You can check the ROM version with a key sequence and then powering on. See here:
https://www.philscomputerlab.com/ultimate-rol … 2-tutorial.html

cyclone3d wrote:

You can check the ROM version with a key sequence and then powering on. See here:
https://www.philscomputerlab.com/ultimate-rol … 2-tutorial.html

Thanks! I forgot about this guide - I discovered it some time ago, but I never found time to watch the whole of it 😁

I'm in the same situation as the original poster. I have two MT-32's. Both "old" versions. But different motherboard revisions (79377310 00 vs 79377310 01).

Are there any known differences between the two motherboard revisions? Which one is "better"?

well if you have 00 + 01 the difference is really obvious if you looked at the pcb. but neither is 'better', they both have quirks. take your pick.
00 is ceramic pga. roms as eeproms
01 is plastic qfp. roms are mask roms.

BloodyCactus wrote on 2020-05-03, 02:52:

well if you have 00 + 01 the difference is really obvious if you looked at the pcb. but neither is 'better', they both have quirks. take your pick.
00 is ceramic pga. roms as eeproms
01 is plastic qfp. roms are mask roms.

Yes the physical differences are obvious. By 'better' I meant which one produces higher quality audio or has less noise on the line outs? Was the 01 revision made to fix problems with the 00 revision? (ie. rev 01 is better) Or was the 01 revision a cost reduced version? (ie. rev 00 is better)

no, the 01 was made to reduce the cermiac PGA roland chip to a QFP chip.

BloodyCactus wrote on 2020-05-03, 12:32:

no, the 01 was made to reduce the cermiac PGA roland chip to a QFP chip.

Roger. In that case I'd say the revision 01 is slightly more desirable (it has mask ROMs for the version 1.07 firmware).

neither is more desireable than the other. mask roms just means you cant erase and reprogram them. I have an 00, it has 1.07.

BloodyCactus wrote on 2020-05-03, 18:06:

neither is more desireable than the other. mask roms just means you cant erase and reprogram them. I have an 00, it has 1.07.

In this case, mask ROMs are definitely superior to EPROMs. The data stored in EPROMs will be corrupted over time. It takes seconds/minuets when exposed to UV light or decades if the UV light exposure is minimized. But it still happens.

the Goat wrote on 2020-05-03, 19:02:

BloodyCactus wrote on 2020-05-03, 18:06:

neither is more desireable than the other. mask roms just means you cant erase and reprogram them. I have an 00, it has 1.07.

In this case, mask ROMs are definitely superior to EPROMs. The data stored in EPROMs will be corrupted over time. It takes seconds/minuets when exposed to UV light or decades if the UV light exposure is minimized. But it still happens.

What EEPROMs could I use to replace the EPROMS in mine ? You got me thinking of getting one of those TL866 programmers, dumping the original ROMs and writing them to EEPROMs .

darry wrote on 2020-05-03, 19:19:

What EEPROMs could I use to replace the EPROMS in mine ? You got me thinking of getting one of those TL866 programmers, dumping the original ROMs and writing them to EEPROMs .

You wouldn't what to use EEPROMs. The MT-32 has no capability or use case to write to the firmware ICs.

The correct EPROMs to use are 27c256.

the Goat wrote on 2020-05-03, 19:33:

darry wrote on 2020-05-03, 19:19:

What EEPROMs could I use to replace the EPROMS in mine ? You got me thinking of getting one of those TL866 programmers, dumping the original ROMs and writing them to EEPROMs .

You wouldn't what to use EEPROMs. The MT-32 has no capability or use case to write to the firmware ICs.

The correct EPROMs to use are 27c256.

I know the MT-32 can't write to EEPROMs, I would be using an external programmer (TL866). The reason I want to use EEPROMs instead of EPROMs is that I would rather not have to buy a UV light source to erase them first (can you even still get new blank EPROMs?). However, it seems that using EEPROMs would requiring swaping some pins, so not great either .

Thanks for the model number .

darry wrote on 2020-05-03, 19:57:

The reason I want to use EEPROMs instead of EPROMs is that I would rather not have to buy a UV light source to erase them first (can you even still get new blank EPROMs?).

