1-Chip with Tim Worthington's YPbPr Kit

[RGB32E]

Wonder how this would work for a launch model SNES? Since its SVideo ( at least on my personal original model console) is superior to its RGB>Component output through a 2100 clone, this solution, seeing that it uses the SNES Luma, might give the best picture possible for the console.

Order the kit and try it! I had a CSY-2100 a number of years back before buying a Kramer FC-14 and found that with a CPU1, using stock S-Video was overall similar to the CSY-2100 output. So yes, you should get superior component output using Tim's kit. Given that you are a component user with your F4500, I'd highly recommend giving this a shot.

Also, since the original models come with proper SVideo by default, maybe the signal levels are perfect

The issue being that the 1CHIP outputs RGB signals that are too high. Since they are routed to the RGB encoder, the resulting RGB, S-Video, and composite outputs are also too high. I generally found the original consoles output too low for my taste, so I'm fine with the higher than standard levels.

...and wouldnt require the resistor and caps?

The luma output from the RGB encoder on the SNES mini (or any other console for that matter) generally requires a 75 ohm resistor followed by a 220uF capacitor before output. Not to be confused with Skips assertion that the value of the resistor could be higher than 75 ohms (or adding an additional resistor on 1CHIP-0x revisions). The problem originates from the PPU/1CHIP outputting a higher than standard RGB output, so the fix would really be to attenuate the RGB levels BEFORE they are fed to the RGB encoder. If you're curious, here is a diagram for how luma is wired to the output:



So, it's not about with or without a resistor. The SNES revisions with stock S-Video do not require an additional resistor. The SNES mini requires a resistor as it's a required component that doesn't exist in the system.

Very interesting. I'd love to know how it compares to RGBs on the original model consoles.

In general I've found that extreme reds, greens, and blue look a little muted compared to RGB when converted to component. However, without a direct comparison, both can look great.

[Skips]

Lets make this easy for everyone now that I found my bag of resistors and had time to dick with this today.

**EDIT*
I noticed that although the contrast was fixed with the 120 ohm resistor the color was still way to intense compared to the other game systems in my setup. I have edited this post with updated instructions and a new schematic.

Here is a diagram for installing it in a 1-Chip/SNES 2 going directly off the video encoder.



I found that 120 ohms brought the luma levels down to acceptable (75 ohms in the 1-CHIP's circuit is too low) so I did not have to adjust any contrast levels on the display/device when switching back to the SNES from other consoles. THIS IS ONLY for the single PPU systems (1-Chip and SNES 2), the older 2 PPU models are fine to just tap luma straight from the video port (no additional resistors needed). Color is also a tad to intense so to fix this we need to remove the two resistors marked in the picture above and place 90 ohm resistors on the Pb anr Pr out lines. This will bring the colors down so you don't have to adjust chroma/color on your CRT displays when switching back to other systems.

The install will be slightly different for the 2 PPU revisions however I don't have one of these to play with so people will have to figure those out for themselves. The kit WILL work on the older models, I just do not have one to mod and do a diagram/write up for you guys.

I decided to just take some straight up CRT pics this time.




And yes RGB32 you can get away with the stock 75ohm resistor the 1-Chip has however you will have to drop contrast levels on whatever display you are using to compensate. For people like me who sell modwork that is not acceptable, the exact resistance is needed since most clients won't want to adjust their display/devices between game consoles. A higher resistance is needed (hence the 120 ohm resistor) to bring it down to the levels of most other game consoles.

[Voultar]

Sure, the 1CHIP outputs a higher than standard level RGB (0.85Vpp instead of 0.7Vpp IIRC), but it doesn't in the end make THAT much of a difference after dialing in the appropriate settings.

The difference is quite noticeable. Terminating the RGB outputs with 75ohm resistors is a little too peaked for my taste (and most of the customers I've performed the mod for).

I'd much rather terminate with 92ohm resistors or something similar than adjusting my display exclusively for my 1CHIP SNES. But again, that's somewhat of a personal preference.

Luma is also a little too hot with the standard termination, IMO. From my experience, the picture quality loses contrast and gets a little too crispy unless you bring it down in hardware.

