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

Sony A7s - New Low-Light Camera - Wow!

This topic has been archived.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

#126 chasing photons

chasing photons

    Mariner 2

  • -----
  • Posts: 287
  • Joined: 19 Mar 2013

Posted 21 June 2014 - 06:01 PM

I believe the A7R has already opened up (high resolution) astro-imaging for under 60 sec exposures, at least it has for me (NM Album). The A7S can only improve upon this. The A7R gives me the same image quality at IS06400 as my Canon 60D at ISO1600. If the A7S pushes this to ISO25600, I'll be shooting 7.5 sec subs, or going deeper with 15-30 sec subs, or using a longer scope, whatever. If the A7S performance at ISO51200 is as good as the marketing hype, I'll be totally impressed .

Astro-imaging is the wild west at the moment; probably has been since the invention of the digital camera...



bwa,

The images in your album are truly impressive, especially considering the stacked subs are 30 seconds or less, unguided, through a 65mm f/6.5 refractor using the high res Sony A7R. I can see now why you are champing at the bit to get your hands on the A7S! It could certainly be a game changer for near real time observing if the on the fly software applications that are springing up can eventually control the A7S and stack/process a handful of short subs while you watch the image 'develop' on the computer screen over a few minutes (or maybe a lot less) time.

Could you post one or two 30 second subs to show the typical amount of visible data and noise you experience with the A7R preprocessed - to get a feel for how much work the stacking/processing software would need to perform? Do you use any darks and flats? How much vignetting do you experience with the 65EDQ/A7R combo?

#127 chasing photons

chasing photons

    Mariner 2

  • -----
  • Posts: 287
  • Joined: 19 Mar 2013

Posted 21 June 2014 - 07:34 PM

This is my plan for the A7S. I already have tried this with a GH3 and A7. The added sensitivity of the A7S over the A7, A7R, GH3, & GH4, should dramatically improve real-time video. And the A7S' improved 1080p video codec should make for cleaner real-time video over the A7/A7R cameras.



jdbastro,

I've been a fan of your You Tube videos with the intensifier/GH3 combo since the beginning. I really smile when you demonstrate it is live video by slewing your mount. Good stuff! I will definitely keep an eye out for your results with the A7S.

#128 bwallan

bwallan

    Viking 1

  • -----
  • Posts: 563
  • Joined: 08 Jul 2010

Posted 21 June 2014 - 09:27 PM

...

bwa,

The images in your album are truly impressive, especially considering the stacked subs are 30 seconds or less, unguided, through a 65mm f/6.5 refractor using the high res Sony A7R. I can see now why you are champing at the bit to get your hands on the A7S! It could certainly be a game changer for near real time observing if the on the fly software applications that are springing up can eventually control the A7S and stack/process a handful of short subs while you watch the image 'develop' on the computer screen over a few minutes (or maybe a lot less) time.

Could you post one or two 30 second subs to show the typical amount of visible data and noise you experience with the A7R preprocessed - to get a feel for how much work the stacking/processing software would need to perform? Do you use any darks and flats? How much vignetting do you experience with the 65EDQ/A7R combo?


I should note all the images in the referenced album were processed from JPG subs since I didn't have any means to load Sony A7R Raw subs at the time. The subsequently drizzled images were processed from Raw subs (converted out of Lightroom 5.4).

I've combined a number of things in the following image:
Image Profile, 100% Crop & Final Drizzled Galaxy
Nothing other than an AutoStretch in ImagesPlus was done to the base image or 100% single sub crop in the above image.
*** BE SURE TO ZOOM IN ON THE IMAGE FOR FULL DETAIL ***

I get no noticeable vignetting with the AT65EDQ scope as is obvious from the image and the line profile. I did try a 0.75x reducer on the scope w/ the A7R and the results were not acceptable (at least to me); black corners.
The 100% single sub crop shows what one has to deal with noise-wise.
The 100% "Drizzled" crop shows the end result of aligning and stacking 30x30 sec. subs. Using 30 subs yields about a 5.5x theoretical improvement in signal-to-noise ratio. Drizzling does not lose detail but does have a smoothing effect.

