QHY294C - Sensor Analysis Issue

[gwsnyder]

I've been trying to run the Sensor Analysis tool on Sharpcap for my QHY294C camera and have had mixed results. If I run the tool with the Colour Space set at RAW8, I do get expected results. However, when I run the tool with the Colour Space set at RAW16, I am not getting any results that resemble the sensor charts for the camera found on the QHY site. Additionally, the only way I can get the tool to start in RAW16 is if I set the RBG color balance values to 16 (the sensor analysis tool sets these at 64 when I hit the 'Start' button.

The issues seem to start right away when going through the e/ADU measurement process. The green crosses on the measurement graph are more of a scatter chart instead of the straight line noted in the instructions pane.

Details: SharpCap 3.2.6421, QHY SDK for SharpCap 20.7.26.1

Am I missing something here, or is the SharpCap SDK still buggy?

[admin]

Hi,

yes, you're correct that the default values should be 16 for the 294 camera – I even have a fix for this in the code for SharpCap 3.3, but that fix isn't in SharpCap 3.2. Unfortunately the problem is that you don't get good results without the correct neutral colour balance values and those values vary from model to model according to no particular pattern with the QHY cameras (and with no documentation). With the ZWO cameras you know that the default colour balance is 50/50 regardless of the model.

The fact that the colour balance values are four times too high immediately means that SharpCap would likely detect this is a 12 bit sensor rather than 14 bit, which would then throw a lot of the later calculations off.

I will put the fix for the default white balance values into the next update to SharpCap 3.2, but I'm not really sure what to suggest about the e/ADU graph being incorrect – I've seen a number of glitches with different types of camera from not responding correctly to small changes in exposure to doing completely weird things like producing a bright image even when the exposure is set near to the minimum value. Usually working out what is going on takes an awful lot of trial and error and isn't something that I can really do without having the camera to play with.

Cheers, Robin

[gwsnyder]

Thank you for your reply. I don't envy you trying to keep all the different manufacturers/cameras straight with their inevitable different development practices (sounds like QHY might have autonomous development groups for each camera). Oh well.

I'll try again when SharpCap 3.3 is available.

[admin]

Hi,

I was actually tinkering in that part of the code the other evening so I pulled the change for the white balance defaults across. Look out for the next update for SharpCap 3.2.

Cheers, Robin

[gwsnyder]

So, I updated to the latest version of SharpCap (3.2.6433.0) and also upgraded the QHY SDK to their latest version. Good news is that I can muddle my way through the sensor analysis - but only in the RAW8 color mode. When I try the analysis at RAW16, the very first readings (e/ADU) are charted all over the place, not in the straight line as seen in the RAW8 sensor analysis or in the notes on the sensor analysis screen. Even though I can get the tool to complete successfully in RAW16, the results are all blank and/or "0". No graph is produced at all.

Any ideas?

[IZ1IVA]

So, I updated to the latest version of SharpCap (3.2.6433.0) and also upgraded the QHY SDK to their latest version. Good news is that I can muddle my way through the sensor analysis - but only in the RAW8 color mode. When I try the analysis at RAW16, the very first readings (e/ADU) are charted all over the place, not in the straight line as seen in the RAW8 sensor analysis or in the notes on the sensor analysis screen. Even though I can get the tool to complete successfully in RAW16, the results are all blank and/or "0". No graph is produced at all.

Any ideas?

I'm afraid I don't have an answer for that, instead I'd like to make sure I understood: you've updated SC to 3.2.6433.0 and then updated the SDK? I vaguely recall Robin warning against simply swapping SDKs, but I might be wrong.

For reference, I've updated to 3.2.6433.0 and the sensor analysis still doesn't work in RAW16. I have also tried selecting "force still mode" beforehand but SC crashed. I haven't tried RAW8.

[admin]

Hi,

thanks for the updates - I think that I am out of ideas for the moment, but will see if I can get hold of a camera to test with.

Robin

[gwsnyder]

Well, I got it to work in the 14 Bit mode. It does make a difference the order for installing updates as SharpCap has a specific SDK version for the QHY cameras that is installed when SharpCap is installed. So, my suggestion is to update the QHY SDK's first using their tool. Next, install/re-install SharpCap (latest version) which will in turn install the specific SharpCap SDK for QHY cameras in the SharpCap directory.

After this is done, run the Sensor Analysis tool in SharpCap. Note that after the e/ADU measurements, you may need to cover the sensor temporarily as SharpCap may run the exposure level all the way to the minimum which forces the the tool to fail. I cover the sensor until I see a consistent stepping up of the exposure time in SharpCap, then uncover.

[IZ1IVA]

So, my suggestion is to update the QHY SDK's first using their tool. Next, install/re-install SharpCap (latest version) which will in turn install the specific SharpCap SDK for QHY cameras in the SharpCap directory.

I'd think SC would overwrite the SDK...

Note that after the e/ADU measurements, you may need to cover the sensor temporarily as SharpCap may run the exposure level all the way to the minimum which forces the the tool to fail.

Thanks for the tip, I'll try to replicate your method.

[IZ1IVA]

thanks for the updates - I think that I am out of ideas for the moment, but will see if I can get hold of a camera to test with.

I'm willing to help. If you're not able to source one locally, I could ship mine to you.