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

ASI1600MM Camera Performance

This topic has been archived.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

#26 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 24,854
  • Joined: 10 Jan 2014

Posted 17 June 2016 - 12:02 AM

Yeah, I think 300s @ unity is plenty good enough. I did not see any additional stars appear with 600s, which surprised me a bit. The stars that were barely visible @ 300s were just a tiny bit brighter. With all the extra noise from dark current and the glows, however, I suspect the overall SNR was about the same, it not worse. 



#27 syscore

syscore

    Aurora

  • *****
  • Posts: 4,711
  • Joined: 07 Sep 2015

Posted 17 June 2016 - 12:07 AM

Remember when we were first looking at the super short exposure examples with other cameras? Now we know what happened to the stars. :)


Edited by syscore, 17 June 2016 - 12:07 AM.


#28 FirstC8

FirstC8

    Apollo

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

Posted 17 June 2016 - 12:08 AM

Jon, you seemed to say amp glows were the issue as exposures increased, but the glows can be removed by darks.

But as the glows increase along with exposure time, other noises also increase, and the increased other noises cannot be calibrated out.

Did I understand it correctly?

Edited by FirstC8, 17 June 2016 - 12:10 AM.


#29 syscore

syscore

    Aurora

  • *****
  • Posts: 4,711
  • Joined: 07 Sep 2015

Posted 17 June 2016 - 12:12 AM

You can't subtract noise. It is just like the example on your M57 thread. If the noise rises faster than the faint signal then it starts to obscure the faint details.


Edited by syscore, 17 June 2016 - 12:12 AM.


#30 Peter in Reno

Peter in Reno

    Voyager 1

  • *****
  • Posts: 11,847
  • Joined: 15 Jul 2008

Posted 17 June 2016 - 12:59 AM

Also I don't think you can use dark scaling with amp glows. Dark scaling is very useful to use only one master dark regardless of sub-exposure times of lights. I have been using 30 minute Master dark to calibrate with various sub-exposures of lights using PixInsight dark scaling from my QSI660wsg CCD camera without issues.

 

Peter



#31 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 24,854
  • Joined: 10 Jan 2014

Posted 17 June 2016 - 01:00 AM

@FirstC8: Dark current, glows, read noise...those are all noise terms. Skyfog, even though it's also a photonic signal, is also a noise term. If you look at the SNR formula:

ObjectSignal/SQRT(ObjectSignal + SkyFog + DarkCurrent + Glow + ReadNoise^2)

^-- Signal  | Noise Terms -^--------^----------^---------^----------^

So, let's say you have only dark current, and no glow. Let's say you have a 9e- object signal. If you had a noiseless camera, then your SNR formula would be the pure signal over the square root of the signal:

9/SQRT(9) = 3:1

You have a 3:1 SNR. This is the maximum potential SNR for a 9e- signal. However, what if you also had a 4e- skyfog signal, a 2.4e- dark current signal with 1.5e- read noise:

9/SQRT(9 + 4 + 2.4 + 1.5^2) = 9/SQRT(9 + 6.4 + 2.25) = 9/SQRT(17.65) = 2.14:1

This is about 72% of the perfect SNR. Not too bad. However if you have an additional 30e- glow signal on top of the base dark signal (about 0.1e-/s):

9/SQRT(9 + 4 + 2.4 + 30 + 1.5^2) = 9.SQRT(9 + 36.4 + 2.25) = 9/SQRT(47.65)  = 1.3:1

Now your down to 43% of the perfect SNR for a 9e- signal.

 

Based on my previous linearity testing of dark current, it seems the glows grow in such a manner as to produce non-linear, accelerated growth in the dark signal. From 5s to about 120s, the dark signal curve was pretty flat at ~320 ADU (16-bit...which is ~20 ADU in 12-bit; the standard unity-gain bias offset is 21 ADU), then it started to accelerate at 300s (although it was only slightly higher than 5-120s). By 600s it grew to over 400 ADU, and by 1800s it was over 600 ADU. With that kind of accelerated glow signal growth, you aren't going to be able to get much better SNRs at 1800s than at 300s. Slightly better, but far from what you could get if you did not have the extra glow signal. It would be easy enough to compensate for the difference by stacking a few more 300s subs, though.


Edited by Jon Rista, 17 June 2016 - 01:03 AM.


