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

Fast telescopes v/s high resolution

  • Please log in to reply
188 replies to this topic

#26 Noah4x4

Noah4x4

    Vanguard

  • *****
  • Posts: 2470
  • Joined: 07 Apr 2016
  • Loc: Colchester UK

Posted 03 November 2018 - 05:12 AM

Don't forget that the ICX814 has very high Q.E. and relatively low read noise. I've seen numerous images at scales between 0.2 and 0.3 arcsec/px with the ASI1600. It has ultra low read noise but more average Q.E. and DR and the results were pretty good. 

The Atik Horizon uses the same sensor as the ASI1600. As I mentioned earlier (IMHO) it is superb on Hyperstar at f/1.9 on my Evolution 8". However, I am a mere novice at imaging and don't know anything beyond the fact  it produces (IMHO) nice pictures in mere seconds beyond the quality of anything I ever achieved in two years with my 24 megapixel DSLR and long traditional exposures. 

 

What has surprised me is that the camera experts have not commented on the merits (or otherwise) of my end to end 4k UHD system on Hyperstar (Post #17). Interestingly, I visited Atik a couple of weeks ago to pick up a warranty repair and I got the impression from their technical support team that they had not heard of anybody else hooking up a 16 megapixel large sensor small pixel low read noise cooled CMOS camera on Hyperstar to a 4k UHD monitor by USB3 and Thunderbolt display port via an equally high resolution Intel NUC with Iris 640 Plus Graphics.  I put this together instinctively (and extravagantly as it wasn't cheap) as I couldn't see the merit of having a high-resolution camera and powerful graphics mini-computer (originally purchased to dabble with remote/wifi)  without a commensurate 4k display.  Hence, upgrading the display to a 4k UHD monitor was just another step into the usual money pit that is astronomy. But have I stumbled on a technological way that novices can succeed in producing great results without polar alignment, wedges, autoguiding; instead using CMOS camera advantages and a fast set up?  Frank, you have the Ph.D in optical Sciences. Any thoughts? 



#27 freestar8n

freestar8n

    Vendor - MetaGuide

  • *****
  • Vendors
  • Posts: 8695
  • Joined: 12 Oct 2007

Posted 03 November 2018 - 05:41 AM

Hi-

 

In theory - a large aperture f/2 system will have exactly the same diffraction limited performance as an f/10 system - and with small enough pixels they would perform identically.  But in practice, it is hard to have a very fast system show the same performance as a slow one - with spherical and chromatic aberration being the main problems - along with design tolerances, collimation - etc.

 

So - in practice if you want really high res detail you are probably better off with a slower system.  And there is always a benefit of pixels that are very small in arc-seconds - if your fwhm's are in the low 1" arc-seconds.

 

That doesn't mean you need to go to f/40, though planetary work often does go that far - and with pixels around 0.1" or so.  But EdgeHD11 f/10 with 3um pixels is a good combo if the seeing, focus, and guiding are good.

 

Frank


  • OleCuss and 555aaa like this

#28 555aaa

555aaa

    Vendor (Xerxes Scientific)

  • *****
  • Vendors
  • Posts: 1361
  • Joined: 09 Aug 2016
  • Loc: Lynnwood, WA, USA

Posted 03 November 2018 - 12:16 PM

I think Frank makes a really good point about the issues with very fast telescopes in terms of their capability to reach theoretical limits. What I think typically happens is that you are looking at a wider field image in a faster system because usually they are shorter focal length. The image is filled with stars. Say for X aperture these are at a limiting magnitude of 16. People are wowed by how many stars they are and how "pinpoint" they are. Now, take another image that is limited to 16th magnitude with exactly the same performance in terms of resolution, but with 1/4 of the field of view. You see a lot less field stars and they look bigger. In fact, you are seeing exactly the same information, just over a narrower field of view. So the images seem to be 'deeper' in the wider field when in fact they are identical.

 

I think people really neglect the impact of guiding, focusing, and seeing on their system performance. I put together this little video showing some asteroid imaging with a 16" f/10 Meade SCT with a KAI-11002 sensor binned to about 1.4 arc seconds per pixel. There are six individual 180 second exposures in this video. You can see that a couple of them have dead on guiding, where the star FWHMs are about 2.8", and a couple have not as good guiding where the FWHM is probably about 3.5". The tool I have, does PSF fitting to the star images so it is very good at measuring and fitting. You can see in this video that with the best guiding, there are field stars down to 21st magnitude which are detectable, but with the 0.7" worse FWHM, I lose almost a magnitude of "equivalent sensitivity" (magnitude for the same SNR).

 

https://www.youtube....h?v=V-xoFumXprw

 

The downside I have in my system is that sometimes my target isn't in the FOV and if I had a "faster" system really meaning wider angle at the same aperture, there would be more total information and I would bin less. But I would still have to achieve the same resolution optically across the image AND the same accuracy in guiding. There is no relief for guiding performance as you go to a wider field in terms of sensitivity. That is a myth. What happens is that you end up with a more abberrated system with more blurry stars, but you have a lot of them and so no one notices the reduction in sensitivity.

