Jump to content

  •  

CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.

Photo

ASI1600MMC Beta test

This topic has been archived. This means that you cannot reply to this topic.
1767 replies to this topic

#301 tolgagumus

tolgagumus

    Vendor

  • *****
  • Vendors
  • topic starter
  • Posts: 3,687
  • Joined: 21 Jul 2014

Posted 09 May 2016 - 11:43 PM

 

 

If you use the XAGYL filter wheel, and the stock t thread adapters it comes with along with the camera, you produce 22 or so millimeters of distance between the filter and the CMOS sensor using the 6mm spec for the camera and estimating the filter wheel dimensions. If you use the 1mm adapter from Precise Parts to marry the camera to the filter wheel instead of the one that comes with the filter wheel package, you can save about 5mm of back focus and get the chip closer to the face of the filter wheel to about 17 mm. I am thinking that if you are using stock nautilus filer wheel adapters along with the camera you are going to be even futher than 22 mm from the chip to the fiter wheel, and if you are not seeing vignetting with your setup, then 22 mm should be okay. I want to avoid haveing to purchase the Precisoin Parts adapter.

That's what I am saying. You don't need any adapters. The threads on the FW is T Thread. Adapters are from T thread to xyz. Just add 6 mm to whatever the FW's face to filter distance

 

Now I get it. The 2" adapter on the camera comes off and it has male T thread right on the face which will screw straight to the FW

 

https://astronomy-im...ds/P1000832.jpg


 

#302 Phil Hosey

Phil Hosey

    Vanguard

  • *****
  • Posts: 2,311
  • Joined: 29 Oct 2008

Posted 10 May 2016 - 05:31 AM

Here is some of the data 

 

https://www.dropbox....EICDuRK6Sa?dl=0

 

Tolga,

Thanks for uploading the data.  I've been playing with it a bit this morning by running through my normal manual calibration steps in PI, but without knowing exactly how the master dark and master bias were created I can't be sure that my process is 100% accurate.  I calibrated each light frame with the master bias, master dark and master flat.  I assumed that the flat had already been calibrated and I assumed that the master dark had not already been calibrated with bias.  One question I have is, were these images taken with the firmware upgrade that reduces the amp glow?  The master dark looks pretty bad to me and I think this may be contributing to the calibration issues.  I don't know if that's all amp glow or if you have some kind or light leak going on.

Attached Thumbnails

  • integration_with_dark.jpg
  • MasterDark_ISO0_60s.jpg

 

#303 buckeyestargazer

buckeyestargazer

    Vendor - Buckeyestargazer.net

  • *****
  • Vendors
  • Posts: 5,139
  • Joined: 12 Jan 2008

Posted 10 May 2016 - 07:59 AM

I'm getting essentially the same results as Phil.  What puzzles me is that the master dark shows severe amp glow in the right two corners, but this glow is not present in the light subs as far as I can tell.  Also, the master flat doesn't appear to match well with the lights (at least those lights in the dropbox folder).  Highly stretching the flat doesn't reveal the dust motes that I would expect base on your previously posted stacked light image.  

 

Please don't take any of that as criticism, we're just all trying to figure things out.  Those flat/dark issues are really scondary to the lights that I'm seeing.  Those one minute subs are pretty impressive.


 

#304 Phil Hosey

Phil Hosey

    Vanguard

  • *****
  • Posts: 2,311
  • Joined: 29 Oct 2008

Posted 10 May 2016 - 08:06 AM

I'm getting essentially the same results as Phil.  What puzzles me is that the master dark shows severe amp glow in the right two corners, but this glow is not present in the light subs as far as I can tell.  Also, the master flat doesn't appear to match well with the lights (at least those lights in the dropbox folder).  Highly stretching the flat doesn't reveal the dust motes that I would expect base on your previously posted stacked light image.  

 

Please don't take any of that as criticism, we're just all trying to figure things out.  Those flat/dark issues are really scondary to the lights that I'm seeing.  Those one minute subs are pretty impressive.

Joel,

I did one version of the calibration with bias and flat only, no darks and the amp glow was present in the integrated image.  I'll post it this afternoon when I get home, or you could re-calibrate without the darks and see for yourself.  I'm worried the amp glow is going to be a problem if it remains at the level seen here.