#32 leemr

leemr

    Messenger

  • -----
  • Posts: 499
  • Joined: 21 Mar 2014

Posted 17 June 2016 - 02:58 AM

Hey Jon, do you mind sharing what driver version you are using in the amp glow tests? Sam said he'd made changes which affected the amp glow, so your results might not match others' depending what version they're running. Someone please correct me if I'm wrong, but I think, from memory, Sam said that a change was made in 1.0.2.7 to reduce the amp glow, then reverted in 1.0.2.8 due to stability issues?

 

Edit: I guess we should probably also mention USB version and transfer speed as well in these discussions, since (again, from memory) the amp glow in USB 2 is impacted by the traffic limit, but not (so much?) in USB 3.

 

Tedious, perhaps, but without the context the information becomes less useful.


Edited by leemr, 17 June 2016 - 03:05 AM.


#33 A. Viegas

A. Viegas

    Gemini

  • *****
  • Posts: 3,071
  • Joined: 05 Mar 2012

Posted 17 June 2016 - 07:00 AM

Hey Jon, do you mind sharing what driver version you are using in the amp glow tests? Sam said he'd made changes which affected the amp glow, so your results might not match others' depending what version they're running. Someone please correct me if I'm wrong, but I think, from memory, Sam said that a change was made in 1.0.2.7 to reduce the amp glow, then reverted in 1.0.2.8 due to stability issues?

 

Edit: I guess we should probably also mention USB version and transfer speed as well in these discussions, since (again, from memory) the amp glow in USB 2 is impacted by the traffic limit, but not (so much?) in USB 3.

 

Tedious, perhaps, but without the context the information becomes less useful.

 

USB2 or USB3 issues should have nothing to do with amp glow.  Again if you are getting messed up frames in download, try using a USB 2.0 connection only and lowest traffic setting.  I have been using the 1600MM now since NEAF and I have never had any issues with light frames downloading dark or otherwise problems when using USB2.0 only.     Using USB3  Oiy-vey!  No Way.... its the Sith lord of problems with the ASI1600 :jedi:



#34 Gucky

Gucky

    Viking 1

  • *****
  • Posts: 537
  • Joined: 18 May 2015

Posted 17 June 2016 - 08:20 AM

Thanks John. Very informative and valuable!!



#35 leemr

leemr

    Messenger

  • -----
  • Posts: 499
  • Joined: 21 Mar 2014

Posted 17 June 2016 - 08:22 AM

 

Hey Jon, do you mind sharing what driver version you are using in the amp glow tests? Sam said he'd made changes which affected the amp glow, so your results might not match others' depending what version they're running. Someone please correct me if I'm wrong, but I think, from memory, Sam said that a change was made in 1.0.2.7 to reduce the amp glow, then reverted in 1.0.2.8 due to stability issues?

 

Edit: I guess we should probably also mention USB version and transfer speed as well in these discussions, since (again, from memory) the amp glow in USB 2 is impacted by the traffic limit, but not (so much?) in USB 3.

 

Tedious, perhaps, but without the context the information becomes less useful.

 

USB2 or USB3 issues should have nothing to do with amp glow.  Again if you are getting messed up frames in download, try using a USB 2.0 connection only and lowest traffic setting.  I have been using the 1600MM now since NEAF and I have never had any issues with light frames downloading dark or otherwise problems when using USB2.0 only.     Using USB3  Oiy-vey!  No Way.... its the Sith lord of problems with the ASI1600 :jedi:

 

 

You might be confusing me with someone else. I'm not having issues with messed up frames. I'm simply stating that Jon's informative posts would be made even more informative with some extra information.

 

According to the vendor, using USB 2 can indeed impact the amp glow, as seen in this post.

 

Last time I tried, things were working fine for me with USB 3 and the latest driver.


Edited by leemr, 17 June 2016 - 08:23 AM.


#36 A. Viegas

A. Viegas

    Gemini

  • *****
  • Posts: 3,071
  • Joined: 05 Mar 2012

Posted 17 June 2016 - 08:59 AM

 

 

Hey Jon, do you mind sharing what driver version you are using in the amp glow tests? Sam said he'd made changes which affected the amp glow, so your results might not match others' depending what version they're running. Someone please correct me if I'm wrong, but I think, from memory, Sam said that a change was made in 1.0.2.7 to reduce the amp glow, then reverted in 1.0.2.8 due to stability issues?

