separate audio stream per midi channel

[kmatze]

Hi,

i use your formidable tsf.h without tml.h.

with

        tsf_channel_note_on(g_TinySoundFont, chn, bData1, 1.0 / 127 * bData2);
	tsf_channel_note_off(g_TinySoundFont, chn, bData1);

i play midi note (on/off) and with

        int SampleCount = (size / (2 * sizeof(short))); 					//2 output channels
	tsf_render_short(g_TinySoundFont, (short*)stream, SampleCount, 0);

i get the actual audio stream for the whole tsf backend.

Is it possible to get an separate audio stream for a (midi) channel? So I can use my own audio effects per MIDI channel and then mix them finally.

thx - kmatze

[WindowsNT]

I'm doing it in my code to support effects per channel.

   tsf* f  = ...;
   struct tsf_voice* v = f->voices, * vEnd = v + f->voiceNum;
		for (; v != vEnd; v++)
		{
			if (v->playingPreset != -1)
			{
				TSF_MEMSET(buffer2, 0, sizeof(float) * SamplesMix);
				tsf_voice_render(f, v, buffer2, samples);

```               ....

[kmatze]

Thx, that`s a good entry, but i get only crazy noise.

So far I had used the following code:

		SFGetSample(sndSF2, waveBufLen);	                                                 // get synth data streams ...	
		tal_effect(sndSF2,   waveBufLen, sfxSF2);                                                 // ... process them with effects
		tal_mixer(soundbuf[i], snd1, sndSF2, snd3, waveBufLen);               	// ... an mixing together

void SFGetSample1(void *stream, long size) {
	int SampleCount = (size / (2 * sizeof(short))); 					//2 output channels
	tsf_render_short(g_TinySoundFont, (short*)stream, SampleCount, 0);
}

i change it to:

void SFGetSample(void *stream, long size) {
	int SampleCount 		= (size / (2 * sizeof(short))); 	//2 output channels
	struct tsf_voice *v    	= g_TinySoundFont->voices;
	struct tsf_voice *vEnd 	= v + g_TinySoundFont->voiceNum;
	for (; v != vEnd; v++) {
		if (v->playingPreset != -1) {
			TSF_MEMSET(stream, 0, sizeof(float)* SampleCount);
			tsf_voice_render(g_TinySoundFont, v, stream, SampleCount);
		}
	}
}

and get crazy noise

i don`t understand the logic of voice and channels, is it the same? Can you explain it please and the use of effect support.

thank you very much - greetings - kmatze