[Skips]

I edited my above post with some changes that made it a bit better.

[RGB32E]

And yes RGB32 you can get away with the stock 75ohm resistor the 1-Chip has however you will have to drop contrast levels on whatever display you are using to compensate. For people like me who sell modwork that is not acceptable, the exact resistance is needed since most clients won't want to adjust their display/devices between game consoles. A higher resistance is needed (hence the 120 ohm resistor) to bring it down to the levels of most other game consoles.

Considering that you're not lowering the RGB levels before the encoder would lead to having the component chroma signals being too high of level as well. But for whatever reason you didn't compensate for that as well? Have you actually confirmed that a 120 ohm resistor is actually appropriate other than eyeballing it, Mr. Modwork?

The difference is quite noticeable. Terminating the RGB outputs with 75ohm resistors is a little too peaked for my taste (and most of the customers I've performed the mod for).

I'd much rather terminate with 92ohm resistors or something similar than adjusting my display exclusively for my 1CHIP SNES. But again, that's somewhat of a personal preference.

Luma is also a little too hot with the standard termination, IMO. From my experience, the picture quality loses contrast and gets a little too crispy unless you bring it down in hardware.

Just didn't find it to be a big deal with the XRGB-mini. I've discussed in the past of using a higher than 75 ohm resistor with the SNES mini, just not 100 ohms or greater. This being an example of 100 ohm resistors being used with a mini (EDIT) RGB mod:



ZSNES capture for comparison:

[Pasky]

And yes RGB32 you can get away with the stock 75ohm resistor the 1-Chip has however you will have to drop contrast levels on whatever display you are using to compensate. For people like me who sell modwork that is not acceptable, the exact resistance is needed since most clients won't want to adjust their display/devices between game consoles. A higher resistance is needed (hence the 120 ohm resistor) to bring it down to the levels of most other game consoles.

Considering that you're not lowering the RGB levels before the encoder would lead to having the component chroma signals being too high of level as well. But for whatever reason you didn't compensate for that as well? Have you actually confirmed that a 120 ohm resistor is actually appropriate other than eyeballing it, Mr. Modwork?

Isn't that all you did was eyeball it? Screenshots don't show much because amarec messes with the color conversion in YUV2 as does the PEXHDcap.

The proper way to do this is find out the RGB and Luma Vp-p levels coming off of the NESRGB encoder and then match those levels. Tim's board uses an OPamp which mixes the colors with those intended vp-p's in mind. Altering the input levels to something that doesn't match the NESRGB's will give unintended colors.

[Skips]

And yes RGB32 you can get away with the stock 75ohm resistor the 1-Chip has however you will have to drop contrast levels on whatever display you are using to compensate. For people like me who sell modwork that is not acceptable, the exact resistance is needed since most clients won't want to adjust their display/devices between game consoles. A higher resistance is needed (hence the 120 ohm resistor) to bring it down to the levels of most other game consoles.

Considering that you're not lowering the RGB levels before the encoder would lead to having the component chroma signals being too high of level as well. But for whatever reason you didn't compensate for that as well? Have you actually confirmed that a 120 ohm resistor is actually appropriate other than eyeballing it, Mr. Modwork?

The difference is quite noticeable. Terminating the RGB outputs with 75ohm resistors is a little too peaked for my taste (and most of the customers I've performed the mod for).

I'd much rather terminate with 92ohm resistors or something similar than adjusting my display exclusively for my 1CHIP SNES. But again, that's somewhat of a personal preference.

Luma is also a little too hot with the standard termination, IMO. From my experience, the picture quality loses contrast and gets a little too crispy unless you bring it down in hardware.

Just didn't find it to be a big deal with the XRGB-mini. I've discussed in the past of using a higher than 75 ohm resistor with the SNES mini, just not 100 ohms or greater. This being an example of 100 ohm resistors being used with a mini s-video mod:

No RGB32 I didnt fucking test it, I am just talking out my ass. Of course I tested it, I spent a good chunk of last night playing with this which you should have been able to tell just from the fact that I took a picture of it and mentioned chroma was still too high resulting in a revised schematic. And yes I am aware the RGB levels were too high. Instead of using three resistors on the RGB lines to compensate I decided to use two on the Pb and Pr lines to bring the chroma back down to acceptable, it saves a bit on parts when you end up having a lot of these mods to do for people. I also test stuff on an actual monitor, not a video processor since that is what most my clients actually tend to use. The XRGB is nice but it does tend to give slightly different results than an actual CRT monitor. When I get an XRGB in June I plan to keep my 14L5 just for the sake of testing modwork. Also what Pasky said about the capture software is also true.

