For my measurements I would like to use larger measurement buffers than the software allows at this moment. I have quite some lag in my system, and I would like to compare the immediate response being sent out (=without lag) with the DUT’s response in -preferably- one go.
For me it would be ideal to use a measurement buffer that allows for ~20 seconds of recording at 192 ksps. This is really large compared to the current maximum buffer. I would not use that big FFT`s, but I would need a buffer of 2^22 samples (maximum now is 2^18 samples).
Would this be possible to add? I would really appreciate this if it is anyhow possible!
Hi @Elmar, the current software will have trouble with buffer sizes this size in that certain operations (cursors and 1/n octave filtering). There is software being worked on the background (no date known) that will easily cope with large data sizes. I know that doesn’t help you right now, but I just wanted to acknowledge the issue.
I would not use that big FFT`s,
Is your issue that you have a huge delay, and you just need to test tones with a 20 second delay? If that is the case, there’s a setting called Latency Compensation that can compensate for system delays. Current it maxes out at 2.73 seconds, but I think that could be increased without issue.
Actually I’m working on a frequency transforming system, which records in realtime and plays back slower. The problem is that I would a high samplerate stimulus, which is slowed down significantly and then played back. (where I could loose the samplerate, but would need the bufferlength).
I would handle the buffer only via an API call but I do understand you might not be able to put it in the main SW branch .
.