Description
Hi, I am using PRESTO to process some data and look for single pulses. The files have the following specifications:
Number of frequency channels: 2048
time resolution: 0.01 seconds
lenght: 298.68 seconds
bandwidth: 1200-1800 MHz
Number of samples: 29868
I realized that when using prepsubband prepsubband -nobary -lodm 300 -dmstep 10 -numdms 20 -nsub 2048 -runavg -o prep_output 2023-10-07_18_16_51_752602noise.fil the number of samples written is less than the original number of samples of the file
So when I use single_pulse_search.py to search for transients it doesn't find anything after the sample 24000. I lose ~6000 samples which correspond to one minute of my five minute files.
I have processed data with a higher time resolution and I noticed this happens as well but due to the higher resolution the time lost is not as much as in my lower resolution case. For example for data with the following specifications:
Number of frequency channels: 512
time resolution: 54.613 microseconds
length: 3599.99913984 seconds
bandwidth: 1210-1510 MHz
Number of samples: 65917953
I lose approximately 25000 samples which correspond to 1.5 seconds of an observation of one hour.
A "solution" to this would be to make the files larger so I dont lose a significant amount of data but still a full minute is a lot of time lost.
I have seen something similar in the PRESTO_search_tutorial.pdf where the samples written is less than the original ones.
This is taken from slide 11 and as you can see the total number of points is 531000 and the data written points is 530000
Is this an expected behaviour? I am not sure why this is happening.
Is there a way to make it process the full file with its current resolution?
Any indication is much aprecciated. :)