 

Edit: I guess we should probably also mention USB version and transfer speed as well in these discussions, since (again, from memory) the amp glow in USB 2 is impacted by the traffic limit, but not (so much?) in USB 3.

 

Tedious, perhaps, but without the context the information becomes less useful.

 

USB2 or USB3 issues should have nothing to do with amp glow.  Again if you are getting messed up frames in download, try using a USB 2.0 connection only and lowest traffic setting.  I have been using the 1600MM now since NEAF and I have never had any issues with light frames downloading dark or otherwise problems when using USB2.0 only.     Using USB3  Oiy-vey!  No Way.... its the Sith lord of problems with the ASI1600 :jedi:

 

 

You might be confusing me with someone else. I'm not having issues with messed up frames. I'm simply stating that Jon's informative posts would be made even more informative with some extra information.

 

According to the vendor, using USB 2 can indeed impact the amp glow, as seen in this post.

 

Last time I tried, things were working fine for me with USB 3 and the latest driver.

 

Amp glow with USB 2 can be minimized using high value in USB traffic, but in any case for anyone who is running long-run cables and does not have a computer at the scope, all I am saying is that I have found USB2 to be 100% reliable over any distance exceeding 10 meters.  Less than 10 meters I have had success with USB 3.   Sorry for the confusion



#37 Thirteen

Thirteen

    Gemini

  • *****
  • Posts: 3,186
  • Joined: 12 Jul 2013

Posted 17 June 2016 - 09:06 AM

Sam's recommendation is USB3 if at all possible because USB2 speeds can indeed impact amp glow on readout.

#38 FirstC8

FirstC8

    Apollo

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

Posted 17 June 2016 - 10:24 AM

Jon I understand some noises inherent in the chip cannot be removed by calibration, but there are exceptions, fixed pattern noises such as amp glow, hot/cold pixels and line pattern noises can be removed.

To what extent they can be removed is the question. For my camera which is a 1/3 chip designed for planetary, long exposure noises are difficult to calibrate.

Some amp glow would remain despite dark calibration, until the exposure reaches certain length, then if the exposure time continues to increase, the glow would be over corrected and leaving dark patches, the opposite of the residual glow.

I am working with QHY on such issue and they blame it on the stacking software.

Aside from DSS I am using, what would be the next stacking software you would recommend?

In the ideal situation, those FPNs can be calibrated, am I correct? In fact when Emilio first made his subs available for us to play with, I used DSS to successfully remove all amp glows, most hot pixels, leaving maybe a few cold pixels only. Although those were relatively short exposure subs.

So I am not convinced DSS was to blame but willing to try different stacking software.

But back to your point, if indeed amp glow can be theoretically removed, the fact the glow increases with exposure time is not a showstopper for the 1600, much like vignetting and dust moles are not deal breakers as long as they can be removed by flats.

Edited by FirstC8, 17 June 2016 - 10:54 AM.


#39 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 24,854
  • Joined: 10 Jan 2014

Posted 17 June 2016 - 11:16 AM

Signals can be removed, their noise cannot be. So, amp glow and dark current. They behave like signals, and they have their own intrinsic noise. If dark current grows at a rate of 0.01e-/s and you expose for 600s, you will have a 6e- dark current signal. You can offset that signal, subtract it out...however, noises add in quadrature. So you cannot remove it's noise, it's mixed in with all the other sources of noise. So, if you subtract 6e- signal from your image, you are still going to have the SQRT(6e-) noise in there, contributing to the total noise alongside read noise, light pollution, and the object signal itself. Hence the reason the total noise term in my previous post was SQRT(ObjectSignal + SkyFog + DarkCurrent + Glow + ReadNoise^2)...all the noises from each of those things combines together in quadrature...and inseparably. Noise is all random offsets from the "correct" mean value. If you don't know exactly what the noise is for a given pixel, and by it's very nature you cannot know, then you cannot simply "remove" it. The only thing you can remove is what you know for sure. The original signals themselves, that added the extra noise...the dark current, the glow, the skyfog. You can identify what those are, say with a master dark frame, or with a mean background level measurement (for skyfog), and subtract the necessary amount of signal from each pixel to remove the signals. But that only changes the offsets, it does not change the noise. 

 

So yes. You can subtract out the amp glow SIGNAL, but you cannot eliminate it's NOISE from the image. You can actually completely eliminate the glow itself. I've verified that on many occasions. However I also verified that in the areas where the glows were before their subtraction, the noise is higher. 

