OSSC (DIY video digitizer & scandoubler)

[citrus3000psi]

Of course, my perfect world there would be digital audio input AND output on the OSSC.

This should be possible using something like a AK4117 ic. Hmmm. Interesting idea.

[FBX]

Followup on my hair-pulling struggle trying to find optimal H. samplerate on 480p for XBOX:

It turns out the default value of 858 for 720x480p really was correct for XBOX after all. Strange as it may seem, the software itself is stretching the graphics of games, causing internal linear interpolation of some of the graphics. In Capcom vs SNK EO for example, the original 640 horizontal graphics are extended to 654, and pre-drawn graphics are getting partially interpolated. This shows up really nice and ugly on the S-N-K groove meters that have grid sections. They were pre-drawn at 640 res, and then the 656 stretch makes them look like shit.

I wish there was some way to turn this off and 'force' XBOX to output 1:1 pixel ratios, but at least now I know I'm no going crazy with why my OSSC techniques weren't working on the XBOX 480p mode.

[strygo]

I thought I would share a couple of quick pictures of a doujindance RGB/240p modded 3DO connected through the OSSC. The results are stellar. I had commented elsewhere that the mod suffers from the RGB alignment issue that previous mods have had. FWIW, while it was noticeable on my CRTs and PVM (and easily fixable on the displays), it wasn't noticeable through the OSSC. I'm not sure if the image is isn't distorted or whether the pillars mask it, but in any case I'm happy with it.

https://imgur.com/a/qGSzn

Pictures were taken with an iPhone.

[BuckoA51]

doujindance have a bit of a rep for shoddy work don't they? Or is that someone else I'm thinking of?

[strygo]

He did, although he has taken feedback from the community and has more recently improved his work.

In the case of the 3DO, he developed a board that sits on top of the primary chip. You can see a photo of it here:



I plan to open my unit up when I pick up a MNEMO board. I'll take some shots of the mod work.

[Kez]

doujindance have a bit of a rep for shoddy work don't they? Or is that someone else I'm thinking of?

They do but in the past few months they were called out by the community. There is a lot of evidence to suggest they took it all on board and are now producing significantly higher quality work.

[citrus3000psi]

He did, although he has taken feedback from the community and has more recently improved his work.

In the case of the 3DO, he developed a board that sits on top of the primary chip. You can see a photo of it here:



I plan to open my unit up when I pick up a MNEMO board. I'll take some shots of the mod work.

I wonder if he has fixed his horizontal alignment issues and is still using Luma.

***edit*** just read up top that it still suffers from alignment issues.

I have a board that builds csync H/V and fixes the alignment issues, just waiting pcbs to come in to test.
Also has some other basic power filtering on board to help with noise etc.




I also have a QSB for MNEMO's usb board found here: https://oshpark.com/shared_projects/5u23CCSW

[BuckoA51]

Wow awesome, I was going to look at a similar thing in the new year, you're the man Citrus, 2018 will be a year of RGB modded 3DOs

[Galdelico]

BTW, playing around more with the 256x240 optimal timing line4x modes, I discovered the 4:3 setting is perfectly aligned. So it's only the 8:7 mode that doesn't line up with the active game graphics when H. active is set to 256. One side is pinched while the other has overscan (and shifting the backporch just switches the side pinched). I've not checked this on the other linexx modes, just 4x at this point.

Since I got my new RGB cables from RGC.uk, I took the chance and tried that out myself... Yep, as far as I can tell there's no mode that pinches both sides in 256p optim. 8:7. They seem to behave kinda differently, though, and it doesn't look like they even pinch correctly one side (one has overscan, the other seems a bit cut off to me).

Really hope this bug will get fixed with the next firmware update, as it also compromises Line2x 320p optim. (at 4:3), which I especially like since its subtle softness works perfectly with scanlines.

Oh, and huge props to Retrogamingcables. They've definitely stepped up their game, since the last time I ordered a cable from them. The C-Sync RGB cable I bought for my Japanese Mega Drive 2 got rid of the jailbars I was seeing almost completely (seriously, 99,9% gone).

