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

Has the $1200 large chipped ASI 1600 made existing modestly priced CCDs obsolete?

  • Please log in to reply
372 replies to this topic

#251 vdb

vdb

    Surveyor 1

  • -----
  • Posts: 1,531
  • Joined: 08 Dec 2009

Posted 09 May 2016 - 02:11 AM

2000 subs, that is a lot of subs to have to process, even for a big computer. If the fits size is 32 megs like any other 16 meg chip, then you are talking 64,000,000,000 bytes total, that is 64 gigs.

Am I missing something here?

Blueman

 

 

Yes and No, you don't use advanced stacking techniques found in PI and the like, you use software that is used for planetary work ...

Emil has written it's own and is very well regarded.



#252 FirstC8

FirstC8

    Apollo

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

Posted 09 May 2016 - 09:56 AM

From what I have gathered so far, if chip size is not the issue, and small FOV is what I am ok with, the 178mm BSI (even the 290mm BSI) would be a better choice than the 1600?

The followup question is then, when a 14 bit 1600 size chip with BSI arrives, assume at a reasonable price, would that not spell the end of CCD?

How soon do we see that happen from ZWO? Six months? A year?

Edited by FirstC8, 09 May 2016 - 10:02 AM.


#253 Jon Rista

Jon Rista

    ISS

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

Posted 09 May 2016 - 11:15 AM

 

From what I have gathered so far, if chip size is not the issue, and small FOV is what I am ok with, the 178mm BSI (even the 290mm BSI) would be a better choice than the 1600?

The followup question is then, when a 14 bit 1600 size chip with BSI arrives, assume at a reasonable price, would that not spell the end of CCD?

How soon do we see that happen from ZWO? Six months? A year?

 

That's a matter of personal choice and what you want. Personally my opinion is ZWO is on the lowest end of the market so advanced imagers will not likely use them as their main imaging camera. The migration will occur when a suitable larger chip is adopted by the mainstream trusted camera vendors such as SBIG, FLI, and the like. When that occurs the transition has begun. There experiments are all really cool but most of us have no desire to shoot video and process gigantic files with lucky imaging, so the camera doesn't have a big appeal. It's greatest feature is the low noise. Astro cameras come with chips already available on the market, ZWO doesn't develop them, nor any vendor the cost would be prohibitive for such a tiny market. Again, I am not bashing CMOS in anyway, that's where the entire market will go in time without a doubt, my doubt is the astro market is nowhere close yet. 

 

 

Well, obviously the astro market IS close...because ZWO is a part of the astro market, regardless of where you personally choose to place them within it (personally I wouldn't put them at the lowest end...ASI cameras are better than similar QHY cameras, IMO, with more responsive customer support). QHY is also using more CMOS sensors in more cameras, as well as an sCMOS camera in a prototype camera they are designing. ZWO's ASI1600 is the beginning. As I've said before, it isn't a CCD killer...however it may indeed be a DSLR killer between now and whenever a higher grade CMOS sensor finds it's way into an astro camera. 

 

When did Sony say they were ending CCD manufacture? Few years here, tops? Any manufacturer not already working on CMOS technology of some kind is going to be left high and dry for their highest sensitivity parts when the flow of Sony CCDs ends. It can take a good number of years to design an entirely new product around an entirely new kind of sensor architecture, and built it to meet the same quality standards you had before. I can't imagine that QSI, FLI, SBIG, etc. aren't at least looking into CMOS technology already, if they don't already have prototypes in the works. Personally I am hoping they are looking into sCMOS rather than your run of the mill consumer grade DSLR/Mirrorless CMOS sensors. I think sCMOS is where this particular market needs to go in the long run. I don't think we are "nowhere close yet"...I think we are just around the corner. 



#254 Peter in Reno

Peter in Reno

    Voyager 1

  • *****
  • Posts: 10,885
  • Joined: 15 Jul 2008
  • Loc: Reno, NV

Posted 09 May 2016 - 11:36 AM

I think Sony is planning to end making or delivering CCDs by 2025 give or take a few years.