 

People here talk a lot about their guiding performance but I rarely see actual measures of their star FWHM.


Edited by 555aaa, 03 November 2018 - 12:18 PM.


#29 jhayes_tucson

jhayes_tucson

    Fly Me to the Moon

  • *****
  • Posts: 7021
  • Joined: 26 Aug 2012
  • Loc: Bend, OR

Posted 03 November 2018 - 01:23 PM

Noah4x4,

I have to disagree with the guy with a Ph.D in optical Sciences (lol.gif lol.gif ).  There is not always "a benefit of pixels that are very small" and here's why.  It's reasonable to conclude that a camera with 3 micron pixels might work well with a diffraction limited system in space, but let's look at a real system under a real sky.  In most locations, it is very difficult to experience 1" seeing conditions but that's a good number to use as a guess at the "best possible" conditions using long exposures for many locations.  For a C11 at F/10, the diameter of a 1" blur spot in the focal plane will be 13.5 microns.  The sampling rate for 3 micron pixels will be 4.5 samples across the spot.  With the C14, F/11 system that the OP asked about, the spot size will be about 19 microns so the same camera would provide a sampling rate of 6.3x across the spot.  So under the very best conditions, a C11 will be sampled by 4.5x and a C14 will be sampled at 6.3x.  Under most common conditions, the sampling rate with a 3 micron camera will be even higher.

 

So what's wrong with over-sampling?  First, the SNR of the measurement due to photon noise is given by the square root of the signal strength.  If you compare the signals give by two identical square pixels with dimensions of 3 microns and 9 microns, the 9 micron pixel will produce a signal that is 9x stronger and an image with 3x better SNR.   No problem you say!  The read noise is so low with the 3 micron CMOS camera that we can simply bin the data to get the same result as the camera with the larger pixels.  Sure that works, but a hypothetical 16 Mpx 4096 x 4096 camera with 3 micron pixels then becomes a 1365 x 1365, 1.8Mpx camera with a field of view that's 1/3 that of a 4096 x 4096 camera with 9 micron pixels.  Both the C14 and the C11 have outstanding field correction and can produce very sharp imaging out to a 52 mm circle.  The penalty of using such small pixels is that you waste most of that field.   Ah, but you say, I've got much better sampling so I'll be able to extract more information from the data to produce a sharper image.  Unfortunately, that's unlikely to work for long exposure imaging so let's consider that issue next.

 

When we talk about sampling, we are really talking about how much information can be recovered from the image about the original object.  The foundation of sampling theory goes back to Fourier theory and for a diffraction limited optical system it's pretty easy to work out the optimum sampling rate (I'll spare the math here.)  In the case of a system with atmospheric blurring, it's a bit harder to come up with an exact value, but it's relatively easy to show that a rough rule of thumb using about 2-3 pixels across the point spread function (PSF) of the optical system works well.  In the case of an atmospherically blurred system, the form of the PSF becomes close to something called a Moffat distribution.  The Moffat distribution does a better job of describing the wings of a blurred star profile than a simple Gaussian function and you can read about it here:

 

https://en.wikipedia...at_distribution

https://www.ltam.lu/.../star_prof.html

 

Whether the distribution is Moffat or Gaussian is a minor point in determining the PSF for a seeing blurred system, which is what leads to some uncertainty about the best possible sampling rate.  In reality, achieving a higher sampling rate to better define the form of the PSF will do very little to improve resolution because the frequency content difference between these various functions is extremely small.  Just remember that the form of the image is given by the irradiance of the object convolved with the PSF, which is always a smoothing operation so oversampling the PSF does virtually nothing to improve image resolution.  The only way to effectively improve resolution is to reduce the extent of the PSF itself, which is the goal with short exposure, lucky imaging.  We can put aside the math and simply apply the simple rule of thumb for sampling to the C11 to find that the optimum pixel size should be 4.5- 6.7 microns.  For the C14, it comes out to 6.3 - 9.5 microns.  In my experience at DSW, I've never once been able to reach 1" seeing and I've found that the best sampling limit is virtually always closer to 2x than 3x across an estimated 1" seeing blur disk.  Remember that smaller pixels, produce a lower SNR so unless you are really gaining something in resolution, it is better to error on the side of larger pixels.  On the other hand, if you are using lucking imaging with very short exposures to filter out seeing effects, it's better to error on the side of smaller pixels even though it produces more noise.  You'll just have to gather more data to gain an acceptable SNR level.  Ultimately there is no free lunch.  To get higher sampling (and maybe better resolution,) you sacrifice SNR and vice versa.

