1
Image Issue Q&A / bit binning quantization problem QHY600 pro
« on: June 14, 2021, 05:12:52 PM »
Hello.
I have a QHY600 pro. and I bought the camera to perform photometry for exoplanets.
Performing a binning test, I found an error that I can't know where it comes from.
According to its specifications, the camera in 2x2 binning has 18 bit in 3x3 19 bit and in 4x4 20 bit.
When trying to carry out the tests with 2x2 binning, the socket would have to have 262,143 values, that is to say "65535 x4" but I find that the histogram continues to show me 65535.
I have tested it with various capture software and they all dial the same.
sharcap NINA etc.
You would also have to save the data in 32bit format and it is saved in 16 bit format, therefore there is an error.
Perform a simple test on a white light frame and the results are as follows
binning 1x1 time exp 49.1 ADU 6617 FACTOR +1
binning 2x2 time exp 49.1 ADU 25972 FACTOR +4
binning 3x3 time exp 49.1 ADU 58770 FACTOR +9
all correct, but the accounts do not rise from 65535 which are 16 bit for longer than the shot.
where is the problem,
In the SDK?
In the acquisition software?
I have a QHY600 pro. and I bought the camera to perform photometry for exoplanets.
Performing a binning test, I found an error that I can't know where it comes from.
According to its specifications, the camera in 2x2 binning has 18 bit in 3x3 19 bit and in 4x4 20 bit.
When trying to carry out the tests with 2x2 binning, the socket would have to have 262,143 values, that is to say "65535 x4" but I find that the histogram continues to show me 65535.
I have tested it with various capture software and they all dial the same.
sharcap NINA etc.
You would also have to save the data in 32bit format and it is saved in 16 bit format, therefore there is an error.
Perform a simple test on a white light frame and the results are as follows
binning 1x1 time exp 49.1 ADU 6617 FACTOR +1
binning 2x2 time exp 49.1 ADU 25972 FACTOR +4
binning 3x3 time exp 49.1 ADU 58770 FACTOR +9
all correct, but the accounts do not rise from 65535 which are 16 bit for longer than the shot.
where is the problem,
In the SDK?
In the acquisition software?