What's new

Tsunoni

Student
Joined
Oct 10, 2016
Messages
64
Reaction score
22
Location
New Jersey
The S-ROM is fine its the C-ROM that is messed up. I wish I had photos right now but I tried fixing it and I dont think there will be anything else I can do. I would like to send it to someone to fix it f possible.
It looks like all the sprites are moved half way off the screen except for moving ones which are in their correct positions. when the game starts the NEO GEO logo is ontop of itselfand half way up the screen. the boards look clean including the battery area.
I replaced the bios and its exactly the same.
 
the diagnosis bios makes troubleshooting so much easier. And the companion z80 test cart can really help too.

These sorts of glitches are probably going to be either a bad address line, or a bad chip multiplexing the address lines.

Are you comfortable using a logic probe? That can really help too.
 
the diagnosis bios makes troubleshooting so much easier. And the companion z80 test cart can really help too.

These sorts of glitches are probably going to be either a bad address line, or a bad chip multiplexing the address lines.

Are you comfortable using a logic probe? That can really help too.
I will burn that diagnosis bios chip today and try it out and bring some pictures back too.
I have basically 0 experience on using a logic probe. I just have a cheap multimeter that I use for continuity checks which may be able to perform some of the tasks
 
I will burn that diagnosis bios chip today and try it out and bring some pictures back too.
I have basically 0 experience on using a logic probe. I just have a cheap multimeter that I use for continuity checks which may be able to perform some of the tasks
Logic probes are pretty simple, all they do is measure a "high" or "low" voltage by clipping onto the circuits ground and then probing various pins. The complicated part is figuring out what the state should be on which pin on which component. So for example, the data pin on a chip should in most cases pulse high when active. Indicating that it is sending binary data as a series of high and low voltage states.
 
I am uploading a video of the spin masters game which is what was loaded on the Multi
IMG_20221126_152722.jpg
MVS
IMG_20221126_152734.jpg
IMG_20221126_152752.jpg
IMG_20221126_154118.jpg
IMG_20221126_155414.jpg
 
Address lines between LSPC and Fast RAM damaged / shorted / pulled up / down ?
Fast RAM dead ? Fairly common.
Since diagbios can detect it pretty sure it's a lower (and only, hopefuly) board problem.
Be sure to socket that RAM, you could swap their positions to see if that has a effect on address or data diag test.
GL !
 
OK, that's a really useful error.

The VRAM2K is sometimes also called FAST VRAM, or HIGH (speed) VRAM - but for today lets call it VRAM 2K

This site will show you where abouts on the MV6 board those 2K VRAM chips are located (in the top right)
https://github.com/jwestfall69/neogeo-diag-bios/blob/master/docs/ram_locations/mv6.md
So if you want to pull the 2K VRAM chips and replace them with known good ones you can do that, but in my experience it has never once helped. For me not has always been either bad addressing or rotten traces in/out.

But swapping the chips easy to do (they are DIP, not surface mount), and good practise, so maybe give that a go.

After that doesn't work we could try some other stuff

The NeoGeo Dev wiki shows the graphics pipeline here
https://wiki.neogeodev.org/index.php?title=Graphics_pipeline
Confusingly the 2K VRAM is called HIGH (Speed) VRAM in this diagram, just one of those things we are expected to know.

Anyhoo, the point is that the 2K VRAM talks only to a chip called LSPC which does all the line sprite control

There are two versions of the LSPC chip, but the 6-slot boards used one called LSPC-A0. You can check the pinout for LSPC-AO here
https://wiki.neogeodev.org/index.php?title=LSPC-A0

The pins that we are most interested in on the LSPC are the ones associated with the Fast VRAM address bus (see! Now the 2K VRAM has another name now, how annoying is that?!) and the Fast VRAM data Bus
Address bus = FVA0 to FVA10
Data bus = FVD0 to FVD15
We can look those up on the pinout and see the
the FVA lines are pins 139 to 149 on the LSPC
the FVD lines are pins 60 to 75 on the LSPC

Then we can look up the pinout for the 2k VRAM and see where those address and data lines go

Of course to look up the pinout for the 2K VRAM we first need to know what chip it is, You could visually inspect it, or look at this list here
https://wiki.neogeodev.org/index.php?title=RAM_chip#MV-6
And see that FAST VIDEO = CXK5814P-35L
So you could look up the pinout here
https://html.alldatasheet.com/html-pdf/47128/SONY/CXK5814P-35L/86/1/CXK5814P-35L.html
And see that
there are 11 address lines - these match FVA0 to FVA10 and are common across both VRAM 2K chips
there are 8 data lines - the first 8 match FVD0-7 on one of the 2k VRAM chips, and the other 8 match FVD8-15 on the other 2K VRAM chip

At that point you can beep out all of the lines on the LSPC and make sure they are going to the right place with continuity on the 2KVRAM chips
And without a logic probe that's all you can do

If you do get a logic probe you can check that the signals are being sent correctly
Address bus = FVA0 to FVA10 = should all be PULSE all the time
Data bus = FVD0 to FVD15 should all be PULSE all the time
Chip enable should pulse on boot then sit LOW
Write enable should pulse on boot then sit LOW
Output enable should pulse on boot then sit LOW

At that point we will know if the problem is indeed With the 2k VRAM per the diagnostics bios. or if this is just a symptom and we need to look further up the graphics chain for the root cause
 
I bought some good sockets and chips so I plan on using them but ill look at the board before I put them in.
 
I finally tried the new ram chips and I think at least one worked because the 6 Slot MVS turned on and showed more graphics but half of stuff was missing so I tried the other chips but now it just wont go past the cross hatch press b to reset screen
 
I tried some more of the VRAM chips that I bought and figered that some were just bad and thats why it would eieth go back to the cross hatch screen, or a white screen. I got two working VRAM chips in the new sockets and the whole system is up and running like normal again!
now I just need to get a magnet for the monitor to fix the purity.
 
Outstanding. Flawless victory! Really pleased to hear that this magnificent beauty is up and running again. I have pretty much every single MVS ever made, but not the elusive MV6 - am most jealous :)
 
Back
Top