GPCAM3 385C sensor analysis and high gain range

[Tibox]

Hello,

I'm learning and enjoying all features of SC 3.2 for a couple of weeks, that's a brilliant piece of software!
Today, I try to understand my camera (GPCAM3 385C) thanks to the Sensor Analysis tool.

I followed the procedure and got this table at the end:

sensor_analysis_Altair385C3_4.png (82.65 KiB) Viewed 4515 times

I understood the way SC calculates the gain value with HGC and LGC dual-mode of the Altair cameras. I would say the unity gain is around 590, (=2.95 HGC in AltairCapture). However, I don't understand why SC jumps from 1000 to 10000 in one step. I saw other sensor analysis outputs and the measures were more granular till the maximum value. Why does mine behave like this?

In comparison with the values reported by Altair on their website, except for the gain 100, the curve quite matches.

Altair_IMX385_Gain_Analysis-2.jpg (61.92 KiB) Viewed 4515 times

I just cannot compare very precisely as I miss some intermediate measures.

altair_vs_mine.png (44.42 KiB) Viewed 4515 times

Thank you for your feedback!

[admin]

Hi,

could you retry with build 3.2.6433 to see if that measures any more intermediate points between gain 1000 and gain 10,000? There was an issue to do with unintentional rounding of the logarithms of the gain values when trying to calculate the list of values to test that should now be resolved and I hope it will fix this issue.

Cheers, Robin

[Tibox]

Hello Robin,

Thanks for the quick answer! I downloaded the new build 6433 to update (I thought actually that the option "check for update" would notify me about this new build, but now I'll do a regular manual check on the website looking for new versions).

Unfortunately the scale fo gain testing is still the same:

2020-09-09 15_01_11-SharpCap Pro (v3.2.6391, 64 bit) - ALTAIR385C3 - C_____CAPTURES.png (91.54 KiB) Viewed 4474 times

Thank you.
Best regards,

Yorick

[Tibox]

EDIT: it seems that my version was not updated correctly despite the new download and install, sorry for this, I'll redo the tests!
I misdownloaded the 32 bits version, where I was using the 64bits version before, that explains the mistake.

[Tibox]

Hello Robin,

With the build 6433 it's all good!

2020-09-09 19_12_03-SharpCap Pro (v3.2.6433, 64 bit) - ALTAIR385C3 - C_____CAPTURES.png (99.65 KiB) Viewed 4469 times

I now have consistent values!

e/ADU is very similar to the values provided by Altair.
However, although Read Noise follows the same evolution, the values are not as good as it should... Could it be due to the test conditions, or is it only dependent on the quality of the sensor and camera?

2020-09-09 19_17_39-sensor_analysis.xlsx - Excel.png (49.2 KiB) Viewed 4469 times

Thanks a lot Robin for the support!

[admin]

Hi,

I just ran the test over my 385C here and I get very similar results to you – I think that the Altair ones may be erroneous as they also seem to have a maximum full well depth of only 10,000 electrons at minimum gain which disagrees with the values we are getting (and also those on the ZWO website for their version of the camera). My experience is that there is relatively little difference between different examples of the same sensor model – that is that the Sony quality control on the sensor is pretty good.

You will see random differences between the results if you run the analysis a few times – I would say that variations of up to 10% would be considered possible in some circumstances. Of course you could reduce these random variations by taking more measurements during the analysis procedure, but that would make the whole procedure take significantly longer...

Cheers, Robin

[admin]

And here is the missing screenshot of the results that I got

Capture.PNG (96.97 KiB) Viewed 4468 times

[Tibox]

Thanks again Robin, this is what I call a premium service! For sure after my first year offered by Altair, I'll renew the Pro license over and over gain!