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

Shorten total integration time with new equipment

astrophotography equipment
  • Please log in to reply
133 replies to this topic

#51 sn2006gy

sn2006gy

    Viking 1

  • ****-
  • Posts: 915
  • Joined: 04 May 2020
  • Loc: Austin, TX

Posted 17 August 2020 - 12:23 PM

Which 8300 were you using? The FLI and newer SBIG cameras download frames in less than 1 second. The FLI 16200 (which is the outlier in terms of performance) is ridiculously good. 
 
https://www.astrobin.../full/380925/J/
 
These are 5 minute narrowband subs with the FLI ML16200 and I would take similar sub exposures with the 6200 I have. You are making these cameras out to have like 30e of noise or something, which just is not the case. The FLI 16200's have 5-6e noise with a 6um pixel size and when cooled have negligible levels of dark current. They are still exceptional cameras and the new Sony chips dont change that at all. The FLI ML16803 has a 100k full well and 8e noise with 9um pixels. The frames are monstrously large (36x36 IIRC) and download within a reasonable amount of time for such large frames. That camera is still rocking around 14 stops of DR.
 
Cost is what I would hang on the CCD cameras. The cost of them is much higher than similar sized CMOS chips. Lets not act like everyone with a CCD is using KAI1100 chips though. Those you can beat on for high noise, high dark current, etc.


Ok, for the newer KAF the USB download may be quick for a kodak. The 6200 (or qhy600) is still faster smile.gif

entirely besides the point.

Which i think is still lost here.

If you treat the 6200 like you did a KAF, you are wasting all of its benefits. May as well use the KAF if you have experience and money to do so... bin away to get your sampling the way you need it.

 

no further discussion needed...

 


 


Edited by sn2006gy, 17 August 2020 - 12:23 PM.


#52 rockstarbill

rockstarbill

    Cosmos

  • *****
  • Posts: 7,564
  • Joined: 16 Jul 2013
  • Loc: Snohomish, WA

Posted 17 August 2020 - 12:28 PM

Ok, for the newer KAF the USB download may be quick for a kodak. The 6200 (or qhy600) is still faster smile.gif

entirely besides the point.

Which i think is still lost here.

If you treat the 6200 like you did a KAF, you are wasting all of its benefits. May as well use the KAF if you have experience and money to do so... bin away to get your sampling the way you need it.

no further discussion needed...



We'll have to agree to disagree here. The suggestion here is to treat the 6200 like it's a Panasonic chip, which is where the logic described here comes from. If folks want to do that, they are welcome to, but I'll respectfully pass on that.

#53 sn2006gy

sn2006gy

    Viking 1

  • ****-
  • Posts: 915
  • Joined: 04 May 2020
  • Loc: Austin, TX

Posted 17 August 2020 - 12:37 PM

We'll have to agree to disagree here. The suggestion here is to treat the 6200 like it's a Panasonic chip, which is where the logic described here comes from. If folks want to do that, they are welcome to, but I'll respectfully pass on that.

I'll be frank...

 

"respectfully passing on that" means just not even having to reply :)

 

You do you. Whatever you're doing works for you man. It was never about that.

 

If you want to treat a panasonic like a kodak, go for it! I have nothing against that.



#54 rockstarbill

rockstarbill

    Cosmos

  • *****
  • Posts: 7,564
  • Joined: 16 Jul 2013
  • Loc: Snohomish, WA

Posted 17 August 2020 - 12:41 PM

I'll be frank...

"respectfully passing on that" means just not even having to reply :)

You do you. Whatever you're doing works for you man. It was never about that.

If you want to treat a panasonic like a kodak, go for it! I have nothing against that.


I'm not suggesting anyone treat a Panasonic like a Kodak....

#55 sn2006gy

sn2006gy

    Viking 1

  • ****-
  • Posts: 915
  • Joined: 04 May 2020
  • Loc: Austin, TX

Posted 17 August 2020 - 01:00 PM

I'm not suggesting anyone treat a Panasonic like a Kodak....

Ok, so we finally cleared that up.

 

Can we get back on discussion about cameras and performance/speed of imaging? Where the 6200 is faster :D

 

(in case its not clear, when i re-read what you said in the context of you saying I was wrong it sounded like you dropped the mic and said i'll do it how i wanna do it...)

