCART->BNC to compare *shrug*

What's really exciting though is that this Loewe Calida 32" Q2300, while being a horrible 100Hz set like the Q2400 next to it, it actually disables 100Hz on the RGB input! I've never seen or heard of this before on any of the 100Hz models, and while I'm about to install a VGA card into it it's a handy feature to have which I wasn't expecting!

[Zets13]

I think using the RGB amp suggested in the RetroRGB guide helps to remove the bar that appears in the center of the screen on many SNESs.

[Pasky]

No, the vertical bar is caused by the DRAM refresh. You can't fix it unless you put a lowpass filter on the vcc lines of the 1chip (or the 2 GPU's & CPU on a regular SNES). You can band aid it by putting a big cap over the regulators 5V and GND but it'll still be visible if your mini/regular SNES has it, but just faintly.

People amplify the mini's RGB off from the encoder because they claim the mini's RGB levels are higher than a normal SNES if you grab it from the encoder.

I've measured both with my scope on the MINI and a regular 1CHIP snes and the Vp-p levels are identical. So it's all just bleh.

[niall]

Trinitron mystery solved - the UMSA doesn't output the RGB signalling voltage on pin 16 of its SCART output.

[RGB0b]

I think using the RGB amp suggested in the RetroRGB guide helps to remove the bar that appears in the center of the screen on many SNESs.

Yes, that is the intention. Also, I find that mod much easier than the alternative, which requires you to solder wires to the tiny pins on the S-RGB chip. It takes a bit longer, because you have to assemble the amp circuit, but overall it's much easier. Also, the quality is identical, so it's not like there's a negative side-effect.

I provide links to the other way of doing it on the top of the page, in case anyone would just prefer to use the internal amp method.

No, the vertical bar is caused by the DRAM refresh. You can't fix it unless you put a lowpass filter on the vcc lines of the 1chip (or the 2 GPU's & CPU on a regular SNES). You can band aid it by putting a big cap over the regulators 5V and GND but it'll still be visible if your mini/regular SNES has it, but just faintly..

I've personally done the RGB amp mod on a few SNES systems (both "fat" 1CHIP and mini) that had the vertical line problem. After bypassing the internal amp and using an external, the line went away.

I've only tested this on a handful of systems though, so it's possible it's just a coincidence. I seriously doubt it, but who knows?

People amplify the mini's RGB off from the encoder because they claim the mini's RGB levels are higher than a normal SNES if you grab it from the encoder.

That doesn't really matter...no matter where you get RGB from, if it's too bright, just use slightly higher resistors. I find 100ohm to be perfect, but 75, 121 and 130 will also work.


Sorry, I didn't mean to butt into this discussion, I just don't want people thinking there's bad info on my SNES Mini RGB mod page. I've done so many SNES Mini mods using both methods for all of my friends, all of their friends, etc and they all work flawlessly and look great.

[niall]

Sorry, I didn't mean to butt into this discussion, I just don't want people thinking there's bad info on my SNES Mini RGB mod page. I've done so many SNES Mini mods using both methods for all of my friends, all of their friends, etc and they all work flawlessly and look great.

No worries man - nothing at all wrong with your instructions - I just got confused between amping CPU vs. resistors on the existing signals. In the end it turned out to be a faulty UMSA, was missing 5v on pin 16 SCART so while I saw 4.5v on sync I got worried when I couldn't find ~1v on RGB. I still need to investigate that, maybe my multimeter skills need improving, but it's all working great now.