So that's two diskwrite modi:

- each channel
- n busses, where the user can assign any channel to a bus

Quote:

- n busses, where the user can assign any channel to a bus

nah, you can do that in your cubase environment anyway

I think the easiest is to dump all audio channels. Since its audio it shouldn't be too CPU intensive and it doesn't require that much disk space either. Of course it would be nice to select channels but that complicates things I think.

norakomi: I was referring to YOUR suggestion of grouping certain channels for diskwriting. It would be a bus in the emulator. You assign any channel to a bus, and the bus will be written to a unique wave file.

That is of course also an option. A third option which some people actually asked for is to be able to make the mixer more advanced so you can split channels. Then it would be quite easy to have a button (like mute or solo buttons in Cubase) that enables/disables saving on a particular channel.
This would be the most elegant solution I think but it would also take a looong time to implement. Quickest and still quite useful is to always save all individual channels.

I think the only ppl who would ever diskwrite channels would be musicians trying to remix or rearrange, for those: all individual channels are the most useful. The rest is nice for bonus but not super-essential.

If we use mono channels for each wav, at least at 16/44, then that makes about 10 channels to write for the average Microcabin tune (6FM + drums + 3 PSG). Are you sure all that disk access is possible? Because in the worst-case scenario ppl are going to diskwrite Moonblaster stereo tunes at 9+9+1+1 (fm1, fm2, adpcm, psg)

Yep I know and thats a concern. It could be a bit too much disk writes but should be doable. I don't really have anything to compare with. When I use Cubase I can only record 4 channels at the time because I don't have any cool multichannel soundcards. But I heard its no problem with 40+ channels on a modern PC.

cool
I'd be happy to hear more good news.
Talk to you later !!

Quote:

Because in the worst-case scenario ppl are going to diskwrite Moonblaster stereo tunes at 9+9+1+1 (fm1, fm2, adpcm, psg)

No. Worst case is when someone start recording full-spec SME3 songs PSG + FM-PAC + 2x MusMod + 2x SCC + 1 PCM on turboR. 43 channels-o-noise... Now that, my howling friend, is worst case.

I consider that chance close to zero. It might technically be a worse situation, but it's so rare... who carez..!

Quote:

I consider that chance close to zero. It might technically be a worse situation, but it's so rare... who carez..!

Anyone who wants to record "Verkeersinformatie" into separate tracks

Technically worse then would be playing 18 channels of OPL4 FM + 24 channels of OPL4 Wave while SME3 is blasting its 43 channels of noise
But I have to admit that the chance of that happening could even be below 0

The openMSX sound logging is based on EmuTime nowadays, which means it will not skip samples even if the PC cannot produce them fast enough. So there is a chance you will have to wait 5 minutes to record a 3 minute song, but the recorded WAVs would be correct.

I'm not sure how hard it is to record individual channels, because the channel mixing is done in the sound chip emulation code rather than in the generic mixer. Currently a sound chip delivers a single mono or stereo sound stream, that would have to become multiple streams if we want to record each individual channel.

If the sound emu mixes it all in a non-common way then what overriding the muted channels with volume 0, each int ?


ps. lost your password or what?

Don't you want to log all channels in one go? (to separate WAVs, of course) If you would record one channel at a time, it would be hard to get the channels synced later.

About the other account: I didn't have my password saved on my notebook and I don't use it often enough to know it by heart.

uh yes.. , what about running -in case of FM-Pac- 9 instances, and each with a uniquely solo'ed channel (by v0'ing the others) ?