 

I've got other things to do.

 

But i am expecting a response because we can't just let things be can we? hehe



#56 ChrisWhite

ChrisWhite

    Fly Me to the Moon

  • *****
  • Posts: 6,142
  • Joined: 28 Feb 2015
  • Loc: Colchester, VT

Posted 17 August 2020 - 01:15 PM

Byron, I'm not sure why you insist short exposures are superior?  With a 12bit camera like the ASI 1600 bit depth recovery through frame count was important.  Not so with a 16bit camera. 

 

It's pretty simple really...  Expose as long as you can without clipping pixels.  This might mean long exposures even with the ASI6200 on some imaging systems.

 

Once you clip pixels you start losing data, and exposing longer really has little benefit... it can actually be a detriment. 

 

But as for short exposures... simply insisting that short exposures are the only way to go is not very logical.  Why not use the FWC of the camera to your advantage.  I would MUCH rather have fewer longer exposures than many super short exposures.  

 

There are only two considerations i have with respect to exposure length.  1) Am I clipping stars?  2) Am I adequately swamping read noise?  In this context, I expose as LONG as I can without clipping too many stars.


  • Midnight Dan and rockstarbill like this

#57 Midnight Dan

Midnight Dan

    Voyager 1

  • *****
  • Posts: 13,682
  • Joined: 23 Jan 2008
  • Loc: Hilton, NY, Yellow Zone (Bortle 4.5)

Posted 17 August 2020 - 01:26 PM

Hi Byron:

 

 

 

 6200 in your case would beat the 071 in apples/apples comparison at native or binned resolutions.

 

Yep. I think that's blatantly obvious. :-)  The 6200 is a much better camera than the 071, and is priced to prove it.

 

But that's not my question.  My current reference point is my 071 at f/7 on the EdgeHD 8.  I'm reasonably happy with the performance there except for the aberrations the FR causes. So I want to dump the FR.

 

I want to know if the 6200 at 2xBin will give me equivalent or better performance at f/10 on the same scope.   And again, by performance I mean exposure (total integration time) and resulting SNR.  In other words can I get the same or better SNR with the same or less integration time, comparing the following 2 setups:

 

EdgeHD 8 at f/7 (using focal reducer) with 071 camera:

Image scale: 0.66"/pixel

Horizontal FOV: 54.6'

 

EdgeHD 8 at f/10 with 6200 camera and 2xBin:

Image scale: 0.73"/pixel

Horizontal FOV: 58.2'

 

The main unknown, at least to me, was the effect of the 2xBin.  Everything else is easy to calculate including the effects of pixel size, f/ratio, QE, etc.  So to quantify the SNR effect, my tests were done with the exact same camera, the 071, comparing no binning to 2x binning to see what the effect was on SNR.  And what I found was that with an OSC camera, apparently using the standard Bilinear Debaying obfuscates most of the 2x binning effect.

 

-Dan


Edited by Midnight Dan, 17 August 2020 - 01:41 PM.


#58 rockstarbill

rockstarbill

    Cosmos

  • *****
  • Posts: 7,564
  • Joined: 16 Jul 2013
  • Loc: Snohomish, WA

Posted 17 August 2020 - 01:28 PM

So, one solution I was considering was to ditch the FR, and move to a full frame sensor.  The FOV at f/10 is about the same as at f/7 with the APS-C sensor.  And if I can get to the same pixel scale, I should have the same etendue per pixel, so the same exposure time and SNR - in theory.  Binning the ASI6200 pixels gets me to 0.73"/pixel, which is pretty close to what I want.

 

The question I was trying to answer for myself was: at f/10 will 2x-binned pixels with the ASI6200 FF sensor get me about the same performance as my ASI071 ASP-C sensor at f/7.  And by performance I mean exposure time and resulting SNR.

 

-Dan

 

Ditching the FR is a good idea. The Edge FR's are not the best in the world. For smaller chips, many people have good experiences with them. Once you get into larger chips, the problems with them really start to stick out like a sore thumb. 