**Edit** On a side note I discovered the PEXHDCAP will do component video via the DVI port. Was not aware of that.

[Skips]

Just some pictures of the SNES in PEXHDCAP. I matched the PEXHDCAP settings up to the CRT the best I could. No video processor, system connected straight to the PEXHDCAP. This is using the revised circuit I posted a few posts back.

[RGB32E]

No RGB32 I didnt fucking test it, I am just talking out my ass. Of course I tested it, I spent a good chunk of last night playing with this which you should have been able to tell just from the fact that I took a picture of it and mentioned chroma was still too high resulting in a revised schematic.

Whoa there buddy! No need to be that defensive, just sharing a few points and observations from my experience with this mod and the SNES mini RGB mod.

And yes I am aware the RGB levels were too high. Instead of using three resistors on the RGB lines to compensate I decided to use two on the Pb and Pr lines to bring the chroma back down to acceptable, it saves a bit on parts when you end up having a lot of these mods to do for people.

If you have a ton of orders for this mod, and aren't interested in trying to get the right levels fed to the RGB encoder to fix the original issue, fine, so be it!

I also test stuff on an actual monitor, not a video processor since that is what most my clients actually tend to use. The XRGB is nice but it does tend to give slightly different results than an actual CRT monitor. When I get an XRGB in June I plan to keep my 14L5 just for the sake of testing modwork. Also what Pasky said about the capture software is also true.

So, you're stating that all CRTs are the same, and that XRGB units are a completely different beast? I still have my PVM-2030 use it occasionally. However, I've used a number of PVMs over the years an they aren't all calibrated to give the same picture. Further, every different console and revision is going to give different output levels. Not to say that the SNES mini doesn't stray towards a corner case, but ultimately different consoles will always need to be tweaked in terms of brightness and contrast, and there's no perfect way to get things absolutely consistent across all consoles.

There was the one user who built a bunch of custom switch boxes with pots for RGB, so I guess there's that route!

Isn't that all you did was eyeball it? Screenshots don't show much because amarec messes with the color conversion in YUV2 as does the PEXHDcap.

That's part of it, but in my experience, the PEXHDCAP captures very similar luminance levels between RGB and component when I was testing the N-C, so the capture cards are valid for the sake of comparison.

http://shmups.system11.org/viewtopic.ph ... 8#p1065958

The proper way to do this is find out the RGB and Luma Vp-p levels coming off of the NESRGB encoder and then match those levels. Tim's board uses an OPamp which mixes the colors with those intended vp-p's in mind. Altering the input levels to something that doesn't match the NESRGB's will give unintended colors.

Yes, in case you didn't get the point before, that compensating for the higher levels after the conversion are less than ideal, especially since the fix is similar to the extra modification of the N-C Skips is proposing.

I also added a comparison to ZSNES in my previous post from someone's 100ohm solution for the SNES mini levels and it illustrates that 100 ohm resistors are way too high (if zsnes can be considered a reference ). I believe had calculated out that if you put 15 ohm series resistors before the RGB encoder that the output would be correct.

There are SMD 0.1uF caps between the SNES mini 1CHIP IC and the RGB encoder IC, so it really wouldn't be much more effort at all to correct the levels instead of messing with the NESRGB-COMPONENT board in the attempt to compensate after the fact.

[Skips]

First of all if you don't want people to hop on your shit try to word your responses so you don't come off like a passive aggressive prick (the unneeded comment about FBX earlier being a good example of what not to do). You make me look like a saint sometimes and I am a complete and utter asshole.