I don't use darks, flats or bias subs. I did create, maintain and use a darks library prime to purchasing the Canon 60D but haven't use darks since AND the A7R's sensor is so clean I haven't even considered using darks with it. In fact, I personally believe the errors (temperature differences, ISO goofs, etc.) in applying darks greatly overrides the benefits.

I also do NOT use Long Exposure or High ISO noise control, i.e.: all forms of noise control that I can access on the Canon 60D and Sony A7R are turned OFF! I much prefer to handle noise control in postprocessing where I have some human control over the process.

Hope this helps.

bwa

#129 RandyC

RandyC

    Viking 1

  • *****
  • Posts: 665
  • Joined: 01 Apr 2013

Posted 21 June 2014 - 10:50 PM

bwa, Do you know if the BionZ is a stacker. There is a dramatic difference in quality between Raw and Jpg. As long as you take jpg as output, you get this benefit. Sure would be great to keep it simple and let the camera do all the work. It will be so great when the day comes, the laymen can get involved without it being too complex. I mean $2500 is a lot, but worth it if it's got so many options to maximize photons. Those A7R images are really great, congrats that's fantastic, way shorter exposures than typical.

#130 bwallan

bwallan

    Viking 1

  • -----
  • Posts: 563
  • Joined: 08 Jul 2010

Posted 22 June 2014 - 01:40 AM

bwa, Do you know if the BionZ is a stacker. There is a dramatic difference in quality between Raw and Jpg. As long as you take jpg as output, you get this benefit. Sure would be great to keep it simple and let the camera do all the work. It will be so great when the day comes, the laymen can get involved without it being too complex. I mean $2500 is a lot, but worth it if it's got so many options to maximize photons. Those A7R images are really great, congrats that's fantastic, way shorter exposures than typical.


Not too sure what you mean by "There is a dramatic difference in quality between Raw and Jpg. As long as you take jpg as output, you get this benefit."? I think the benefit is in the Raw images with their greater bit depth, malleability, lower noise, etc. Yes, the JPG images are preprocessed BUT may not be preprocessed the way everyone wants them to be... In my case, for astro-imaging, the reduction to 8 bit data reduces the image quality, i.e.: loses information, compression loses data I may want/need and introduces artificial noise and the whole process limits what I can retrieve from an image. I do prefer Raw images BUT I do see where you're heading WRT to JPG's being easier for the average person to get their head around...

Exposure time and ISO are inversely proportional. If you can retain the image quality at double the ISO, you can cut the exposure by half. If I can retain image quality with an A7S at ISO32000, I can cut my next round of New Mexico imaging to 30x3 seconds; however, what I would probably do is shoot 150x6 sec. or 90x10 (if the sky cooperates) and go deeper into the void, with the same total time of 900 seconds and improved signal-to-noise ratio. Furthermore, with the A7S's silent shutter mode, I can do this without any wear and tear on its mechanical shutter and no shutter vibration.

bwa

#131 Moromete

Moromete

    Viking 1

  • *****
  • Posts: 711
  • Joined: 15 Feb 2012

Posted 22 June 2014 - 03:03 PM

Question for Sony A7/A7R users: is one able to shoot RAW with A7/A7R when these are set in APS-C crop mode for shooting stills? What will be the size of the RAW file in this case?

#132 bwallan

bwallan

    Viking 1

  • -----
  • Posts: 563
  • Joined: 08 Jul 2010

Posted 23 June 2014 - 12:44 AM

Question for Sony A7/A7R users: is one able to shoot RAW with A7/A7R when these are set in APS-C crop mode for shooting stills? What will be the size of the RAW file in this case?


In the directory image below the 1st three *.ARW files are full frame off the A7R. The remainder of the *.ARW files are APS-C Raws off the A7R.

A7R - FF Raw vs. APS-C Raw File Size

The *.JPG's are small format, simply to allow me to see what the Raw files contain.

bwa

#133 Moromete

Moromete

    Viking 1

  • *****
  • Posts: 711
  • Joined: 15 Feb 2012

Posted 23 June 2014 - 03:16 AM

Thank you Allan!

So, it seems that the Sone A7 line is capable of shooting RAW in APS-C mode. WOW!