 

We can answer your question pretty easily though:

 

Assuming the usual etendue calculation of A^2 * S^2 * QE (where A = Aperture, S = Scale, and QE = Quantum Efficiency) we can see the following in your configurations:

 

Dan_Compare.PNG

 

The Etendue of the bin 2x full frame chip is almost twice as high (94% increase) mostly due to the poor quantum efficiency of the IMX071 sensor. This calculation doesnt concern itself with read noise, and is not intended to tell you what you would need to expose to in order to hit your swamping target. But in general the F10 system would be almost twice as "fast".


  • Midnight Dan likes this

#59 ChrisWhite

ChrisWhite

    Fly Me to the Moon

  • *****
  • Posts: 6,142
  • Joined: 28 Feb 2015
  • Loc: Colchester, VT

Posted 17 August 2020 - 01:34 PM

Bill, according to ZWO spec, the QE of the 6200 is 91%.   Might want to correct that in your calculation. 


  • rockstarbill likes this

#60 Midnight Dan

Midnight Dan

    Voyager 1

  • *****
  • Posts: 13,682
  • Joined: 23 Jan 2008
  • Loc: Hilton, NY, Yellow Zone (Bortle 4.5)

Posted 17 August 2020 - 01:37 PM

Ditching the FR is a good idea. The Edge FR's are not the best in the world. For smaller chips, many people have good experiences with them. Once you get into larger chips, the problems with them really start to stick out like a sore thumb. 

 

We can answer your question pretty easily though:

 

Assuming the usual etendue calculation of A^2 * S^2 * QE (where A = Aperture, S = Scale, and QE = Quantum Efficiency) we can see the following in your configurations:

 

attachicon.gifDan_Compare.PNG

 

The Etendue of the bin 2x full frame chip is almost twice as high (94% increase) mostly due to the poor quantum efficiency of the IMX071 sensor. This calculation doesnt concern itself with read noise, and is not intended to tell you what you would need to expose to in order to hit your swamping target. But in general the F10 system would be almost twice as "fast".

Hi Bill:

 

thanks for the info.  But ... wow, I didn't realize that the 071 has a QE of only 50.  I have looked around and not been able to find that spec.  Where did you dig it up?  I was taking a wild guess and setting it at 65.  

 

-Dan



#61 rockstarbill

rockstarbill

    Cosmos

  • *****
  • Posts: 7,564
  • Joined: 16 Jul 2013
  • Loc: Snohomish, WA

Posted 17 August 2020 - 01:37 PM

Bill, according to ZWO spec, the QE of the 6200 is 91%.   Might want to correct that in your calculation. 

Man I have changed this from 0.8 to 0.87 then back to 0.8. Now its 91%? If true thats huge.



#62 Midnight Dan

Midnight Dan

    Voyager 1

  • *****
  • Posts: 13,682
  • Joined: 23 Jan 2008
  • Loc: Hilton, NY, Yellow Zone (Bortle 4.5)

Posted 17 August 2020 - 01:39 PM

Bill, according to ZWO spec, the QE of the 6200 is 91%.   Might want to correct that in your calculation. 

If you look at the specs on ZWO's site, both the mono and color versions are listed as 91%.  Since the bayer filters cut down the light significantly, I can't believe the color version is that high.  So I expect that they just incorrectly used the same spec on both version.

 

It seems really hard to get a valid QE spec on the OSC cameras.

 

-Dan


Edited by Midnight Dan, 17 August 2020 - 01:39 PM.

  • rockstarbill likes this

#63 rockstarbill

rockstarbill

    Cosmos

  • *****
  • Posts: 7,564
  • Joined: 16 Jul 2013
  • Loc: Snohomish, WA

Posted 17 August 2020 - 01:40 PM

Hi Bill:

 

thanks for the info.  But ... wow, I didn't realize that the 071 has a QE of only 50.  I have looked around and not been able to find that spec.  Where did you dig it up?  I was taking a wild guess and setting it at 65.  

 

-Dan

https://astronomy-im...ct/asi071mc-pro

 

The graphic near the top shows QE = 50%

 

Updated the Sony QE, and yeah it pulls ahead pretty far.

 

Dan_Compare.PNG

 

I wonder how accurate that measurement was from ZWO!