Furthermore this board does not use an RGB encoder, it uses an OPamp as pasky said. So dropping the RGB levels or the Pb Pr levels are both acceptable methods. If it were an actual encoder you would want to drop the RGB levels, its not so either method will work just fine. I went with the method that requires less components is all. If you want to do a write up/diagram for people that drops the RGB levels then do so. It wouldn't hurt, its just an alternate method of doing it.

[RGB32E]

First of all if you don't want people to hop on your shit try to word your responses so you don't come off like a passive aggressive prick (the unneeded comment about FBX earlier being a good example of what not to do). You make me look like a saint sometimes and I am a complete and utter asshole.

Haha! Who's to say FBX wasn't being a prick to begin with, and why the pissing match if you weren't being defensive? Make you look like a saint?! If you want to hop on posts, then feel free to turn this forum into the likes of NG and Shoryuken!

Furthermore this board does not use an RGB encoder, it uses an OPamp as pasky said. So dropping the RGB levels or the Pb Pr levels are both acceptable methods. If it were an actual encoder you would want to drop the RGB levels, its not so either method will work just fine. I went with the method that requires less components is all. If you want to do a write up/diagram for people that drops the RGB levels then do so. It wouldn't hurt, its just an alternate method of doing it.

Considering that the N-C uses RGB output from the RGB encoder, it in fact does use the RGB encoder. Hence the suggestion to correct the original issue. Adding a point about the N-C using an op-amp doesn't prove or dis-prove that.

[Skips]

First of all if you don't want people to hop on your shit try to word your responses so you don't come off like a passive aggressive prick (the unneeded comment about FBX earlier being a good example of what not to do). You make me look like a saint sometimes and I am a complete and utter asshole.

Haha! Who's to say FBX wasn't being a prick to begin with, and why the pissing match if you weren't being defensive? Make you look like a saint?! If you want to hop on posts, then feel free to turn this forum into the likes of NG and Shoryuken!

Furthermore this board does not use an RGB encoder, it uses an OPamp as pasky said. So dropping the RGB levels or the Pb Pr levels are both acceptable methods. If it were an actual encoder you would want to drop the RGB levels, its not so either method will work just fine. I went with the method that requires less components is all. If you want to do a write up/diagram for people that drops the RGB levels then do so. It wouldn't hurt, its just an alternate method of doing it.

Considering that the N-C uses RGB output from the RGB encoder, it in fact does use the RGB encoder. Hence the suggestion to correct the original issue. Adding a point about the N-C using an op-amp doesn't prove or dis-prove that.

My point was that dragging your past issues with him into this thread was a dick move and had nothing to do with what was being posted here (which is why I used it as an example). It was 100% unnecessary as he had not even posted in this thread before you made that post, hence my original point. Funny that you mention turning the place into Neo-geo.com because that is just the type of crap that goes on there (and one of the many reasons I picked up and moved shop elsewhere). The pissing match comes from the fact that when I post outside of retromods.net I am doing it to provide useful information I find to others who may not be registered there. I really don't give a shit about the unneeded stabs and sarcastic remarks you like to lace your posts with (which is exactly what goes on at one of the sites you just bashed). Anyway I am done on this subject, I am not going to drag this out into a pissing match.

And yes the board using an OPamp does matter. When using a video encoder to convert RGB to other video signals you want to feed it the correct level. With the OPamp either way is acceptable and is up to the users preference of method. How you worded it made the method I provided sound incorrect when indeed it is not. I am merely correcting your assertion that dropping RGB levels is the correct way of doing it when either way would be correct. In this case it is a matter of preference. Like I said, if you want to go with dropping the RGB levels write it up and post it instead of arguing about it. My guess is it would be somewhere between 25-45 ohms on each RGB out line on the SNES encoder before going into the component PCB. I don't have anymore that small to test it with unfortunately since I used them all in another project this morning.

[RGB32E]

Well, shoot, now you're just beating a dead horse about an off-hand comment. That doesn't justify your behavior.

Considering that the RGB encoder is used to generate the luma signal and the N-C is used for the chroma signals, it isn't a mutually exclusive idea to get the correct signal levels. Stating that the chroma signals are ok to compensate for doesn't fix that the luma signal isn't correct either. Chill out - this is still a work in progress!