 

In my view, if you are doing long exposure imaging, you are better off starting with a sensor that covers the whole field of your telescope with appropriately sized pixels.  For the C11 and C14 at the native Cassegrain focal plane, the optimum pixel size is in the range of 6-9 microns.  I have direct experience on my C14 clearly demonstrating that 9 micron pixels work much better than 6 micron pixels--in terms of resolution, SNR, and FOV.   Unfortunately, these large sensors are expensive so cost is a major consideration and that's the real reason that most folks hang small sensor cameras on these slow telescopes.   It works but it's not optimum.

 

John


Edited by jhayes_tucson, 03 November 2018 - 01:48 PM.

  • Shiraz and OleCuss like this

#30 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 23381
  • Joined: 10 Jan 2014
  • Loc: Colorado

Posted 03 November 2018 - 02:17 PM

John, I am curious to know what you think of a faster f/4 system combined with small pixels. We have talked about f/2, and f/10|11, but not much about f/4 which is kind of in the middle there. I agree with you that going beyond about 0.5"/px image scale does not make much sense for most people, assuming seeing is between 1.5-2" (which seems to be the case for me most of the time). So assuming we stick with that...you could pair a KAF-16200 with an 11" corrected SCT at f/10, or an IMX183 with a 10" corrected Newt at f/4. Aberrations should not be a problem for either system. 

 

The KAF/SCT system would have the following traits:

 

Pixels:       6 micron

Q.E.:         ~60%

Read Noise:   6-10e-

FWC:          40,000e-

Focal Length: 2800mm

Aperture:     280mm

F-ratio:      f/10

Image Scale:  0.442"/px

Field Width:  33.15'

Field Height: 26.52'

 

The IMX/Newt @ Unity Gain system would have the following traits:

 

Pixels:       2.4 micron

Q.E.:         ~84%

Read Noise:   1.7e-

FWC:          4100e-

Focal Length: 1000mm

Aperture:     250mm

F-ratio:      f/4

Image Scale:  0.495"/px

Field Width:  45.35'

Field Height: 30.30'

 

The two systems are pretty comparable here. Diffraction is going to be very similar since the apertures are close. The image scales are very close. The newt system actually has a bigger FoV (by 12x4'), the SCT system has more dynamic range (0.7 stops). In terms of SNR, the two systems should produce similar SNR in similar time. However, one of these systems is going to be significantly cheaper than the other.

 

SCT: $16,300

 

Scope: $3500

Camera: ~$8000 (give or take)

LRGB Filters: ~$1200 (Astrodon)

NB Filters: ~3600 (Astrodon)

 

Newt: $4700

 

Scope: $1700

Camera: $1000

LRGB Filters: $500 (Astrodon)

NB Filters: $1500 (Astrodon)

 

This is the intriguing question for me. I've been using the IMX183 for a year now at an image scale of 0.82"/px, so a little smaller than the newt system here. So far, it has performed very well. I am able to get very small stars (usually about 2" in the final integration, as I am using 10 minute sub exposures and my seeing is not always ideal, but I've had FWHM measurements down to nearly 1.6" in individual subs).


Edited by Jon Rista, 04 November 2018 - 12:23 PM.


#31 jhayes_tucson

jhayes_tucson

    Fly Me to the Moon

  • *****
  • Posts: 7021
  • Joined: 26 Aug 2012
  • Loc: Bend, OR

Posted 03 November 2018 - 04:37 PM

Jon,

I think that for faster systems, smaller pixels certainly make more sense but you still have to look at the seeing limited blur disk size in the focal plane. Obviously, the requirements for a F/4, 30" scope will be very different than for a F/4, 4" scope.  I think that keeping the angular sampling rate at approximately 0.5"/pixel (as you've done) will provide the best results under most common conditions.  There's no doubt that if you can get a good coma corrector, a fast Newtonian provides a lot of bang for the buck.  Bart has posted a lot of outstanding images with his 16", F/4 Newt' (as I recall) showing how well it works.

 

John



#32 freestar8n

freestar8n

    Vendor - MetaGuide

  • *****
  • Vendors
  • Posts: 8695
  • Joined: 12 Oct 2007

Posted 03 November 2018 - 05:09 PM

I'm not citing credentials as the basis for my claims - I cite evidence and empirical results - along with a fairly common sense explanation.

 

A long focal length is only theoretically equivalent to a short focal length of the same aperture - but theory isn't what determines the final result.  Reality and all the other factors it introduces determines the final result.

 

Sampling theory, especially in the Nyquist sense, involves recovering an underlying signal from discrete samples.  In that context the discrete samples are put through a low pass filter to create a continuous signal - and if you do it right that continuous signal will exactly equal the original continuous signal - even though it was only sampled at discrete points.