[Naiera]

I bought one of those really expensive Neo Geo AES RGB cables from Retro Gaming UK. I still get the shadows I talked about several pages back

[Thomago]

Regarding RGC:
I recently got a Packapunch SNES C-Sync and a Packapunch PS2 C-Sync cable. In comparison to my previous cables (a Nintendo DOL-013 for the SNES and a Wolfsoft "modded" Sync-on-Luma cable for the PS2) the difference is minmal, but it's there (minimally cleaner picture on the SNES and minimally less noise on the PS2).
To be honest, I hoped for more (no more audio buzz on the SNES and no more "brightness bleeding" into the overscan area on the PS2), but well... it won't get better than that I guess.

[Galdelico]

Regarding RGC:
I recently got a Packapunch SNES C-Sync and a Packapunch PS2 C-Sync cable. In comparison to my previous cables (a Nintendo DOL-013 for the SNES and a Wolfsoft "modded" Sync-on-Luma cable for the PS2) the difference is minmal, but it's there (minimally cleaner picture on the SNES and minimally less noise on the PS2).
To be honest, I hoped for more (no more audio buzz on the SNES and no more "brightness bleeding" into the overscan area on the PS2), but well... it won't get better than that I guess.

I intentionally wanted to give their regular C-Sync cables a shot first, because - judging by the comments I've read, and also by the amount of new info on the website - I was under the impression they're the ones that took the most advantage, in terms of buid quality (new resistors, better caps... You know, after the amp-armageddon), compared to the previous line.
And - at least with my Japanese MD2 - I can report huge improvements. Jailbars got wiped off almost entirely: literally, if I glue my eye to the screen, on certain colours, I still notice a very faint 'shadow' (which I believe it can only get fixed through a proper mod), but the columns effect is gone and, from a normal viewing distance, the image is now crystal clean, both in 320 and 256 optim. modes. I'm actually surprised and, needless to say, very happy with the results.

[fafangus]

I bought one of those really expensive Neo Geo AES RGB cables from Retro Gaming UK. I still get the shadows I talked about several pages back

You have to find a xneo1

[Naiera]

Even when the exact same Neo worked perfectly with the XRGB-mini? (aside from the losing sync stuff that made me lose interest)

[fafangus]

What's your aes révision ?

[Naiera]

Same as it was when I had an XRGB-mini. It's one of the "good RGB" ones.

[paulb_nl]

With the OSSC you will see every detail of your source signal so either the shadows are in the output of your Neo Geo or the OSSC is causing the shadows but then you would see them with every source.

[FBX]

Same as it was when I had an XRGB-mini. It's one of the "good RGB" ones.

I take it you went by the site that had serial number ranges? It's actually bad info. Open the cart slot and look to the extreme right edge of the board inside. You should see the revision on the right edge. Use a flashlight if it's too dark to see it.

Edit: Also on this:

It took a LONG time for this setting to go from 1170 to 384

It sounds like you forgot to set the output mode for whatever lineX you were using to "Optimal Timing 320x240". It doesn't make you have to lower all the way down from 1170.

[Naiera]

I take it you went by the site that had serial number ranges? It's actually bad info. Open the cart slot and look to the extreme right edge of the board inside. You should see the revision on the right edge. Use a flashlight if it's too dark to see it.

I've gone by two different sets of instructions and found that it had "good RGB". And, again, it looked great when I had the XRGB-mini.

I figured out the OSSC advanced settings thing later. It didn't remove the shadows.

[Naiera]

With the OSSC you will see every detail of your source signal so either the shadows are in the output of your Neo Geo or the OSSC is causing the shadows but then you would see them with every source.

I sincerely doubt that you are right. I know what good "retro console RGB" looks like, and it was great with the XRGB-mini from the same AES console.

There are two white lines of text in the opening sequence for King of Fighters 2001 where it's particularly obvious. If it's still the cable then I wasted a lot of money. If it's the unit I need to get a new one.

[FBX]

I take it you went by the site that had serial number ranges? It's actually bad info. Open the cart slot and look to the extreme right edge of the board inside. You should see the revision on the right edge. Use a flashlight if it's too dark to see it.

I've gone by two different sets of instructions and found that it had "good RGB". And, again, it looked great when I had the XRGB-mini.

Have you tried video LPF set to 9Mhz? As others have stated, the OSSC reveals far more than the Framemeister does on every little nuance of the image signal.

Also did you try what I suggested by looking at the inside right edge of the board from the cartridge bay? As I mentioned sites that go by serial numbers are only guessing on what's inside AES consoles. There's quite a bit of board-swapping that went on with that console over the years.

At any rate, it very likely is that your console needs servicing (like a cap replacement job).