[Voultar]

Isn't that all you did was eyeball it? Screenshots don't show much because amarec messes with the color conversion in YUV2 as does the PEXHDcap.

This is absolutely correct. A video processor such as the XRGB and/or a capturing device don't provide a pure native representation of the output.

The proper way to do this is find out the RGB and Luma Vp-p levels coming off of the NESRGB encoder and then match those levels. Tim's board uses an OPamp which mixes the colors with those intended vp-p's in mind. Altering the input levels to something that doesn't match the NESRGB's will give unintended colors.

Correct again.

[Josh128]

My point was that dragging your past issues with him into this thread was a dick move and had nothing to do with what was being posted here (which is why I used it as an example). It was 100% unnecessary as he had not even posted in this thread before you made that post, hence my original point. Funny that you mention turning the place into Neo-geo.com because that is just the type of crap that goes on there (and one of the many reasons I picked up and moved shop elsewhere). The pissing match comes from the fact that when I post outside of retromods.net I am doing it to provide useful information I find to others who may not be registered there. I really don't give a shit about the unneeded stabs and sarcastic remarks you like to lace your posts with (which is exactly what goes on at one of the sites you just bashed). Anyway I am done on this subject, I am not going to drag this out into a pissing match..

Skips, first, thanks for your contributions and sharing your experiences here-- it is appreciated. For a big, white, immortal gorilla, I find your forum etiquette to be impeccable.

I also find the personal jabs and shit stirring annoying and altogether unnecessary. Ive taken my share of them and I also know Ive pissed a few people off here-- Im no saint either, but I always make it a point to try to not make things personal or get drawn into these pissing matches. Im just here to share my experiences and learn from others. Lifes too short and Im getting too old.

[Skips]

My point was that dragging your past issues with him into this thread was a dick move and had nothing to do with what was being posted here (which is why I used it as an example). It was 100% unnecessary as he had not even posted in this thread before you made that post, hence my original point. Funny that you mention turning the place into Neo-geo.com because that is just the type of crap that goes on there (and one of the many reasons I picked up and moved shop elsewhere). The pissing match comes from the fact that when I post outside of retromods.net I am doing it to provide useful information I find to others who may not be registered there. I really don't give a shit about the unneeded stabs and sarcastic remarks you like to lace your posts with (which is exactly what goes on at one of the sites you just bashed). Anyway I am done on this subject, I am not going to drag this out into a pissing match..

Skips, first, thanks for your contributions and sharing your experiences here-- it is appreciated. For a big, white, immortal gorilla, I find your forum etiquette to be impeccable.

I also find the personal jabs and shit stirring annoying and altogether unnecessary. Ive taken my share of them and I also know Ive pissed a few people off here-- Im no saint either, but I always make it a point to try to not make things personal or get drawn into these pissing matches. Im just here to share my experiences and learn from others. Lifes too short and Im getting too old.

LOL the name actually isnt from the regular show, it was a nickname some gave me as a child. I stuttered a bit (and still do if I get angry enough). Was something that took a bit to correct and grow out of.

[Josh128]

LOL the name actually isnt from the regular show, it was a nickname some gave me as a child. I stuttered a bit (and still do if I get angry enough). Was something that took a bit to correct and grow out of.

Thats cool. I used to work with a guy who had a similiar issue growing up, but like you he overcame it/corrected it as an adult and only very rarely had the minor slip. Dude was extremely bright, valedictorian of his high school class and a hell of a baseball player.

I saw your avatar on another forum and recognized it right away, my kids and I actually watch the Regular show, I think its one of the funniest modern cartoons out there.

[viletim]

The NESRGB-Component board is parasite on the video encoder chip. It was designed specially for the output pins of a BH7236AF, but there's no reason why it can't be attached to any encoder that provides RGB and Y driver outputs simultaneously (such as CXA1645 and friends). The video encoder in the older Super Nintendo consoles doesn't have RGB driver outputs, so that one's out. The later models have appear to have a better encoder.

The component board has a built in 75 ohm resistor and coupling capacitor. If 100 ohms series resistance is best used for RGB (As an example - I don't have this hardware) on the output of the PR and PB you would add 27 ohm resistors (27 + 75 is close enough to 100). Ideally, you'd want to measure it with an oscilloscope to be sure. It can be misleading to use the display device as a measurement tool.

