View Single Post
  #11  
Old 30-03-2008, 02:04 AM
g__day's Avatar
g__day (Matthew)
Tech Guru

g__day is offline
 
Join Date: Dec 2005
Location: Sydney
Posts: 2,809
Glad you like the 40D - I am seriously tempted!

BTW - I keep feeding features and queries of what I would like to be in a digitial camera for astronomy to Canon research (via my brother) - but I guess anyone can. The latest feedback he gave me from the labs developing this Camera and software is below - along with their contact e-mail:

Matt


(My bro) My brother is a hobbyist astronomer who is considering a Canon 40D in sophisticated ways, and has some questions that he hopes some people here might be able to answer. (There isn't a Camera/Photography mailing list, is there?)

(The Camera techs) digphoto@cisra is probably what you want, but here's some answers anyway. This is all publicly available info.

(CiSRA) is Canon Systems Research Australia - with alomst 400 PhDs - second in size only to CSIRO...

(my Bro) 1. I read it displays video at 30 frames per second - can this be adjusted down for low light conditions to say 3 frames a second - to as slow as one frame per 5 seconds?

(Lab guys) The exposure of live view *is* adjusted automatically to try and make it bright, but in very low light it gets very grainy. I don't know if the effective count time is reduced. I was reasonably impressed with how low we could push the light and still get a recognisable image out.

(My Bro) 2. Is the Live View capability only operational on models 40D and better because they have the Digic 3 chip? Would it operate on a 400D using the Digic 2 chip if they were inclined to release this software?

(Lab guys) There was an old specialist astronomy camera as it happens, called the 20D-A, and as far as I know it was the first camera to introduce the LiveView feature. The camera was only released in Japan and was DigicII based.

So I would guess in theory it could be done, but there's a lot of software involved doing the mirrorlock etc. I think the 20D-A only supported back-of-camera view, not remote live view (although someone such as dpreview.com may know otherwise?). It also had the IR filter removed.

Since Astronomy is what liveview is originally designed for, I would postulate that it's not too hard to use it to focus on a star at night through a telescope.

(My bro) Again for an astronomer - the Remote Live View - would be brilliant for framing a shot from a remote controlled observatory.

(Lab guys) The default software that comes on the CD with the 40D allows you to view a stream of the liveview image and change zooms, focus, etc. It's quite good.

(My bro) 3. Does Live View cut out when the shutter is depressed? I presume so - that its a prism flip, not a dual CCD chip like S-BIG astro photography CCDs that have two CCDs on camera - one to image an one to guide!

(Lab guys) Yes, but only for the shutter movement. The mirror/prism does not have to move - you can configure it to keep the mirror up while the shot is taken and return to liveview straight away. The viewfinder is not usable in this state of course, nor is autofocus (both of which happen in the top of the camera). The recently announced 450D incorporates a "compact camera" style autofocus operation while in live view. It's not as fast as the true autofocus, but it works. That camera isn't out yet unfortunately.

(My bro) 4. Lastly all commerical DSLRs tend to have IR filters, but do they filter out above the Hydrogen alpha line or below it? This is one of the key features limiting a DSLR for astro-photography (that and water or peltier cooling to minimise dark currents in the CCD). They use filters that tend to cut out the main light source from galaxies, if they were set just a tad lower astro photographers would be excited! I'd much rather spend $6K on (just the body) an all purpose DSLR that could be switched into night mode that would work light the really expensive dedicated astro photography camera S-BIG and Fingers Lake Instruments. Maybe the DSLRs guys should look at this because if they put their minds to it it's a big market with folks spending lots and their needs are not too different from normal users. The ideal astrophotography camera would have:

1. High quantum efficiency - low dark current
2. Different IR filters that sit below the Hydrogen Alpha line
3. Optional ability to guide [See below re Guiding]
4. Ability to store dark frames (shots of different duration with the lens closed, possibly at different recorded temperatures) which is removed from images taken.

(Lab guys) Like I said above, the 20D-A had the IR filter removed (or at least customised), but it was only a limited release. Google would know more.

The newer cameras (350D onwards) can do auto dark-frame subtraction - It's called "long exposure noise reduction", and it basically takes a dark frame of the same length straight after your exposure. On 30s exposures this means you've got 30s downtime between images, which is actually fairly annoying. Doing it manually is less painful and you get more control. I suggest reading up on what causes the dark count and
the different types of "hot" pixels to properly characterise them and subtract them more accurately. There are several papers on the subject.

I would guess you can do guiding using the Liveview image feed to a PC, but then you'd need to obtain the SDK (available from Canon Australia under an NDA) and write some software to do it. Shouldn't be too hard
if you're good at software and image processing
Reply With Quote