Neo Diagnostics ROMs

TravistyOJ

Another Striker
Joined
Oct 15, 2005
Posts
326
What does it mean if all tests pass, and I still can't get past the crosshatch. I've checked multiple carts and multiple slots and I don't see any broken traces.
 

Xian Xi

JammaNationX,
15 Year Member
Joined
Dec 1, 2005
Posts
27,748
If possible, please make the NDR work with larger eproms.

Would like to stack the image and put it on a larger rom and toss that on the neo.

Last time I tried: No go. Black screen.

You just need to double, triple, quadruple the image if you want to use a larger eprom. You can do that in DOS or romwaks.
 

Xian Xi

JammaNationX,
15 Year Member
Joined
Dec 1, 2005
Posts
27,748
I have an FZ that says Lower BRAM Unwritable. The HC32 is fine.
 

smkdan

Galford's Armourer
Joined
Dec 30, 2009
Posts
452
What does it mean if all tests pass, and I still can't get past the crosshatch. I've checked multiple carts and multiple slots and I don't see any broken traces.

The current tests don't account for bad P ROM access which causes the crosshatch problem. It would need a new version with updated test cart. If the tests all pass, it atleast tells you that the board is probably fine once you figure out what's causing the crosshatch.

If it happens with every slot, it could be the 68k address / data (see A1~A23 and D0~D15 pins on a pinout) or a control signal from one of the custom chips. On your MV4, they should all lead to some 74244/74245 chips and the pinout posted in your thread should help figure out the top board connections too.

@Xian: How is SRAMWEL from NEO-C1? Pins 97 + 98 should go to one of those HC32s. The 1F schematic PDF (1st page in that dotted box) shows the circuit the 1FZ may be using as well. The PDF shows how it uses "SRAMWEN" and "SRAMCS" to get the actual BRAM write enables so those connections between the HC32s have to be good too.
 

Xian Xi

JammaNationX,
15 Year Member
Joined
Dec 1, 2005
Posts
27,748
Got another FZ, non responsive SM1. All connections are good and swapped the SM1 with same results. If I press start after the results I get crunchy sound and it says YM2610 IRQ error.
 

GadgetUK

Ace Ghost Pilot
Joined
Sep 27, 2013
Posts
1,323
Just wanted to say thanks for creating the diag ROMS - awesome piece of work and they helped me fix my MV-1FZ. I will send a donation when I get next paid. Very grateful, you are a smart helpful guy =)
 

smkdan

Galford's Armourer
Joined
Dec 30, 2009
Posts
452
Should've gotten back sooner but I've been a bit busy lately.

Just wanted to say thanks for creating the diag ROMS - awesome piece of work and they helped me fix my MV-1FZ. I will send a donation when I get next paid. Very grateful, you are a smart helpful guy =)

Cool, glad it helped. There's plenty left to do to make it better but that'll come whenever time permits.

Got another FZ, non responsive SM1. All connections are good and swapped the SM1 with same results. If I press start after the results I get crunchy sound and it says YM2610 IRQ error.

That's weird. The bad response from the SM1 and screwy sound makes me think it could be a D0 issue or more likely just a YM issue. Bad YM can make the Z80 act strangely and it can get in the way of the SM1 code working properly. Does it get anywhere with a regular game or even with glitched sound?

Also, I had a chance to try the MV-1 recently. I can't get it to reset, show a black screen etc. with or without the sound test. I use ST 120ns chips all around. Are you using any 150ns chips on these older boards? The first gen board layouts look like they'd be fussier with chip speeds and 150ns would end up being borderline. Can we rule that out? I've reset the board dozens of times and I can't get it to mess up on my end. I also have access to a NEO-AES system now that I'll get socketed soon so that should cover 1st gen testing pretty well.
 

smkdan

Galford's Armourer
Joined
Dec 30, 2009
Posts
452
I've added a lot more documentation on the home page for the new error codes. I've been meaning to do this a lot earlier but never got around to it. Most of the specific errors have been explained with relevant chips and traces to test and investigate.

This sort of info can also used as a reference on why a board isn't working after you've tested the obvious address/data connections to RAMs etc.. A lot of the 'glue' logic that supplements all of that and how it actually affects a system doesn't seem to be well documented. I'll try to keep it updated as more info gets put on the dev wiki or when I have time to probe my own boards.
 