[Skips]

The NESRGB-Component board is parasite on the video encoder chip. It was designed specially for the output pins of a BH7236AF, but there's no reason why it can't be attached to any encoder that provides RGB and Y driver outputs simultaneously (such as CXA1645 and friends). The video encoder in the older Super Nintendo consoles doesn't have RGB driver outputs, so that one's out. The later models have appear to have a better encoder.

The component board has a built in 75 ohm resistor and coupling capacitor. If 100 ohms series resistance is best used for RGB (As an example - I don't have this hardware) on the output of the PR and PB you would add 27 ohm resistors (27 + 75 is close enough to 100). Ideally, you'd want to measure it with an oscilloscope to be sure. It can be misleading to use the display device as a measurement tool.

With the older models cant you amp the rgb lines with a ths7314 then connect that to the component board? One ofthe 1-Chips I have is one I got cheap from a client who broke the rgb out legs off the encoder. I used a Ths7314 to restore RGB on it and connected the component board to the THS7314 before the resistors. I got the same results with it that I got with using the 1Chips's encoder on my other 1-Chip. In theory shouldnt the THS7314 work for these older models as well?

[Pasky]

The NESRGB-Component board is parasite on the video encoder chip. It was designed specially for the output pins of a BH7236AF, but there's no reason why it can't be attached to any encoder that provides RGB and Y driver outputs simultaneously (such as CXA1645 and friends). The video encoder in the older Super Nintendo consoles doesn't have RGB driver outputs, so that one's out. The later models have appear to have a better encoder.

The component board has a built in 75 ohm resistor and coupling capacitor. If 100 ohms series resistance is best used for RGB (As an example - I don't have this hardware) on the output of the PR and PB you would add 27 ohm resistors (27 + 75 is close enough to 100). Ideally, you'd want to measure it with an oscilloscope to be sure. It can be misleading to use the display device as a measurement tool.

With the older models cant you amp the rgb lines with a ths7314 then connect that to the component board? One ofthe 1-Chips I have is one I got cheap from a client who broke the rgb out legs off the encoder. I used a Ths7314 to restore RGB on it and connected the component board to the THS7314 before the resistors. I got the same results with it that I got with using the 1Chips's encoder on my other 1-Chip. In theory shouldnt the THS7314 work for these older models as well?

Yes, the THS7314 can drive 75Ω loads. You'll need to feed it 0.7Vp-p signals because the gain is 6dB (doubles the voltage peak to peak) so that the output is 1.4Vp-p and terminated to .7Vp-p with a 75Ω load. You'll want to use pots to adjust the voltage levels on the inputs because any higher and the colors will get crushed, assuming they don't go above as I'm unsure how high they are at the multi av out. You'll need an oscilloscope to get it accurate, but I'm sure you could eyeball it with pot adjustments on the fly. IIRC the old snes's amplified the RGB signals with transistors and the encoder only transcoded it to S-video/Composite. At least the BA6592F variants did if I'm remembering correctly.

[Skips]

The NESRGB-Component board is parasite on the video encoder chip. It was designed specially for the output pins of a BH7236AF, but there's no reason why it can't be attached to any encoder that provides RGB and Y driver outputs simultaneously (such as CXA1645 and friends). The video encoder in the older Super Nintendo consoles doesn't have RGB driver outputs, so that one's out. The later models have appear to have a better encoder.

The component board has a built in 75 ohm resistor and coupling capacitor. If 100 ohms series resistance is best used for RGB (As an example - I don't have this hardware) on the output of the PR and PB you would add 27 ohm resistors (27 + 75 is close enough to 100). Ideally, you'd want to measure it with an oscilloscope to be sure. It can be misleading to use the display device as a measurement tool.

With the older models cant you amp the rgb lines with a ths7314 then connect that to the component board? One ofthe 1-Chips I have is one I got cheap from a client who broke the rgb out legs off the encoder. I used a Ths7314 to restore RGB on it and connected the component board to the THS7314 before the resistors. I got the same results with it that I got with using the 1Chips's encoder on my other 1-Chip. In theory shouldnt the THS7314 work for these older models as well?

