MoonSound Music Studio?

Page 1/2
| 2

By Dhampird

Hero (585)

Dhampird's picture

07-04-2013, 03:51

I´m wondering if the software "MoonSound Music Studio"was finished or if only today still beta version. Anyone knows about this?

Login or register to post comments

By syn

Prophet (2069)

syn's picture

07-04-2013, 07:52

It is still in beta but it also still in active development. If you want more info/specifics, you could try contacting guyver800 either through mail or just drop by at the #msxdev chatroom where he can be found occasionally.

By Grauw

Ascended (10174)

Grauw's picture

13-03-2014, 00:07

From another thread:

GuyveR800 wrote:

As for playing the music on a MoonSound, a MoonSound is capable of producing all sounds that MSX-MUSIC (FM-PAC) can. For example, MoonSound Music Studio emulates the MSX-MUSIC perfectly when importing eg. MB-STEREO music.

So users could be provided with the exact same music experience as FM-PAC on their MoonSound.

Hmm, interesting. So the exact settings for the OPLL built-in instruments are available? I mean I guess they are out there as there are emulations and FPGA implementations of the MSX-MUSIC, but I always figured those were using approximations of the instruments, rather than exact reproductions?

I wonder how people figured those out. Dissecting the IC? Some OPL variation or driver that supported those exact same hardware instruments for backwards compatibilty, but whose values could be read? :)

By Manuel

Ascended (18254)

Manuel's picture

13-03-2014, 08:21

I guess the OPL3 part is used for that emulation...?

By msd

Paragon (1462)

msd's picture

13-03-2014, 09:12

I wouldn't use the word emulation for this. The opl chips are all more a less the same family of chips. The opll is only a stripped version of one of the opl chips. (I think a stripped opl2 version). There are some differences in adsr times and such, but you can get very close. IIRC the main reason way the msx-audio can't do the opll sounds is because it doesn't have the extra wave form. Opl2 has 4 and opl3 has 8 waveforms. So in other words, if you know the settings used for the hardware voices of the opll you can make the same sounds on the opl2 or opl3.

By anonymous

incognito ergo sum (116)

anonymous's picture

13-03-2014, 11:33

msd is correct.

Grauw is also correct in that the hardware voices of the OPLL are still approximations in FPGA and software emulators. This is easily verified by inputting their register data into a real OPLL and hear the massive difference! Obviously, this points to errors in the emulations themselves, because those "approximate" voice data do sound similar to a real OPLL in the emulations.

As for MS2's OPLL import, while software voices are 100%, the hardware voices are also "beta". However, given my experience in FM instrument design and editing, I am confident I will be able to reverse-engineer the instrument settings more accurately than current approximations. The final say on this will come when the YM2413 decapitation project is finished, which will hopefully allow the internal ROM to be read.

Frankly, I wonder wtf ppl are doing to get those approximations so wrong... For example, emulation authors recently figured out that the KSL bits on the OPLL are different than on all other OPL chips. This fact was known to me since the early 1990s, because I noticed this while designing FM instruments in MB1.4 at the time...

This also means a lot of software instruments sound wrong on OPLL in MoonBlaster. But rest assured, MS2 makes sure MoonBlaster music sounds identical (ie. equally wrong) when imported in MS2!

By Grauw

Ascended (10174)

Grauw's picture

13-03-2014, 13:51

GuyveR800 wrote:

Grauw is also correct in that the hardware voices of the OPLL are still approximations in FPGA and software emulators. This is easily verified by inputting their register data into a real OPLL and hear the massive difference! Obviously, this points to errors in the emulations themselves, because those "approximate" voice data do sound similar to a real OPLL in the emulations.

So it is as I thought. Interesting and somewhat amusing to hear that those emulation errors sort of double backed on themselves in the voice data, once again producing something similar to the hardware. Smile Downside is that this means taking the hardware instrument settings from an emulator won’t do me much good.

I’m asking you see because eventually in Synthesix I’d like to support loading OPLL-patches on the OPL1 and OPL4, and vice versa, as long as they don’t use any features that aren’t supported by the other chip (such as the OPLL’s 2nd waveform).

GuyveR800 wrote:

The final say on this will come when the YM2413 decapitation project is finished, which will hopefully allow the internal ROM to be read.

That’d be sweet! Then there’d finally be a definitive answer.

Are there any details / threads about this project, where I could follow the progress?

GuyveR800 wrote:

For example, emulation authors recently figured out that the KSL bits on the OPLL are different than on all other OPL chips.

Good tidbit to know!

GuyveR800 wrote:

This also means a lot of software instruments sound wrong on OPLL in MoonBlaster. But rest assured, MS2 makes sure MoonBlaster music sounds identical (ie. equally wrong) when imported in MS2!

Perfect! Smile

By Retrofan

Paragon (1264)

Retrofan's picture

13-03-2014, 14:02

@GuyveR800: Very nice! Hope to see a next release or beta of MS2 very soon! Smile

Would be cool if this feature also could be done at the hardware side, so all software with MSX-MUSIC (and MSX-AUDIO) support would be automatically be compatible with the OPL4. Only problem is the ADPCM part I think (which MS2 can convert to PCM, right?) Maybe this can be done with the new MSX computer of Tecnobytes with a additional PCB...
Then the MSX-MUSIC and MSX-AUDIO aren't needed anymore stand-alone as they are supported with the OPL4... (without background noise!)

By Grauw

Ascended (10174)

Grauw's picture

13-03-2014, 14:54

Retrofan wrote:

Would be cool if this feature also could be done at the hardware side, so all software with MSX-MUSIC (and MSX-AUDIO) support would be automatically be compatible with the OPL4. Only problem is the ADPCM part I think (which MS2 can convert to PCM, right?) Maybe this can be done with the new MSX computer of Tecnobytes with a additional PCB... Then the MSX-MUSIC and MSX-AUDIO aren't needed anymore stand-alone as they are supported with the OPL4... (without background noise!)

OPL4 is already 100% I/O compatible with OPL1. MSX-AUDIO is using the Y8950 which has ADPCM in addition to the OPL1 so indeed you won’t be able to hear the samples, but otherwise it should just work. Though I think you need to change some wires or dip switch in the MoonSound to make it listen to the MSX-AUDIO ports (I forgot). Some people have done this, it came up in this forum just recently.

For the MSX-MUSIC (OPLL), things are not so easy though, as OPL4 is not I/O compatible with it. That’d need some kind of translation layer, best done in software.

Also, note that even though OPL4 appears to have enough channels to support both MSX-MUSIC and MSX-AUDIO, if both of them are configured to have the 5 drum channels (uncommon I guess), OPL4 won’t be able to support that.

By Manuel

Ascended (18254)

Manuel's picture

13-03-2014, 15:01

So, it would be almost possible to make a MoonSound-like cartridge that implements MSX-AUDIO and MSX-MUSIC using an OPL4 as implementation ;-) An MB-stereo cart, as it were... (A bit like MIDI-PAC, but then routing stuff to the OPL4.)

By anonymous

incognito ergo sum (116)

anonymous's picture

13-03-2014, 15:09

@Grauw: Yeah, that's one reason to be thankful for MoonBlaster's limitations Smile It only supported drum mode on MSX-MUSIC, not on MSX-AUDIO. Other MSX music programs that did support MSX-AUDIO drums never did so in combination with MSX-MUSIC.

BTW, one thing that makes a difference in the sound is the rarely used feature of Modulation Depth, which OPLL did not support. So when you use it, the OPLL instruments sound different compared to a real OPLL.

Page 1/2
| 2