Canon 350D v's 20D

[Gama]

Originally Posted by CometGuy

The information I have indicates that the D100, D70 and D50 all use the Sony ICX413AQ CCD.

Actually they are not..
Here is a couple of quotes i cut and pasted from the web sites.

"
The CCD sensor the D70 uses is unique. The rumor is that it is made by Sanyo for Nikon using a hybrid Sony/Nikon design. It's not one of the Sony-produced 6-megapixel sensors that the D100, Pentax *ist, and other cameras have started using, though it may be derived from the same technology. The D70 sensor consists of a 3008 x 2000 array, and it has massive (compared to consumer cameras such as the Coolpix) 7 micron pixels. That's just one reason why the D70 can produce 12-bit RGB images with rich color and low noise. Like virtually all digital cameras, a filter array is placed over the D70's CCD. This filter has four purposes"
From link here :> http://www.bythom.com/D70REVIEW.HTM about half way down.

Then this quote as well:

"The CCD sensor the D50 uses is unique. No, it's not the one used in the D100 or D70 models, though it is certainly derived from the same technology. The D50 sensor consists of a 3008 x 2000 array, and it has massive (compared to consumer cameras such as the Coolpix) 7 micron pixels. That's just one reason why the D50 can produce 12-bit RGB images with rich color and low noise. Like virtually all digital cameras, a filter array is placed over the D50's CCD. "
This quote from here http://www.bythom.com/D50REVIEW.htm about half way as well.
Any way enuf said..

[Merlin66 (Ken)]

Check out www.astrosurf.com/buil
He has a lot of info on digital camera testing etc etc.

[Gama]

Well, i have some good news.
I will be testing a Nikon D200 very soon, as one is going to be sent to me for Astronomical evaluation.
Once i get some results i'll let you all know the progress.
But from initial results i have seen around, i would put off any purchase on the Canon 20D, 20Da, and the 5D for a while.
Also with a little support from the Astronomical Community, they may bring out a version with a better Ha response like the 20Da.
.....

[Jonathan]

I'm looking forward to your results Gama. I wonder if Hutech are going to bring out a modified version for astrophotography.

[Striker (Tony)]

I also look forward to the results.

I hope they make some more future camera's with the live view like the 20da....focus is made so easy with it.

[CometGuy]

Gama,

I look forward to seeing some more on the D200. I have only seen brief snippets of info so far.

The D200 is getting a bit up there in the price range, however there are also some people reporting that the D50 is a very good performer (Teds sell the D50 for $999 with kit lens - a great bargain IMO). I wouldn't be suprised if the D50 outperformed the D200 for deep sky on account of its much larger pixels.

Another interesting recent note from a Nikon user on the Digital_astro list claimed is apparently it is it possible to cause 'mode 3' by reducing the power supply voltage to a point it causes an 'undervoltage fault' then recovering the power supply. If this is true one could supply the camera with an external power supply that can stepped down using a control signal. This would be quite a breakthrough I think for getting true - or near true - raw images.

Terry

[Gama]

Yes, the D50 or actually any other larger pixel based camera will pickup up a better photon collection, but the problem with the D50 is internal Noise. Its just too much noise, thus the signal get washed out. But still i can Bin the D100 which would eat the D50 alive.
The best so far is the D100, or the D70 with the filter removed.
Price wise, the D200 is sort of the Canon 30D now, but it has a better sensor, and with winter here now, the camera is going to be around 10 deg or less, so noise will be much better..
But like i said earlier, doesnt matter too much with the noise, as the sensors QE is so much better that you could afford a little loss to noise. Just wander how much better it would be if the D200 had no or modified filter..
That would just kick so much arse.

[CometGuy]

If you bin the pixels in the D200 you will increase noise as square of the binning since you can only do this in software. i.e bin 2x2 the noise will be 4x as much, and you will have only 4x as much signal. You will not gain anything. In an monochrome astrocam if you bin in hardware, you increase noise less than 4x since you are reading out all 4 pixels at once.

Anyway, I have asked somebody who owns both a D200 and 350D/XT to do some side-by-side and he has agreed. So I'll report back.