Yes, the THS7314 can drive 75Ω loads. You'll need to feed it 0.7Vp-p signals because the gain is 6dB (doubles the voltage peak to peak) so that the output is 1.4Vp-p and terminated to .7Vp-p with a 75Ω load. You'll want to use pots to adjust the voltage levels on the inputs because any higher and the colors will get crushed, assuming they don't go above as I'm unsure how high they are at the multi av out. You'll need an oscilloscope to get it accurate, but I'm sure you could eyeball it with pot adjustments on the fly. IIRC the old snes's amplified the RGB signals with transistors and the encoder only transcoded it to S-video/Composite. At least the BA6592F variants did if I'm remembering correctly.

Correct, the older models amped with transistors and converted RGB to s-video and composite with the encoder. l know the GPM-02 encoder did not do RGB out and did Pb and Pr instead. To my knowlege Tim's board has better quality than those old encoders do but I have not tried amping Pb and Pr on them with a THS7314 yet. I know colors were faded using the transistor method of amping Pb and Pr with those old encoders, I will have to try amping those with a THS7314 later as well to see how it compares to the kit.

[CkRtech]

Correct, the older models amped with transistors and converted RGB to s-video and composite with the encoder. l know the GPM-02 encoder did not do RGB out and did Pb and Pr instead. To my knowlege Tim's board has better quality than those old encoders do but I have not tried amping Pb and Pr on them with a THS7314 yet. I know colors were faded using the transistor method of amping Pb and Pr with those old encoders, I will have to try amping those with a THS7314 later as well to see how it compares to the kit.

Those built-in Pb Pr signals from the native SNES encoder are rather odd. It seems like a lot of people have messed with transistors on them and done a lot of "guessing" to try to find the right signal strengths. I've done two SNES "component mods" over the last few years just for kicks. The first needed the transistors and never looked right (can't remember the model). The second, more recent one is sitting next to me right now - It is a 1991 GPM-01 with S-ENC and is running Pr Pb straight out to the panel mount RCA jacks. Supposedly, this one didn't need any extra components, but the colors are still incorrect.

Obviously, weak component color signals paired with a strong Y signal are going to produce weird results due to how colors are derived.

I figured I would just take Tim's RGB Component board, throw it into the system, and use its Pr Pb output instead of trying to make the S-ENC circuit happy. I can't recall the circuits suggested across the net circa 2011, but I assume using a THS7314 is basically going to yield the same result just with fewer components unless the community has a better, proper amp circuit design by now.

[Skips]

Correct, the older models amped with transistors and converted RGB to s-video and composite with the encoder. l know the GPM-02 encoder did not do RGB out and did Pb and Pr instead. To my knowlege Tim's board has better quality than those old encoders do but I have not tried amping Pb and Pr on them with a THS7314 yet. I know colors were faded using the transistor method of amping Pb and Pr with those old encoders, I will have to try amping those with a THS7314 later as well to see how it compares to the kit.

Those built-in Pb Pr signals from the native SNES encoder are rather odd. It seems like a lot of people have messed with transistors on them and done a lot of "guessing" to try to find the right signal strengths. I've done two SNES "component mods" over the last few years just for kicks. The first needed the transistors and never looked right (can't remember the model). The second, more recent one is sitting next to me right now - It is a 1991 GPM-01 with S-ENC and is running Pr Pb straight out to the panel mount RCA jacks. Supposedly, this one didn't need any extra components, but the colors are still incorrect.

Obviously, weak component color signals paired with a strong Y signal are going to produce weird results due to how colors are derived.

I figured I would just take Tim's RGB Component board, throw it into the system, and use its Pr Pb output instead of trying to make the S-ENC circuit happy. I can't recall the circuits suggested across the net circa 2011, but I assume using a THS7314 is basically going to yield the same result just with fewer components unless the community has a better, proper amp circuit design by now.

Im skeptical it will yeild different results as well but I just did not want to throw the idea out until I had the chance of trying it.

[Josh128]

The video encoder in the older Super Nintendo consoles doesn't have RGB driver outputs, so that one's out. The later models have appear to have a better encoder.