 

The big problem with these glows is a) they potentially can leave behind more noise in the areas where they were than the total noise in other areas of the image where there is no glow, and b) they are non-uniform across the field, so you end up with different amounts of noise in different areas of the image, which greatly complicates noise reduction and recovery of faint details. 

 

Now with shorter exposures, 300s, the overall dark current and glows is pretty small. Easy enough to manage.You can subtract them and they do not leave behind much noise. At 600s the glows are more significant, and start to affect the entire field non-uniformly. Once your up to 1200-1800 seconds, the entire field has very non-uniform glow around it and the two bright spots are extremely bright. Compared to say IFN, they are likely to be much brighter. In my measurements (16-bit), the background level at 1200-1800s can be around 450-500 ADU, however the glows can be 1300 ADU or so. The bias offset level is still ~320 ADU. I think you are unlikely to get even 200 ADU on IFN, let alone 1000 ADU, even in 1800 seconds, with these smaller pixels. 



#40 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 24,854
  • Joined: 10 Jan 2014

Posted 17 June 2016 - 11:32 AM

Hey Jon, do you mind sharing what driver version you are using in the amp glow tests? Sam said he'd made changes which affected the amp glow, so your results might not match others' depending what version they're running. Someone please correct me if I'm wrong, but I think, from memory, Sam said that a change was made in 1.0.2.7 to reduce the amp glow, then reverted in 1.0.2.8 due to stability issues?

 

Edit: I guess we should probably also mention USB version and transfer speed as well in these discussions, since (again, from memory) the amp glow in USB 2 is impacted by the traffic limit, but not (so much?) in USB 3.

 

Tedious, perhaps, but without the context the information becomes less useful.

 

Great question! I should have said it before. I am still using the 1.0.2.5 driver. The ASCOM driver actually DOES affect amp glow. Sam removed the glow bubble on the left edge with 1.0.2.5. They added it back in 1.0.2.7, but that seemed to introduce another problem. I am not sure what the current state of 1.0.2.8 is, as I think they have been working on it. 



#41 futuneral

futuneral

    Apollo

  • *****
  • Posts: 1,382
  • Joined: 27 Dec 2014

Posted 17 June 2016 - 12:46 PM

So yes. You can subtract out the amp glow SIGNAL, but you cannot eliminate it's NOISE from the image. You can actually completely eliminate the glow itself. I've verified that on many occasions. However I also verified that in the areas where the glows were before their subtraction, the noise is higher. 

 

The big problem with these glows is a) they potentially can leave behind more noise in the areas where they were than the total noise in other areas of the image where there is no glow, and b) they are non-uniform across the field, so you end up with different amounts of noise in different areas of the image, which greatly complicates noise reduction and recovery of faint details. 

 

 

 

Hey Jon,

 

Is my understanding correct here? You can't remove the noise from an image via subtraction, you only do this to signal. But wouldn't averaging multiple images help to eliminate this noise?

 

The noise introduced by the amp glow, after averaging, would become a complex non-uniform map of offsets, right? Wouldn't subtracting the SQRT(glow signal image) from the integrated (averaged) image allow to remove this "noise map" introduced by the amp glow?



#42 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 24,854
  • Joined: 10 Jan 2014

Posted 17 June 2016 - 01:17 PM

Averaging will reduce noise, because of it's random nature averaging many random numbers together will result in convergence onto the "correct" value, but it cannot eliminate noise. The reduction in noise is relative to the square root of the number of frames stacked. So, integrate four frames, reduce noise by a factor of 2x. Integrate 16 frames, reduce noise by a factor of 4x, integrate 64 frames, reduce noise by a factor of 8x, integrate 256 frames, reduce noise by a factor of 16x. The next step is 1024 frames, then 4096 frames. So, beyond 256 frames, you rapidly enter the realm of diminishing returns. 