 

This thread is mainly about ASI1600 camera and the sensor is NOT made by Sony. It's made by Panasonic. And this CMOS sensor is NOT small so it's already has similar FOV (21.9mm diagonal) as KAF-8300 CCD (22.5mm).

 

I remember when amateur astro CCD cameras first came out (was it Meade DSI?), they were horrible, tiny FOV, noisy, uncooled and expensive as compared to today. CMOS introduction in astro market is far better and much cheaper than when amateur astro CCD camera was first introduced.

 

Astro camera manufacturers already know what the market wants so DSO imaging with CMOS cameras will get pretty darn good sooner than some people think.

 

Peter


Edited by Peter in Reno, 09 May 2016 - 11:37 AM.


#255 TimN

TimN

    Gemini

  • *****
  • Posts: 3,209
  • Joined: 20 Apr 2008
  • Loc: Ontario, Canada

Posted 09 May 2016 - 12:26 PM

Sony is stopping production of all CCD's in March 2017. However, they will ship them until March 2020.

 

https://www.alliedvi...production.html



#256 FirstC8

FirstC8

    Apollo

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

Posted 09 May 2016 - 12:57 PM

Sony is stopping production of all CCD's in March 2017. However, they will ship them until March 2020.

https://www.alliedvi...production.html


Are they going to ship them in steep discount?

If so, in the next few years the high end CCD products can meet "low end" demand and still have a fighting chance, while the high end manufactures work on sCMOS as replacement.

#257 Jon Rista

Jon Rista

    ISS

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

Posted 09 May 2016 - 01:27 PM

At the current 20,000 deep space long exposure images are not possible at that well depth.


Long exposure images are not necessary to get the same SNR, though. The read noise is 3.5e- vs. 9e- in the case of the ASI1600 vs. KAF-16803. The total noise term added to other noise terms is then 12.25e- vs. 81e-. At the same image scale, that is a huge difference. Long exposures would be totally unnecessary with the ASI1600. Now, I am not saying that the ASI1600 is a replacement for the KAF-16803. Certainly not. However it is a little naive to say that it is not possible and "therefor bad" (in the implication) that these sensors with smaller pixels can't be used properly for astrophotography.

Considering the small pixel size of 3.8 microns, having a 20,000e- well depth is actually quite impressive. Compared to the 5.4 micron pixels of the KAF-8300, which has 25,500e- FWC. In terms of pixel area, the KAF-8300 is over 2x larger, yet only has 1.275x more well capacity. It also has significantly more read noise (7-8e- in specs, 9-10e- in actual real-world measurements). If the KAF-8300 with a "mere" 25,500e- FWC, or a Sony ICX834 with a mere 9000e- FWC (!!), are more than capable for astrophotography, then the 20,000e- FWC of the ASI1600 is also more than capable. It is not CCD killing, but more than capable.

Combined with the low read noise, and you simply do not need long exposures. I am expecting to do 5-10 minute narrow band exposures at a gain setting of 60 or at unity gain with the ASI1600. At unity the read noise is only 1.75e-. You simply do not need 30, 45, 90 minute exposures in that case for most NB objects. There might be a few exceptions..OU4...however even in that case, I suspect the ASI1600 can get away with shorter exposures with the same SNR, or use exposures just as long for even better SNR, than most existing CCD cameras for that object (at a given image scale).

Edited by Jon Rista, 09 May 2016 - 01:34 PM.

  • hytham and Gucky like this

#258 bobzeq25

bobzeq25

    ISS

  • *****
  • topic starter
  • Posts: 20,136
  • Joined: 27 Oct 2014

Posted 09 May 2016 - 02:32 PM

 At the current 20,000 deep space long exposure images are not possible at that well depth.

Why not?  You're neglecting the effect of low read noise.  What's important is not well depth, it's dynamic range.   Once you've "buried the read noise" in skyfog, longer exposures have little value, unless you need to limit the number of subs.

 

Many very serious people have taken very serious DSO images with 8300 chips, which have only a bit more than 11 stops of dynamic range. less than the 1600.

 

The proper comparison with the $1200 1600 is the 8300 or similar, not $5K+ cameras.