The component board has a built in 75 ohm resistor and coupling capacitor. If 100 ohms series resistance is best used for RGB (As an example - I don't have this hardware) on the output of the PR and PB you would add 27 ohm resistors (27 + 75 is close enough to 100). Ideally, you'd want to measure it with an oscilloscope to be sure. It can be misleading to use the display device as a measurement tool.

Excuse my ignorance, but I was assuming I would just be able to use the RGB and Y from the multi-out to feed the component board. So this would not be acceptable, the signal levels would be too hot on the RGB pins?

Not specifically for Tim to answer, but if the older consoles encoder doesnt output RGB, where does the RGB and S present on the multi out originate from?

[Pasky]

The video encoder in the older Super Nintendo consoles doesn't have RGB driver outputs, so that one's out. The later models have appear to have a better encoder.

The component board has a built in 75 ohm resistor and coupling capacitor. If 100 ohms series resistance is best used for RGB (As an example - I don't have this hardware) on the output of the PR and PB you would add 27 ohm resistors (27 + 75 is close enough to 100). Ideally, you'd want to measure it with an oscilloscope to be sure. It can be misleading to use the display device as a measurement tool.

Excuse my ignorance, but I was assuming I would just be able to use the RGB and Y from the multi-out to feed the component board. So this would not be acceptable, the signal levels would be too hot on the RGB pins?

Not specifically for Tim to answer, but if the older consoles encoder doesnt output RGB, where does the RGB and S present on the multi out originate from?

The PPU itself, splits off to the encoder and gets amplified via transistors before hitting the multi out.

[Skips]

The video encoder in the older Super Nintendo consoles doesn't have RGB driver outputs, so that one's out. The later models have appear to have a better encoder.

The component board has a built in 75 ohm resistor and coupling capacitor. If 100 ohms series resistance is best used for RGB (As an example - I don't have this hardware) on the output of the PR and PB you would add 27 ohm resistors (27 + 75 is close enough to 100). Ideally, you'd want to measure it with an oscilloscope to be sure. It can be misleading to use the display device as a measurement tool.

Excuse my ignorance, but I was assuming I would just be able to use the RGB and Y from the multi-out to feed the component board. So this would not be acceptable, the signal levels would be too hot on the RGB pins?

Not specifically for Tim to answer, but if the older consoles encoder doesnt output RGB, where does the RGB and S present on the multi out originate from?

You don't want to pull RGB from the multiout (Luma is fine though) because the system has resistors inline that will drop the RGB levels below what the component board is expecting. When used with an NESRGB is it pulling RGB from the encoder before the 75ohm resistors so placing it after them would make it look wrong.

[RGB32E]

I figured I would just take Tim's RGB Component board, throw it into the system, and use its Pr Pb output instead of trying to make the S-ENC circuit happy. I can't recall the circuits suggested across the net circa 2011, but I assume using a THS7314 is basically going to yield the same result just with fewer components unless the community has a better, proper amp circuit design by now.

Given that the SNES systems without the S-RGB/S-RGB A encoders (BA6595F/BA6596) buffer RGB from the PPU to multi-out something like this:



It would make sense to bypass the circuit and use a THS7314 or equivalent to buffer the RGB signals for a 75 ohm load for the N-C. I performed this type of bypass on a 1CHIP and the results were fine aside from the 1CHIP outputting >0.7 Vpp and looking slightly too sharp/peaked compared to the 1CHIP-01 SNES.

[cfx]

.

[BONKERS]

I'm thinking of doing this on my 93' GPM-02 with the S-ENC (6592).

Tim mentioned

The video encoder in the older Super Nintendo consoles doesn't have RGB driver outputs, so that one's out.

Does this prevent this from working? I know some people have pulled component straight from the encoder but quality isn't so great.

Would bypassing the S-ENC and using RetroRGB's THS7314 make a difference or? (Possibly provide even better quality?) http://retrorgb.com/1chiprgb.html

[RGB0b]

Would bypassing the S-ENC and using RetroRGB's THS7314 make a difference or? (Possibly provide even better quality?)

That guide won't work on older SNES systems, like your GPM-02.