It's worth explaining that a SC-50 is in fact a SC-55mkII minus MT-32 tones. And that's what this unit is, a SC-50 with sequencer - well, probably!
I don't believe the service notes/manual are available for this device, so it's hard to be sure that it uses the UPD63200 DAC, and so on. It could be a hybrid SC-50 with some (lesser) 55 parts. If you want to confirm which UPD DAC is used, that would be VERY helpful and appreciated!
But, *most likely*, it's what I said in the first line, and what you implied.

The MT-200, from reading the manual's specs, is based on the SC-55. Lower polyphony and less GS instruments, but has the MT-32 bank.
You're right though, misinformation abounds about Sound Canvases, which is why I (somewhat naively) tried to make a project about cataloguing them. Google first page of results for the MT-200 had some confusing the specs with the 120, some with the SC-55mkII..

Just had a look and it uses the UPD6736 DAC which itself appears to be the 16-bit only baby brother of the UPD63200. It's pin-compatible, with the only difference being that pin 3 that selects between 16/18-bit on the UPD63200 is not connected on the UPD6736. I imagine you could 'upgrade' it if desired, but to be absolutely honest I've struggled to hear any difference between my MT-120 and my SC-55ST, and looking at the spec sheets between the two chips shows very little difference in performance characteristics.

Here's a list of the rest of the chips if interested:

Hitachi HD 63266F
Roland R15239186
NEC D70325GJ-8
Roland R15239187
Hitachi 65256BLFP-12T
SRM2A256SLM10
Roland LCGS-2A R15209486
Roland R15199827
Roland R15239229
Roland R15279824

The format which Roland and many japanese things in general outputs is Left Justified which, unlike I2S and Right Justified (LSB aligned), requires receiving side to explicitly know how many bits is transferred or you get messed up output since bits are actually misaligned and all zero crossings get lost because the sign bit gets misinterpreted which results in not so nice things 🤣.

I2S and RJ formats (MSB aligned) can output as many bits as they want with receiving side not having to know how many there really are. If a device outputs 32bits and the DAC can only do 16, it can and it will still sound as good as DAC can do without any explicit knowledge about how many bits were being output, since it captures highest bits first and ignores any lower ones. Unfortunately japanese seemed to have hated these formats...

Because of this, you cannot just put the 18bit DAC into a thing that was using 16bit one, while the DAC is confed to 18bits. TC6116AF based devices can output either 16 or 18 bits, depending on software conf, so in theory it is possible to do the upgrade if software gets hacked to enable the right kind of output.

RichB93 wrote on 2024-05-11, 23:41:

Just had a look and it uses the UPD6736 DAC which itself appears to be the 16-bit only baby brother of the UPD63200. It's pin-compatible, with the only difference being that pin 3 that selects between 16/18-bit on the UPD63200 is not connected on the UPD6736. I imagine you could 'upgrade' it if desired, but to be absolutely honest I've struggled to hear any difference between my MT-120 and my SC-55ST, and looking at the spec sheets between the two chips shows very little difference in performance characteristics.

Ah, interesting. So it uses the lower quality version. It's 6376, not 6736, btw. A couple SC variants use it, although the original 55 uses the 6379 (not sure what differences exist between 6/9).
Generally speaking the UPD63200 is highly regarded, it's in the JV-1080 which people love the sound of.

The SC-55ST doesn't use the UPD63200, it uses the UPD6379, same as the SC-55, and presumably very close to the 6376. So it's definitely expected that they would perform similarly.

One interesting point from comparing the datasheets was how much lower the noise rating is for the 18-bit chip, about 5dB less noise, crosstalk, distortion and improved dynamic range.

Interestingly, looking at the 6376, it looks like it might outperform the 6379, but they are extremely close, with the 63200 a bit better than either.

6379 is the lowest cost version, and its main issue is both channels use same VREF internally and it increases stereo crosstalk and distortion a little. 6376 and 63200 also have separate analog and digital power rails which helps to improve performance even more compared to 6379.