[Naiera]

As others have stated, the OSSC reveals far more than the Framemeister does on every little nuance of the image signal.

This is not the case here. Seriously. Fot the Flame Meister to have obscured this issue, the picture would have to be a complete blur.

[strygo]

What revision is it? What serial number is on the bottom?

As FBX mentioned earlier, the resources that list the best models aren't actually accurate. Notably, the 3-6 revision is actually one of the best, because the 'mod' needed is a simple trace cut.

[ApolloBoy]

Anyone use an Audio Authority 1362 with their OSSC? Just bought one on eBay so I can finally use composite and S-vid consoles on my new setup and I was curious as to how it performs.

[Nrg]

Anyone use an Audio Authority 1362 with their OSSC? Just bought one on eBay so I can finally use composite and S-vid consoles on my new setup and I was curious as to how it performs.

Looking at the manual I'm *guessing* that it'll do framerate conversion, meaning it'll introduce frame drops and/or frame repeating, causing stuttering/jerkiness of the signal. If i'm right about that, i wouldn't use that device myself..

[BuckoA51]

Anyone use an Audio Authority 1362 with their OSSC? Just bought one on eBay so I can finally use composite and S-vid consoles on my new setup and I was curious as to how it performs.

if it's composite video/S-Video to VGA it has to be an upscaler, so not what you want with OSSC.

[Omnigamer]

Is there any major difference in sync or output processing in the most recent hardware revisions versus older OSSC units? I've gotten several reports from folks using the newest model with audio embedding that it has some odd behavior specifically with SNES. The reports have it breaking sync on minor resolution swaps and extended (30+ second) resync times for some displays and capture devices. I cannot reproduce any of this activity with the same games and scenes on my OSSC, which is one of the original run units with DVI output. This only applied to SNES signals; other consoles did not exhibit these symptoms and worked fine. For reference, the affected units are running 0.78a, while my unit is 0.78.

This is fairly important to narrow down for us, as we're considering replacing our older units with the audio embedding ones to simplify AV flow. If the newer units have different compatibility issues, we'll need to figure out a backup plan.

[marqs]

Is there any major difference in sync or output processing in the most recent hardware revisions versus older OSSC units? I've gotten several reports from folks using the newest model with audio embedding that it has some odd behavior specifically with SNES. The reports have it breaking sync on minor resolution swaps and extended (30+ second) resync times for some displays and capture devices. I cannot reproduce any of this activity with the same games and scenes on my OSSC, which is one of the original run units with DVI output. This only applied to SNES signals; other consoles did not exhibit these symptoms and worked fine. For reference, the affected units are running 0.78a, while my unit is 0.78.

This is fairly important to narrow down for us, as we're considering replacing our older units with the audio embedding ones to simplify AV flow. If the newer units have different compatibility issues, we'll need to figure out a backup plan.

There should be no difference in functionality if TX_MODE is set to DVI on newer boards. Anyway, pretty much all NES & SNES 240p related audio/video stability issues come down to their jittery sync as described here.

[H6rdc0re]

Is there any major difference in sync or output processing in the most recent hardware revisions versus older OSSC units? I've gotten several reports from folks using the newest model with audio embedding that it has some odd behavior specifically with SNES. The reports have it breaking sync on minor resolution swaps and extended (30+ second) resync times for some displays and capture devices. I cannot reproduce any of this activity with the same games and scenes on my OSSC, which is one of the original run units with DVI output. This only applied to SNES signals; other consoles did not exhibit these symptoms and worked fine. For reference, the affected units are running 0.78a, while my unit is 0.78.

This is fairly important to narrow down for us, as we're considering replacing our older units with the audio embedding ones to simplify AV flow. If the newer units have different compatibility issues, we'll need to figure out a backup plan.

There should be no difference in functionality if TX_MODE is set to DVI on newer boards. Anyway, pretty much all NES & SNES 240p related audio/video stability issues come down to their jittery sync as described here.

Any chance these problems can be fixed through software? Or do I have to wait for someone like Voultar to make a RGB board?

[Galdelico]

Just out of curiosity, Marqs. Any idea about when you'll release the firmware update featuring fully adjustable scanlines? I'm spending quite a long time to tune up my CRT profiles for 480i sources, then I read the Wiki and realized it could all be pretty much useless, should the new firmware land relatively soon.

Thanks in advance! ^_-