Is video on the Neo Geo supposed to be this awful?

Well it didn't fix my spare first-PCB AES' horrific dot crawl. Tomorrow I'll trace it out. If I find it tunes the oscillator, you're a very bad man.

 

Sorry, man.

 

Well I traced the circuit and it is there just to get the oscillator oscillating, it does absolutely nothing for picture clarity. It must have had a placebo effect :lol: You shouldn't believe anything you're told on N-G.com

 

It's not a placebo effect. Turn it it a little slows down the scrolling bars. Turn it a tiny bit more, they slow down even more. If you find the sweet spot, the rolling bars will become stationary.


I'm not doubting that your technical knowledge is better than mine, but I have no reason to lie about this. It stops the flickering/scrolling lines/bars as easily seen in health bars in fighting games.

 

Dot crawl sucks indeed. I suffer of it on my supergun (CXA1645 based) and it varies with the resolution of the source. A PSX based arcade board works with it flawlessly (No dot crawl) *if* the game runs using the PSX normal interlaced video mode. If it switches to 640x480 (turn on video interlace) the screen becomes dot crawl nightmare.
That is extremly annoying with the MVS and mildly annoying with the CPS2...

Tinkering with the chroma subcarrier adjust (yes I put a variable capacitor on the circuit) does change the behavior of the "dot crawl cloud" just as DiamondDave mentioned. I even got into the extreme of hooking a frequency meter on it to see what was going on.
Sadly nothing I did got rid of the dot crawl.
Using Super Video reduced it fairly but it's stil noticeable. At leas the image gains a lot of sharpness while S-Video is used.

I'm now convinced the only real cure to dot crawl (the only thing that annoy me atm with my own supergun design) is switching to RGB/Component.

DiamondDave:
While I do not disagree from your view, I should add that any kind of changes/adjusts made to the mentioned circuits will only affect the composite/ S-video outputs on the neo geo.

RGB output will not be affected by anything you change in the metioned asjusts.
You could even operate the AES Neo in RGB without the 3.579545 crystal as long the 24mhz one is working properly.

Edit: When I hook my supergun hardware to a BT878 based TV capture card on the PC, the dot crawl vanishes.

 

I still disagree, touching the varicap will cause the oscillator to turn on/off with switch bounce. The new phase can get the alignment more desirable (faster less noticeable rainbowing and "hanging dots"/horizontal crawl) but as soon as you power cycle or wait a while for drift it's back again. Since on the Neo colorburst isn't in sync the master clock, adjusting the phase is also temporary.

..

Drr I remembered that's what the later PCB revisions try to do, get a period of the master clock aligned the subcarrier, in order for the phase to stick. So, OK OK, on later revisions the varicap will adjust the crawl. (Even so, adjusting does not LOSE the artifacts as they're the result of the Neo's timing, they just won't resonate with the pixels making all color edges shimmer.)

I don't see how a proper S-video circuit under any circumstances will still have the hanging dots... I tried shunting the subcarrier to get a B&W image and there weren't any.

 

So, apparently the issue was settled by using the following combinations, of performing the Neo Geo AES RGB NEO-AES3-6 Vertical Lines fix:
http://www.mmmonkey.co.uk/snk-neo-geo-aes-rgb-neo-aes3-6-vertical-lines/
Adjuring the little screw on the upper left corner next to the oscillator and AV-Out fixed the dot-crawl of the RCA Composite output and a high quality RGB Scart Stereo cable displays beautifully.

Interesting.

 

i tried mmmonkey guide on my AES 3-6 using a rgb cable and its a lot better, just open the console, flip it upside down and cut the trace, but be careful as they are tiny, and in the picture they look big