 

But that has little application to deep sky imaging, where you sample discretely - and then look at those discrete samples as pixels.  You also use those pixels to find centroids of stars - and align based on those centroids.  And as you align you shift and interpolate to create the stack.  This is all a million miles from simple sampling theory - and instead relies on software and heuristics for the final result.

 

Not to mention the additional challenges of getting a well focused and small, in arc-seconds, star spot in a fast vs. slow system.  It is only in theory that the resulting Moffat PSF would have the same profile in a fast system as a slow one.

 

Does anyone think that Hyperstar with 1um pixels would have exactly the same star profiles as 5um at f/10?  All the aberrations would be identical?  If it's easy to make them equivalent at f/2, why not go to f/1?

 

Read noise is harder to quantify in terms of its impact on an image - but my approach has been to approximate it on a per area basis - in order to compare different pixel sizes.  If you bin N pixels in software, the aggregate read noise will go up as sqrt(N) - so 2um pixels with a read noise of 3e would bin 2x2 to yield 6e.  A cmos camera with 3.8um pixels and 2e read noise would bin to 7.6um pixels with 4e read noise - which is much better than many equivalent larger-pixel ccd's.  So if you were imaging with the cmos camera and found the seeing to be very good - those smaller pixels would be a resolution win.  And if you binned them in software - they would still be a read noise win.

 

But the OP has expressed interest in high res in conditions of good seeing.  So the conclusion is straightforward, and smaller will be better - with diminishing returns at a size that samples the fwhm very well.

 

Frank


  • OleCuss likes this

#33 Noah4x4

Noah4x4

    Vanguard

  • *****
  • Posts: 2470
  • Joined: 07 Apr 2016
  • Loc: Colchester UK

Posted 03 November 2018 - 07:42 PM

Must confess I am lost now as regards the optical science. I just enjoy the nice on screen images that  my 4k UHD system produces. 



#34 OleCuss

OleCuss

    Vanguard

  • *****
  • Posts: 2336
  • Joined: 22 Nov 2010

Posted 03 November 2018 - 08:01 PM

. . .  I just enjoy the nice on screen images that  my 4k UHD system produces. 

For us amateurs, that is the ultimate test as to whether you have a good system.

 

If you enjoy doing the imaging and you enjoy the results - then it's great!


  • Noah4x4 likes this

#35 james7ca

james7ca

    Fly Me to the Moon

  • *****
  • Posts: 7204
  • Joined: 21 May 2011
  • Loc: San Diego, CA

Posted 04 November 2018 - 02:06 AM

One issue that seems to have had little discussion here is pixel count. IMO, mega pixels matter in terms of producing a compelling image and this is one reason why recent CMOS sensors like the IMX183 or even the smaller IMX178 have allowed users to produce surprisingly good images for not too much money.

 

It's certainly true that image scale matters in terms of sampling and resolution, but if you don't have a sufficient number of pixels and consequently a sufficient number of resolvable elements in the final result then it becomes that much more difficult to produce an impressive image. For most of us, it's the final image that matters and sometimes raw pixel count can be just as important as a resolution or a FWHM measured in arc seconds.

 

That said, I have a long history here on CN of imaging at 0.75 arc seconds per pixel and have posted many images that were done at that scale using a 5" Tele Vue NP127is with the Sony IMX178 sensor that has 6 million pixels at 2.4 microns each. Recently, however, I've been doing more and more imaging using small, very short focus lenses that can produce a very large field when combined with the relatively inexpensive IMX183 sensor that has a modest physical size but a very high 20 million pixel count. In fact, even the 6 mega pixel IMX178 is capable of producing images with small lenses/scopes that can come close to matching the visual impact that use to require larger and more expensive systems.

 

True, it's a trade off since high pixel counts generally mean smaller pixels and thus less signal per pixel for any given integration time, but that's where these "faster" and shorter focal length systems are showing their strengths, at least when combined with sensors similar to the IMX178 and IMX183.

 

I also do a fair amount of planetary work and while you certainly don't need a large pixel count to image Jupiter, Saturn, or Mars I think sensors like the IMX178 and IMX183 have produced something of a revolution in lunar imaging since they can reach critical sampling at around f/10 while simultaneously providing amazingly wide and high resolution fields with a suitable flat field instrument (like Celestron's EdgeHD). This obviously plays toward similar lucky imaging techniques for deep space imaging which is another area where I have been active here on CN.

 

Given the above, it is my opinion that much of the future in astrophotography belongs to smaller, shorter optical systems combined with sensor with smaller pixels. Of course, the "ultimate" system might always be the largest scope that you can get combined with a large pixel camera but that can be economically impractical for the vast majority of people, particularly when you consider the current trend toward smaller and smaller pixels as being driven by market forces outside of astrophotography.

 

Certainly, there are lots of differing ways to make compelling images and in a practical sense the tools that are available today generally outstrip most of our own talents or capabilities, so I sometimes think there is a bit too much debate and not enough doing here on CN (but I guess that is why we have "cloudy nights" -- as in overcast wink.gif  ).


