- Thread starter Evan
- Start date
chacal231077
Grand Master
Hi! 
I had the same thing 3 days ago with my KOF99.
many welds were broken.
I redid everything and zero problem
I had the same thing 3 days ago with my KOF99.
many welds were broken.
I redid everything and zero problem
Just so you know theres 2 different CMC variants and they aren't compatible with each other.
I've never swapped one before but I've been succesful swapping the roms instead, reflowing the pins didn't fix my issue.
https://wiki.neogeodev.org/index.php?title=NEO-CMC
I've never swapped one before but I've been succesful swapping the roms instead, reflowing the pins didn't fix my issue.
https://wiki.neogeodev.org/index.php?title=NEO-CMC
Hey guys. I too have a KOF 2003, purchased as "faulty" and was hoping I could request a nice close up photo of this section (see attached) of the CHA board with all the 0603 resistors. My board is in a serious mess, post previous surgery, by the looks, and I have a feeling a couple of these resistors have been accidentally switched. They're mostly 560 Ohm (marked 561) but there's a 160 Ohm (161) currently at R18 on my board that I feel may belong at R2... Point of reference, the images on MVS Scans aren't quite large enough to make out the values with certainty.
While we're on this grid of resistors, do any of the brainiacs on the forum know how this array works with regard to the Neo-CMC function?
A side question on the 64Mbit maskroms - is there a clever way to read/test these? I have GQ 4x4 (w/16bit adapter) and XGecu T56 programmers on hand.
Thank ya kindly for any help offered.
While we're on this grid of resistors, do any of the brainiacs on the forum know how this array works with regard to the Neo-CMC function?
A side question on the 64Mbit maskroms - is there a clever way to read/test these? I have GQ 4x4 (w/16bit adapter) and XGecu T56 programmers on hand.
Thank ya kindly for any help offered.
Attachments
cyberwillis
Beginner
Hi...
That resistor you see of 161OHMs is OK also all other 561OHMs if they still measure that value. If you can check any partial image on the screen your CMC50 is probably OK, and probably is just some lines between C ROM and CMC50 has a faulty connection. But in case everything is a mess on the screen then I would ask you to exchange the M1 Eprom because it is the center for "CMC50" decrypting all the Cx ROMs and S1.
*UPDATE* : In case you need a new M1, find one 271-m1c.bin with this hash (crc32: f5515629) don't try to put a PCB version on it because is different
That resistor you see of 161OHMs is OK also all other 561OHMs if they still measure that value. If you can check any partial image on the screen your CMC50 is probably OK, and probably is just some lines between C ROM and CMC50 has a faulty connection. But in case everything is a mess on the screen then I would ask you to exchange the M1 Eprom because it is the center for "CMC50" decrypting all the Cx ROMs and S1.
*UPDATE* : In case you need a new M1, find one 271-m1c.bin with this hash (crc32: f5515629) don't try to put a PCB version on it because is different
Attachments
Last edited:
cyberwillis
Beginner
Hi,
The CMC42 and CMC50 are compatible with each other (they have the same pinout). The problem itself is that the M1 from an encrypted game (CMC50 board) cannot be decrypted by a CMC42 and an unencrypted M1 (CMC42 board) cannot decrypt a Game that is not prepared for it. because every latest game (CMC50 board) was specifically prepared to pre-decrypt the game and the code to encrypt a generic game to use a different M1 is not out there and neither MAME source has it. But a person could with success transplant the ASIC from a board that has the same part number. Also, notice that there is a small detail, the latest CMC50 boards (2003.07.24) have some extra line(s), and the previous CMC50 board (1999.06.14) is not compatible with that. Metal Slug 5, SNK vs Capcom, Samurai Shodown 5, Kof 2003 and Samurai Shodown 5 Special boards can probably receive a game from another CMC50 board but the other way around will probably not work as expected.
Last edited:
Hey Cyberwillis
You Sir, are a true gentleman! Many thanks for your input, and apologies for the delay. I assumed I'd receive an email notification of replies, but didn't...Your close up pic of that SMD resistor section confirms what I thought. I did need to switch the 160 Ohm resistor back to position "R2", and went on to straighten and clean up much of the array, as they were in a real mess. All values are correct upon testing. Your insight on the M1 decryption responsibilities is super helpful. Mine is the EPROM version, seen in the first post. In fact, I wonder if it might actually be the very same copy of KOF 2k3... Anyway, it checked out fine after dumping and comparing it on ROMIDENT. In the end, I did need to replace the badly re-installed CMC50 it came with, with one I had spare from a KOF 2k. The game is working beautifully now! Really stoked to add this cart to the collection
Can I ask if you know how to read/dump/test the 64Mbit C roms on these later boards? I have no way to check them, which adds a painful slice of uncertainty when trouble shooting GFX issues. The same goes for 32/64Mbit V roms on PROG boards. I would love to be able to test these for proper function, too. The programmers I have on hand are GQ 4x4 and XGecu T56. Any clarity would be brilliant!
Thanks again
T
cyberwillis
Beginner
Hello again,
Well, if you extract the C/V Roms from the board for reading, they are MASK ROMs of 64Mbit (8M x 8 / 4M x 16) compatible MX23C64010 64Mbit (4M x16) from Macronix MASK ROM.
The others you can follow this :
Original reference: https://servimg.com/view/15705655/320
Regarding the P1 and P2 Mask Roms, in general, they are 32Mbit (4M x 8 / 2M x16) but can be 64Mbit or 32Mbit, depending on the manufacturing time. For example, I do have a Samurai Showdown 5 Special that was supposed to come with 2x 32Mbit P1 and P2 but instead arrived with 1x 64Mbit (P1) and 1x 32Mbit(P2), at the 0x40000 was filled out with 00s.