#64 rockstarbill

rockstarbill

    Cosmos

  • *****
  • Posts: 7,564
  • Joined: 16 Jul 2013
  • Loc: Snohomish, WA

Posted 17 August 2020 - 02:01 PM

Here are some comparisons I was doing of systems I plan to use the Sony IMX455 camera on

 

Bill_Compare.PNG

 

In my neck of the woods, 0.8-1" scale is about the best I can do in terms of sky quality but we can see that I can get that across almost all of the scopes I have either in bin 1 or bin 2 mode. The RCOS is a bit on the extreme side at 0.67" but that becomes a more respectable 1.01"/px if I bin 3 (and the Etendue jumps to 60,099). On a really good night, or at a different imaging site I have, I could possibly pull off the 0.67" at bin 2.

 

Very versatile camera that matches up nicely across a wide range of different scopes. The E-160ED in this list is on pre-order, as its the re-released brand new version. Folks should check it out though, as (according to Tak) it has a 3 micron spot size across its 44mm imaging circle. Seems the wizards at Tak were behind the scenes reworking the E-160 to have a spot size more acceptable for CMOS imaging.


Edited by rockstarbill, 17 August 2020 - 02:10 PM.


#65 Midnight Dan

Midnight Dan

    Voyager 1

  • *****
  • Posts: 13,682
  • Joined: 23 Jan 2008
  • Loc: Hilton, NY, Yellow Zone (Bortle 4.5)

Posted 17 August 2020 - 02:16 PM

https://astronomy-im...ct/asi071mc-pro

 

The graphic near the top shows QE = 50%

So it does! :-) That must have been added.  When I bought the camera a few years ago, the rating was "TBD".

 

Again, I would not believe the 91% rating on the 6200 OSC.  That's the same as the Mono 6200 and that's not possible.

 

-Dan



#66 rockstarbill

rockstarbill

    Cosmos

  • *****
  • Posts: 7,564
  • Joined: 16 Jul 2013
  • Loc: Snohomish, WA

Posted 17 August 2020 - 02:22 PM

So it does! :-) That must have been added.  When I bought the camera a few years ago, the rating was "TBD".

 

Again, I would not believe the 91% rating on the 6200 OSC.  That's the same as the Mono 6200 and that's not possible.

 

-Dan

You are using the Color camera? If so, you would undoubtedly gain significant improvements in performance if you were to move to mono and step into the 6200 at the same time or wait until later this year and step into the APS-C mono that has the same performance, just a smaller chip.


Edited by rockstarbill, 17 August 2020 - 02:23 PM.

  • Midnight Dan likes this

#67 Midnight Dan

Midnight Dan

    Voyager 1

  • *****
  • Posts: 13,682
  • Joined: 23 Jan 2008
  • Loc: Hilton, NY, Yellow Zone (Bortle 4.5)

Posted 17 August 2020 - 03:20 PM

You are using the Color camera? If so, you would undoubtedly gain significant improvements in performance if you were to move to mono and step into the 6200 at the same time or wait until later this year and step into the APS-C mono that has the same performance, just a smaller chip.

Ok, that woke me up! I was thinking an APS-C mono might be just the ticket.  Something like a mono version of the 2600.  Has something been announced?  I must have missed it.

 

-Dan


Edited by Midnight Dan, 17 August 2020 - 03:22 PM.


#68 rockstarbill

rockstarbill

    Cosmos

  • *****
  • Posts: 7,564
  • Joined: 16 Jul 2013
  • Loc: Snohomish, WA

Posted 17 August 2020 - 03:34 PM

QHY268M has been announced, grumblings around about ZWO doing a 2600MM but nothing official yet. Later this year.

#69 Peregrinatum

Peregrinatum

    Surveyor 1

  • *****
  • Posts: 1,506
  • Joined: 27 Dec 2018
  • Loc: South Central Valley, Ca

Posted 18 August 2020 - 11:30 AM

An ASI2600MM would be sweet, perhaps priced at half that of the ASI6200MM?  Hopefully with no micro lensings issues!!

 

popcorn.gif


Edited by Peregrinatum, 18 August 2020 - 11:31 AM.