In my case this means that I can use an A7 with a C11+Alan Gee telecompressor (designed for T2 accesories) without needing to do the crop in post for each RAW file and avoid vignetting or other optical aberrations revealed by a FF sensor but benefiting of the large pixels and high ISOs of the FF sensor in the same time. Cool! I think you can't do this in-camera with a Canon 6D. Maybe Nikon D610 allows this in-camera RAW cropping, I don't know.

Recently I discovered the IRIS astro software which allows one to bin any image(Jpeg, Raw, etc.) in 2x2,3x3,4x4 modes and keep the colors even with a Bayer matrix like in DSLRs!

Since you have a fantastic 36Mpx sensor in your A7R, I warmly recommend you to try the IRIS software (it's free) and let us know how it works with A7R files under night sky. Link http://www.astrosurf...s/iris/iris.htm

Does anyone know a similar software like IRIS which allows binning and preserves colors?

#134 ccs_hello

ccs_hello

    Voyager 1

  • *****
  • Posts: 10303
  • Joined: 03 Jul 2004

Posted 23 June 2014 - 06:08 AM

Re: A7s in APS-C mode

Just like to point out APS-C vs. FF size area-wise is its 43%. I.e., a 12M pixel FF in APS-C sized term is just a 5.16M pixel sensor. (Of course nowadays no such cameras are produced commercially :( due to megapixel race.)

My main point is as opposed to a typically $500 APS-C sized camera, you will pay premium on a FF (typically $2000.) The A7s fetches additional premium (special, sensitive, super pricey :) ...)
To fit that premium, you buy expensive optics (with fast optics $$$$$) for near-realtime experiences.
If not going that path but using a 0.63x focal reducer at 0.63x, you'll have a smaller image circle best fit for APS-C format. To me personally with limited astro budget, it seems to be a great amount of money wasted.

I bet a 5 or 6M pixel APS-C mirrorless using a modern CMOS sensor (e.g., SONY EXMOR) will be a HOT sell. Will SONY, Nikon, Pentax (or even Samsung) consider this? Time will tell.

P.S. Four-Thirds is only 28% of the FF area. So a mFT equivalent of A7s will only have 3.36M pixels!

Clear Skies!

ccs_hello

#135 chasing photons

chasing photons

    Mariner 2

  • -----
  • Posts: 287
  • Joined: 19 Mar 2013

Posted 23 June 2014 - 08:18 AM

I bet a 5 or 6M pixel APS-C mirrorless using a modern CMOS sensor (e.g., SONY EXMOR) will be a HOT sell. Will SONY, Nikon, Pentax (or even Samsung) consider this? Time will tell.

P.S. Four-Thirds is only 28% of the FF area. So a mFT equivalent of A7s will only have 3.36M pixels!



I am afraid that, in the not too distant future, nearly every new mirrorless and dslr camera will have to support 4K video to please the mass market. At 3840 x 2160 resolution, this is 8.3MP in a 16:9 format. This may become the absolute minimum in pixel resolution for any size image sensor for mirrorless and dslr cameras very soon. Not only will the pixel wars continue, but the infantry will be buried. Of course, the industrial, science and surveillance markets will still deal with CCD and CMOS sensors in the few mega pixel range, but I doubt commercial cameras will ever dip that far down. :(

#136 RandyC

RandyC

    Viking 1

  • *****
  • Posts: 665
  • Joined: 01 Apr 2013

Posted 23 June 2014 - 10:11 AM

Hi, Brian posted a comparison of .5x vs. .8x focal reduction using a C11 on the Canon EOS50D. Do you expect a similar result with the A7s? It sure would be great to use the entire fov, and also make gotos easier while maintaining magnification. Anyway you can keep FF versus APS-C seems to be the best optical efficiency.

#137 Moromete

Moromete

    Viking 1

  • *****
  • Posts: 711
  • Joined: 15 Feb 2012

Posted 23 June 2014 - 01:48 PM

Re: A7s in APS-C mode

Just like to point out APS-C vs. FF size area-wise is its 43%. I.e., a 12M pixel FF in APS-C sized term is just a 5.16M pixel sensor. (Of course nowadays no such cameras are produced commercially :( due to megapixel race.)

My main point is as opposed to a typically $500 APS-C sized camera, you will pay premium on a FF (typically $2000.) The A7s fetches additional premium (special, sensitive, super pricey :) ...)
To fit that premium, you buy expensive optics (with fast optics $$$$$) for near-realtime experiences.
If not going that path but using a 0.63x focal reducer at 0.63x, you'll have a smaller image circle best fit for APS-C format. To me personally with limited astro budget, it seems to be a great amount of money wasted.

I bet a 5 or 6M pixel APS-C mirrorless using a modern CMOS sensor (e.g., SONY EXMOR) will be a HOT sell. Will SONY, Nikon, Pentax (or even Samsung) consider this? Time will tell.

P.S. Four-Thirds is only 28% of the FF area. So a mFT equivalent of A7s will only have 3.36M pixels!

Clear Skies!

ccs_hello


On the whole I agree with you but you can use the FF dslr for normal photography and images have more pop even in daylight compared to an APS-C sensor. Still the A7S is very expensive, at least for me.

On the other hand there are no APS-C sensors with 5-6Mpx and new technology. The actual astro CCDs with dslr sensors have too old technology even if they have 8.4um pixels and also cost north of $2000.

Let's hope Canon will present this year at Photokina a dslr at under $2000 (as a homage to amateur astronomer customers) with this sensor http://www.youtube.c...h?v=eHnUtmenGQ0 which has 19um pixels and around 2Mpx! ;)

#138 bwallan

bwallan

    Viking 1

  • -----
  • Posts: 563
  • Joined: 08 Jul 2010

Posted 23 June 2014 - 02:24 PM

Thank you Allan!

So, it seems that the Sone A7 line is capable of shooting RAW in APS-C mode. WOW!

In my case this means that I can use an A7 with a C11+Alan Gee telecompressor (designed for T2 accesories) without needing to do the crop in post for each RAW file and avoid vignetting or other optical aberrations revealed by a FF sensor but benefiting of the large pixels and high ISOs of the FF sensor in the same time. Cool! I think you can't do this in-camera with a Canon 6D. Maybe Nikon D610 allows this in-camera RAW cropping, I don't know.

Recently I discovered the IRIS astro software which allows one to bin any image(Jpeg, Raw, etc.) in 2x2,3x3,4x4 modes and keep the colors even with a Bayer matrix like in DSLRs!

Since you have a fantastic 36Mpx sensor in your A7R, I warmly recommend you to try the IRIS software (it's free) and let us know how it works with A7R files under night sky. Link http://www.astrosurf...s/iris/iris.htm

Does anyone know a similar software like IRIS which allows binning and preserves colors?


I don't know what the reduction is with your reducer; however, anything greater that about 0.75x is probably going to give some vignetting even with an APS-C sensor (see HERE). If you happen to have one of the newer astrographs, i.e.: scopes designed for astro-imaging and large format sensors, you could probably use full frame and your reducer. I have an Astro-Tech 8" RC (oversize secondary mirror and a 3-1/3" back) which works very well with a full frame camera. I've never tried it with a reducer (to this point). The only problem is, the larger the sensor the better the collimation has to be to keep the edges of the image under control... :(.

I've never tried IRIS but I know ImagesPlus has pseudo binning functionality. I say "pseudo" because anything not done on the actual sensor is not true binning; at its best it can only use bayered data off the camera.

bwa

#139 bwallan

bwallan

    Viking 1

  • -----
  • Posts: 563
  • Joined: 08 Jul 2010

Posted 23 June 2014 - 02:38 PM

Re: A7s in APS-C mode

Just like to point out APS-C vs. FF size area-wise is its 43%. I.e., a 12M pixel FF in APS-C sized term is just a 5.16M pixel sensor. (Of course nowadays no such cameras are produced commercially :( due to megapixel race.)

My main point is as opposed to a typically $500 APS-C sized camera, you will pay premium on a FF (typically $2000.) The A7s fetches additional premium (special, sensitive, super pricey :) ...)
To fit that premium, you buy expensive optics (with fast optics $$$$$) for near-realtime experiences.
If not going that path but using a 0.63x focal reducer at 0.63x, you'll have a smaller image circle best fit for APS-C format. To me personally with limited astro budget, it seems to be a great amount of money wasted.

I bet a 5 or 6M pixel APS-C mirrorless using a modern CMOS sensor (e.g., SONY EXMOR) will be a HOT sell. Will SONY, Nikon, Pentax (or even Samsung) consider this? Time will tell.

P.S. Four-Thirds is only 28% of the FF area. So a mFT equivalent of A7s will only have 3.36M pixels!

Clear Skies!
ccs_hello


Yes, traditionally dedicated astro-imaging cameras (CCD's) have had a relatively small pixel count. I have several planetary CCD cameras and a mono ATIK 428 CCD; all under 3 MPixel. There are high-end APS-C and larger sensor, cooled, dedicated astro-imaging cameras available BUT in the order of $4000++. If an A7, A7R, A7S, Canon 6D, etc. will function as an astro-imaging camera and you want/need full frame coverage, it is a bargain. And with a full spectrum conversion and the appropriate filters it will do triple duty for astro-imaging, IR and daylight.

bwa

#140 bwallan

bwallan

    Viking 1

  • -----
  • Posts: 563
  • Joined: 08 Jul 2010

Posted 23 June 2014 - 03:00 PM

I bet a 5 or 6M pixel APS-C mirrorless using a modern CMOS sensor (e.g., SONY EXMOR) will be a HOT sell. Will SONY, Nikon, Pentax (or even Samsung) consider this? Time will tell.

P.S. Four-Thirds is only 28% of the FF area. So a mFT equivalent of A7s will only have 3.36M pixels!



...
Of course, the industrial, science and surveillance markets will still deal with CCD and CMOS sensors in the few mega pixel range, but I doubt commercial cameras will ever dip that far down. :(


Yes, there will still be dedicated CCD/CMOS cameras for at least astronomy with sensors having 0.5++ MPixels... and when you add in cooling, filter wheels, inherent guiding, etc. these cameras quite often top the price of an A7S! Furthermore, most of these cameras have their gains (ISO) fixed at the optimum value for their particular sensor and circuitry. From my testing this gain equates to about ISO400-640 on a DSLR.

Now if I can get a good clean image off a DSLR at ISO6400 rather than at the equivalent of ISO640 off a dedicated CCD camera I gain a WHOLE LOT of advantages! One tenth the exposure time to start with (this is not totally true if you're talking really short exposures but close). Reduced need for totally accurate tracking and probably being able to do w/o guiding. Less concerns about polar alignment (and field rotation). Less concern about equipment flexure and temperature induced focus changes. Maybe being able to use a less expensive Alt-Az mount in place of an EQ mount. Lots of considerations to optimize against using a DSLR vs. a CCD, OR even a fast DSLR vs a slow one...

Just thinking out loud...

bwa

#141 bwallan

bwallan

    Viking 1

  • -----
  • Posts: 563
  • Joined: 08 Jul 2010

Posted 23 June 2014 - 03:12 PM

Hi, Brian posted a comparison of .5x vs. .8x focal reduction using a C11 on the Canon EOS50D. Do you expect a similar result with the A7s? It sure would be great to use the entire fov, and also make gotos easier while maintaining magnification. Anyway you can keep FF versus APS-C seems to be the best optical efficiency.


Sadly, the results with a full frame will be worse (UNLESS you have an astrograph scope that was designed to cover a large sensor) since the sensor is considerably larger. Scope like the Astro-Tech nn RC series or the latest 190 Mak-Newt all have larger secondary mirrors and larger backs, and display a larger image circle. Even the newer C11's have 2" backs. My old C11 had a 1-1/4" back.

bwa

#142 bwallan

bwallan

    Viking 1

  • -----
  • Posts: 563
  • Joined: 08 Jul 2010

Posted 23 June 2014 - 03:27 PM

Let's hope Canon will present this year at Photokina a dslr at under $2000 (as a homage to amateur astronomer customers) with this sensor http://www.youtube.c...h?v=eHnUtmenGQ0 which has 19um pixels and around 2Mpx! ;)


It would be nice BUT I suspect we may all have to continue dreaming about an "homage to amateur astronomers" for under $2000 having this technology. Furthermore such a camera would only be good for HD video... Doesn't seem to be trending?

Maybe Sony also saw the video and decided to push the A7S to market? One never knows the reasons behind the appearance of some products!?

bwa

#143 RandyC

RandyC

    Viking 1

  • *****
  • Posts: 665
  • Joined: 01 Apr 2013

Posted 23 June 2014 - 05:36 PM

I have a 2" back on my C11 but not sure how that affects the light cone? I guess native F10 would get you a good piece of the FF. At least enough to center the object and then switch to APS-C.

#144 bwallan

bwallan

    Viking 1

  • -----
  • Posts: 563
  • Joined: 08 Jul 2010

Posted 23 June 2014 - 06:37 PM

I have a 2" back on my C11 but not sure how that affects the light cone? I guess native F10 would get you a good piece of the FF. At least enough to center the object and then switch to APS-C.


Since the light off the secondary in a SCT scope is coming in almost parallel a 2" (50.8mm) back should light a full frame (43mm diagonal) sensor with no problems. The problems arise when you throw a reducer into the light train. Then you run into problems as to how much outside edge you lose simply because of blocking by the outside tube of the scope...

bwa

#145 ccs_hello

ccs_hello

    Voyager 1

  • *****
  • Posts: 10303
  • Joined: 03 Jul 2004

Posted 23 June 2014 - 07:23 PM

The following calculation should help (cell: B29). The imager is an APS-C sized sensor (BTW, it's the famous but now aged ICX453/ICX413.)

Only 64% of the imager is fully illuminated by on-axis light.

Clear Skies!

ccs_hello

Attached Thumbnails

  • 6598195-Meade63FR.jpg


#146 ccs_hello

ccs_hello

    Voyager 1

  • *****
  • Posts: 10303
  • Joined: 03 Jul 2004

Posted 23 June 2014 - 07:32 PM

Count in off-axis light, I'd say 2" image circle output from an OTA cascaded with a 0.63x FR in its normal setting (0.63x) would work on an APS-C sized imaging device.

Again my only point is "need premium optics" when fast (f-ratio) and large (imager size) are both needed.
$$$$$, fast, and large, can only pick two.


P.S. please do not get me wrong, it took quite awhile to see a FF DSLR/mirrorless price dropping to a level $2000-ish. It is still a good news for regular shooting, so long as we are not going extreme for low f number lenses for photon-starvation situations.
I have to admit, with so many years of waiting, my lens collections are now mainly for APS-C sized.


Clear Skies!

ccs_hello

#147 bwallan

bwallan

    Viking 1

  • -----
  • Posts: 563
  • Joined: 08 Jul 2010

Posted 23 June 2014 - 08:38 PM

Count in off-axis light, I'd say 2" image circle output from an OTA cascaded with a 0.63x FR in its normal setting (0.63x) would work on an APS-C sized imaging device.

And it does; probably the reason Celestron and others opted for a 0.63x ratio...

Again my only point is "need premium optics" when fast (f-ratio) and large (imager size) are both needed.
$$$$$, fast, and large, can only pick two.

Yes, too bad $$$$$ are in this equation!

A nice compromise are the Samyang/Bower/Rokinon 8, 14, 24, 35 & 85mm fast prime lenses.

P.S. please do not get me wrong, it took quite awhile to see a FF DSLR/mirrorless price dropping to a level $2000-ish. It is still a good news for regular shooting, so long as we are not going extreme for low f number lenses for photon-starvation situations.
I have to admit, with so many years of waiting, my lens collections are now mainly for APS-C sized.

Clear Skies!

ccs_hello

As for lenses, I find my collection is going the other way; paring down the APS-C and adding full frame. I find the full frame lenses give superior performance on an APS-C camera and, in most cases, work adequately with full frame sensors (although a few have edge/corner problems). Maybe I should be looking at medium format lenses and adapting them to full frame!?

bwa

#148 chasing photons

chasing photons

    Mariner 2

  • -----
  • Posts: 287
  • Joined: 19 Mar 2013

Posted 23 June 2014 - 09:02 PM

Let's hope Canon will present this year at Photokina a dslr at under $2000 (as a homage to amateur astronomer customers) with this sensor http://www.youtube.c...h?v=eHnUtmenGQ0 which has 19um pixels and around 2Mpx!