  • Jon Rista, Noah4x4 and Adun like this

#36 freestar8n

freestar8n

    Vendor - MetaGuide

  • *****
  • Vendors
  • Posts: 8695
  • Joined: 12 Oct 2007

Posted 04 November 2018 - 03:17 AM



 

Certainly, there are lots of differing ways to make compelling images and in a practical sense the tools that are available today generally outstrip most of our own talents or capabilities, so I sometimes think there is a bit too much debate and not enough doing here on CN (but I guess that is why we have "cloudy nights" -- as in overcast wink.gif  ).

Well - here's my m8 bowtie at 0.2" per pixel with cmos:

 

get.jpg?insecure

 

Here it is with a bit less detail at f/7 with larger pixel ccd and longer exposure: 

 

get.jpg?insecure

 

Neither image has been processed other than applying levels.  No sharpening.

 

The OP is interested in high res and detail with skies around 1" fwhm.

 

I did a comparison of hyperstar with f/10 here:

 

https://www.astrogee...20/ngc2020.html

 

The hyperstar goes very deep much faster than f/10, but the difference in resolution is significant.

 

This result shows star profiles at 0.28" per pixel, and the fwhm is only about 4 pixels:

 

https://www.astrogee...88/ngc6188.html

 

Here is Eta Car with fwhm around 1.4" and 0.57" per pixel - and you can see how blocky the stars are:

 

https://www.astrogee...car/etacar.html

 

So - if your skies are 3" or you are just using short focus lenses that don't have much resolution, 0.2" per pixel may seem crazy.  But if you have good seeing and things are working well - and you are striving for high res - 0.5" per pixel would represent a crazy limitation on what you could otherwise achieve - and you should go much smaller.

 

Frank


  • Noah4x4 likes this

#37 Noah4x4

Noah4x4

    Vanguard

  • *****
  • Posts: 2470
  • Joined: 07 Apr 2016
  • Loc: Colchester UK

Posted 04 November 2018 - 03:59 AM

................it is my opinion that much of the future in astrophotography belongs to smaller, shorter optical systems combined with sensor with smaller pixels. 

 

Certainly, there are lots of differing ways to make compelling images and in a practical sense the tools that are available today generally outstrip most of our own talents or capabilities, so I sometimes think there is a bit too much debate and not enough doing here on CN (but I guess that is why we have "cloudy nights" -- as in overcast wink.gif  ).

My gear certainly "outstrips my talents and capabilities" in that I have suddenly leapt from being pretty useless with a DSLR into producing credible images albeit I would never win any awards. Hyperstar means I no longer struggle with wedge, polar alignment and guiding. The huge FOV it offers means my target is always present. Tiny pixels and high resolution means I can <zoom> and hence enlarge, swoop and centre it and results are fairly jaw-dropping, albeit the 'experts' would remain critical.

 

I think manufacturers are recognising that there is there are now two distinct groups. There are the experts that so elegantly debate stuff here and those (like me) that for much of the debate havn't a clue what they are talking about. But before, it was all must master the complexities and dark arts of photography or suffer frustration. Now, technology, high pixel count and faster equipment permits more of us doing more and less debating. That can't be a bad thing. It won't diminish the status of the experts, it is simply that they will have to work harder to maintain the quality differential as ever cheaper technology allows novices to produce more for less effort.


  • reddog1972 likes this

#38 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 23381
  • Joined: 10 Jan 2014
  • Loc: Colorado

Posted 04 November 2018 - 12:27 PM

My gear certainly "outstrips my talents and capabilities" in that I have suddenly leapt from being pretty useless with a DSLR into producing credible images albeit I would never win any awards. Hyperstar means I no longer struggle with wedge, polar alignment and guiding. The huge FOV it offers means my target is always present. Tiny pixels and high resolution means I can <zoom> and hence enlarge, swoop and centre it and results are fairly jaw-dropping, albeit the 'experts' would remain critical.

 

I think manufacturers are recognising that there is there are now two distinct groups. There are the experts that so elegantly debate stuff here and those (like me) that for much of the debate havn't a clue what they are talking about. But before, it was all must master the complexities and dark arts of photography or suffer frustration. Now, technology, high pixel count and faster equipment permits more of us doing more and less debating. That can't be a bad thing. It won't diminish the status of the experts, it is simply that they will have to work harder to maintain the quality differential as ever cheaper technology allows novices to produce more for less effort.

To distill it all down, it is possible to achieve similar throughput, efficiency and resolution with smaller and more cost-effective systems these days, than used to be possible in the past. While LARGE systems with BIG pixels may still have the ultimate advantage and will still deliver the ultimate creme of the crop in terms of image quality, you can often achieve 80-90% of what a big and expensive system can do, for 1/4, 1/5, or even less cost. That is a big bonus for the average pretty-picture astrophotographer, as you do not need to be willing to sink as much money as a car (and in some cases, a really really nice car) into your hobby anymore. 


