Having trouble running the sensor analysis on my QHY268c

Post Reply
DiligentSkies
Posts: 52
Joined: Wed Dec 02, 2020 10:46 pm

Having trouble running the sensor analysis on my QHY268c

#1

Post by DiligentSkies »

Hello All,

I have the SharpCap Pro License in place. Version 3.2.6442.0
My OTA is a William Optics GT81 with the adjustable 0.8 reducer.
My filter drawer selections are an Optolong L-eNhance, Optolong S-II, Astronomik l-2 UV/IR and a visual blank drawer.

To the best of my knowledge I have this imaging train correctly back focused to its prime back focus length.

So I have been trying to run the SharpCap Sensor Analysis tool with no results.
Today was my third attempt. I took some screen shots.

This a screen capture of SharpCap with the Histogram tool enable on a 6 second exposure at zero gain.
There is not something quit right about the histogram. I would expect much higher values on the histogram.
Capture1.PNG
Capture1.PNG (239.36 KiB) Viewed 1710 times
This a screen capture of SharpCap with the Sensor Analysis tool enable.
First I am unable to select any portion of the image and the analysis reporting back, "That exposure time is getting quite long, etc... means the sensor analysis is unable to proceed."
Capture2.PNG
Capture2.PNG (268.87 KiB) Viewed 1710 times
Both where taken with no filters in my filter drawer just an empty filter drawer holder in place.
And, I was just capturing from a uniform lite blue sky.

The QHY268c being a newer camera from QHY has three readout modes.
I been trying to get the sensor analysis tool to preform its tests so I can optimize my exposure and gain settings.

However from just reading off the QHY spec charts for the QHY268c...

The sweet spot for the camera seems to me to be at a gain of 60 for the High Gain readout mode and a gain of 30 for the PhotoGRaphic DSO readout mode. That is judging from the readout noise, dynamic range and full well capacity specs of the QHY268c spec charts at these gains to make this determination.

Therefore the sensor analysis data is very important to me since I want to have the brain mode analysis to target in on the optimal exposure time for my subs at these gain settings and then calculate the total length of imaging time on a target(aka number of subs to stack) via the brain modes analysis of my seeing conditions (aka my light polluted urban sky.)

Sincerely,
Mark
User avatar
admin
Site Admin
Posts: 13177
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Having trouble running the sensor analysis on my QHY268c

#2

Post by admin »

Hi Mark,

I think the first problem is that the selection area is (for some reason) a tiny box at the top left of the frame. In your first screenshot you can see the little red arrow pointing to it in the top left corner of the white image. You need to make this selection area bigger, and in SharpCap 3.2 the way to do that is to zoom in on that corner of the image (use CTRL+mouse wheel with the mouse hovered over that top left corner of the image). Then you will finally see a small red box when you zoom in far enough - make this box bigger. Hopefully once you have a reasonable area selected (say 200 by 200 pixels) the histogram will return to normal and everything will proceed properly in the analysis.

cheers,

Robin

PS. In SharpCap 3.3, double clicking on the image when the selection is active will make the selection appear centered on the double click position, about half the size of the visible area of the image - much easier!
DiligentSkies
Posts: 52
Joined: Wed Dec 02, 2020 10:46 pm

Re: Having trouble running the sensor analysis on my QHY268c

#3

Post by DiligentSkies »

Hello Dr. Robin,

Your response was the solution to unlocking the Sensor Analysis within SharpCap 3.2 and unleashing the brain function of the smart histogram functionality.

I am very much pleased, and while this is my first outing with SharpCap Pro fully unleashed. I am most impressed.

To have a veteran enthusiast astronomer of nearly fifty years come up to me at recent gathering, and "Say, watching what you are doing is worth ruining my night vision and commenting that is the best imagine of Andromeda that he has ever scene for the size of scope(a 3 inch refractor) says much."

To put it into words, "As in a recent text messaging session with a fellow traveler..."