Edited by bobzeq25, 09 May 2016 - 02:37 PM.

  • hytham, josh smith and Jon Rista like this

#259 AstroCatinfo

AstroCatinfo

    Ranger 4

  • *****
  • Posts: 360
  • Joined: 22 Jan 2016
  • Loc: Prades, Catalunya - Spain

Posted 09 May 2016 - 02:40 PM

Today's image with the ASI1600. 2000 x 1 second exposures. Dobsonian 16" on EQ platform. AWESOME!

 

By Emil Kraaikamp from Netherlands.

 

Full details: 

http://www.cloudynig...cool/?p=7205950

 

image.jpeg


Edited by AstroCatinfo, 09 May 2016 - 02:47 PM.

  • R Botero, saguaro and Jon Rista like this

#260 Jon Rista

Jon Rista

    ISS

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

Posted 09 May 2016 - 02:47 PM

Very detailed! That is what I'm looking forward to. I have such bad skies most of the time. Really hoping some shorter exposure imaging will help.

#261 TheRock

TheRock

    Messenger

  • *****
  • Posts: 416
  • Joined: 13 Jan 2008
  • Loc: Bergen County, NJ

Posted 09 May 2016 - 03:24 PM

Guys, let's be honest, nobody is going to replace their $5,000-$10,000 cameras with the ASI1600. The ASI1600 is for folks (like me) who are currently on the outside looking in and have had to settle for either a sensitive camera -or- a camera with a wide FOV, but not both because cost is the limiting factor. My price point is $1000-$2000, not something two, three or even five times that! I'm also not thrilled at the possibility of having to upgrade my mount for longer exposures, so if a camera like this allows for stacking more subs at shorter exposures, then even better.

 

Regarding the sweet spot for the ASI1600... I agree with others that it will likely be a first step-up from DSLRs but I also suspect it will eat into some of ZWO's own sales of their $600-$700 cameras.


Edited by TheRock, 09 May 2016 - 03:45 PM.

  • josh smith and dapalmer like this

#262 tolgagumus

tolgagumus

    Vendor

  • *****
  • Vendors
  • Posts: 3,687
  • Joined: 21 Jul 2014
  • Loc: NJ

Posted 09 May 2016 - 07:59 PM

OK gang as promised I went out to my dark site last night to test the new camera. I did not bring my Paramount. I wanted to demonstrate using short exposures with an AVX. I did use a pretty fast refractor. I am so spoiled with using paramounts, it took me forever to polar align. LOL. I shot 125 60 second subs but lost 15 due to tracking errors. I am pretty impressed with the results.

I know I also promised to make all data available but It is ALOT. Each file is 32 mg. Between the cal frames the folder is 20 gigs. If anyone has an idea how to upload it somewhere, I'll make it available

I only did basic preprocessing (darks,flats) and masked stretch in PI. No noise reduction,

 

Ascom driver Unity preset

get.jpg


Edited by tolgagumus, 10 May 2016 - 12:00 AM.


#263 buckeyestargazer

buckeyestargazer

    Vendor - Buckeyestargazer.net

  • *****
  • Vendors
  • Posts: 5,136
  • Joined: 12 Jan 2008
  • Loc: IN, USA

Posted 09 May 2016 - 08:27 PM

Dangit Tolga, now I might just have to consider this camera for myself.  I'm pretty set on a 16200 camera, but I've been following these ZWO threads too.  So far I have not been very impressed, but this widefield showing IFN with only 1min subs is impressive.  Still, I did see a lot of amp glow in a previous post and it does appear to me that some glow is still present in the final image.  How about making available the master unprocessed images (master flat, dark, light)?


  • tolgagumus likes this

#264 tolgagumus

tolgagumus

    Vendor

  • *****
  • Vendors
  • Posts: 3,687
  • Joined: 21 Jul 2014
  • Loc: NJ

Posted 09 May 2016 - 08:31 PM

Dangit Tolga, now I might just have to consider this camera for myself.  I'm pretty set on a 16200 camera, but I've been following these ZWO threads too.  So far I have not been very impressed, but this widefield showing IFN with only 1min subs is impressive.  Still, I did see a lot of amp glow in a previous post and it does appear to me that some glow is still present in the final image.  How about making available the master unprocessed images (master flat, dark, light)?