  • rockstarbill, GA-HAMAL and Noah4x4 like this

#39 Noah4x4

Noah4x4

    Vanguard

  • *****
  • Posts: 2470
  • Joined: 07 Apr 2016
  • Loc: Colchester UK

Posted 04 November 2018 - 01:31 PM

Time saving is also material with regard to new technologies. In past times I was frequently denied opportunity by clouds rolling in soon after I had set up on wedge, polar alignment etc. The set up process took too long under fickle UK skies. Now, without the weight of the wedge I can carry and keep my scope fully assembled (other than adding camera/Hyperstar). Hence I can usually squeeze in some fun EAA/AP under mixed conditions even though its not quite "Instamatic". That has enhanced my enjoyment. 

 

 I probably represent the "average pretty picture astrophotographer" and whilst I have invested as much money as a small car in this hobby, I didn't actually need to with hindsight. I now have boxes of redundant expensive eyepieces and other unused gear such as wedge, focal reducers etc.  Severe light pollution means EAA/AP is now my only viable way forward in my back yard and it is rare I get opportunity to travel to dark skies. These forums were hence detrimental to my wallet. Hyperstar has made life easy and wish I had discovered it earlier. It was recently called "cheating" in a live broadcast on U -Tube. Fair cop! 



#40 freestar8n

freestar8n

    Vendor - MetaGuide

  • *****
  • Vendors
  • Posts: 8695
  • Joined: 12 Oct 2007

Posted 04 November 2018 - 06:15 PM

it is simply that they will have to work harder to maintain the quality differential as ever cheaper technology allows novices to produce more for less effort.

I make a point of doing work with mid-range equipment - which specifically goes against the general notion that you need a premium mount to get good results.  Right now the EdgeHD11 on CGX costs around $5k as a combo - and that is less than half the price of many high end mounts by themselves.  And once you have it working well, there is no particular burden in imaging at f/10 for high res. results.  All it takes is software, experience, and the right technique.

 

So - I'm all for novices getting high end results with mid-range equipment and low cost cameras.  I'm trying to make it happen more by describing my approach and why I do things a certain way - and backing it with regularly posted, minimally processed results - along with free software and other analysis tools.

 

But in terms of high res. results, which the OP is after, I don't see it happening very soon with large aperture, fast systems at low cost - because it is much harder to have a very fast system diffraction limited across the spectrum.  And you would need perhaps 1um pixels.  But I would be happy to be proven wrong if a system like that comes along and can be shown to yield fwhm's in the low 1" - as is possible now with EdgeHD11 at f/10 - at relatively low cost - and with no significant extra effort.

 

Frank


  • OleCuss likes this

#41 TareqPhoto

TareqPhoto

    Skylab

  • -----
  • Posts: 4411
  • Joined: 20 Feb 2017
  • Loc: Ajman - UAE

Posted 06 November 2018 - 12:27 AM

Nice thread or topic



#42 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 23381
  • Joined: 10 Jan 2014
  • Loc: Colorado

Posted 06 November 2018 - 12:37 AM

Jon,

I think that for faster systems, smaller pixels certainly make more sense but you still have to look at the seeing limited blur disk size in the focal plane. Obviously, the requirements for a F/4, 30" scope will be very different than for a F/4, 4" scope.  I think that keeping the angular sampling rate at approximately 0.5"/pixel (as you've done) will provide the best results under most common conditions.  There's no doubt that if you can get a good coma corrector, a fast Newtonian provides a lot of bang for the buck.  Bart has posted a lot of outstanding images with his 16", F/4 Newt' (as I recall) showing how well it works.

 

John

Oh sure, with a 30" aperture. That is so far beyond most peoples means, though... Even the wealthy, that is really expensive. Heck, 30" @ f/6 is wicked expensive. 

 

Getting a good corrector is the difficulty. I've been looking for the right 8-10" newt setup for a while now, and the corrector is the sticking point. I've talked to quite a few people, and a lot of those who stuck with their newts have gone through just about every corrector on the market to find a good one. Seems the larger Paracorr is very good, but you lose a bit of speed (f/4.5). 



#43 jhayes_tucson

jhayes_tucson

    Fly Me to the Moon

  • *****
  • Posts: 7021
  • Joined: 26 Aug 2012
  • Loc: Bend, OR

Posted 06 November 2018 - 02:00 AM

Oh sure, with a 30" aperture. That is so far beyond most peoples means, though... Even the wealthy, that is really expensive. Heck, 30" @ f/6 is wicked expensive. 

 

Getting a good corrector is the difficulty. I've been looking for the right 8-10" newt setup for a while now, and the corrector is the sticking point. I've talked to quite a few people, and a lot of those who stuck with their newts have gone through just about every corrector on the market to find a good one. Seems the larger Paracorr is very good, but you lose a bit of speed (f/4.5). 

