Question

Is it possible to write the audio buffers with an offset(delay) to generate a flat echo effect ?

The following piece of code outputs my audio buffers :

for(s=0; s<inNumberFrames; s++) {
    ioBuffer[s] = audioBuffer[audioBufferReadPos];
}

Is it possible for me to do something like this within the for loop :

tempBuffer[s] = audioBuffer[audioBufferReadPos];
--- Then somehow offset tempBuffer[] as bufferWithOffset[] --- 
ioBuffer[s] = audioBuffer[audioBufferReadPos] + bufferWithOffset[];

Any guidance in this regard will be highly appreciated.

Thanks.

Answer 1

Finally got it to work, thanks to Hollance on RW Forums for explaining all of this to me.

[ http://www.raywenderlich.com/forums/viewtopic.php?f=2&t=2864 ]

I still have a lot of crashing issues, caused most probably by memory leaks. But the logic works.

---

Initialized a temporary buffer with 22050 zero samples :

(SInt16 *)tempBuffer = (SInt16*)calloc(22050, sizeof(SInt16));

Initialized a long counter with zero :

long d=0;

Then within the for loop fed the temporary buffer mixed with the current sample :

for(s=0; s<inNumberFrames; s++) {
    ioBuffer[s] = tempBuffer[d] + audioBuffer[audioBufferReadPos];

Added the current sample to the tempBuffer :

    tempBuffer[d] = audioBuffer[audioBufferReadPos];
    d++;

Reset the counter to zero if the limit of the tempBuffer is reached :

    if(d >= 22050)
    d=0;
}

Assuming that the sampling rate is 44100 Hz, this will create a delay of 0.5 seconds.

---

UPDATE :

Changed

    ioBuffer[s] = tempBuffer[d] + audioBuffer[audioBufferReadPos];

To

    ioBuffer[s] = 0.6*tempBuffer[d] + 0.4*audioBuffer[audioBufferReadPos];

And that fixed the crashing.