 

As for the noise introduced by amp glow. The amp glow itself is the offset. You do map that, by generating a master dark and subtracting it. ;) Once you remove the offset, you no longer have anything to map. You simply have randomness around a mean, however the randomness varies differently across the frame. You might have a standard deviation of ~30 ADU in the middle of the frame, 60 ADU in the midframe, and 180 ADU around the glows, however the mean might be a consistent 320 ADU (the bias offset). There is no brighter or darker average...the average across the frame is the same. However the standard deviation is now non-uniform. There really isn't any great way to deal with that. It will always be the case. Even if you average a thousand frames together, you might then end up with a standard deviation in the middle of ~1 ADU, in the midframe of ~2 ADU, and in glow areas of around 5-6 ADU. The non-uniform nature of the noise after the glows have been removed remains, no matter how much you stack. However...if you stack enough that your faint signal strengths are considerably larger than the worst of the noise, then, it wouldn't matter as much. That starts to happen around 81 subs...and gets pretty good at 256 subs, for signals around the same magnitude as the noise. If you are going for even fainter signals...then you might need to stack even more. 

 

The non-uniformity in the noise at 300 seconds is very low. It wouldn't be a problem once you stacked enough subs, which is probably around 49-64 or more. So, it is not really a problem. Hence the reason I recommend sticking to 300s subs for NB. It's when you get to 600s and particularly beyond that the glows become so significant that the non-uniform effect they have on noise can no longer really be ignored. If you are going for really faint details, then there might actually be some value in stacking a few hundred 120-150 second NB subs at gain 200. Read noise is even lower, so would better support shorter subs. You would lose a bit more DR, but the glows would barely be as large as read noise, so rather trivial. The noise they would add would also be rather trivial. 



#43 FiremanDan

FiremanDan

    Aurora

  • *****
  • Posts: 4,869
  • Joined: 11 Apr 2014

Posted 17 June 2016 - 01:19 PM

Oh man, I didn't think to re-do calibration frames after driver updates. 

It doesn't seem to have caused an issue with my image though. 
Something to keep in mind though! 



#44 mikefulb

mikefulb

    Surveyor 1

  • *****
  • Posts: 1,884
  • Joined: 17 Apr 2006

Posted 17 June 2016 - 01:23 PM

Jon - say we're going after something really faint and small and 300s isn't adequate - are the glows restricted to the periphery of the chip sufficiently that if we cropped down to the central 50% of the sensor we could maintain a steady increase in S/N for that area?  I'm thinking interacting galaxies or OII/SII data for the outer shells of M57.



#45 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 24,854
  • Joined: 10 Jan 2014

Posted 17 June 2016 - 01:51 PM

Jon - say we're going after something really faint and small and 300s isn't adequate - are the glows restricted to the periphery of the chip sufficiently that if we cropped down to the central 50% of the sensor we could maintain a steady increase in S/N for that area?  I'm thinking interacting galaxies or OII/SII data for the outer shells of M57.

 

If you crop down to the central 50%, you wouldn't have the severe glows. However, I've been trying to test on M57, and I did not really see any additional faint details with it dead center of the frame at 600s than I did at 300s. I am trying to integrate a 36x300s set of subs I acquired right now. I honestly don't know how it will pan out. I've been having problems with both focus and dithering in SGP, and I am pretty sure this data is just as much junk as all my previous attempts. I've had severe correlated noise in all my previous attempts to integrate, because dithering is resulting in a consistent diagonal movement of the stars across the frame, like drift...so dithering is effectively broken for me right now. I've been screwing with it for four nights now...I'm about at my wits end. I've decided to sell the RC, it's been nothing but trouble and in 16 months I've basically gotten nothing out of it, and I am not sure I want to keep fiddling with it to try and get all these variables in sync so it works right, even for one night, let alone consistently for months on end. 

 

So I don't know if I can actually demonstrate visually yet. I am hoping I can try to do just that with my 600mm lens when OU4 rolls around in a few weeks here. It's a very faint object, and I am going to see what's possible with 300s subs, or maybe even shorter ones at gain 200. 



#46 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 24,854
  • Joined: 10 Jan 2014

Posted 17 June 2016 - 02:54 PM

I apologize for the softness here. It's really bad, frustrating the crap out of me. SGP cannot seem to keep up with the cooling rate of my scope, and I have been unable to set proper offsets and temp compensation. I managed to get it closer, so I actually had enough frames to integrate and sharp enough stars to even register at all this time, but the data here is significantly blurrier than I've been able to get with SharpCap in the past. Anyway, this is 35x300 (one of the subs was bad), BIN 1x1 (I think previous comparisons were binned 2x2!!!) so 2h55m, f/8, AT8RC, unity gain, ASCOM driver .5, SGP .15. Calibrated with 64x 300s unity darks, 25x 25s Ha unity flats (which were just bias calibrated with a 256x unity bias):

 

M57_300s_3h.jpg

 

So, both the inner and outer shells are visible. The outer shell ring is there, but it's substance is not quite visible. A lot of that is probably due to the fact that the information was spread out over several times the pixel area it should have been due to the poor focus, and the signal when properly focused should be a good deal better for this integration time since it would be more concentrated onto fewer pixels. 