In other words, "The technology mumbo jumbo quickly becomes meaningless techno porn."
I have spent nearly six months reading about signal to noise ratios as related to ADU's of CMOS sensors as it relates to, "What gain setting to use for my astrophotography camera?"
And come to the healthy conclusion Mr. Glover is spot spot, "In shaking this technology tree to its core."

His dedication to improving SharpCap and leveraging it towards the evolving CMOS community as whole on the technology front is cutting edge."

In truth what Mr. Glover is saying, "The CMOS technology is now coming into its own to replacing CCD sensors, such to the point that a rigorous, scientific based analysis is now required to move the technology from the black arts to into a rightful, rigorous and scientific and constituent application of measurable outcomes."

I was further very impressed with the histogram stretching and colour balancing of SharpCap's live stacking routines. Such, that when I went to Deep Sky Stacker the next day on my captured data and did my own stacking and imagine histogram stretching the results was very much aesthetically inclined toward an indifference of the ascetic. The impression I have is, "The routes Dr. Glover is providing via live stacking and the focus on Sigma clipping is very much pushing post-processing into a thing of the past.

The sum total is I have is, "Dr. Glover is providing a means that breaks away from the past and ushering in CMOS sensor technology in its own right."

This is a very bold statement based on only one imagining session BUT..., "The internet is awash in home spun and pseudo understandings and implementations of the past of tried and true techniques that refuses to except the realities that all CMOS camera sensors ARE ANALOG DEVICES devices at heart." "Mr. Glover's efforts are scientifically based on providing optimized outcomes that can be systemically and consistently applied in my opinion."

My best appreciation of understanding is summed up by what Dr. Glover is saying, "Within a given CMOS camera there is a limits of trade offs on setting Exposure, Gain and Offsets values that only amplifies the background noise that either leads to optimizations or chasing your tail."

Last night I proved that to myself.

Thank you Dr. Glover.

Sincerely, Mark at DiligentSkies@outlook.com

P.S...

As a owner of the QHY268c. I have only the desire to provide feedback to that leads towards actionable feedback on your quest to move the SharpCap software platform forward.
Please, understand this as an invitation to reach out to me for real world data.
"The type of things, that only your very specific focus and your highly refined understandings about CMOS sensors is geared towards

"I will attempt provide to you the experimentation of whatever is required. Just spell out what you would require."
You can move the discussion to the email provided.

As a targeted focus, for the next two months I can provide a dedicated premiant observation platform of feedback on the greater Onion Complex in all it glories.
Not just its main observational features but its many nuisance outliners that lie outside the wider feedback around the main Orion Complex.

January/February is setting up to be the optimal time frame to doing this towards my location constraints.

As such, solely based on my observational limits of being able to polar align with zero views of Castral North, nor any any West or East views of from the Southern meridian, and that I am totally able to refine my mount's polar alignment to achieve reasonable polar alignment through Celestron All-star polar alignment to within 50 arc-seconds of true polar alignment in that PHD2 guiding can guide on with +/- within the four scale. ALL I CAN SAY is my DATA IS GOING TO BE VERY SOLID.

I am saying my guiding and understanding of the details of how that impacts subs exposure as to miniating is within the constraints of the norms of SharpCap's analysis of the night sky brightest that leads to directly analysis of optimal exposure through the BRAIN Histogram.
Basically, let see what set of actionable data I can provide, that refines SharpCap Brain Histogram analysis as to what EXPRURE, GAIN AND OFFSET PARAMETERS ARE the OPTIMUM.

To recap, I am proposing a real measure based on SharpCap's sensor analysis and brightness measurements as to the contributing factors of the given night sky brightness that leads to the best dynamic range of the SharpCap's measuremnts of CMOS sensor analysis. That says in the end there comes a point in time when increasing exposure lengths for a given GIAN only amplifies the background skies noise in the signal to noise ratio.

Basically what , Dr. Glover is saying per my understanding there is a trade off on this quest of signal to noise quest. Try to improve the signal in one requard leads to returns that work against you in another. There is optimization at work. And, Dr. Glover's quest IN SharpCap is to measure this CMOS factoid.