 

#305 fetoma

fetoma

    Gemini

  • *****
  • Posts: 3,051
  • Joined: 26 Sep 2006

Posted 10 May 2016 - 08:07 AM

The excitement is building and we are all about to explode!!! :bomb:


 

#306 josh smith

josh smith

    Fly Me to the Moon

  • *****
  • Posts: 5,366
  • Joined: 13 Oct 2013

Posted 10 May 2016 - 08:19 AM

I'm getting essentially the same results as Phil. What puzzles me is that the master dark shows severe amp glow in the right two corners, but this glow is not present in the light subs as far as I can tell. Also, the master flat doesn't appear to match well with the lights (at least those lights in the dropbox folder). Highly stretching the flat doesn't reveal the dust motes that I would expect base on your previously posted stacked light image.

Please don't take any of that as criticism, we're just all trying to figure things out. Those flat/dark issues are really scondary to the lights that I'm seeing. Those one minute subs are pretty impressive.


You bring up an important point! This is a beta test camera. What's beta testing for if not figuring out the bugs and getting them sorted out before the camera is in everyone else's hands. No one is being critical of Tolga or the camera. In fact none of the previous discussion has been critical of the camera save one or two select people. It's all in the spirit of being critical thinkers and asking relevant questions. There is no need for the divisiveness and animosity this camera seems to be causing. Everyone is excited about it and there is no reason not to ask real questions about it and make it the best it can be. It's an interesting phenomenon to see how much people feel they need to defend it.

On topic, I also don't see the dust mote in the flat. Tolga, were the lights your provided calibrated or just raw?
 

#307 buckeyestargazer

buckeyestargazer

    Vendor - Buckeyestargazer.net

  • *****
  • Vendors
  • Posts: 5,139
  • Joined: 12 Jan 2008

Posted 10 May 2016 - 08:29 AM

 

I did one version of the calibration with bias and flat only, no darks and the amp glow was present in the integrated image.  I'll post it this afternoon when I get home, or you could re-calibrate without the darks and see for yourself.  I'm worried the amp glow is going to be a problem if it remains at the level seen here.

 

Ok that makes sense.  So the puzzle is why the amp glow didn't calibrate out very well.

 

I've dealt with amp glow before, with my current camera (SX694 Trius).  There was some amp glow at 6 points around the edge and thankfully there was an easy hardware fix that SX instructed me to do from home that eliminated the amp glow.  Before I knew what amp glow was I tried calibrating and it was rather difficult, and the glow wasn't nearly as pronounced as in this case.  I share Phil's concern about the amp glow, but I am also confident that there should be hardware and software fixes, and processing fixes, that can eliminate it.


 

#308 Thirteen

Thirteen

    Gemini

  • *****
  • Posts: 3,158
  • Joined: 12 Jul 2013

Posted 10 May 2016 - 08:40 AM

Without looking at the raw data, and coming from someone that deals with amp glow regularly, I think this glow looks rather mild considering a one minute exposure at unity gain, but from the calibration on the image it would seem that there is a disparity in the darks and lights. Is it possible that they were at a different gain or something?

Also, probably my biggest concern is that dark optimization DOES NOT work properly with amp glow gradients like this in my experience. Please everyone be sure that scaling is not occurring in either DSS or PI.

Edited by Thirteen, 10 May 2016 - 08:41 AM.

 

#309 Phil Hosey

Phil Hosey

    Vanguard

  • *****
  • Posts: 2,311
  • Joined: 29 Oct 2008

Posted 10 May 2016 - 08:47 AM

Without looking at the raw data, and coming from someone that deals with amp glow regularly, I think this glow looks rather mild considering a one minute exposure at unity gain, but from the calibration on the image it would seem that there is a disparity in the darks and lights. Is it possible that they were at a different gain or something?

Also, probably my biggest concern is that dark optimization DOES NOT work properly with amp glow gradients like this in my experience. Please everyone be sure that scaling is not occurring in either DSS or PI.

 

I made sure that I did not use dark optimization in the images I posted.  I did however try it and the results were obviously bad.


 

#310 Phil Hosey

Phil Hosey

    Vanguard

  • *****
  • Posts: 2,311
  • Joined: 29 Oct 2008

Posted 10 May 2016 - 08:48 AM

I think the missing piece of info for me is whether or not the flats where calibrated with optimized dark or not.  Knowing exactly how the master flat was made would help in knowing how to calibrate the lights.


