I've made the driver source code available on my website. khalek, the only files that should really concern you are CSynthMT32.cpp, SynthMT32.h and mathutils.h. All the other files are Windows/DDK specific and shouldn't help you at all in writing your Linux driver.
Check the 'Not handling expression control?' thread again; I posted an accurate set expression, reset all controllers, and active sense(though you'll probably want to swap the doxbox timer with one from directx).
Oh yeah... I knew I was forgetting something. I still haven't found any software that uses expression on the MT-32 though to actually test the thing out. Sorry.. was so focused on the driver code I forgot to include your additions. Oh well.. a new version should be out in a few days.
In case you missed my post, CapTVK has a CM-64 that he's pulled from storage and is asking what to do to extract the samples. You mentioned an easier way. He's been posting about it in the 'well, I lost my ebay bids' thread.
wrote:In case you missed my post, CapTVK has a CM-64 that he's pulled from storage and is asking what to do to extract the samples. You mentioned an easier way. He's been posting about it in the 'well, I lost my ebay bids' thread.
I'm really sorry. I've been really busy and have been focusing on getting my driver up and running. As for the CM-64, I was thinking it would be possible to rip using lasynth (available at Quest Studios) but it doesn't support the CM-32/64. I haven't been able to find any games or midi files that support the CM-32/64. Without any support, it will be hard to do any development with regard to emulating the CM-32.
Function not matching description?
"Editor/Librarian A Windows 3x/9x/ME based patch editor and librarian for the MT-32, LAPC-1, CM-32L, CM-64, and D-110 synths. Saves and loads standard SysEx banks. "
ps. quest's link is dead, so will have to find it someplace else.
Perhaps put up the copy you have?
found it here:
http://www.xs4all.nl/~giovanni/midisoft.html
Originally posted by ih8registrations
"Editor/Librarian A Windows 3x/9x/ME based patch editor and librarian for the MT-32, LAPC-1, CM-32L, CM-64, and D-110 synths. Saves and loads standard SysEx banks. "
True, it supports the CM-32 because the CM-32 is MT-32 compatible. On running the software, however, it becomes apparent that it only supports either a sysex MT-32 mode or the D-110 sysex mode.
wrote:I've made the driver source code available on my website. khalek, the only files that should really concern you are CSynthMT32.cpp, SynthMT32.h and mathutils.h. All the other files are Windows/DDK specific and shouldn't help you at all in writing your Linux driver.
Thanks will have a look at it tonight
Um, the cm-64 is a combination of a cm-32l + u110 in one module.
The u110 part can be cascaded off of the cm-32l part for a sort of 64 note polyphony, but that doesn't mean you can't just access the cm-32l part. I ran lasynth and saw they only had mt-32 or d-110 in the pulldown menu but I'll bet they're using mt-32 as generic classification to select mt-32/cm-32l/lapc-i/cm-64(accessing just the cm-32l part). As far as the additional sound effects go, they are appended to the drum bank, notes 76-108, so I can easily imagine how a even a 'mt-32 only' editor could be used to access them. And if lasynth doesn't work, going off of the tricks you describe, a editor that's works with the cm-64 capable of the same procedure(s) could be searched for. It wouldn't hurt to try would it?
wrote:As far as the additional sound effects go, they are appended to the drum bank, notes 76-108, so I can easily imagine how a even a 'mt-32 only' editor could be used to access them.
Hmm... are you sure this information is correct? If so, it actually would be very easy to emulate a CM-32, since all the MT-32's drum bank is comprised of are 30 preset sound patches. They aren't specific PCM samples but are only preconfigured instruments just like the 128 preset instruments and 64 user configurable instruments. If all the CM-32 has is an additional 32 patches, adding them should be a piece of cake. Accessing them through lasynth, however, is impossible. There are no sysex commands that allow you to read the percussion patch parameters directly. They will have to be extracted from the firmware ROM the old fashioned way.
I'm not trying to be confrontational, but with my new job I have much less time to spend on the MT-32 emulation project. As such, I'm budgeting my time where I think the emulator will see the most improvement. This includes refining the sound quality and making an efficient, optimized driver. Adding CM-32 capability, on the hand, distracts me from that goal and will prove to be an exceptionally difficult task (with little reward) since support for it is so rare.
LAPC-I doc from quest studios:
http://www.queststudios.com/roland/docs/lapc1.txt
As for the time involved, captvk is the one who has the cm-64 and he'd be doing the extraction. What's being asked is to describe what it was you were refering to about tricks to getting clean samples.
khalek, if you want to collaborate on the alsaseq client, let me know. I did some preliminary work but I didn't know there were plans to actually separate the emulation code from Dosbox. 😀 Now it should be much easier.
wrote:LAPC-I doc from quest studios:
http://www.queststudios.com/roland/docs/lapc1.txtAs for the time involved, captvk is the one who has the cm-64 and he'd be doing the extraction. What's being asked is to describe what it was you were refering to about tricks to getting clean samples.
Sorry about the late reply. Interesting that those are drum patches. That means there are no new PCM samples to download. Unfortunately, that also means that they are unobtainable without actually dumping the firmware inside the lapc-1. This is because the preset percussion instruments are 4 partial patches just like all the other instruments. Only Roland did not leave any way in Sysex to read or write to the percussion patches like they did the instrumental tibres. Hmmm...
wrote:khalek, if you want to collaborate on the alsaseq client, let me know. I did some preliminary work but I didn't know there were plans to actually separate the emulation code from Dosbox. 😀 Now it should be much easier.
Sounds good (no pun intended). I haven't done much so far, mainly investigating other soft synths and the way they work. I've had increasingly little time to devote to things lately, hance the rather lagged response. Also I didn't have email notification of replies on which I do now...