Last edited:

Xian Xi

JammaNationX,
15 Year Member
Joined
Dec 1, 2005
Posts
27,748
I've added a lot more documentation on the home page for the new error codes. I've been meaning to do this a lot earlier but never got around to it. Most of the specific errors have been explained with relevant chips and traces to test and investigate.

This sort of info can also used as a reference on why a board isn't working after you've tested the obvious address/data connections to RAMs etc.. A lot of the 'glue' logic that supplements all of that and how it actually affects a system doesn't seem to be well documented. I'll try to keep it updated as more info gets put on the dev wiki or when I have time to probe my own boards.

That looks really helpful dude. Thanks.
 

Xian Xi

JammaNationX,
15 Year Member
Joined
Dec 1, 2005
Posts
27,748
What causes the Diag bios to say it fails the CRC check when running a test?
 

smkdan

Galford's Armourer
Joined
Dec 30, 2009
Posts
452
What causes the Diag bios to say it fails the CRC check when running a test?

A bad EPROM burn is the simplest cause. If it verifies OK or it's the only system it happens on, then it's possibly a lower address line to the BIOS but it's not as likely. It'll probably be A12 or A11 if any for the 68k side. If it's an upper address line (A13~A15), it should catch that as a separate error.
 

Xian Xi

JammaNationX,
15 Year Member
Joined
Dec 1, 2005
Posts
27,748
A bad EPROM burn is the simplest cause. If it verifies OK or it's the only system it happens on, then it's possibly a lower address line to the BIOS but it's not as likely. It'll probably be A12 or A11 if any for the 68k side. If it's an upper address line (A13~A15), it should catch that as a separate error.

Address lines were fine. When I ran a BRAM test I got some weird error indicating a bunch of bits were jacked up, mostly upper. Replaced the upper BRAM then it changed to AA9A, replaced the lower BRAM and now it's AA55 for the read.
 

Xian Xi

JammaNationX,
15 Year Member
Joined
Dec 1, 2005
Posts
27,748
Hey Dan, is it possible to add some Z80 type loop tests for RAM, SM1, YM2610 and Z80? Is there any that can be made/added?
 

smkdan

Galford's Armourer
Joined
Dec 30, 2009
Posts
452
Hey Dan, is it possible to add some Z80 type loop tests for RAM, SM1, YM2610 and Z80? Is there any that can be made/added?

A future version could just add a few of those in the sub menu but I'd need to update the SM1 for the first time. Only the SP1 ROM has changed up until now. Have you had to replace a YM2610 that had audio/muting issues when running for a while after power on?
 

Xian Xi

JammaNationX,
15 Year Member
Joined
Dec 1, 2005
Posts
27,748
A future version could just add a few of those in the sub menu but I'd need to update the SM1 for the first time. Only the SP1 ROM has changed up until now. Have you had to replace a YM2610 that had audio/muting issues when running for a while after power on?

Yes, it was a thermal issue with the 2610.
 

Footix

n00b
Joined
Apr 13, 2014
Posts
19
Hello,

any help on this one? :)

Bought this AES version 3-6 out of ebay with the following details: gives a green screen when started without cartridge.

I hoped for an easy fix... it appears to be a nightmare...

symptoms:
1) does not play any cartridge
2) gives the solid green screen

here is what I did so far
1) changed the bios for smkdan's diagnostic bios. => I get a palette bank 0 error (5555), written 5555, read 5556 at 400000
2) verified all traces from the 2 palette rams, all ok
3) changed the two 245 driving the rams, no joy
4) changed the two rams, no joy
5) changed the neo-g0, no joy
6) changed the neo-c1, no joy

I hoped a change of g0 or C1 would fix the non booting issue, but it remains so.

I also tried an unibios and I always get a solid white screen (no cart inserted).

I traced back the ROMOE ROMOEL and ROMOEU signals and they are stuck at logical high (+5V), leading to the disabling of the cartridge P1 rom.

To be honest, I am kind of lost now and I am starting to be desperate.

Any ideas?

Thanks!

Arnaud


[EDIT] I just fixed the non booting issue -> nasty gauged trace between the 68k and the Prog board. (D0). I am now getting a 5555 written, 5554 read...
 

Xian Xi

JammaNationX,
15 Year Member
Joined
Dec 1, 2005
Posts
27,748
Why would you change the G0 and the C1? Since only the first bit is off, check the Neo-B1.
 

Footix

n00b
Joined
Apr 13, 2014
Posts
19
Why would you change the G0 and the C1? Since only the first bit is off, check the Neo-B1.

Hello.

Thanks for your help.