 

Jon,

I was simply making a comparison.

 

Still there are a lot of amateurs who operate very large scopes.  20" - 28" scopes on DOB mounts are not unheard of.  Ken Crawford operates a 24" scope and there are more folks that you might realize who are willing to dump $100k or more into this hobby.  Yes, an AP capable 30" is unreasonable for 99% of us but there are quite a few scopes operating in the range between 16" and 30".  CDKs are certainly a good way to go as the aperture gets larger.

John



#44 freestar8n

freestar8n

    Vendor - MetaGuide

  • *****
  • Vendors
  • Posts: 8695
  • Joined: 12 Oct 2007

Posted 06 November 2018 - 02:21 AM

Jon,

I think that for faster systems, smaller pixels certainly make more sense but you still have to look at the seeing limited blur disk size in the focal plane. Obviously, the requirements for a F/4, 30" scope will be very different than for a F/4, 4" scope.  I think that keeping the angular sampling rate at approximately 0.5"/pixel (as you've done) will provide the best results under most common conditions.  There's no doubt that if you can get a good coma corrector, a fast Newtonian provides a lot of bang for the buck.  Bart has posted a lot of outstanding images with his 16", F/4 Newt' (as I recall) showing how well it works.

 

John

I think that both of you (John and Jon) tend to refer to what is best for "most common conditions."  That is perfectly fine - but should not be the mindset of someone who strives for max detail and resolution - such as the OP in his first post.  If you aim to optimize everything for typical conditions at a given site, you will be stuck with giant pixels on those rare nights when the seeing is exceptional.  And those are the nights that are most valuable - if you strive for high resolution.

 

And if you do have small pixels but the seeing is not exceptional - you can always bin or low pass after the fact - with little or no penalty as long as read noise is not a strong factor - and nowadays it usually isn't.  And even if the image is a bit noisier due to smaller pixels - you can image longer to achieve the same result - after binning.  Small pixels do not represent a fundamental limit to what can be achieved - whereas large pixels do.

 

With large pixels (and in many situations, 0.5" is quite large) you are just stuck with what you got - and the great seeing and amazing optics you paid for - are being wasted.

 

As always - if you don't actually care about max detail, or if you seeing is never great - 0.5" may be a decent minimum pixel size.  But the OP in his first post mentioned a desire for high res, and seeing around 1".  To me - both the mindset and the seeing point to 0.2" pixels.

 

Frank


  • dhaval likes this

#45 dhaval

dhaval

    Vendor

  • *****
  • Vendors
  • topic starter
  • Posts: 1673
  • Joined: 21 Jul 2008
  • Loc: Round Rock, TX

Posted 06 November 2018 - 10:37 AM

I think that both of you (John and Jon) tend to refer to what is best for "most common conditions."  That is perfectly fine - but should not be the mindset of someone who strives for max detail and resolution - such as the OP in his first post.  If you aim to optimize everything for typical conditions at a given site, you will be stuck with giant pixels on those rare nights when the seeing is exceptional.  And those are the nights that are most valuable - if you strive for high resolution.

 

And if you do have small pixels but the seeing is not exceptional - you can always bin or low pass after the fact - with little or no penalty as long as read noise is not a strong factor - and nowadays it usually isn't.  And even if the image is a bit noisier due to smaller pixels - you can image longer to achieve the same result - after binning.  Small pixels do not represent a fundamental limit to what can be achieved - whereas large pixels do.

 

With large pixels (and in many situations, 0.5" is quite large) you are just stuck with what you got - and the great seeing and amazing optics you paid for - are being wasted.

 

As always - if you don't actually care about max detail, or if you seeing is never great - 0.5" may be a decent minimum pixel size.  But the OP in his first post mentioned a desire for high res, and seeing around 1".  To me - both the mindset and the seeing point to 0.2" pixels.

 

Frank

Frank,

I think that is a fair point that you make - being able to use the best nights to their maximum and still having the option to do decent imaging on those nights when the seeing is middling. 

 

I am hoping to be able to get out to the observatory and get everything set up. The proof will be in the pudding as they say! 

 

CS!



#46 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 23381
  • Joined: 10 Jan 2014
  • Loc: Colorado

Posted 06 November 2018 - 11:15 AM

I think that both of you (John and Jon) tend to refer to what is best for "most common conditions."  

Yeah, I guess this is true. I agree, if you want the best resolution possible, and have good seeing often enough, then configuring a system for best conditions is better.



#47 TareqPhoto

TareqPhoto

    Skylab

  • -----
  • Posts: 4411
  • Joined: 20 Feb 2017
  • Loc: Ajman - UAE

Posted 06 November 2018 - 11:22 AM

Yeah, I guess this is true. I agree, if you want the best resolution possible, and have good seeing often enough, then configuring a system for best conditions is better.