#70 Midnight Dan

Midnight Dan

    Voyager 1

  • *****
  • Posts: 13,682
  • Joined: 23 Jan 2008
  • Loc: Hilton, NY, Yellow Zone (Bortle 4.5)

Posted 18 August 2020 - 11:49 AM

Just did some searching and found a QHY video showing their upcoming lineup.  It showed both the color and mono version of the QHY268 on their roadmap, and both at the same price of $2095. Of course that's subject to change.

 

I also found a thread where someone messaged ZWO to ask about their plans for a mono ASI2600, due to the challenge from QHY.  They replied simply that it is on their plan.

 

-Dan



#71 Peregrinatum

Peregrinatum

    Surveyor 1

  • *****
  • Posts: 1,506
  • Joined: 27 Dec 2018
  • Loc: South Central Valley, Ca

Posted 18 August 2020 - 11:59 AM

I took a look at the etendue for my current setup, and also with the possibility of a mono IMX571 chip, the difference would be significant for me in terms of optical speed and a reduction in imaging time:

 

3,649  ASI1600MM C925HD F10

4,814  ASI2600MM C925HD F10

 

4814/3649 = 1.32

 

My current Ha exposure is 10' (mount limitations), but with the IMX571 I could get a similar signal using a 7' exposure... this is a big deal to me!


Edited by Peregrinatum, 18 August 2020 - 11:59 AM.


#72 rockstarbill

rockstarbill

    Cosmos

  • *****
  • Posts: 7,564
  • Joined: 16 Jul 2013
  • Loc: Snohomish, WA

Posted 18 August 2020 - 07:31 PM

Byron, I'm not sure why you insist short exposures are superior?  With a 12bit camera like the ASI 1600 bit depth recovery through frame count was important.  Not so with a 16bit camera. 

 

It's pretty simple really...  Expose as long as you can without clipping pixels.  This might mean long exposures even with the ASI6200 on some imaging systems.

 

Once you clip pixels you start losing data, and exposing longer really has little benefit... it can actually be a detriment. 

 

But as for short exposures... simply insisting that short exposures are the only way to go is not very logical.  Why not use the FWC of the camera to your advantage.  I would MUCH rather have fewer longer exposures than many super short exposures.  

 

There are only two considerations i have with respect to exposure length.  1) Am I clipping stars?  2) Am I adequately swamping read noise?  In this context, I expose as LONG as I can without clipping too many stars.

Just to add to how wrong that position is, here is some work done by this guy named Roland.

 

https://www.astrobin...6uf7/C/?nc=user



#73 freestar8n

freestar8n

    Vendor - MetaGuide

  • *****
  • Vendors
  • Posts: 9,806
  • Joined: 12 Oct 2007

Posted 19 August 2020 - 03:07 AM

I took a look at the etendue for my current setup, and also with the possibility of a mono IMX571 chip, the difference would be significant for me in terms of optical speed and a reduction in imaging time:

 

3,649  ASI1600MM C925HD F10

4,814  ASI2600MM C925HD F10

 

4814/3649 = 1.32

 

My current Ha exposure is 10' (mount limitations), but with the IMX571 I could get a similar signal using a 7' exposure... this is a big deal to me!

This thread is going several directions - but for comparing these two systems specifically in terms of exposure time - it's important to distinguish optical speed from etendue and from pixel etendue.  When you use two different sensors that only differ in physical size, but the pixel size is the same, you know that the optical speed is the same (f/10) and the pixel etendue is the same - if the pixels are also the same size.  But the etendue of the overall system will be larger due to the larger sensor.  It will have greater "optical throughput" with the larger sensor.

 

But that doesn't directly translate to exposure time unless you view images from the sensors at the same physical size on the monitor - in which case the galaxy would appear smaller with the larger sensor.  By making it smaller you are effectively concentrating the signal as seen by your eye - but if you compared the images on the same pixel scale there would be no difference - theoretically.

 

And if you went ahead and cropped the larger image it would be identical to the 1600 - if all those parameters are the same.

 

There may be a slight QE difference - but that is small compared to going from f/10 to f/5 or something.