 

Anyway...300s subs at unity gain. Three hours of integration. Zero noise reduction of any kind, this was just integrated, stretched, and I ran HDRMT on the core to reveal it's (very blurry!) structure. Not too bad, I don't think, for less than three hours of integration. I honestly don't know if I'm going to get another chance to try with better focus, because I am honestly not sure I can get better focus, and I am hoping to have this darn RC and all the related accessories sold soon here. :\


Edited by Jon Rista, 17 June 2016 - 03:00 PM.


#47 mikefulb

mikefulb

    Surveyor 1

  • *****
  • Posts: 1,884
  • Joined: 17 Apr 2006

Posted 17 June 2016 - 03:09 PM

That is a perfect test Jon - thank you for sharing.  I had gathered 100 minutes (10 x 600s) with my SXV-H9 at f/5 on my AT8RC recently and it is interesting to compare to your image.  Your image is soft but for the outer shells hopefully that smoothing isn't going to reduce contrast too much in order to get an idea of what the camera can do.  I don't think I will be sacrificing much, if anything, in sensitivity (my SXV-H9 is an original one with 10 e- read noise!).

 

Looks like it ought to be possible to get some fainter details with the ASI1600MM.  Now I just hope the vendor is correct and mine will ship in two weeks!



#48 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 24,854
  • Joined: 10 Jan 2014

Posted 17 June 2016 - 03:18 PM

I am sure the oof blur is blurring out details that should be there, and hurting SNR in the process. Light that should be concentrated on 1-2 pixels is being spread out over 8 pixels or more. It is supposed to be clear again tonight. I am going to try to get some data with SharpCap and see if I can get it better focused. It's just been ridiculous trying to get my focus stuff figured out in SGP. Some 60% or more of the time, the AF routine can't seem to  get consistent results, and the HFD will jump up and down, giving me bunk results. This has been the case both with the temp compensation tool, which is supposed to figure out, accurately, your temperature compensation shift, as well as when running AF on different filters to figure out the filter offsets from L. That should make it possible to focus in L, then switch back to one of the other filters, and be in focus. The inconsistencies have left me with FWHMs around 4-5" or larger, which is ludicrous considering I was getting FWHMs as low as 1.23" with SharpCap, and on one night for over an hour they were as low as 0.9"!! O_o

 

Blarg. Four really nice, clear nights, completely wasted futzing around with wonkey focus routines. *sigh* Incredibly frustrating. Well, at least it is the weekend tomorrow. No reason I can't stay up until 4am and just manually keep things focused with SharpCap, and maybe I can get some better results. Assuming the seeing holds...it was a bit worse last night, and transparency was poor. Tonight it looks like higher jetstream speeds and more bad layers according to MeteoBlue. They are saying .3-.4" seeing, but with the higher jetstream velocity and more bad layers, I doubt it will be that good, probably more like .8-1" if past experience says anything. 


Edited by Jon Rista, 17 June 2016 - 03:20 PM.


#49 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 24,854
  • Joined: 10 Jan 2014

Posted 17 June 2016 - 03:43 PM

Here is an example of how noise remains higher in the glow areas after calibration:

 

ASI1600-CalibratedGlowNoise.jpg

 

Note the lower noise near the left and the higher noise near the right. The background level is the same...the noise however is not. At 600s it would be about twice as bad, and it just gets worse with longer subs. 

 

This is after the integration of 35x300s subs. The difference is not huge, which is good, but even with 300s subs, it is still there. 


Edited by Jon Rista, 17 June 2016 - 03:44 PM.


#50 FirstC8

FirstC8

    Apollo

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

Posted 17 June 2016 - 04:46 PM

Thanks Jon for the visuals, I can see what you meant by noise and signal now.

OK consider unwanted signals as a form of noise isn't a real bad idea though, just a type of removable noise.

If the 1600 chip exposure time should be limited to 300s, as a result of amp glow, then there is one disadvantage compared to the 8300.


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