Hey y'all,
I'm pretty much brand new to midi, but I bought one of these for fun and I've been learning a lot. I'm basically using it as a floppy disk boombox, and generally it's been working. One issue I've run into though is that the MT-120 (seemingly arbitrarily) is not able to play many of the standard midi files I download.
Take the dark side of the moon for example:

These songs WILL play:
Breathe (In the Air): https://freemidi.org/download4-7858-breath-in … -air-pink-floyd
Time: https://freemidi.org/download4-7866-time-pink-floyd
The Great Gig in the Sky (VERSION 2): https://freemidi.org/download4-24951-great-gi … ky-2-pink-floyd
Us And Them: https://freemidi.org/download4-7867-us-and-them-pink-floyd
Brain Damage (VERSION 1): https://freemidi.org/download4-5697-brain-damage-pink-floyd
Eclipse: https://freemidi.org/download4-24950-eclipse-pink-floyd

But these songs WON'T:
On the Run: https://freemidi.org/download4-13439-on-the-run-pink-floyd
The Great Gig in the Sky (VERSION 1): https://freemidi.org/download4-11374-great-gi … -sky-pink-floyd
Money: https://freemidi.org/download4-5702-money-pink-floyd
Brain Damage (VERSION 2): https://freemidi.org/download4-24949-brain-da … ge-2-pink-floyd

I feel like the issue has to be something to do with the midi files. I suspect that the files which don't work do not conform to "General MIDI System Level 1 Specification" in some way, but I simply can't figure out specifically how, and whether I could potentially fix these files so that they would be playable. The MT-120 manual is available here: https://www.polynominal.com/Roland-mt120/inde … #prettyphoto/4/

It's interesting to note that for the songs The Great Gig in the Sky and Brain Damage, one version works while the other doesn't- but they don't seem radically different when I listen to them. Seems like these would be the best clue as to what is going wrong.

I would much appreciate any help that any of you can provide in solving this mystery!

Thanks!

For anyone curious, I figured it out finally. Per the manual, I had to turn down the PPQN to 120. You can do it in Reaper. So now they all play. I am having a new issue though where sometimes specific portions of a track are not playing 🤣. I haven't figured that out quite yet. The files play correctly on my computer though, so it seems to be something specific to MT-120. Any ideas on this issue would be appreciated too!

I haven't checked the songs above, but it could be the polyphony limit being reached. The MT-120 has a 28 note polyphony, the same as the SC-55 (as it uses the same chip). Some instruments also have multiple elements, meaning that one note played actually uses multiple voices, further lowering the available polyphony. For example, Synth Brass 1 uses two voices per note, so playing a three note chord would use up 6 voices.

georgefestoon wrote on 2024-11-18, 01:27:

For anyone curious, I figured it out finally. Per the manual, I had to turn down the PPQN to 120. You can do it in Reaper. So now they all play. I am having a new issue though where sometimes specific portions of a track are not playing 🤣. I haven't figured that out quite yet. The files play correctly on my computer though, so it seems to be something specific to MT-120. Any ideas on this issue would be appreciated too!

Which tracks? And what does "specific portions" mean, certain instruments or certain parts of a song entirely?
It appears some tracks have meter changes, so maybe that's having an effect with the MT-120 similar to the tempo being off 120.

Okay, I'm still having trouble with this 2nd issue I mentioned. I've done a lot of investigative work, and I realized that basically, whatever is not audible is whatever is assigned to channel 1. and I realized from looking at the manual that channel 1 is the channel that can go out to an external keyboard. I pressed the sound button and sure enough, "part 1" is showing "Ext Equipment" as shown on page 35 of the manual I linked above. This is the case DESPITE the fact I have the Piano switch set to INT on the back, as shown on page 13 of the manual.

I even tracked down this Factory Reset Manual from Roland: https://www.rolandus.com/assets/press_media_r … ctory_Reset.pdf

and successfully factory reset the device. This has not changed the fact that Part 1 is still stuck on "Ext Equipment"

Any ideas for how to fix this??

I've confirmed it is a hardware issue using test 6 in the diagnostics/reset menu (shown here: https://youtu.be/I_JbyT0cz2g?si=-ZT-9AvTnmgacF_U&t=675).

All the front buttons work fine but start/stop doesn't work. Also the headphones port and outputs ports on the back don't work either. On the other hand, the input ports DO work, as does the guitar/line switch and both the input volume AND the metronome volume.

So, generally it seems like the left side of the back works, but the right side does not work. the only exception is the metronome volume, which is on the right side of the board but does not work.

I'm not talented enough at electrical engineering to diagnose the actual circuitboard. Is there anything obvious to try, or do I need to take it somewhere?