Sensor Analysis on QHY165C

bwells
Posts: 18
Joined: Sat Sep 08, 2018 1:01 am

Re: Sensor Analysis on QHY165C

Post by bwells » Fri Oct 05, 2018 2:46 pm

I tried using the latest SharpCap 3.2.5781.0 (BETA) 9/28/2018 software.

I turn off all illumination and run the sensor analysis.

Gain goes to 100 and offset goes to 204. And I get the same message about light levels being too high. I have been shooting with an offset of 4 and gain of 180 which "works", but I am not sure how the noise levels are at those settings for the 2 minute exposure that i am shooting.

It seems like setting offset to 204 is a lot since in theory, offset 0 should put the lowest light level at zero on the histogram (?).

For reference, at a gain of 100, I need to set the offset to 65 to get the histogram near the center of the graph. And to get the histogram near the center with offset at 204, I must set the gain to 34.

What do you recommend I try next?

Thanks

bwells
Posts: 18
Joined: Sat Sep 08, 2018 1:01 am

Re: Sensor Analysis on QHY165C

Post by bwells » Fri Oct 05, 2018 3:49 pm

I just discovered if I select "Colour Space" of RAW8 instead of RAW16, the sensor analysis works. It still sets the initial offset to 204 and gain to 100 for SharpCap 3.2, but the sensor analysis was successful. I got a Mean ADU measurement graph that is close to a diagonal line from upper right to lower left.

I had similar results using SharpCap 3.1 from September 15th; that is, Raw8 sensor analysis ran without an error. But I got different results from Sensor Analysis between SharpCap 3.1 and SharpCap 3.2. Here are my results:

Report for SharpCap 3.1: Image

Report for SharpCap 3.2: Image


Thanks

bwells
Posts: 18
Joined: Sat Sep 08, 2018 1:01 am

Re: Sensor Analysis on QHY165C

Post by bwells » Fri Oct 05, 2018 4:45 pm

As soon as SharpCap 3.1 finished running with RAW8, I switched to RAW16 and re-ran the Sensor Analysis (I did not touch anything else in the GUI).

This time, it ran as expected up to the point where it asked me to cover the sensor for dark measurements. When I covered the sensor, the "Proceed" button would not enable because gain was at 255 and offset was at 204, so the light was too strong. I lowered the gain to 1 which enabled the "Proceed" button. The highest gain I could select and still enable the "Proceed" button was 40. Then I clicked the "Proceed" button and the analysis finished.

Here is the analysis results from Sharp 3.1 for RAW16, with the one caveat I mentioned above. Can you tell me what I should do now? Did changing the gain to 1 in order to get the "Proceed" button enabled cause the sensor analysis to be invalid? Or do you think this data looks correct to you and I can now use this data with the Histogram tool?

Image

I tried the "Start" button again on the analysis, to just re-run the analysis again. But this time, I lowered the gain to 40 to enable the "Proceed" button and these are the results I got which are pretty much the same as when I lowered the gain to 1 so the "Proceed" button would enable:

Image

For some reason, if I run SharpCap Sensor Analysis first with RAW8, and then with RAW16, it worked successfully.

Thanks
Last edited by bwells on Fri Oct 05, 2018 5:19 pm, edited 1 time in total.

User avatar
admin
Site Admin
Posts: 2088
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Sensor Analysis on QHY165C

Post by admin » Fri Oct 05, 2018 5:18 pm

Hi,

thanks for the update - interesting that 3.1 succeeds and 3.2 doesnt :(

Do you think you could dig out the log files for those sensor analysis runs (all of them - 3.1 and 3.2?) The saved log files will be in

Code: Select all

C:\Users\<your user name>\AppData\Local\SharpCap\logs
and you should be able to identify the correct files by their timestamp (the sensor analysis ones will be larger than average because a lot of info gets logged during the analysis).

I'd like to get 3.2 fixed up properly as you can see from the 3.1 results that almost all the gain is done by a value of about 100. In fact an awful lot of the gain seems to happen between 1 and 20.

I wouldn't say you will get good results from using those figures yet in smart histogram - that is because there are not enough points in the low gain region.

I have been discussing this problem with QHY - apparently this sensor has different gain ranges depending on whether it is in still or video mode and the gain values are not yet mapped from the raw values that the chip expects to dB, which explains why we are having problems.

cheers,

Robin

bwells
Posts: 18
Joined: Sat Sep 08, 2018 1:01 am

Re: Sensor Analysis on QHY165C

Post by bwells » Fri Oct 05, 2018 5:44 pm

Robin:

Here are the only two log files I see created today when I just ran these tests. The 14204 seems to be the 3.2 and the 13208 is 3.1. But I did 2 or even 3 runs and those must have appended to the same log file.

Thank you for helping to make this work!

Bryan
Attachments
Log_2018-10-05T10_31_52-14204.log
(138.42 KiB) Downloaded 12 times
Log_2018-10-05T11_26_40-13208.log
(285.19 KiB) Downloaded 11 times

bwells
Posts: 18
Joined: Sat Sep 08, 2018 1:01 am

Re: Sensor Analysis on QHY165C

Post by bwells » Tue Oct 16, 2018 1:27 am

I thought I would check back to see if any progress has been made with QHY and SharpCap. Thank you!

User avatar
admin
Site Admin
Posts: 2088
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Sensor Analysis on QHY165C

Post by admin » Tue Oct 16, 2018 6:22 pm

Sorry, this one slipped off my radar with preparations for the IAS show over the last week or so :(

Robin

User avatar
admin
Site Admin
Posts: 2088
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Sensor Analysis on QHY165C

Post by admin » Tue Oct 16, 2018 8:37 pm

Hi,

in 16 bit mode I can see that the camera has an offset range of 0..2047. SharpCap then tests values at 0, 0.1*max, 0.5*max, 0.9*max and max offset. For minimum gain, this works out fine, but for higher gains the higher offset levels (0.1*max = 204 or higher) all white out.

So, here is a question for you... At higher gains (ie 255), what is the biggest usable offset value - say one that gives a 50% histogram level for a black image. How does the offset needed to get a mid level (50%) histogram from a dark vary with gain? I need to try to work out if there is some logic to how the offset values work on this camera or if it is just weird!

cheers,

Robin

bwells
Posts: 18
Joined: Sat Sep 08, 2018 1:01 am

Re: Sensor Analysis on QHY165C

Post by bwells » Wed Oct 17, 2018 1:13 pm

I ran a few tests and see a gain of 255 and exposure of 1.0 ms, I need an offset of 30 to get a 50% histogram. An offset of 60 brings the histogram to near the top.
pic1.jpg
pic1.jpg (305.87 KiB) Viewed 229 times
pic1.jpg
pic1.jpg (305.87 KiB) Viewed 229 times
Attachments
pic2.jpg
pic2.jpg (404.92 KiB) Viewed 229 times

bwells
Posts: 18
Joined: Sat Sep 08, 2018 1:01 am

Re: Sensor Analysis on QHY165C

Post by bwells » Thu Nov 01, 2018 10:44 pm

I am home again for a week, so I thought I would see if any progress had been made to get SharpCap Sensor Analysis to support my QHY165C.

Thanks
Bryan

Post Reply

Who is online

Users browsing this forum: No registered users and 1 guest