I was going to but the folder is huge, like 20 gigs. I can't put it all to dropbox



#265 Jon Rista

Jon Rista

    ISS

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

Posted 09 May 2016 - 08:33 PM

Tolga, have you considered Amazon Cloud Drive?



#266 buckeyestargazer

buckeyestargazer

    Vendor - Buckeyestargazer.net

  • *****
  • Vendors
  • Posts: 5,136
  • Joined: 12 Jan 2008
  • Loc: IN, USA

Posted 09 May 2016 - 08:35 PM

I didn't mean all the images, just three images (stacked light image, master flat and master dark).  That should be less than 100mb.



#267 FirstC8

FirstC8

    Apollo

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

Posted 09 May 2016 - 09:30 PM

Yeah Jon after read your post I signed up with Amazon unlimited cloud drive, sweet deal only $60 a year.

Tolga that is the answer. Thanks for doing the testing for us.
  • Jon Rista and tolgagumus like this

#268 Jon Rista

Jon Rista

    ISS

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

Posted 09 May 2016 - 09:58 PM

I only found out about it because Amazon sent me an email for a free year, about six months ago. Then a month ago, they gave me another free year. :p Figured, how could I possibly pass up free unlimited storage for 18 months! 



#269 tolgagumus

tolgagumus

    Vendor

  • *****
  • Vendors
  • Posts: 3,687
  • Joined: 21 Jul 2014
  • Loc: NJ

Posted 09 May 2016 - 10:51 PM

Dangit Tolga, now I might just have to consider this camera for myself.  I'm pretty set on a 16200 camera, but I've been following these ZWO threads too.  So far I have not been very impressed, but this widefield showing IFN with only 1min subs is impressive.  Still, I did see a lot of amp glow in a previous post and it does appear to me that some glow is still present in the final image.  How about making available the master unprocessed images (master flat, dark, light)?

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


  • StarDust1 likes this

#270 vdb

vdb

    Surveyor 1

  • -----
  • Posts: 1,531
  • Joined: 08 Dec 2009

Posted 09 May 2016 - 11:52 PM

Hi Tolga, what was the gain setting?



#271 tolgagumus

tolgagumus

    Vendor

  • *****
  • Vendors
  • Posts: 3,687
  • Joined: 21 Jul 2014
  • Loc: NJ

Posted 09 May 2016 - 11:57 PM

Hi Tolga, what was the gain setting?

I used Unity 



#272 Phil Hosey

Phil Hosey

    Vanguard

  • *****
  • Posts: 2,311
  • Joined: 29 Oct 2008
  • Loc: Billings, Montana

Posted 10 May 2016 - 05:57 AM

I made an attempt to calibrate Tolga's data on the other thread.



#273 hytham

hytham

    Surveyor 1

  • *****
  • Posts: 1,990
  • Joined: 25 Dec 2012
  • Loc: Northern Tundra

Posted 10 May 2016 - 12:34 PM

 

At the current 20,000 deep space long exposure images are not possible at that well depth.


Long exposure images are not necessary to get the same SNR, though. The read noise is 3.5e- vs. 9e- in the case of the ASI1600 vs. KAF-16803. The total noise term added to other noise terms is then 12.25e- vs. 81e-. At the same image scale, that is a huge difference. Long exposures would be totally unnecessary with the ASI1600. Now, I am not saying that the ASI1600 is a replacement for the KAF-16803. Certainly not. However it is a little naive to say that it is not possible and "therefor bad" (in the implication) that these sensors with smaller pixels can't be used properly for astrophotography.

Considering the small pixel size of 3.8 microns, having a 20,000e- well depth is actually quite impressive. Compared to the 5.4 micron pixels of the KAF-8300, which has 25,500e- FWC. In terms of pixel area, the KAF-8300 is over 2x larger, yet only has 1.275x more well capacity. It also has significantly more read noise (7-8e- in specs, 9-10e- in actual real-world measurements). If the KAF-8300 with a "mere" 25,500e- FWC, or a Sony ICX834 with a mere 9000e- FWC (!!), are more than capable for astrophotography, then the 20,000e- FWC of the ASI1600 is also more than capable. It is not CCD killing, but more than capable.