That said, here is my Sensor analysis output...
Mind you my QHY268c sensor has three modes of reading...
The first of which is the DEEP SKY Object settings and the second is a HIGH MODE reading, while the third output is described as being scientifically dedicated to base line luminance measurements of criteria of the a given star's spectrum.
AS such here is the read modes analysis for the frist two readout modes...

DSO the first of three read out modes...

Gain e/ADU Read Noise (e) Full Well (e) Relative Gain Rel. Gain (db) Dynamic Range (Stops)
0 0.852042979798767 7.92465611170674 55839.488724092 1 0 12.7826497803429
10 0.648703606202429 7.75204000025692 42513.4395360824 1.31345497643632 2.36830380631095 12.4210553489959
20 0.449159606023125 7.25880002825806 29436.1239403315 1.89697151830456 5.56121620621793 11.9855690215987
30 0.247313058983138 2.71923611511891 16207.9086335189 3.4452001172201 10.7442890664129 12.5412088980764
40 0.181123981403533 2.61472251052125 11870.1412452619 4.70419749608125 13.449710928274 12.1483916341053
50 0.115160772462463 2.53057623673941 7547.17638409998 7.39872581244185 17.3831386641525 11.5422553383324
60 0.0485394585069036 2.33837598312681 3181.08195270844 17.5536152649413 24.8873315076826 10.4097949077139
70 0.00943031422665178 1.90049267596995 618.025073157851 90.3514940563422 39.1187067693857 8.34514809353112
80 0.00501320186976759 1.80536075485407 328.545197737089 169.959838429221 44.6069261911912 7.50766089317049
90 0.00347140484783354 1.79604605240954 227.501988107619 245.446157146008 47.7991247382229 6.98491099995755
100 0.0025658355809503 1.72355363224846 168.154600633159 332.072322219181 50.4246535849333 6.60825824855208


The High AGAIN MODE READOUT MODE is...

Gain e/ADU Read Noise (e) Full Well (e) Relative Gain Rel. Gain (db) Dynamic Range (Stops)
0 0.513738913716527 3.64042991665466 33668.3934493263 1 0 13.174998425863
10 0.46385180759401 3.58998474053114 30398.992062481 1.10754966415089 0.887264194662071 13.04775815687
20 0.414082714331907 3.5552211340756 27137.3247664559 1.24066737377678 1.87310723196811 12.8980515865298
30 0.364020203909631 3.49856668618806 23856.4280834216 1.4112923079513 2.9923394895817 12.7353264393294
40 0.313943034551805 3.44908201061584 20574.5707123871 1.63640806508084 4.27783222527121 12.5423622758183
50 0.264037309021846 3.42732195079693 17303.9490840557 1.94570576264289 5.78154330320778 12.3017319848989
60 0.20285700665775 1.65749504368502 13294.4367883223 2.53251747218808 8.07104900600625 12.9695304827049
70 0.153581846194341 1.61197328745397 10065.1398721924 3.34504973371948 10.4880515835977 12.6082517639588
80 0.104433213503609 1.57328189381249 6844.13508017253 4.91930580780962 13.8380764248239 12.0868753289688
90 0.0546637228851244 1.48893588064827 3582.44174299951 9.39816914402532 19.4608651398842 11.2324459012864
100 0.00352255758585827 1.10675379372518 230.854333946808 145.842587720636 43.2776872301659 7.70450468816297

Yes there are alignment TAB erros but not to distract from the observational fact interpolating between these measurements is the outliner observation of fact that increasing the over all gain leads to a reduction of dynamic range in ADUs for a given exposure sum total provide it is understood that there is a component that reaches an optimum of trade off. AS Dr glover would suggest, "There comes a point in time, where increasing the amount of sub exposures leads to diminished returns. Put in another way," There comes a point in time where increasing the gain leads to a drastic reduction in dynamic range as the back ground noise columinates as you try to increase the signal to noise ratio by increasing the exposure lengths.
That is the trade off. And the optimization problem as Dr. Glover's understand it in his constraint algorithms of the histogram brain function.

