Closed sudorook closed 9 years ago
This is one of those “subjective” decisions…not many downsample by more than a factor of two. What rate are you aiming for/wanting?
On Mar 4, 2015, at 16:21, anselg notifications@github.com wrote:
https://github.com/RTXI/rtxi/blob/qt4/plugins/data_recorder/data_recorder.cpp#L517 https://github.com/RTXI/rtxi/blob/qt4/plugins/data_recorder/data_recorder.cpp#L517 Why is the maximum downsampling rate 2?
— Reply to this email directly or view it on GitHub https://github.com/RTXI/rtxi/issues/53.
Yogi A. Patel @yapatel525 https://twitter.com/yapatel525
When I record the RT benchmarks module for a couple minutes at 50 kHz, the files get massive. When I plot the data with R and ggplot, their voraciousness for memory makes the issue worse, and my system starts thrashing. I want to downsample it by, say, 100-200. Enough for me to record for 30 min. to an hour without the *.h5 files becoming unmanageably large.
I suppose that 100-200 is a bad idea when collecting experimental data, but I don't think that RTXI should block it. The 1.4 version lets you downsample between 1-1000.
How about we set the max to 500?
https://github.com/RTXI/rtxi/blob/qt4/plugins/data_recorder/data_recorder.cpp#L517
Why is the maximum downsampling rate 2?