Regarding the P3 is a normal EPROM I suppose, can be 27C800 or 27C160.
Now if you ask me about testing the proper functions, I don't know exactly what you mean, but in old times we just checked to see if they had the original CRC32 and we kept track of that. For Kof2003 "MVS version" should be:
If any of your Rom readers cannot read the correct version of it, I suggest you try to create some kind of adapter between the device your reader is capable of and the original MASK ROM. Also because all the ROMS are 5V you don't need to worry about it. On the latest games, more precisely in SVCChaos, Mslug 5, and KOF 2003 cartridges (boards PROGBK3S / PROGBK3R), you will see a 3.3v power regulator (48M033F), that is required only for the NEO-PVC chip and previous boards don't have it.
Regards.
Well, if you extract the C/V Roms from the board for reading, they are MASK ROMs of 64Mbit (8M x 8 / 4M x 16) compatible MX23C64010 64Mbit (4M x16) from Macronix MASK ROM.
The others you can follow this :
Original reference: https://servimg.com/view/15705655/320
Regarding the P1 and P2 Mask Roms, in general, they are 32Mbit (4M x 8 / 2M x16) but can be 64Mbit or 32Mbit, depending on the manufacturing time. For example, I do have a Samurai Showdown 5 Special that was supposed to come with 2x 32Mbit P1 and P2 but instead arrived with 1x 64Mbit (P1) and 1x 32Mbit(P2), at the 0x40000 was filled out with 00s.
Regarding the P3 is a normal EPROM I suppose, can be 27C800 or 27C160.
Now if you ask me about testing the proper functions, I don't know exactly what you mean, but in old times we just checked to see if they had the original CRC32 and we kept track of that. For Kof2003 "MVS version" should be:
Code:
b1dc25d0 271-c1c.bin
d5362437 271-c2c.bin
0a1fbeab 271-c3c.bin
87b19a0c 271-c4c.bin
704ea371 271-c5c.bin
20a1164c 271-c6c.bin
189aba7f 271-c7c.bin
20ec4fdc 271-c8c.bin
f5515629 271-m1c.bin
530ecc14 271-p1c.bin
fd568da9 271-p2c.bin
aec5b4a9 271-p3c.bin
ffa3f8c7 271-v1c.bin
5382c7d1 271-v2c.binIf any of your Rom readers cannot read the correct version of it, I suggest you try to create some kind of adapter between the device your reader is capable of and the original MASK ROM. Also because all the ROMS are 5V you don't need to worry about it. On the latest games, more precisely in SVCChaos, Mslug 5, and KOF 2003 cartridges (boards PROGBK3S / PROGBK3R), you will see a 3.3v power regulator (48M033F), that is required only for the NEO-PVC chip and previous boards don't have it.
Regards.
cyberwillis
Beginner
Now if you ask about how to test the ROMs in real time without extracting from the board now it's a trip, because every game has a different setup in a different board release.
Regarding the Program Board, you could use UNIBIOS from Razoola (http://unibios.free.fr/whatisit.html), to verify ROM integrity, if something is wrong you probably will have to repair them.
About the Character Board, it's clear on screen, if anything happens to the Cs you will not be able to decrypt, if anything happens to the M1 you will not have sound (CMC42), or not have decryption on screen and sound together (CMC50).
If you have some scrambled image information and can hear some sound from the game, but not all the graphics be ok on screen, then you will probably have some lines between the CMC and C Roms that may be suffering from bad welding that decayed over time or the Rom itself get burned.
But in general, each case is different. People tend to jump to the conclusion that the problem is the CMC because there is no easy substitution (in an easy way) to replace the C Roms without sacrificing the original MASK ROM. If somebody can find a company still selling the Industrial OTP or MASK ROM (pre-programmed) for MX23C6410PC this would be a nice replacement, for the damaged C Rom or even V Rom and rare cases P Rom.
In my humble opinion, if really matters to have the original MASK ROM saved I would suggest taking them all out, and replacing them with an easy alternative like the SOP44 to DIP42 adapter and having a bunch of them saved for further repair. And in the future, if there is a time to sell this card to somebody else, put the original MASK ROMs back and sell it.
BTW your Kof2003 is completely original, indeed the M1 is the correct one from manufacturing (EPROM 27C4001 - https://www.mvs-scans.com/index.php/King_of_Fighters_2003 )
Regarding the Program Board, you could use UNIBIOS from Razoola (http://unibios.free.fr/whatisit.html), to verify ROM integrity, if something is wrong you probably will have to repair them.
About the Character Board, it's clear on screen, if anything happens to the Cs you will not be able to decrypt, if anything happens to the M1 you will not have sound (CMC42), or not have decryption on screen and sound together (CMC50).
If you have some scrambled image information and can hear some sound from the game, but not all the graphics be ok on screen, then you will probably have some lines between the CMC and C Roms that may be suffering from bad welding that decayed over time or the Rom itself get burned.
But in general, each case is different. People tend to jump to the conclusion that the problem is the CMC because there is no easy substitution (in an easy way) to replace the C Roms without sacrificing the original MASK ROM. If somebody can find a company still selling the Industrial OTP or MASK ROM (pre-programmed) for MX23C6410PC this would be a nice replacement, for the damaged C Rom or even V Rom and rare cases P Rom.
In my humble opinion, if really matters to have the original MASK ROM saved I would suggest taking them all out, and replacing them with an easy alternative like the SOP44 to DIP42 adapter and having a bunch of them saved for further repair. And in the future, if there is a time to sell this card to somebody else, put the original MASK ROMs back and sell it.
BTW your Kof2003 is completely original, indeed the M1 is the correct one from manufacturing (EPROM 27C4001 - https://www.mvs-scans.com/index.php/King_of_Fighters_2003 )
Last edited:
