Software Support > GroovyMAME
Artifacting on Nanao MS8 29FSG w/ CRT Emu Driver (SOLVED!)
BumbleChump:
All good to know. My monitor chassis isn't grounded to anything for sure. And dang, walking!? Glad to hear you're better now.
I've done a few things now. I connected the VGA from the computer directly to the monitor by twisting the syncs together like you mentioned. No JPAC. Here's some comparisons between raw VGA and JPAC. The first image is raw VGA, the second is JPAC.
Blegh, Bison's gums are a mess. Based on this, it looks like the computer is outputting those white ringing spots even without the JPAC. It's especially noticeable around the edges of M.Bison's cape. The JPAC is just amplifying the ringing. So I made a low pass filter at 220 microfarads and 75 ohms and I get this:
The image just turns blue? You can faintly see the icons on the computer desktop. I made sure there wasn't any wire issues with the red or green. As I removed each capacitor, it restores that color. If I remove all 3 caps, the image looks normal. I also tried higher and lower capacitances. The higher capacitance didn't change anything. But if I went lower, the image would become slightly less blue. I went down to 0.1 microfarads and it looked like someone smeared the screen like a hot knife on buttered toast. I also tried flipping the capacitors +/-, but that didn't change anything.
I connected the input red, green, and blue to the left of those resistors, and the output RGB to the right of the resistors. I cut a very heavily shielded VGA cable in half, and spliced this low pass filter into it. All the capacitors were grounded to the main ground. I tried video ground too, but nothing changed. (ignore that one flipped capacitor in the photo)
I'm thinking I need to "clean" the signal from the computer with some external device, I'm just not sure what yet. And I'm sure some might think that the image looks fine, I'm just being picky. And you're right. But the Neo Geo outputs a fantastic signal with no ringing so I want to get that from the computer as well!
Thanks again for the help so far, will continue to figure this out.
Zebidee:
So yeah, you still get an image without the JPAC but just fainter/weaker signal. Your JPAC pics look sharper too, but that may just be because your camera will auto-adjust aperture/exposure settings to available light.
Breadboards can be finicky beasts for video signals. I suggest you simplify your breadboarding slightly by putting the negative (-) ends of the capacitors on the negative (-) rail because then they'll all be connected. Then you can remove those bridging wires and just run a single ground wire from the negative - rail. Ground to video ground rather than main ground (although I bet you'll find they are connected if you test with a multimeter). Then you can take the video output from the rail that runs between your resistor and capacitor + end.
The value of 75 ohm for the resistors works well for video signals, but the 220uF value for the capacitors is mostly educated guesswork. TBH it is all guesswork. So I decided to stop guessing and, rather than hurting my too much head with the math, I found an online low-pass (RC) filter calculator (http://sim.okawa-denshi.jp/en/CRlowkeisan.htm) and plugged some value in. According to that, if we are using 75 ohm resistors and want to filter out frequencies above 60hz (and not below) with this then 33uF capacitors would seem like the way to go. But I guess you've already tried that value.
Good idea to check your VGA cables for resistance end-to-end with your DMM. Occasionally I've found VGA cables that have significant resistance, either by design (eg sometimes happens with VGA cables scavenged from dead monitors) or by poor quality.
Also suggest that you try another computer or at least another video card.
Eventually I'll get around to doing that video amp circuit with the THS7314 chip again. I've still got a few stashed away. Now that I've just had to scrap my international travel plans for next week I might even have some time to do so!
BumbleChump:
Thanks for the suggestions. I did already try another computer, and GPU. Same thing.
I got the parts to put together a THS7314 RGB amp/low pass filter.
https://www.retrorgb.com/thsamps.html
Did exactly what it said. 75 ohms, 0.1 microfarad ceramic capacitor. I pulled 5 volts and ground from Jamma, and I got this:
It's way worse! I understand that it's gonna be bright, but I was hoping it would at least get rid of the ringing, but the ringing is even more visible now. I did read that someone had a similar issue as me, and got this multi-scart adapter which has a low pass filter in it:
https://lotharek.pl/productdetail.php?id=159
But it's pricey, especially for something that I'm unsure will fix this issue. Is there anything else I can do?
BumbleChump:
Made a discovery!
After reading the manual for the THS7314, I realized that the RGB website failed to mention that an additional 22-100 micro farad capacitor is to be used in parallel with the 0.1 microfarad capacitor.
So I added a 100 micro farad capacitor, as well as three 47k ohm resistors on the input side of the chip, and it completely removed the ringing! (Edit: Actually, I tried it without the additional 100 microfarad capacitor. It looks the same, so I guess it's actually not necessary. Using those three 47k ohm resistors at the input side is what fixed the ringing)
The issue now is that it's still too bright, and if I use higher resistance capacitors the image becomes blurry and colors start to bleed. (I tried three 100k ohm resistors at the input side) But at least now I know that a low pass filter is the solution to my problem. Now to figure out what filter to use, or how to pull down the signal because the THS7314 combined with the JPAC is amplifying the signal too much.
Zebidee:
That's good news!
Do you actually need to use the JPAC? You can bypass it with the video, but still use it for the controls if needed.
I once got around this issue completely once (when using PC->JPAC->monitor) by using resistors to ground on the RGB inputs (between the jpac and the monitor) to create a voltage divider (between that resistor and the 75ohm termination on the Sony PVM I was using), reducing the voltage level and therefore brightness. I can't remember what value I used, but probably 75R - 220R.
You could try doing similar, maybe with pots to get a value that suits (200R or 500R pots should be sufficient).
Navigation
[0] Message Index
[#] Next page
[*] Previous page
Go to full version