Jump to content


Photo

How to turn an 8" F4.0 into a 16" F4.0?

  • Please log in to reply
7 replies to this topic

#1 mpgxsvcd

mpgxsvcd

    Surveyor 1

  • -----
  • Posts: 1951
  • Joined: 21 Dec 2011
  • Loc: Raleigh, North Carolina

Posted 12 October 2012 - 10:02 AM

I decided to try out the lossless crop mode on my modified GF1 last night. It basically takes a 12 megapixel sensor down to 2-3 megapixels. That gives you a 2x zoom without any light loss. However, it only does this for jpg images so you cannot shoot RAW in this mode.

There is no free lunch though. It reduces the resolution but do you really need 12 megapixels for AP? If your monitor or TV is no greater than 1920x1080 it doesn't really make much sense to record images with that many megapixels.

This is a single exposure 105 second ISO 800 jpg image @ F4.0 with an 800mm scope and a 2x crop factor(1600mm in 35mm terms). Also note that I did not guide with this at all. This is simply a CG-5 polar aligned very accurately with a polar scope.

It took me about 30 minutes to go from have the scope and mount inside my house to producing this single exposure image in my backyard.

Attached Files



#2 mega256

mega256

    Viking 1

  • *****
  • Posts: 925
  • Joined: 10 May 2007
  • Loc: N of Tampa

Posted 12 October 2012 - 11:49 AM

Travis very nice indeed.....Great Pic for that short exposure...Very nice tracking..

#3 mpgxsvcd

mpgxsvcd

    Surveyor 1

  • -----
  • Posts: 1951
  • Joined: 21 Dec 2011
  • Loc: Raleigh, North Carolina

Posted 12 October 2012 - 11:59 AM

Thanks mega. I actually had to buy another polar finder to replace my original one. I accidentally loosened all of the screws in the polar finder when trying to align it with the housing.

The little piece of glass in it turned slightly when the screws were loosened. That meant that it was no longer calibrated and that was what was causing my tracking errors before.

I tried the all star polar align with the old finder scope and it moved Polaris way off of the circle in the polar finder. Then I tried it with the new polar finder and it put Polaris exactly where it should be when you looked through the Polar finder scope.

Now that I know the Polar Finder is accurate I can just skip the All Star Align process which saves me about 5-10 minutes of setup time.

The CG-5 was putting everything dead center after I switched to the newer Polar Finder that I had not messed up.

#4 ccs_hello

ccs_hello

    Fly Me to the Moon

  • *****
  • Posts: 6626
  • Joined: 03 Jul 2004

Posted 13 October 2012 - 08:46 AM

Just wish to mention that Panny's ETC (Extra Tele Conversion) mode is a center crop of the camera sensor. For still photo capture in a 12 Mpxl (4000x3000 4:3 aspect-ratio sensor) in a 2x ETC, only the center part (2000x1500) is used.
One can envision that it's effectively cutting the sensor into 4 equal parts and only use one part of it for shooting.

BTW, none of these affect telescope optics (or camera lens) specification. If the OTA is a 8" F4, with 2x center-crop recording, it is still 8" F4.
The "effective focal length" change is just an illusion. This "crop factor" discussion happened since the days that digital cameras started to use smaller sensors. Remember the APS-C DLSR's (Canon 1.6x crop and Nikon's 1.5x crop stories :) ...)

Clear Skies!

ccs_hello

#5 nytecam

nytecam

    Voyager 1

  • *****
  • Posts: 11542
  • Joined: 20 Aug 2005
  • Loc: London UK

Posted 14 October 2012 - 05:22 PM

Very nice pic :bow: but cropping [very sensible for CN TOS] doesn't convert a 8" @ f/4 to 16" @ f4 eg there's no x4 increase in flux ;)

#6 mpgxsvcd

mpgxsvcd

    Surveyor 1

  • -----
  • Posts: 1951
  • Joined: 21 Dec 2011
  • Loc: Raleigh, North Carolina

Posted 15 October 2012 - 09:12 AM

Yes I completely agree. That is what the "No Free Lunch" was about. However, it does offer some of the benefits of using a larger telescope without some of the drawbacks(ie: Even if I could afford a 16" telescope I couldn't track as well with it without spending many thousands of dollars on a better mount).