 

#311 bigeastro

bigeastro

    Apollo

  • *****
  • Posts: 1,340
  • Joined: 20 Feb 2015

Posted 10 May 2016 - 09:10 AM

 

I'm getting essentially the same results as Phil. What puzzles me is that the master dark shows severe amp glow in the right two corners, but this glow is not present in the light subs as far as I can tell. Also, the master flat doesn't appear to match well with the lights (at least those lights in the dropbox folder). Highly stretching the flat doesn't reveal the dust motes that I would expect base on your previously posted stacked light image.

Please don't take any of that as criticism, we're just all trying to figure things out. Those flat/dark issues are really scondary to the lights that I'm seeing. Those one minute subs are pretty impressive.


You bring up an important point! This is a beta test camera. What's beta testing for if not figuring out the bugs and getting them sorted out before the camera is in everyone else's hands. No one is being critical of Tolga or the camera. In fact none of the previous discussion has been critical of the camera save one or two select people. It's all in the spirit of being critical thinkers and asking relevant questions. There is no need for the divisiveness and animosity this camera seems to be causing. Everyone is excited about it and there is no reason not to ask real questions about it and make it the best it can be. It's an interesting phenomenon to see how much people feel they need to defend it.

On topic, I also don't see the dust mote in the flat. Tolga, were the lights your provided calibrated or just raw?

 

Beta Smeta,

 

The production units are being shipped unless of course Sam has taken my advice and not ship the bananas until they are just about ripe.....  Green bananas will give you diarrhea unless they are cooked.  Hopefully we can cook out the amp glow.   Whatever we are seeing at this point in time is whatever we are going to get in a few days.


 

#312 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 24,324
  • Joined: 10 Jan 2014

Posted 10 May 2016 - 09:13 AM

Based on the levels in each of the masters vs. the lights, I think the Master Dark was already calibrated. The dark has a mean level (according to the Statistics tool) of 0.00063, while the bias has a mean level of 0.00429. The light has a mean level of 0.04283. So, if the dark was calibrated with the bias, I believe that would have resulted in an improper calibration. 

 

I tried a simple manual calibration using PixelMath. I renamed the Master Dark to 'dark', Master Bias to 'bias', Master Flat to 'flat' and one of the lights to 'light'. I used the following formula:

((light + mean(bias) - bias) - dark) / flat

That produced the following image:

 

Fully Calibrated.jpg

 

I then applied the following formula:

light / flat

Which produced this image:

 

Flat Calibrated.jpg

 

I then subtracted the first image from the second, which produced the difference:

 

Difference Map.jpg

 

It seems clear that the glows were indeed removed by the full calibration, given how they show up in the difference map. Whatever remains would have to be a gradient of some other kind, and it looks like your standard light pollution gradient to me. 

 

 


Edited by Jon Rista, 10 May 2016 - 09:14 AM.

 

#313 Phil Hosey

Phil Hosey

    Vanguard

  • *****
  • Posts: 2,311
  • Joined: 29 Oct 2008

Posted 10 May 2016 - 09:14 AM

 

 

I'm getting essentially the same results as Phil. What puzzles me is that the master dark shows severe amp glow in the right two corners, but this glow is not present in the light subs as far as I can tell. Also, the master flat doesn't appear to match well with the lights (at least those lights in the dropbox folder). Highly stretching the flat doesn't reveal the dust motes that I would expect base on your previously posted stacked light image.

Please don't take any of that as criticism, we're just all trying to figure things out. Those flat/dark issues are really scondary to the lights that I'm seeing. Those one minute subs are pretty impressive.


You bring up an important point! This is a beta test camera. What's beta testing for if not figuring out the bugs and getting them sorted out before the camera is in everyone else's hands. No one is being critical of Tolga or the camera. In fact none of the previous discussion has been critical of the camera save one or two select people. It's all in the spirit of being critical thinkers and asking relevant questions. There is no need for the divisiveness and animosity this camera seems to be causing. Everyone is excited about it and there is no reason not to ask real questions about it and make it the best it can be. It's an interesting phenomenon to see how much people feel they need to defend it.

On topic, I also don't see the dust mote in the flat. Tolga, were the lights your provided calibrated or just raw?

 

Beta Smeta,

 

The production units are being shipped unless of course Sam has taken my advice and not ship the bananas until they are just about ripe.....  Green bananas will give you diarrhea unless they are cooked.  Hopefully we can cook out the amp glow.   Whatever we are seeing at this point in time is whatever we are going to get in a few days.

 

 

I remember seeing someone that was beta testing reported that they had a fix that reduced the amp glow.  I can't remember if it was Tolga or someone else.  