Combined with the low read noise, and you simply do not need long exposures. I am expecting to do 5-10 minute narrow band exposures at a gain setting of 60 or at unity gain with the ASI1600. At unity the read noise is only 1.75e-. You simply do not need 30, 45, 90 minute exposures in that case for most NB objects. There might be a few exceptions..OU4...however even in that case, I suspect the ASI1600 can get away with shorter exposures with the same SNR, or use exposures just as long for even better SNR, than most existing CCD cameras for that object (at a given image scale).

 

 

 

 

 At the current 20,000 deep space long exposure images are not possible at that well depth.

Why not?  You're neglecting the effect of low read noise.  What's important is not well depth, it's dynamic range.   Once you've "buried the read noise" in skyfog, longer exposures have little value, unless you need to limit the number of subs.

 

Many very serious people have taken very serious DSO images with 8300 chips, which have only a bit more than 11 stops of dynamic range. less than the 1600.

 

The proper comparison with the $1200 1600 is the 8300 or similar, not $5K+ cameras.

 

 

 

Science & Critical Thinking > Opinion

 

Great explanations, Bob and Jon!


Edited by hytham, 10 May 2016 - 12:34 PM.

  • TimN, Goofi, Jon Rista and 1 other like this

#274 happylimpet

happylimpet

    Aurora

  • *****
  • Posts: 4,748
  • Joined: 29 Sep 2013
  • Loc: Southampton, UK

Posted 11 May 2016 - 11:29 AM

As an aside, from my urban location, unflitered, I am sky limited at around 15 seconds with my 460ex. This translates to 1 second with the 1600mm, assuming similar QE and factoring in the different pixel size. I can still get 22nd magnitude quasars.

 

1 sec exposures mean I can do lucky DSO imaging, with no loss of SNR (around 8% for the pedantic). In practice I would probably use ROI a lot of the time for smaller targets to manage the data volume.

 

The much tighter star profiles this gives (and I know it works, I used to do 1-2sec exposures with my ASI120MM) mean you reach much fainter magnitudes for point sources.

 

For me this is a no brainer.

 

EDIT 15seconds / (pixel area factor of 0.7) * (1.2/5.2)^2 = 1 second


Edited by happylimpet, 11 May 2016 - 11:30 AM.


#275 JJK

JJK

    Gemini

  • -----
  • Posts: 3,465
  • Joined: 27 Apr 2008

Posted 11 May 2016 - 12:55 PM

 

 

Also I used to have a KAF8300 based camera. I think the ASI1600 will exceed it in performance for imaging. They key reason being lower read noise. You need much (30s vs. 300s) shorter exposures (although total exposure times will be similar but more subs) to generate the same SNR which equals much easier calibration.

 

30 seconds vs 300 seconds?  I'd probably check the math on that to make sure you have reasonable expectations.  There is no chance that is going to happen.  

 

 

I don't think you read the full post. Total exposure times will likely be similar but you will be able to use shorter subs due to the lower read noise.

 

If you use a higher gain setting you can bring the read noise down to 1-2e. At a higher gain setting you will not use the stated full well of the sensor but that is irrelevant. You will use sub exposure times that fill the scaled up sensor well exactly. At that sweet spot you will maximize the extraction of every photon captured by minimizing read noise. The math on this actually works very well.

 

But the real benefit will be easier calibration and less subs (exposure time) lost to mount errors or satellite trails etc.

 

While still a high quality sensor the KAF8300 is a decade old technology. These new cameras like the ASi1600 will take some getting used to. The biggest challenge will be for imagers to get out of a 20 year old imaging paradigm driven by CCDs... which is the most challenging part.

 

I understand why some would want the ability to take relatively short exposures, but a decent mount can track well in excess of 600 seconds, and satellite trails can be removed with processing.

 

The downside to taking many subs to to get the same integration time as a single sub, is the greater storage requirement.




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