 

Frank


  • ChrisWhite likes this

#74 freestar8n

freestar8n

    Vendor - MetaGuide

  • *****
  • Vendors
  • Posts: 9,806
  • Joined: 12 Oct 2007

Posted 19 August 2020 - 03:22 AM

Just to add to how wrong that position is, here is some work done by this guy named Roland.

 

https://www.astrobin...6uf7/C/?nc=user

I'm a big fan of going for long exposures if you can, limited by saturation, but in most practical imaging situations I would not go 60 minutes because there would be loss of yield and fewer frames to dither and reject via stacking.

 

I'm not sure the intent of that comparison - but I expect people will use it as a very general and conclusive example - and I think that's unfortunate.

 

The camera used is a qsi683 which is spec'd at 8e read noise - and this is a narrowband image at f/6 so it is a good situation where read noise is a big factor and pushes subexposures longer.  But if the sensor instead had 1.5e read noise this would be very different.

 

There is no indication of any dithering done or how the stacking/rejection was done in the 10m exposures - and that is a big omission.  If the sensor has pattern noise that is not completely removed by calibration then you know the single exposure image has zero chance or reducing it, whereas the stacked exposure can - but only if it is dithered.

 

So in the year 2020 I hope people can keep track of these basic competing issues.  Long exposure is good but watch out for clipping - and make sure you have enough dithered frames that the stacking can reduce pattern noise in the stack.  There is no need to worry about anything optimal - and you most certainly shouldn't expect a big gain going all the way to 1 hour exposures with a low read noise camera.  But some people seem to think 20s is optimal - and it likely isn't unless you have clipping.  And going from 5m to 10m may have a noticeable benefit - or it may not.  But the basic underlying competing factors aren't that hard to keep in mind.

 

No matter what - I hope people don't conclude from this that in a very general sense - 1 hour exposures are inherently a big win.  My 8300 sensor had more like 12e read noise, so long exposures were a benefit.  But with a 1600 the situation is very different.

 

Frank

 

addendum:  And there is zero reason the stacked image should have higher fwhm.  That implies a problem in the stacking/alignment or a change in seeing conditions.


Edited by freestar8n, 19 August 2020 - 03:30 AM.

  • rockstarbill and sn2006gy like this

#75 ChrisWhite

ChrisWhite

    Fly Me to the Moon

  • *****
  • Posts: 6,142
  • Joined: 28 Feb 2015
  • Loc: Colchester, VT

Posted 19 August 2020 - 09:29 AM

This thread is going several directions - but for comparing these two systems specifically in terms of exposure time - it's important to distinguish optical speed from etendue and from pixel etendue.  When you use two different sensors that only differ in physical size, but the pixel size is the same, you know that the optical speed is the same (f/10) and the pixel etendue is the same - if the pixels are also the same size.  But the etendue of the overall system will be larger due to the larger sensor.  It will have greater "optical throughput" with the larger sensor.

 

But that doesn't directly translate to exposure time unless you view images from the sensors at the same physical size on the monitor - in which case the galaxy would appear smaller with the larger sensor.  By making it smaller you are effectively concentrating the signal as seen by your eye - but if you compared the images on the same pixel scale there would be no difference - theoretically.

 

And if you went ahead and cropped the larger image it would be identical to the 1600 - if all those parameters are the same.

 

There may be a slight QE difference - but that is small compared to going from f/10 to f/5 or something.

 

Frank

Frank,

 

I believe that most of the time when people are talking about Entendue on here they are talking about pixel entendue not sensor entendue.  Most of the basic calculations that I have used dont have any input for sensor size, just pixel size.  In the contest of discussing overall "speed" (or whatever we want to label it) of a system with respect to integration time pixel entendue is really the only relevant part of the discussion... is this accurate?

 

Regarding QE, that has a huge impact on Entendue calculation.   If a theorized 2600mono sensor has the same QE as the 6200 (being the same sensor tech) it might have around 90% QE.  Compare that to the ASI 1600 which has a QE of 60%, that alone would make the 2600 sensor 1.5x "faster" would it not?  That seems very significant to me.  (Pixel size is almost identical between the two cameras).

 

Thoughts?


  • rockstarbill likes this


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





Also tagged with one or more of these keywords: astrophotography, equipment



Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics