Go Back   IceInSpace > Equipment > Astrophotography and Imaging Equipment and Discussions

Reply
 
Thread Tools Rate Thread
  #41  
Old 26-04-2010, 07:20 AM
Hagar (Doug)
Registered User

Hagar is offline
 
Join Date: Oct 2006
Location: Australia
Posts: 4,646
This is one of those discusions that will go round and round in circles for days and really achieve nothing. I am surprised the big boys haven't bought into this one yet with all the specs of mono cameras to justify the debate. Robins examples above are like my examples and Marcs, They really show nothing which can be considered conclusive. The images taken with the mono have a high screen stretch while the OSC have a medium screen stretch which does make them appear darker.
To carry out a test we really need two cameras with the same size CCD, same pixel size, same scope, same night and software and then we might be able to get something which is somewhat conclusive.

The question was asked about the ability of the OSC camera to perform and I think it has been demonstrated reasonably well. There is a few things which must be remembered about OSC cameras which is often overlooked in a discusion like this. The first is the two different sensors can only possibly collect the photons which happen to hit the CCD and in the case of a OSC camera software in the camera does some interpolation to fill all the pixels with a mix of the three different colours and with modern camera CCD's this is done quite effectively. The mono camera on the other hand fills the pixels with actual photons.
Colour cameras are not suitable for scientific work such as photometry for this very reason.

I think what has been demonstrated here certainly stands the OSC camera well for general imaging and for producing pretty pictures which is what most of us use a camera for.
It is a granted fact that the mono camera will be more accurate in it's depiction of an object but then we take the resultant images into photoshop or similar and often completely change the pixel values from those collected. Accuracy just fell out the door.
I think (Just my thoughts) both cameras have a place in astrophotography and in the hands of a skilled operator can produce quite comparable results. I don't believe the void between the mono and OSC is anything like as great as it is made out to be and at no point should the OSC be discounted because of sensitivity reasons as I (again just my thoughts)believe the cameras are quite comparable and in fact OSC cameras are very simple to use and convenient from a time point of view.
Reply With Quote
  #42  
Old 26-04-2010, 11:01 AM
marki's Avatar
marki
Waiting for next electron

marki is offline
 
Join Date: Mar 2008
Posts: 2,427
Doug I dont think anybody is discounting OSC here but you are right in that this discussion will go around in circles. I personally find mono easier to use and have got better results using them from my suburban home. But like all things it depends on the user and the level of equipment they have, the camera is only one part of the imaging train. The best way to compare as you have suggested would be using identical camera's with the only difference being the bayer array on the same scope under the same conditions etc etc. Mono's are more sensitive that is a given...whether that translates into the final picture is another thing all together.

Mark
Reply With Quote
  #43  
Old 26-04-2010, 03:58 PM
Bassnut's Avatar
Bassnut (Fred)
Narrowfield rules!

Bassnut is offline
 
Join Date: Nov 2006
Location: Torquay
Posts: 5,064
IMO the QE difference alone is a killer. If a OSC has a third of the QE then you must expose 3 times longer for the same flux and still you have a worse S/N ratio (3 times the dark noise). This wouldnt matter so much on bright objects though, and carefull dark subtraction.

Ive never seen an explanation why the QE is so low on OSCs. An external filter on a mono has a pass of some 95% or higher at the filtered frequency so dosent affect QE much of the sensor, but the OSCs on-chip filters seem to dump the QE, is the filter "film" or whatever much less efficient on a OSC?, I guess so, must be.

The res difference is something I puzzle about too. OK one quarter of the pixels are used for each colour (approx) on a OSC, but it seems to me, the interpolation is pretty good and its acurracy would be better than seeing at long FLs anyway, so your sort of getting extra exposure for free, albiet a carefully crafted processing trick. Im hard pressed to tell the res difference between a colour and mono small-pixel sensor final RGB pic for the average longish FL pic, despite the experts banging on about the theoretical difference.

NB is just woefull on a OSC, as you would expect, with the internal filter and then another filter at a different frequency after that. You would have to really not be interested in NB long term if you had a OSC, or put up with insane effort for the odd excursion. Bang goes the "easier" moniker right there, for NB.

I mainly do NB or HaRGB, because in urban skies LRGB is very limiting and requires no moon and far more exposure time for a given result to overcome skyglow (again, bang goes "easier").

Ive sometimes been tempted by a OSC, but ONLY for possible feild use in no moon dark skies, for which indeed, OSC capture is easier on brighter objects.

For urban skies, a OSC limits use to low QE LRGB and the difficulties that come with that without, realistically, the option of superior NB S/N performance, HaRGB at least is becoming more and more popular.

A motorised filterwheel makes the capture difficulty the same as a OSC IMO, and the extra processing required for seperate filter subs is so simple and automatic, that the extra work compared to the hrs spent in PS, is trivial. You can take RGB rotating sequences to assure a result if the weather turns bad.

So, in short, if you want a simple mobile dark sky rig for bright LRGB, the OSC is the go. For everything else, monos the answer.
Reply With Quote
  #44  
Old 26-04-2010, 03:59 PM
ericwbenson (Eric)
Registered User

ericwbenson is offline
 
Join Date: Sep 2009
Location: Adelaide, Australia
Posts: 209
Man this subject comes up a lot!

In trying to determine "which one is better" it really helps to have a clear question like "better at what?"

OSC was originally developed for video applications. All other aspects of color sensors are compromises when compared to their mono counterparts. Since color video is rarely a requirement in astronomy (even the best high res planetary imagers use filter wheels with mono cameras) there aren't any technical reasons to use a OSC, except cost at first look, but I believe that is a fallacy. What does a mono sensor give you that a OSC can't?
Higher QE = more signal, reviewing the QE graphs and filter transmission curves tells the story.

Higher QE is the same as getting a bigger telescope with the same focal length with no added mass for the mount to handle. How is that ever a bad thing? What is the alternative? A bigger, faster OTA and a beefier mount to hold it? I think any cost savings of the OSC just got swallowed by the shipping cost of the new mount

The resolution thing is not a big deal if you have some freedom with focal length. Anyways you can think of OSC pixels as 2x2 blocks, hence your effective pixel size is 14.8 um (for the popular 7.4 um OSCs), and use this to calculate the correct sampling.

OSC astro images are not as easy to process as some mono frames, especially in urban conditions, where ironically most OSC are used?!? The effects of light pollution make flat fielding vital, but it's harder to do right in color land.

And of course the kicker is narrowband. Narrowband has the most bang for the buck in light polluted skies (again the irony) but check the QE of a OSC at 653 nm (HII wavelength), not pretty.

So IMHO if someone is interested in taking pretty pics of the brighter nebulae, a DSLR is way cheaper (what's my ancient 350D worth these days?), does an pretty good job, and easier to justify to your better half than an astro OSC.

EB
Reply With Quote
  #45  
Old 26-04-2010, 04:12 PM
ericwbenson (Eric)
Registered User

ericwbenson is offline
 
Join Date: Sep 2009
Location: Adelaide, Australia
Posts: 209
Quote:
Originally Posted by Bassnut View Post
Ive never seen an explanation why the QE is so low on OSCs. An external filter on a mono has a pass of some 95% or higher at the filtered frequency so dosent affect QE much of the sensor, but the OSCs on-chip filters seem to dump the QE, is the filter "film" or whatever much less efficient on a OSC?, I guess so, must be.
Fred, I think the Bayer filter was engineered to match the color response of the eye, which is poor away from green, and of course it's probably cheaper to make them low transmission so a happy coincidence for the manufacturer. So you could say they are bad for astronomers on purpose!

EB
Reply With Quote
  #46  
Old 26-04-2010, 04:25 PM
Bassnut's Avatar
Bassnut (Fred)
Narrowfield rules!

Bassnut is offline
 
Join Date: Nov 2006
Location: Torquay
Posts: 5,064
OK, that makes sense EB. Your post is much on the same page as mine, nice back up, that should do the trick, seems pretty clear IMO
Reply With Quote
  #47  
Old 26-04-2010, 04:53 PM
Hagar (Doug)
Registered User

Hagar is offline
 
Join Date: Oct 2006
Location: Australia
Posts: 4,646
Using the figures and specifications Fred and Eric have used here seem to suggest that using a OSC camera for RGB or NB imaging should incur a huge time penalty imaging in light poluted areas or dark sites but in practice this doesn't seem to be the case. In fact the opposite seems to be the case. An image taken with a OSC camera seems to be able to produce very similar results at about 1/3 of the total exposure time of the mono. The only explanation seems to be the incredible interpolation algorithm used to calculate the different colours into the different pixels. On a scientific note the OSC cannot compete but for general pretty picture imaging they seem to cut the mustard pretty well.
I agree the spec sheet for the two sensors do indicate the figures you mention but in reality something happens to render this difference a non event or at least reduce it to very acceptable levels.

Clear as mud Fred
Reply With Quote
  #48  
Old 26-04-2010, 05:31 PM
Bassnut's Avatar
Bassnut (Fred)
Narrowfield rules!

Bassnut is offline
 
Join Date: Nov 2006
Location: Torquay
Posts: 5,064
I know what you mean Doug, I too have seen many excellent OSC pics that seem to buck the math, and indeed cut the mustard very well.

But, S/N is always the key, the essence to the difference. I have noticed OSC pics are invariably of brighter objects, and often, on zooming in, noise is apparent that is higher than in a comparable mono pic. This is nitpicking of course, if the pic looks great at full view, then zooming is just getting technical and difficult for the sake of it.

I suspect that those that take excellent more difficult pics (eg dimmer) dont admit to the extra time they spend processing, or they expose longer than with mono.

The point is, for a given result, for all but bright objects OSC must be "harder" and take longer exposures than a mono. If your keen and willing to make the effort for the same result, then fine, but the mantra "easier" is a myth.

Im not dumping on OSC, for bright objects in dark skies its cheaper, easier and faster, but the limitations otherwise make OSC a transitory purchase unless your not that serious, are short of cash,or are willing to spend much more time in capture/processing.
Reply With Quote
  #49  
Old 26-04-2010, 06:20 PM
multiweb's Avatar
multiweb (Marc)
ze frogginator

multiweb is offline
 
Join Date: Oct 2007
Location: Sydney
Posts: 22,060
Quote:
Originally Posted by ericwbenson View Post
Man And of course the kicker is narrowband. Narrowband has the most bang for the buck in light polluted skies (again the irony) but check the QE of a OSC at 653 nm (HII wavelength), not pretty.
Nup, Has never been an issue (at all). Sii is another story but even for a mono it's demanding.

Quote:
Originally Posted by ericwbenson View Post
So IMHO if someone is interested in taking pretty pics of the brighter nebulae, a DSLR is way cheaper (what's my ancient 350D worth these days?), does an pretty good job, and easier to justify to your better half than an astro OSC.
I don't think so. A cooled OSC wil perform better with less fuss.
Reply With Quote
  #50  
Old 26-04-2010, 07:13 PM
ericwbenson (Eric)
Registered User

ericwbenson is offline
 
Join Date: Sep 2009
Location: Adelaide, Australia
Posts: 209
Quote:
Originally Posted by multiweb View Post
Nup, Has never been an issue (at all). Sii is another story but even for a mono it's demanding.

Are you imaging nebula with strong HII emission? This comparison is fairly dramatic: http://helixgate.net/XMvsXCM2.html

EB
Reply With Quote
  #51  
Old 26-04-2010, 07:32 PM
multiweb's Avatar
multiweb (Marc)
ze frogginator

multiweb is offline
 
Join Date: Oct 2007
Location: Sydney
Posts: 22,060
Quote:
Originally Posted by ericwbenson View Post
Are you imaging nebula with strong HII emission? This comparison is fairly dramatic: http://helixgate.net/XMvsXCM2.html

EB
I agree with the comparison in this article but it is a sampling issue (bayer vs. mono) ie 1 in 4 instead of 1 in 1 not a QE issue. You mentioned poor QE of OSC for a particular bandwidth. OSC can pick 656nm Ha quite well although it will be only in red so the resolution won't be there but it also depends on the image scale. On widefield you won't feel the pinch. At long FL on M1, well yeah of course. Haven't had any issues with Ha to date and I do a fair bit of NB from home. Oiii works very well as it falls both in 1xblue and 2xgreen, so very very fast. Sii is another story.... tedious.
Reply With Quote
  #52  
Old 26-04-2010, 08:39 PM
darrellx's Avatar
darrellx (Darrell)
Registered User

darrellx is offline
 
Join Date: Dec 2007
Location: Kulgun, Queensland
Posts: 278
Hi All

Well I do appreciate everyones response to my question(s). Rather than seeking a comparison of OSC to Mono CCDs, I was really seeking some insight into any differences in processing the raw data. It sounds like there is not that much of a difference with what I do with my planetaries - stack each channel; do a bit of processing; combine; then do some more processing.

I suppose that the alignment of the separate channels is not that difficult. I have done some practices in PS where I have aligned multiple layers. And getting the layers to align will get better with practice. Aligning a series of say, Jupiter or Saturn will be different to aligning a series from say any DSO that isn't so "distinct" and with fuzzy boundaries.

I have enjoyed reading all the responses though. And it is an interesting topic. I suppose my only contribution I can make to the discussion on which would be the one to get is that since I already have a OSC device (ie my EOS400), I am inclined not to purchase another one.

Darrell
Reply With Quote
  #53  
Old 26-04-2010, 09:02 PM
Bassnut's Avatar
Bassnut (Fred)
Narrowfield rules!

Bassnut is offline
 
Join Date: Nov 2006
Location: Torquay
Posts: 5,064
Darrell. Aligning is typically done in an astro app almost automatically, not PS.
Reply With Quote
  #54  
Old 26-04-2010, 09:27 PM
mill's Avatar
mill (Martin)
sword collector

mill is offline
 
Join Date: Oct 2006
Location: Mount Evelyn
Posts: 2,922
Darrel just get what you want and don't look too much into this religious babble
In the end nobody can deny that with mono you can use a load of nb filters that you are going to have trouble with on OSC camera's.
If it was up to me, i would shoot RGB with a QHY8 and all the rest with an QHY9 mono (these camera's are just an example so don't attack me about the brand name etc).
But i don't have the money to do that so i got a mono camera instead and am happy with it
Reply With Quote
  #55  
Old 26-04-2010, 09:43 PM
darrellx's Avatar
darrellx (Darrell)
Registered User

darrellx is offline
 
Join Date: Dec 2007
Location: Kulgun, Queensland
Posts: 278
Ah, there you go, Fred. I may have been worrying over things for nothing then. With my planetaries, I video; use VirtualDub; Ninox; Registax or K3CCD. But I still use PS (or maybe PaintShop Pro) to combine the four channels into one image - Combine Channel; Combine from RGB. This is the fiddly part - making sure each of the images lines up correctly. Then I finish with PS or AstroArt.

With my current deep sky shots from the EOS400, I essentially go direct to DSS and still finish with PS or AstroArt. But these are already colour so there is no combining involved.

I was under the impression that with mono imaging (deep sky images) you got your three or four "final" images (one from each channel), then overlayed them manually in something like PS as well.

Are you suggesting that I have been doing it the hard way? What astro app will do the aligning for me?

Martin, sure. I am not too concerned over the choice of mono or OSC. I am more interested at this stage in the differences in processing. I think I am like quite a few here - I would like to do some narrow band imaging once I sort out some of the other issues. And yes, money is always an issue.

Thanks
Darrell
Reply With Quote
  #56  
Old 26-04-2010, 10:09 PM
mill's Avatar
mill (Martin)
sword collector

mill is offline
 
Join Date: Oct 2006
Location: Mount Evelyn
Posts: 2,922
If you use Maximdl, it will all be automatic.
Dont know about the other progs
Other people might chime in on this with suggestions and tips.
Reply With Quote
  #57  
Old 27-04-2010, 08:14 AM
Bassnut's Avatar
Bassnut (Fred)
Narrowfield rules!

Bassnut is offline
 
Join Date: Nov 2006
Location: Torquay
Posts: 5,064
All the Astro apps autoalign mono subs as far as I know, even astroart?.

Aligning in PS is time consuming in comparison. Astroaps automatically look for stars to align on and move, rotate, scale and even sometimes warp to align subs all in one go, then if you wish, combine them into a colour image with auto/manual adjustments for colour balance such as white and background black sample selections and RGB weights.

A common practice is to have the astro app align the 4 LRGB sub stacks to each other, then export to PS for colour combining, which is easy since the are already aligned.

Apart from the freebies I havent used (im pretty sure they do too), CCDsoft,CCDstack, Registar, DL.
Reply With Quote
  #58  
Old 27-04-2010, 09:13 AM
darrellx's Avatar
darrellx (Darrell)
Registered User

darrellx is offline
 
Join Date: Dec 2007
Location: Kulgun, Queensland
Posts: 278
Well it sounds like I may have been doing things the long way. I don't want to labour the point to much, but does that mean in an app like deepsky stacker or registax I could simply load all the subs that were taken (eg 50 red, 50 green, 200 blue, no lum) as well as any flats, bias etc, and just run it in one go?

To think, I have been processing each channel at a time, then manually stacking the result.

I may have just increased my productivity here by 300%.
Reply With Quote
  #59  
Old 27-04-2010, 09:31 AM
Bassnut's Avatar
Bassnut (Fred)
Narrowfield rules!

Bassnut is offline
 
Join Date: Nov 2006
Location: Torquay
Posts: 5,064
Hehe, thats right, load subs, darks and flats and click go. Some apps like DL even find the appropriate darks and flats in your (pre-setup) library automatically, and do the whole bunch of stacks in one go.

CCD stack requires you pick the dark and flat for each filter stack, and then align the LRGB stacked subs at the end, but its still a simple proceedure.

Registar also processes all in one go, and can even be scripted for a huge number of automatic operations with one click, including Mosaics.

Registax Im not sure about, but seem to remember its pretty automatic too.

Anyway, all the apps are easy to stack/combine with, thats what they are made for !. The time spent doing this is minor compared to tweaking in PS after that.
Reply With Quote
  #60  
Old 27-04-2010, 10:11 AM
telecasterguru's Avatar
telecasterguru (Frank)
Have scope will travel!

telecasterguru is offline
 
Join Date: Jan 2009
Location: Pitnacree NSW
Posts: 1,490
Fred,

Can I just ask, after stacking in DSS (I have CCD stack but haven't come to terms with it yet) I have been processing my images with fits liberator, combining and aligning then doing curves and levels in PS and from what you are saying DL will combine all the colour images in one go and therefore make most of my processing redundant.

Is the stretching of each individual channel automatic?

I like fits liberator because it is free and it is a PS plug in so there is no changing between processing software.

I use Nebulosity as capture software, would you recommend that I switch to DL. It is around $500US I believe.

Frank
Reply With Quote
Reply

Bookmarks

Thread Tools
Rate This Thread
Rate This Thread:

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump


All times are GMT +10. The time is now 12:03 AM.

Powered by vBulletin Version 3.8.7 | Copyright ©2000 - 2024, Jelsoft Enterprises Ltd.
Advertisement
Bintel
Advertisement
Testar
Advertisement