You definitely can look at this the other way around and say well you could just use a small sensor camera with an 800mm F4.0 and a focal reducer and easily turn that into a true 800mm F2.0 scope. Which is faster and still provides the same field of view.

That would work because the small sensor would not vignette with the focal reducer. However, you can also accomplish the same thing with a large sensor camera that allows you to work in a lossless crop mode.

My point really is that with a large sensor camera you can go wide field with a fast scope or go with a narrow view all without changing the optics. You can actually get very close to the results that you would have if you did change the optics.

For semi-live view observing where you are trying to see as many objects as possible(Think messier marathon) the fewer changes you need to make to the optics the much more likely you will be to get through all of the objects.

I attempted to see as many objects as I could on Friday night. I had a list of 30 messier objects that were possible to see in the 4 hours we had out there. I got through about 15 of them. However, that was not because I couldn’t get them with the equipment I had. It was because I had a crowd of a dozen people all wanting to see the images I was getting that night and had kept on my camera from previous nights.

I use a single 32 GB card in my camera and keep all of the images I have ever shot on it. Even if the conditions are bad or the objects are not visible I still can show them to people as if I had just recorded them.

A fellow astronomer had a Malincam Extreme out there with a nice little scope and the same CG-5 mount I have. He appeared to have everything setup correctly but he struggled to get through the alignment routine in the four hours we were there.

I tried to help but I really didn’t know enough about his optical chain (Focal reducers and optics distances) to provide any real help. I wondered if it was simply that his field of view was too narrow to allow him to find the alignment stars. However, there really wasn’t anything he could do about it since he was already using a focal reducer and his field of view still appeared to be quite small.

#7 mpgxsvcd

mpgxsvcd

    Surveyor 1

  • -----
  • Posts: 1951
  • Joined: 21 Dec 2011
  • Loc: Raleigh, North Carolina

Posted 15 October 2012 - 09:44 AM

Here is an example of a wide field object. You couldn't do this and get the image above without changing the optics with a small sensor camera.

M8 Lagoon Nebula

ISO 800 85 seconds 800mm F4.0 2x crop factor(1600mm in 35mm terms)

Attached Files



#8 ccs_hello

ccs_hello

    Fly Me to the Moon

  • *****
  • Posts: 6626
  • Joined: 03 Jul 2004

Posted 16 October 2012 - 07:18 PM

Hi,

I did an analysis on my GF3 (which should be about identical as your GF1) in ETC mode (shown as "EX" symbol) on sensor crop:

In STILL picture capture:

L (full resolution): 4000x3000 (4:3 aspect-ratio 12 Mpixel sensor)
M (center crop): 2816x2112 (4:3 crop) 1.42x crop
S (center crop): 2048x1536 (4:3 crop) 2x crop

In VIDEO recording mode:

FSH (full resolution - AVCHD 16:9) @1920x1080 resolution
which is the full sensor area using 3840x2160 resolution, but taking a "2-line readout, skipping 2-line readout" method to yield 1920x1080

SH (sensor center crop - AVCHD 16:9) @1280x720 resolution
which is about 3.1x center crop (i.e., 1280 x 3.1 = 3960)
top-and-bottom has more pixels cropped due to (4:3 <-- this is 16:12 sensor with only 16:9 actually being used)

HD mode (MP4 using Quicktime MJpeg codec, 16:9 aspect ratio) also center crop 1280x720, same as SH mode, different codec used

VGA mode (MP4 using Quicktime MJpeg codec, 4:3 aspect ratio) which is a center crop 960x720, then scaled down to 640x480
Note this is 4.2x (960 x 4.2 = 4032, about the sensor native resolution)

QVGA mode (MP4 using Quicktime MJpeg codec, 4:3 aspect ratio) also center crop 960x720, then scaled down to 320x240
This is also 4.2x crop.

Hope this explains the sensor + camera's Image Processing DSP magic.

Clear Skies!

ccs_hello






Cloudy Nights LLC
Cloudy Nights Sponsor: Astronomics