Ahh, I understand.

Well, looks like I should have looked inside my MT-32 first : soldered mask ROMs .
So no need to worry about EPROM erasure. FW is 1.07

the Goat wrote on 2020-05-03, 19:02:

BloodyCactus wrote on 2020-05-03, 18:06:

neither is more desireable than the other. mask roms just means you cant erase and reprogram them. I have an 00, it has 1.07.

In this case, mask ROMs are definitely superior to EPROMs. The data stored in EPROMs will be corrupted over time. It takes seconds/minuets when exposed to UV light or decades if the UV light exposure is minimized. But it still happens.

But bin files to program your own EPROMs are easly available. OTP ROMs were just cheaper. Also it takes a long time for UV EPROMs with exposed window in direct sunlight to be "erased" (hours if not days) and I'm not sure what kind of UV radiation it should be to erase them in seconds - takes at least couple minutes to erase the UV EPROM in the eraser... so by no means it's worse/better than OTP EPROMs.

PS: Also 29C256 Flash is a direct 27C256 UV EPROM replacement.

PS: Board revision 00 is more rare then 01 as far as I know.

HanJammer wrote on 2020-05-03, 21:57:

OTP ROMs were just cheaper.

OTP EPROMs are identical to UV erasable EPROMs, just they don't have the window to let in UV light. The data will still be corrupted over time.

HanJammer wrote on 2020-05-03, 21:57:

Also it takes a long time for UV EPROMs with exposed window in direct sunlight to be "erased" (hours if not days) and I'm not sure what kind of UV radiation it should be to erase them in seconds - takes at least couple minutes to erase the UV EPROM in the eraser... so by no means it's worse/better than OTP EPROMs.

The EPROM doesn't need to be erased to turn the MT-32 (or other device) into a brick. Only a single bit needs to flip from a '0' to a '1', or vice versa, to ruin your day.

Any way to tell whether the ROMs are OTP or MASK Roms ?

EDIT: Without decapping them

darry wrote on 2020-05-03, 22:54:

Any way to tell whether the ROMs are OTP or MASK Roms ?

EDIT: Without decapping them

You can usually guess based on the part number and or company logo on the package. ie. if the chip is branded 'Roland' with a proprietary part number it is probably a Mask ROM. If it is branded AMD or NEC, etc. and has a part number that follows the industry standard for PROMs then you've got a OTP PROM.

the Goat wrote on 2020-05-03, 22:52:

OTP EPROMs are identical to UV erasable EPROMs, just they don't have the window to let in UV light. The data will still be corr […]
Show full quote

HanJammer wrote on 2020-05-03, 21:57:

OTP ROMs were just cheaper.

OTP EPROMs are identical to UV erasable EPROMs, just they don't have the window to let in UV light. The data will still be corrupted over time.

HanJammer wrote on 2020-05-03, 21:57:

Also it takes a long time for UV EPROMs with exposed window in direct sunlight to be "erased" (hours if not days) and I'm not sure what kind of UV radiation it should be to erase them in seconds - takes at least couple minutes to erase the UV EPROM in the eraser... so by no means it's worse/better than OTP EPROMs.

The EPROM doesn't need to be erased to turn the MT-32 (or other device) into a brick. Only a single bit needs to flip from a '0' to a '1', or vice versa, to ruin your day.

I have a really good laugh reading your pesimistic comments about EPROMs failing. While it's true EPROMs degrade over time - personally I never seen one failing because of age (even the early 80s BIOS and BASIC EPROMs from 5150s were all working correctly).
ESD or failure because of some other component failed is another thing. And if it fails - so what. Those are just off the shelf parts, easly available and cheap too. Also socketed in case of rev 00 board.

PS: Unlike the ROMs used on 01 board which are Roland branded and soldered on. And good luck finding those masked ROMs in case it fails (not because of it's limited lifespan but for different reason - ie. ESD when you are fiddling inside of the MT-32).
So 00 revision definitely has a big advantage over 01 here.