Great, so it is not about 10Micron/AP mounts and TEC/Tak scopes and FLI/SBIG cameras and Astrodon filters?



#48 Jon Rista

Jon Rista

    ISS

  • *****
  • Posts: 23381
  • Joined: 10 Jan 2014
  • Loc: Colorado

Posted 06 November 2018 - 11:27 AM

Great, so it is not about 10Micron/AP mounts and TEC/Tak scopes and FLI/SBIG cameras and Astrodon filters?

Well, if you really want to get the best resolution possible and have good seeing, then what Frank is saying here (and I agree) is that you would want a large scope (big aperture), with moderately longer focal length (which will minimize aberrations), paired with a camera that has small pixels. To get the most out of a big scope like that, you would want good filters...so either AstroDon or Chroma I would think. And to handle a big scope like that, you would need a mount capable of handling such a large payload reliably and smoothly. 

 

You would still likely need a big mount. And you would still need an expensive scope. And you would still need good filters. The only thing that really changes is the camera...instead of one that has 9 micron pixels, you might end up with one that has 4 micron or even 3 micron pixels, so that you are sampling 1" stars well.  You would be able to save money on filters...even with AstroDon or Chroma, they would be able to be much smaller with current small pixel CMOS sensors. But those savings would in turn go into the scope and/or mount...

 

If you really do have good seeing on an often enough basis, and you really do want to maximize your resolution, then you are still going to want to spend the money.

 

That said, with tiny pixels, it IS possible to get good resolution with smaller systems. It may not be the "best possible" resolution, but you can indeed still get very good resolution with much, much more affordable systems like the newt system I described above. You would be able to spend significantly less money. You would not achieve the same potential as a higher end system, but you should be able to get a majority of the way there.


Edited by Jon Rista, 06 November 2018 - 11:28 AM.


#49 OleCuss

OleCuss

    Vanguard

  • *****
  • Posts: 2336
  • Joined: 22 Nov 2010

Posted 06 November 2018 - 11:31 AM

Great, so it is not about 10Micron/AP mounts and TEC/Tak scopes and FLI/SBIG cameras and Astrodon filters?

It's not really about any of those.

 

It is about enjoying what you are doing.

 

If the high-end stuff is what it takes to make one happy then I'm all for it.  If one can be happy with lesser equipment then that is at least as awesome.


  • NorthField likes this

#50 TareqPhoto

TareqPhoto

    Skylab

  • -----
  • Posts: 4411
  • Joined: 20 Feb 2017
  • Loc: Ajman - UAE

Posted 06 November 2018 - 11:36 AM

Well, if you really want to get the best resolution possible and have good seeing, then what Frank is saying here (and I agree) is that you would want a large scope (big aperture), with moderately longer focal length (which will minimize aberrations), paired with a camera that has small pixels. To get the most out of a big scope like that, you would want good filters...so either AstroDon or Chroma I would think. And to handle a big scope like that, you would need a mount capable of handling such a large payload reliably and smoothly. 

 

You would still likely need a big mount. And you would still need an expensive scope. And you would still need good filters. The only thing that really changes is the camera...instead of one that has 9 micron pixels, you might end up with one that has 4 micron or even 3 micron pixels, so that you are sampling 1" stars well.  You would be able to save money on filters...even with AstroDon or Chroma, they would be able to be much smaller with current small pixel CMOS sensors. But those savings would in turn go into the scope and/or mount...

 

If you really do have good seeing on an often enough basis, and you really do want to maximize your resolution, then you are still going to want to spend the money.

 

That said, with tiny pixels, it IS possible to get good resolution with smaller systems. It may not be the "best possible" resolution, but you can indeed still get very good resolution with much, much more affordable systems like the newt system I described above. You would be able to spend significantly less money. You would not achieve the same potential as a higher end system, but you should be able to get a majority of the way there.

Yes, that is why i am still happy with my QHY mono camera and cheap filters and i am slowly upgrading to Astrodon but at 1.25" only, and about large aperture or scope i really don't know about that yet, in my mind for my small mount i was thinking about 8" or 10" RC maximum, but the trap of Facebook there are members told me to forget this small and go directly to 14" or 16" RC, i still don't know if 14" RC or 16" are ultimate scopes to be used over 8"-12" RC, and i already planned on a solid mount, so mount and scopes are likely covered or known, but still not sure about this camera pixel resolving story yet.

 

I saw a lot of nice images from ASI1600 and also my camera or similar ones, but i still read here and there about CCD vs. CMOS, so i don't know what i am missing there, and definitely when i see scopes and mounts and cameras i mentioned above combined then i know "Blindly" that the results are outmatched or in another level completely, so i was thinking are cheaper affordable one any good even with so nice amazing results or we only put those high end full setup as minimum and standard?




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