MV-1B Graphic glitches *FIXED*

barakka

n00b
Joined
Mar 17, 2005
Posts
32
I have a board that shows these glitches:


Neo1.jpgNeo2.jpg


Looks like some sprites, but not all, have lines in them and some are out of place.
I've already tried swapping the sub board, so I can rule out a dirty/defective slot.
I've also tried the Diag Bios, but it doesn't find anything wrong with the video RAM.
Is this a problem of the NEO-GRC2?
 
Last edited:

barakka

n00b
Joined
Mar 17, 2005
Posts
32
Found the problem! It was the LO-ROM!

At first I thought it was the NEO-GRC2, so I took a scrap MV-1B (corroded tracks/watchdog) and swapped the graphic chip.
That didn't seem to work as glitches were as before, moreover some tiles had the wrong color and I didn't remember noticing it earlier.

So, after that, there wasn't much left to try. I didn't have another NEO-GRC2 so I tried to swap the LO-ROM with the one from the scrap PCB.
It was noticeably better! There were no more sprites out of place or with horizontal lines, however the graphic was pixelated.
It looked exactly like the effect described in the wiki.neogeodev.org when A0 is stuck, I was using AOFII as a test cart and the picture was almost the same as this:

C_A0.png


But peppered with some tiles with the wrong colour...

I spent a lot of time trying to find the problem with the A0 address line, until I gave up and decided to try another LO-ROM, this time borrowed from a 100% working board.

That solved the pixelation problem! Image was close to perfect, but there was still the problem with some tiles with the wrong colour.

So I tried to test the VRAM memory with the Diag bios, and I had spurious errors: Sometime the test would complete and find nothing wrong, sometimes it would find an error.

Memory addresses would change, but It was always the same 3 errors:

ACTUAL: 4000 EXPECTED:0000
ACTUAL: BFFF EXPECTED:FFFF
ACTUAL: EAAA EXPECTED:AAAA

Again, spent a lot of time tracking the elusive bit error (I though it was RAM6D6 or pin 149), replaced the VRAM ICs but nothing worked.
So I thought that maybe (probably) the problem was the NEO-GRC2, and decided to swap it again with the original IC... :blow_top:

And it worked! A lot of work/time wasted but the board is fixed! :glee:

BTW:
GadgetUK if you are reading this, thanks for your Youtube repair videos!
Thanks to you, I learned how to remove SMD ICs without destroying the PCB in the process.
 
Top