One also needs to remember that pixel size affects image resolution and the bigger the pixel the poorer the resolution. In general, for DSOs, the recommendation is to sample at 1/2 the seeing resolution and for planetary, it is 1/3 the seeing resolution.

Pixel Resolution = 206(pixel size)/(focal length of optics) in arc seconds per pixel where pixel size is in microns and focal length is in millimeters.

For example, let’s say that average seeing limits you to 4 arc seconds resolution. Then you would normally use a pixel resolution of 2 arc seconds or better for DSOs. Now we use an imaging sensor with monster 19 micron pixels. What is the telescope focal length we need for 2 arc seconds resolution when using 19 micron pixels?

Focal length = 206(19)/2 = 1957 mm

Yikes! This means that even a big ol' C14 with 355 mm aperture and 3910 mm focal length dropped down to 1955 mm effective focal length using a .5x focal reducer would only get you down to f/5.5 which isn’t exactly hyper fast optics.

Now what happens if you want to use a smallish, fast refractor with a 19 micron pixel sensor? Let’s use a typical 100 mm aperture, 600 mm focal length f/6 imaging refractor. First off we see that the focal length of the refractor at 600 mm is very short compared to our recommended 1957 mm to reach 2 arc seconds resolution using 19 micron pixels. So what resolution do we end up with using this refractor?

Pixel Resolution = 206(19)/(600) = 6.5 arc seconds per pixel.

Yikes! This means that the image resolution of our refractor ends up being over 3 times worse than the 2 arc seconds we were striving for. Can you say… undersampled? Those faint fuzzies will be brighter, but maybe still a bit fuzzier than you were hoping. And remember, this is just in average seeing conditions. What happens when seeing gets good and your optics can see down to 2 arc seconds? Now the 6.5 arc seconds per pixel resolution is over 6 times worse than the 1 arc second per pixel resolution that is recommended! And what focal ratio do we need to go up to just to get down to 2 arc seconds resolution?

1957/100 = f/19.57 = one slooooooooow refractor! (You'll need a 19.5/6 = 3.25x barlow)

The moral of the story seems to be that really big pixels work best with large aperture, long focal length and not necessarily very fast focal ratio optics - in other words - really big, expensive glass.

Feel free to dissect my logic and math.

#149 ccs_hello

ccs_hello

    Voyager 1

  • *****
  • Posts: 10303
  • Joined: 03 Jul 2004

Posted 23 June 2014 - 09:16 PM

Todd,

No question on your reasoning.
I think Canon will reward you (on that) if you ever buy that super, super expensive (to be custom made) lens, they will throw in that 19um FF HD cam for you for free :) so you'll buy more of their lenses :) :).

Clear Skies!

ccs_hello

#150 coinboy1

coinboy1

    Apollo

  • *****
  • Posts: 1245
  • Joined: 03 May 2011

Posted 23 June 2014 - 09:20 PM

Hello,

I have been following this thread with much interest. I think the A7S will be an excellent astro camera. The large pixels and full frame will attract many. Bwa- I am impressed with your images with the A7R and the AT65EDQ. I have a SharpStar65EDQ and image with a full format Canon 5D classic and it handles full format with nice round stars to the edges no problem.

As for the debate with having telescope optics capable of full format, I found a cheap alternative to utilize full format for our 10"+SCTs. It is the discontinued Lumicon Giant Easy Guider with large 3" aperture corrector. I image with an 10" LX200 with my canon 5D full format. With the Lumicon GEG, vignetting is minimal and I can still guide with my full format DSLR. The lumicon giant easy guider can be picked up used for around $200. The below image was taken in Alt-Az mode as well.

People with 10, 12, 14" LX200s, 11 Nexstars or CPCs, can use the Lumicon GEG with their full format DSLRS and actually use the real estate of the full format CMOS sensor.

I have a Canon 5D classic and its a 12.2MP full format CMOS sensor with 8.2uM pixels. The only thing is its nearly 10 years old and the noise characteristics of the camera is not great. I really look forward to the Sony A7S. I think it will be a breakthrough in imaging...

Attached Thumbnails

  • 6598398-m27.jpg



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