What's new
Marqs solution to get the F3 working on the OSSC is this as far a sync goes, but I guess you guys already know this?!

2018-01-05_10-26-25.jpg


But a circuit that fixes the sync and horizontal shift/position issues on the F3 would be ideal.
 
  • Like
Reactions: RGB
He is feeding Csync on pin#13 (H-Sync) of the VGA head and leaving pin#14 (V-Sync) unconnected.
I'm interested in using a Sync processor/regen to feed a separated H-Sync and V-Sync to both pins13 and 14 (aka the correct VGA wiring only the wrong 15khz not 31khz rate).
 
@ShootTheCore, can you post a pic of the issue you are experiencing? Like what @jugu says, I have issues with alignment. Wonder if once you guys come up with this solution, it fixes the alignment/geometry stuff. The CPS3 does this thing with alignment too. Wondering if what is fabbed here fixes that as well.
Yeah, I'll dig into it again this weekend and take pictures of everything. Sadly, my current scope is only a single-channel, so I can't probe the color channels simultaneously with the sync pulses. A quality scope is next on the Purchase List - just need to research whether to go with a Rigol / Tektronix digital scope, a USB Picoscope, a SainSmart Mini, or an old-school analog HP scope off Ebay. So many options...

I'm also taken aback at how this thread suddenly exploded - it was just crickets chirping when I first posted a few months ago. Thank you for all the interest and help everyone!
 
I'm also taken aback at how this thread suddenly exploded - it was just crickets chirping when I first posted a few months ago. Thank you for all the interest and help everyone!
For those of us that plug into a CGA monitor, the issue manifests itself as a severe alignment issue. The thought of risking life and limb to adjust a K7000's width coil scares me. Something similar happens, but not as severe, with the CPS3 as well. Need a solution.

Plus the level of genius discussion going on around this forum has increased exponentially as of late. Don't want to fall behind. Need to keep up the façade.
 
@ShootTheCore, can you post a pic of the issue you are experiencing? Like what @jugu says, I have issues with alignment. Wonder if once you guys come up with this solution, it fixes the alignment/geometry stuff. The CPS3 does this thing with alignment too. Wondering if what is fabbed here fixes that as well.
Yeah, I'll dig into it again this weekend and take pictures of everything. Sadly, my current scope is only a single-channel, so I can't probe the color channels simultaneously with the sync pulses. A quality scope is next on the Purchase List - just need to research whether to go with a Rigol / Tektronix digital scope, a USB Picoscope, a SainSmart Mini, or an old-school analog HP scope off Ebay. So many options...
A guy from work h4x0red a keygen a while back, pretty pretty pretty cool, to unlock features on Rigol scopes. But there's an easier way to unlock a Rigol nowadays https://hackaday.com/2015/10/22/upgrading-rigols-more-expensive-oscilloscopes/ dunno if you know this already :D don't own a scope (yet) either by the way.
 
I don't have my F3 in front of me but looking at some PCB pics it seems that the Sync pin on the JAMMA edge goes to a ceramic cap. Would be cool to have a board where you could simply remove that cap and solder some wires to the sync board for a totally seamless solution (and one that is easily reversed if you wanted to put the PCB back to factory)
 
@K405 I didn’t know that the lower-end Rigol scopes were firmware-upgradable to the higher models. Thank you for the info on that-I just placed an order for the Rigol DS-1054Z. It’ll be perfect for my projects after a 100mhz firmware upgrade. :thumbsup:
 
I tested the OSSC with the HAS V3.1, CSync Level 75 Ohm i.e. JP2 closed.

First I tested Thunderforce AC on the FM and the OSSC (AV1). No issues.

Then I tested the F3 with Pop'n Pop on the FM, image was unstable until I flipped the regeneration switch on the HAS V3.1.

Connected the OSSC (AV1) with the regeneration switch ON gave: NO SYNC in the display.

Connected the OSSC (AV1) with the regeneration switch OFF got the display to flash with:

261p, 15.43 KHz, 59.14 Hz

The sync LED stayed RED and there was no signal out.

I tried a few combinations powering F3 before OSSC but to no avail.
 
Connected the OSSC (AV1) with the regeneration switch OFF got the display to flash with:

261p, 15.43 KHz, 59.14 Hz

The sync LED stayed RED and there was no signal out.

I tried a few combinations powering F3 before OSSC but to no avail.
I get the exact same thing, but I did get it to work. I started with setting H-PLL pre-coast and post-coast to 4. After that, you can cycle through output modes. For me, every second one works. Tested it out yesterday, just before going to bed, so I don't remember exactly which one worked, but go through line doubling, 3x, 4x, 5x, and some of them should work. The diode would turn green before the image appeared.

This was on a 1.6 OSSC with what I'm assuming was the latest firmware. No sync cleanup, RGB from JAMMA via a Vogatek MKIV supergun, to OSSC's SCART. Worked the same both on my TV and PC screen. For a while I thought I wasn't going to get it to work, but it works fine. Tried Elevator Action Returns, Kaiser Knuckle and Rayforce.

The only problem is slight image warping in the upper part of the screen. The picture, as well as the image border, are shifted a bit. I can post a picture later.
 
Has anyone tried just running the F3 through a simple scart/rgb to component video transcoder before the ossc? I know it's probably not ideal but I'm genuinely curious there. I own some transcoders but no ossc to test... yet :)
 
I have one (SCART to Component transcoder) but no I've never tried it with the OSSC's component input.
I would like to say I could test this for you, I'm beyond busy atm however so it won't be any time soon.
 
I did some more testing after coming home, and I wish I had a definitive, full-proof, step by step method that makes the OSSC + F3 combo work every time, but I don't. There's some randomness to the process. I can produce positive results, but I can't guarantee it will be on the first try.

Here's how I go about it:

I power on the F3, and after that I turn on the OSSC. The best working picture is always on the x3 setting. Everything else is chopped at the top, bottom, or both. Set it to x3, and if you're not getting a picture, turn off the OSSC and turn it on again. For me, it always starts with the test pattern, so I switch to 1 and boom, the picture should be there. I did a video, but it was dark, my phone wouldn't focus and it's generally garbage. At around 1:45 you can see a clear picture, with the warping at the top of the screen, which I get a close-up on.

I've noticed that neither my TV nor my PC screen are picky about the signals, so as long as the OSSC can accept it, both of my screens will display it. If you can't get a picture on your TV doing it the way I did, maybe try a PC screen. Again, it does work, it's just process of repeating the aforementioned steps.

Also, my TV, despite being somewhat modern [modern enough to support 3D, at least], does accept RGB via SCART and it has no problems with the F3 signal. It's just not as good at scaling the picture as OSSC.

I also have a cheapo Chinese RGB to HDMI scaler. I wonder how it would deal with the F3 signal? I might try it later. The image quality leaves a lot to be desired when compared to OSSC, but you get what you pay for.
 
Could someone with a dual-trace oscilloscope could take a look at the sync pulses on one channel and the video output on another (R, G or B, it doesn't matter which)?
This was the suggestion from Chris over at martin-jones.com (<- Interesting read on arcade Sync issues).
Here are the sync pulses and video output sampled from the Red channel on my F3 board. One is from the gradient monitor test screen (the X sweep), and the other is while the game is running.

IMG_4590.jpg
 

Attachments

  • IMG_4568.jpg
    IMG_4568.jpg
    332.6 KB · Views: 176
Back
Top