I proceeded like on the different repair log. I am definately not a hardware guru and i am self learning.

Can you please explain me why you say it is only the first bit that is off? And why it relates to the b1? The connections to the b1 have been tested good.

I would like to understand what is happening down there...

Thx!
 

smkdan

Galford's Armourer
Joined
Dec 30, 2009
Posts
452
AES 3-6 doesn't have any 74245. The closest thing is the 2x 74273 which are also connected to palette RAM but are used for RGB output instead. Did you check NEO-G0 pinout to 68k/from palette RAMs? On AES 3-6, 68k data for palette and memcard goes through NEO-G0.

https://wiki.neogeodev.org/index.php?title=NEO-G0#AES_pinout=

Read 5556 means 68k D0 and D1 are bad. Read 5554 byte means D1 is OK, but D0 is still bad (should be high, not low). You can enter those hex values into MS calculator and convert to binary mode to see how each bit differs. Each bit maps to one pin on 68k D0~D15 starting with the rightmost bit in calc for D0 and the leftmost bit for D15.

The set of traces from 68k D0 -> NEO-G0 D0 -> NEO-G0 PALD0 -> Palette RAM D0 should be tested. You said you tested traces from the palette RAM but what about 68k <-> G0?
 
Last edited:

Footix

n00b
Joined
Apr 13, 2014
Posts
19
Sharp eye!

it is indeed a typo from my side. These are 273 and not 245. They have been replaced by known good ones.

Thanks for explanation! It is much clearer now!

Here are the connections I have for the tracing of the D0 line.

68k (5) -> NEO-G0 (2) for the CPU line

NEO-G0 (62) -> RAM at location G1 (11) ok...

when I probed the CPU line, it appeared that the connection was intermittent.

Thank you sir :)

you are a star!

ps: like I said in the email I sent you, please leave me your paypal and I will do small donation. thanks!
 

DEL 707

Cheng's Errand Boy
Joined
Jan 9, 2014
Posts
120
Hate to post in 2 different places, but I'm hoping someone can help me pin point the fault on my MV1FZS with a Z80 error.

I ran the diagnostic rom (great btw), passed all tests, then I threw in a cart and did the Z80 tests, this is what I got back.

agj3RHq.jpg


pt3XrCu.jpg
 

TravistyOJ

Another Striker
Joined
Oct 15, 2005
Posts
326
Hey guys, I get this consistently on an MV-1:

WRAM DATA (0000)

ADDRESS: 10100C
ACTUAL: 0080
EXPECTED: 0000

No variance if I cycle power. Does this mean I need to replace a RAM, and if so, which one? Thanks!
 

DEL 707

Cheng's Errand Boy
Joined
Jan 9, 2014
Posts
120
New board, new problem.

I have no sound, so threw in the diagnostic ROM and I get this error report.

Z80 Reported Error Code: 14
M1 Bank Error (16K)

:scratch:
 

smkdan

Galford's Armourer
Joined
Dec 30, 2009
Posts
452
Hey guys, I get this consistently on an MV-1:

WRAM DATA (0000)

ADDRESS: 10100C
ACTUAL: 0080
EXPECTED: 0000

No variance if I cycle power. Does this mean I need to replace a RAM, and if so, which one? Thanks!

If the error details are the same every time, then it's probably a bad chip. The chip is marked "42356" on the board and should the one immediately to the left of the battery. It's the one with pin 11 connected to pin 5 of the 68k.

New board, new problem.

I have no sound, so threw in the diagnostic ROM and I get this error report.

Z80 Reported Error Code: 14
M1 Bank Error (16K)

:scratch:

I posted about this in another thread where another poster had the same issue. The page I linked to there has a few pins to test. It's not the worst error you could get since there's a very specific cause for it and it won't take too long to check:

http://www.neo-geo.com/forums/showthread.php?252389-MVS-Z80-error&p=3744533&viewfull=1#post3744533
 

DEL 707

Cheng's Errand Boy
Joined
Jan 9, 2014
Posts
120
Thanks for the reply.

I did the continuity check and it all seemed good
NEO-D0 SDRD0 (45) → CHA SDRD0 (MVS: B47)
Z80 A15 (5) → CHA SDA15 (MVS: A58)

Bit of a loss at what to do, read the other thread but I'm a bit lost.

Can you give any suggestions for my other board? The (SM1 Looks unresponsive) error.
I've got 2 boards here with sound errors, if I can Frankenstein something out of the 2 of them...
 
Top