 

#314 Phil Hosey

Phil Hosey

    Vanguard

  • *****
  • Posts: 2,311
  • Joined: 29 Oct 2008

Posted 10 May 2016 - 09:18 AM

Based on the levels in each of the masters vs. the lights, I think the Master Dark was already calibrated. The dark has a mean level (according to the Statistics tool) of 0.00063, while the bias has a mean level of 0.00429. The light has a mean level of 0.04283. So, if the dark was calibrated with the bias, I believe that would have resulted in an improper calibration. 

 

 

 

Jon, looks like you got it.  I did my processing based on the assumption that the master dark had not had the bias removed. 


 

#315 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 24,324
  • Joined: 10 Jan 2014

Posted 10 May 2016 - 09:18 AM

Tolga reported they had removed the glow on the left edge, which it appears they did comparing the current dark with prior darks. I have not yet heard that they fixed all of the glows...and the two on the right still seem to be there. 


 

#316 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 24,324
  • Joined: 10 Jan 2014

Posted 10 May 2016 - 09:20 AM

 

Based on the levels in each of the masters vs. the lights, I think the Master Dark was already calibrated. The dark has a mean level (according to the Statistics tool) of 0.00063, while the bias has a mean level of 0.00429. The light has a mean level of 0.04283. So, if the dark was calibrated with the bias, I believe that would have resulted in an improper calibration. 

 

 

 

Jon, looks like you got it.  I did my processing based on the assumption that the master dark had not had the bias removed. 

 

 

Yeah. I think that would have basically removed a lot of the dark level, but possibly left a remnant of the glows. That would have resulted in under-correction of the glows. I like to use PixelMath to test calibrations when I'm working with data from a camera I haven't used before. ;) Gives you a lot of insight. :p


 

#317 bigeastro

bigeastro

    Apollo

  • *****
  • Posts: 1,340
  • Joined: 20 Feb 2015

Posted 10 May 2016 - 09:20 AM

Great detective work Jon.  It appears you are also a good cooker.  Tolga indicated in prior posts that a new driver release had reduced the amount of amp glow.


 

#318 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 24,324
  • Joined: 10 Jan 2014

Posted 10 May 2016 - 09:22 AM

I find this interesting. The glow that rings around the left side, top and bottom of the difference map here:

 

Difference Map.jpg

 

That is actually from the bias, from what I can tell. It's like a glow, but I don't think it is thermal in nature. 


 

#319 josh smith

josh smith

    Fly Me to the Moon

  • *****
  • Posts: 5,366
  • Joined: 13 Oct 2013

Posted 10 May 2016 - 09:26 AM

I think what Jon produced looks much better and almost seems to be truly calibrated.  There still is the very large scale circular ring and maybe that isn't from the flat.


 

#320 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 24,324
  • Joined: 10 Jan 2014

Posted 10 May 2016 - 09:30 AM

I agree with Josh, there does appear to be some uncorrected field structure in the image. Here is a more heavily stretched version of the calibrated image:

 

Circular Field Shape.jpg

 

I cannot say what that is caused by. I have experienced a myriad of field shape issues with my own gear over the years. Tracking down the source of them can be very difficult. I've had issues caused by LP filters, light leaks, internal diffuse reflections, bad calibration, improperly created flats, etc. So I cannot assume that the issue is due to the camera or sensor at this point. My best guess would be it is due to either some very faint diffuse reflections or improperly created flats...I don't see any evidence yet to indicate it is a camera or sensor issue. 

 

One thing that I can say...it appears that the filter is not totally centered on the sensor. The circular structures apparent above seem to be slightly offcentered towards the upper right, and the lower left corner is also more vignetted. 


Edited by Jon Rista, 10 May 2016 - 09:32 AM.

 

#321 bigeastro

bigeastro

    Apollo

  • *****
  • Posts: 1,340
  • Joined: 20 Feb 2015

Posted 10 May 2016 - 12:00 PM

Trumpets Please....

 

 

Beta is over.  From ASI Yahoo Groups.......... :waytogo:

 

"I got my tracking number for my ASI1600MC-Cooled. It is scheduled to arrive tomorrow.

I plan to use ASI1600MC-Cooled on my C11 EdgeHD with Hyperstar and on an imaging f/4

10 inch carbon fiber Newtonian."

