Strange results with Sensor Analysis

Anything that doesn't fit into any of the other forums
Post Reply
bluesilver
Posts: 80
Joined: Mon Oct 05, 2020 12:51 am

Strange results with Sensor Analysis

#1

Post by bluesilver »

Hi, I have a asi2600mc and running SharpCap Pro, thought i would give this sensor analysis a try just to see what the results look like.
I have got the results back and they look a bit off.
The main thing was it come back with measured bit depth of 14 bit instead of 16 bit.
The other thing for me at least, it took around 1 1/2 hours to complete.
The laptop that i am using for some reason dropped the frame rates right down to 0.1, very slow.

Below is a screen shot of the results as i tried to copy to the clip board as the icon said, but that didn't work for me. I played around trying to up the USB rate to 100 from 40 and made no difference, also tried turning on and off the USB high speed and no difference.
You can see how many frames were dropped, but the counter resets after 2 hours, ( i had to wander off it was taking so long )

Any advice on what may be going on here, especially the measured bit depth of 14 bit instead of 16 bit.
Attachments
Screenshot.jpg
Screenshot.jpg (197.95 KiB) Viewed 1245 times
bluesilver
Posts: 80
Joined: Mon Oct 05, 2020 12:51 am

Re: Strange results with Sensor Analysis

#2

Post by bluesilver »

I should of also added:
For connection, I ran a 12V supply to the camera and cooled it to -10
I ran a USB cable directly from the camera straight to the Laptop.
I even tried with everything setup ready for imaging and light box on dew shield, result in speed was no different.

To speed up the slow 0.1 frame rate, I am thinking maybe somehow have SharpCap use the native ZWO drivers rather than Ascom? not sure how to do that part yet.

Or, Is this all fine and run as it is even with it reporting 14 bit, as from what i have read, this result from the sensor analysis is what SharpCap now uses.
Or somehow get back to the default settings?
Hope this helps with some extra information on what may be going on here.

I have been using APT for my imaging and that works with the camera fine,
Do i need to go install the latest V.21 Camera driver and V1.0 Camera Firmware Upgrade Tool from ZWOs site,
Plus the lasted download of SharpCap?
Last edited by bluesilver on Wed Apr 12, 2023 8:23 am, edited 1 time in total.
User avatar
turfpit
Posts: 1783
Joined: Mon Feb 13, 2017 8:13 pm
Location: UK
Contact:

Re: Strange results with Sensor Analysis

#3

Post by turfpit »

Or somehow get back to the default settings?
Hold down the <Control> key when connecting the camera. This will return the default camera settings. Then run the tests again.

Dave
bluesilver
Posts: 80
Joined: Mon Oct 05, 2020 12:51 am

Re: Strange results with Sensor Analysis

#4

Post by bluesilver »

Thanks for that tip, Never knew about that one.
The other thing i have just thought about was the option called Colour Space
I currently have mine set at RAW 16 ( my thinking was that the camera is a 16 bit colour camera )
Is this correct or should i be setting it at something else?
Won't get to try any of this until tomorrow though
User avatar
admin
Site Admin
Posts: 13362
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Strange results with Sensor Analysis

#5

Post by admin »

Hi,

I think the first thing I would suggest would be to update to a more recent version of SharpCap - the 4.0.9268 version you are using is about 8 months old now, and there have been lots of small bug fixes, updates from manufacturers like ZWO etc added since then. It's perfectly possible that the weird 14 bit result will vanish when you update. There are other oddities in the analysis, but my guess would be that they all come from the wrong bit depth measurement, so we shouldn't worry about those right now.

As far as I can see you are using the correct version of the camera driver (direct via SharpCap, not ASCOM), so nothing to worry about there.

The image on screen looks very dark for a just finished analysis run... I wonder if the light levels were very low, leading to long exposures (normally the analysis should take a few minutes). If you see the exposure length go beyond 2s at any point then that is a sign that the light level is too low.

If the light levels are fine, but low frame rate is slowing you down, set the camera into ROI mode - say 1600x1200 to improve the frame rate.

cheers,

Robin
bluesilver
Posts: 80
Joined: Mon Oct 05, 2020 12:51 am

Re: Strange results with Sensor Analysis

#6

Post by bluesilver »

Appreciate the advice,
So, I have just downloaded the latest version Of SharpCap Pro,
Re ran the sensor analysis and it went trough very quickly this time, around 10 minutes was the longest time, ( I ran two, one with the capture area at 1280 x 1024 and the other at the default 4680 x 3132
You can see both results below and they both came back saying 14 bit, not 16 bit.

I went to the ZWO down load site, (it appears to be a different download site from last time i was there )
Tried to download the V3.21 Camera driver, This proved to be no good as it kept on getting blocked both by the anti virus software and also by google chrome
The same deal was with the V1.0 camera firmware upgrade tool, this one did download but will not install.

Just looking for information on what may be going on with the sensor analysis still saying 14 bit and not 16 bit.
I did run this sensor analysis about a year or a bit more ago and the result on that one was 16 bit.

I am now guessing if i try to run the brain function to test darkness for optimal gain and exposure times, i will be getting an incorrect result due to the sensor analysis saying the camera is 14 bit?
sharpcap2.jpg
sharpcap2.jpg (192.8 KiB) Viewed 1202 times
Screenshot sharpcap 1.jpg
Screenshot sharpcap 1.jpg (195.09 KiB) Viewed 1202 times
And just for a bit of reference, I found the old link to a sensor analysis i did back a year ago
viewtopic.php?t=5584
bluesilver
Posts: 80
Joined: Mon Oct 05, 2020 12:51 am

Re: Strange results with Sensor Analysis

#7

Post by bluesilver »

I have just also did a test going back to SharpCap 3.2, the results were worst there, they came back as reading Bit12
I tried going back to the early release of SharpCap 4.0 and did the test, pretty much the same results as posted above at Bit 14

Starting to wonder if i have bad camera?
User avatar
admin
Site Admin
Posts: 13362
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Strange results with Sensor Analysis

#8

Post by admin »

Hi,

the bit depth is measured by capturing frames and looking to see if there is a pattern in the pixel values - so, for instance, if all the pixel values are a multiple of 4 then you have a 14 bit sensor that is being scaled up to a 16 bit range by multiplying everything by 4.

If you are regularly getting 14 bit results for the IMX571 then something is adjusting the pixel values before they get to SharpCap - possibly a camera setting from the controls on the right (those adjust ZWO SDK settings). Make sure the colour balance controls are set to default (50/50) and I would also suggest turning *off* high speed mode (although I thought that was ignored except in 8 bit). Of course an SDK bug may be possible too (or a SharpCap bug I suppose), so no guarantee that the settings will fix it.

I just checked the analysis data I have on file for the 2600MC, and that did pick up the 16 bit sensor, so it is looking like something specific to your config.

cheers,

Robin
bluesilver
Posts: 80
Joined: Mon Oct 05, 2020 12:51 am

Re: Strange results with Sensor Analysis

#9

Post by bluesilver »

Ok, Just ran it again and got the correct results this time, it reported back as 16 Bit.
All i did was as you mentioned, just made sure those colour balances were all set at 50 and not on Auto
Changed the high speed to off
I ran it with the cooler on this time at -10c as that is what i image at.
It did take a bit longer to go through this time, I am guessing due to the High Speed turned off.

There was one control that I was not 100% sure on, and that was the Window Heater Power,
I have left it off, but am i correct in saying this is the dew heater built into the asi2600mc, or is it about something else?
sharpcap 16 bit.jpg
sharpcap 16 bit.jpg (206.52 KiB) Viewed 1185 times
User avatar
admin
Site Admin
Posts: 13362
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Strange results with Sensor Analysis

#10

Post by admin »

Hi,

that looks better (although the display stretch you have on is whiting out the image, which puts you at a disadvantage when choosing a good area to do the analysis on). A lower frame rate can slow things down, but another culprit is lower light levels which lead to longer exposures being used. If it is the frame rate, you can do the analysis with the camera set on ROI (ie 800x600).

The window heater is indeed the dew heater - it will not make any difference to the sensor results (actually the cooling doesn't really have any significant effect either).

cheers,

Robin
Post Reply