I am very happy with Dr. Glover's model on CMOS reality.

The Universe once gain imposes there no are, "No Free Lunches."
User avatar
admin
Site Admin
Posts: 13177
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Having trouble running the sensor analysis on my QHY268c

#4

Post by admin »

Hi,

Glad to hear that you have everything working to your satisfaction now and thank you for your kind comments.

Thanks for also pointing out that some QHY cameras have multiple readout modes and that the sensor analysis routine should take account of the chosen readout mode as well as the bit depth when deciding which data to use. I will make a note of that as an improvement to add to SharpCap 3.3.

Cheers, Robin
DiligentSkies
Posts: 52
Joined: Wed Dec 02, 2020 10:46 pm

Re: Having trouble running the sensor analysis on my QHY268c

#5

Post by DiligentSkies »

Dr. Glover,

On the subject of multiple readout modes and SharpCap's sensor analysis algorithm for the QHY268c camera...

In practice, SharpCap3.2, "Does recognize the three readout modes of the QHY268c in the Capture Format and Area control panel (A Good thing)."

So by setting the bit depth and the readout mode in this control panel. The sensor analysis runs for, "Whatever is set and that this is the sensor analysis data that is going to be used by the Brain Smart Histogram routine."

Meaning, if the sensor analysis was run with the PhotoGraphic DSO readout mode(what QHY calls mode# 0) and I want to switch to imaging with the High Gain mode(what QHY calls mode# 1)(A very interesting readout mode that beguiles the fact that readout mode#0 is called Photographic DSO)

So if I want to use the Histogram Brain analysis for that readout mode. Then, "I should just run the sensor analysis for this readout mode and that will be the one the Histogram Brain will start to use."


Additionally, by knowing how and where SharpCap stores the Sensor Analysis Data. Then maybe I can just swap out the Sensor Analysis Data Sets and fore go having to run the sensor analysis each and every time I want to experiment around with these two readout modes.

For example if SharpCap stores the sensor analysis data in the Registry. Then just exporting these Registry Keys will be the quick and dirty work around until SharpCap 3.3.

I just did a Registry search and did not find anything that stood out, nor did I find any file recently created in the install path of SharpCap 3.2.
This is leading me to believe the Sensor Analysis data is NOT persistent between imagining sessions.
Is that the case?

Sincerely,
Mark
User avatar
admin
Site Admin
Posts: 13177
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Having trouble running the sensor analysis on my QHY268c

#6

Post by admin »

Hi Mark,

the sensor analysis is stored in files which you can find in the folder

Code: Select all

C:\Users\<your user name>\AppData\Roaming\SharpCap\SensorCharacteristics
As you suggested, you can switch these files about if necessary when moving between modes.

I'm currently working on making the sensor analysis and smart histogram cope with cameras with different readout modes – this will be in SharpCap 3.3 and will allow you to have separate datasets for the different readout modes the camera and SharpCap will automatically choose the correct one.

Cheers, Robin
DiligentSkies
Posts: 52
Joined: Wed Dec 02, 2020 10:46 pm

Re: Having trouble running the sensor analysis on my QHY268c

#7

Post by DiligentSkies »

Hi Robin,

Once again thank you.

Just wrote a batch file to switch about the files.
Do I need to restart SharpCap to have the switch take effect, or can I simply switch them in and out...
And SharpCap's Brain Histogram then goes about its analysis by reading whatever data is present in the QHY268c~RAW.json file.
The file looks to be nothing more than a structured array.
So my guess is, "It is the later."

Cheers,
Mark
User avatar
admin
Site Admin
Posts: 13177
Joined: Sat Feb 11, 2017 3:52 pm
Location: Vale of the White Horse, UK
Contact:

Re: Having trouble running the sensor analysis on my QHY268c

#8

Post by admin »

Hi Mark,

just switching the file with SharpCap running will not be enough. Closing and re-opening the camera in SharpCap (or closing and re-opening SharpCap) will do the job.

Cheers, Robin
Post Reply