 

Unfortunately it is not me :bawling: .  Someone out there will get an alpha unit within a day or two.  Tolga's unit is now obsolete.  What do you do about this thread?


 

#322 FirstC8

FirstC8

    Apollo

  • -----
  • Posts: 1,118
  • Joined: 10 Mar 2016

Posted 10 May 2016 - 12:04 PM

Jon, if it is indeed sensor issue, would this be an issue more associated with CMOS, or have we noticed similar patterns on CCD chips?

One thing I noticed from my own project and use a 1/3 CMOS chip for long exposures, my 30-min subs and darks could not remove certain curved bands along the edge of the amp glow along the left side of the frame, yet hour-long subs do not have such issue, even though closely examining those 30min or 60min lights and darks did not reveal any unusual phenomenon.

How can this be explained?

There were speculations, moonlight not blocked by dew shield, power supply/voltage issue, lack of flats...each mitigation effort did not solve the banding issue.

I decided to give up just take hour long subs.

At least in the case of the 1600, those different sources of glows were largely removed by darks and flats, as long as the images do not need to be overly stretched.

Edited by FirstC8, 10 May 2016 - 12:13 PM.

 

#323 josh smith

josh smith

    Fly Me to the Moon

  • *****
  • Posts: 5,366
  • Joined: 13 Oct 2013

Posted 10 May 2016 - 12:09 PM

Jon, if it is indeed sensor issue, would this be an issue more associated with CMOS, or have we noticed similar patterns on CCD chips?

One thing I noticed from my own project and use a 1/3 CMOS chip for long exposures, my 30-min subs and darks could not remove certain curved bands along the edge of the amp glow along the left side of the frame, yet hour-long subs do not have such issue, even though closely examining those 30min or 60min lights and darks did not reveal any unusual phenomenon.

How can this be explained?

There were speculations, moonlight not blocked by dew shield, power supply/voltage issue, lack of flats...each mitigation effort did not solve the banding issue.

I decided to give up just take hour long subs.

 

It is possible with such long subs you are swamping the different noises overwhelmingly so that they are unrecognizable.  

 

I think it is very careful we don't say this is a camera issue without a lot further investigation.  First of all, it is very mild.  Second of all it is one set of data that Tolga was trying to get out fast.  It needs to be investigated and explained, but nothing that can be concluded yet.


 

#324 tolgagumus

tolgagumus

    Vendor

  • *****
  • Vendors
  • topic starter
  • Posts: 3,687
  • Joined: 21 Jul 2014

Posted 10 May 2016 - 12:16 PM

Well I hope this thread helped some of you make a decision. I am interested in more people getting into astrophotography who can otherwise may not be able to due to cost. That's why I chose to use it with an AVX instead of bringing my Paramount. I think this camera coupled with a WO Star 71, AVX or Sirius class mount will make a very nice beginners setup for a very reasonable price.
I don't know about what the lucky imaging is all about. I don't know what Emil does, how he does what he does. He does not publish his methods. I still think the traditional processing will be superior. This is my opinion.
 

#325 FirstC8

FirstC8

    Apollo

  • -----
  • Posts: 1,118
  • Joined: 10 Mar 2016

Posted 10 May 2016 - 12:18 PM


Jon, if it is indeed sensor issue, would this be an issue more associated with CMOS, or have we noticed similar patterns on CCD chips?

One thing I noticed from my own project and use a 1/3 CMOS chip for long exposures, my 30-min subs and darks could not remove certain curved bands along the edge of the amp glow along the left side of the frame, yet hour-long subs do not have such issue, even though closely examining those 30min or 60min lights and darks did not reveal any unusual phenomenon.

How can this be explained?

There were speculations, moonlight not blocked by dew shield, power supply/voltage issue, lack of flats...each mitigation effort did not solve the banding issue.

I decided to give up just take hour long subs.


It is possible with such long subs you are swamping the different noises overwhelmingly so that they are unrecognizable.

I think it is very careful we don't say this is a camera issue without a lot further investigation. First of all, it is very mild. Second of all it is one set of data that Tolga was trying to get out fast. It needs to be investigated and explained, but nothing that can be concluded yet.

As I added later in the above post, those different types of glows seemed reasonably removed through proper subtractions, as long as there is no need to overly stretch the images.

Occasionally though, aggressive stretching maybe necessary.
 


CNers have asked about a donation box for Cloudy Nights over the years, so here you go. Donation is not required by any means, so please enjoy your stay.


Recent Topics






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics