Sharpcap freezes during sensor linearity part of sensor analysis

[NeilG]

I managed to get the sensor analysis done for 16 bit RAW on my ZWO ASI%33MC-Pro - it was a bit of an ordeal as it seems to be very sensitive to the initial exposure and Sharpcap would sometime freeze during bit depth, sometimes while doing the e/ADU test and sometimes later in the process, but I got there in the end.
however, I have not been able to find a set of initial setting in terms of exposure (usually somewhere between 8ms and 15ms) that allows the sensor analysis to complete when running in 8 bit RAW.
It always seems that Sharpcap freezes during linearity testing - the number of frames previewed stops and the frame rates drop to zero.

Log file from the latest attempt is attached

Any ideas?

Regards
Neil

[admin]

Hi,

I have recently learned that for certain ZWO cameras the exposures being used are not quite accurate above one second exposure – typically they are rounded to the next lowest 0.1 second rather than being the exposure requested. This can cause the sensor analysis to become stuck when it is looking for the correct exposure level as it expects the image to respond to small changes in exposure, which doesn't happen.

I am planning an update to the sensor analysis engine that will ensure that the procedure should complete without needing to use exposures above one 2nd to avoid this problem, but I haven't got that ready yet. Your best bet with the current version of SharpCap will be to try a slightly brighter light source so that the initial exposure when getting started on the sensor analysis is between 0.5 and 1 ms. At this point ignore any warnings that the exposure may be too short and you should try a dimmer light source.



Cheers, Robin

[NeilG]

Robin
Thanks for the quick reply.

I'll try your suggestion today.
Of course, I've got the 16 bit working so I was able to use that last night to get some test images. Not quite got my head around One Shot Colour for deep sky yet - especially getting the image stretched while live stacking

Regards
Neil

[NeilG]

Finally got it to work @ 3.2ms starting exposure - I re-ran the 16 bit RAW as well, so now I've got a 'matched' set.

thanks for your help
Neil

[admin]

Hi,

Good to hear that you got it working. As I say, my goal is to make this much more reliable in a future version.

Cheers, Robin

[Djibi]

Hello,
I reply on this post because my problem is the same : Sharpcap freezes during sensor linearity, part of sensor analysis.
I have also an ASI 533 camera

I have read also a post on that subject : viewtopic.php?t=2212

What I don't understand is that you are refering to a starting exposure around some miliseconds

And when I launched Sensor Analysis (I have SC version 3.2.63322.0), The box is saying :
" Select an area ... peak at about 65% with exposures of 0,26s to 1s" (!!!)
No need to say that with a 533, to get an histogram peak at 65% with such exposure, I am almost in the dark !!
I can manage to launch the process this way but then, SC freezes and is locked during the "measuring the linearity and setting to 1%' phase"

It seems that you could manage to handle this by setting the exposure at start with 3ms or so but I can not do that : in that case, I have a message saying "exposure too short reduce lighting to an exposure at least 0,26sec".

Any hints for me ?
Thanks for your help

[admin]

Hi,

The older procedure for sensor analysis initialised with the camera set to maximum gain, meaning that you started off with a very short exposure of just a few milliseconds. The problem with this was that some cameras have very wide gain range and that could mean that the exposure is needed for minimum gain would be excessively long. Recently I changed so that the procedure starts at minimum gain and tries to get you to pick and exposure that is

a) under one second and
b) far enough above the camera minimum exposure that SharpCap will be able to measure again range of at least 1000x (50db)

This ensures that the sensor analysis should not need to use any exposures much above one second while running the full procedure. Unfortunately it looks like we may be hitting some teething troubles with this new procedure.

If you need to reduce the light levels getting to the camera then there are a number of ways that you can do so – my favourite approach is to use a relatively cheap CS thread lens with a variable aperture (these can be bought from China for about £12 on eBay or sourced from inside Europe for about £20). A much cheaper approach would be to use a piece of card over the nosepiece at the camera with a hole in it or just placing layers of cloth over the camera to attenuate the light levels.

If you still have difficulty then please post the log file after the procedure has become stuck and also if possible a screenshot. Do make sure that the frame counter in the bottom left-hand corner is still incrementing – it's always possible that an unrelated camera glitches stopping the frames from arriving at some point during the procedure.

Cheers, Robin