Hi everyone,

I'm just expanding my processing experience to the audio world so I'm new to minim library but I have processing programming experience. Here is what I want to do and could veterans on this board help me by commenting on whether I will be able to accomplish what I want? Thank you!

Objective: play a sound periodically and record the playback at the same time. Bats use this way to fly in the dark without hitting obstacles.

Details: I have a sample I want to play. It has less than a millisecond of a high frequency sound (sine wave at 20KHz) followed by silence. The length of the sample is possibly at most 0.1 second long if not shorter. While I play the sample I want to record it on the mike. Theoretically, the mike will record the initial high frequency sound and then another sound, a reflection of the original sound wave from objects near the speaker, a wall, a cup etc. Think about talking in an empty room and you can hear your echo. I want to play a sound and record its echo. The time lag between the two short sounds can be used to find the distance of the objects to the speaker. I think this is how bats navigate in dark caves.

Question 0: Can I get some comment on my objective?

Question 1: I did some search online about recording and playback. It seems that recording has a variable and inevitable latency, which is between you request for recording and recording occurs. I'd like to know how much latency minim has if ever.

Question 2: I also read through several pages at this sound forum and found some people having lags on playback. How bad can that be if all I want is to repeat a sample?

Question 3: Is there a way to play repeatedly on a hardware level so that once I set the sample to play the sound card will play the sample until the sound card is told to stop?

Thank you for your time!!!

I'm by no means a board veteran so feel free to disregard if this seems like a stupid idea...

Couldn't you build in some calibration to take account of the effects on distance readings caused by lag?  i.e. have a variable you can adjust to compensate for lag after having tested against known distances...  Assuming the lag is unavoidable but consistent on a specific setup this might be an acceptable workaround.  In fact given that the lag may differ depending on the hardware the application runs on this might actually be an essential step in order to make it portable to other hardware.

Presumably a bat goes through a similar learning process when it first uses echo-location?