Terry

[iceman (Mike)]

Look forward to reading the reports and seeing the results! Pictures we want pictures!

[avandonk]

This is a comparison of a Nikon D200 vs. Canon 5D. I know that the 5D has about half the noise of a 20D. I will let you draw your own conclusions of how the D200 would perform as an astro camera from these tests or comparisons.
Real astro tests should be interesting.

http://www.digitalcamerainfo.com/con...ad-Review-.htm

Bert

[Gama]

Originally Posted by CometGuy

If you bin the pixels in the D200 you will increase noise as square of the binning since you can only do this in software

Under Maxim DSLR you can Hardware Bin a 20D to 2x2. So i would assume and hope the same for the D200, as i dont have it yet i wont know. Time will tell.

[avandonk]

My prediction is the Nikon D200 is not even in the race against a 20D, let alone a 5D. This is my opinion based on purely signal to noise. The other consideration it is only a small sensor. Nikon have set their future on C size sensors supposedly to take advantage of the 'sweet' spot of their lenses. This way their lenses will inevitably deteriorate to only accommodate small sensors.
It is sad when a fine company finally succumbs to the marketeers.

The best example of this is 10MB sensors (in P&S) that are only 7X5mm at best, with pixels about one or two micron. The noise at even an ISO of 200 is appalling!
The other consideration is that even an F2.8 lens with these sensors has an Airy disk of 3.6 micron,which is bigger than a pixel! It gets worse at F8 the Airy disk is 10.3 micron. This is for green light!
If you need more information just look it up.

Bert

[Gama]

The 20D is not better, proof is out there with heaps of reviews comparing the 20D to the D200.
QUOTE " The D200 trumps the Canon 20D in spades, and every time Nikon announces something Canon one-ups them the next week "
Here is just one site, but theres 1000,'s more http://www.kenrockwell.com/nikon/d200-vs-canon.htm.

Yes the 5D, with a larger sensor its better at some things and the D200 at others, and overall the 5D is a better pick, but at a cost of double $$$. Also the chip size of the D200 is fine for the money, APS size is OK with me. But if you want size, then go for the big pro's, ive seen them in the 30Mpixel plus and 3.5 inch square sensor.. Now thats a sensor !.

The D200 has its drawbacks, of course in terms of CCD heat generated. But like i said, the CCD is so much more sensitive compared to CMOS that even with the extra noise you will pick up (And remove Most with dark frames) your image will be captured more quickly. Their will obviosly be some loss of image quality, as will with the Cmos sensors, but in the end (to me anyway) 10 minutes is better than 40 minutes exposures.
There is a reason why Meade or anyone else are not using Cmos detectors for astronomy, as in this cut throat business every QE percentage is money in their bank.
Cooling down the camera should not be a huge problem, with an enclosure around it and a peltier/fan assembly cooling it should do pretty well, even for the Canon series. Theres some good methods out there to try.
Amazing what we come up with to lower the cursed noise.
Heres a link to a good site for the old Cmos Vs CCD debate.
http://www.dalsa.com/markets/ccd_vs_cmos.asp

But at the end of the day, which ever camera one can afford is usually the best camera for the job.

A question, do you own a 20D ?..
I do, and its not as good as you try to make it out to be.
It really needs alot of work to remove noise, very tedious work.

[CometGuy]

I don't doubt that the D200 is a very fine camera, but for astrophotography (deep sky) I am skeptical it will not perform as well as the larger pixel cameras like the 30D/350D or the D50/D70. This is unless there has been a dramatic improvement in Quantum efficiency over previous cameras (Its hard to tell this in normal reviews as they concentrate on shadow noise, which actually doesn't tell you the full story off how the camera performs for astrophotography). Anyway, we shall see.

Gama, I have the 300D (modified) and 350D (standard) and I find the 350D has very low noise levels and is sensitive. As a test I mounted the 350D on a tripod, with a 50mm FL lens set at f2.8 and aimed around Crux I could see stars down to mag 11.0 in just 10 seconds of exposure! A 3 minute exposure with a 100mm f2.8 lens shows stars easily to mag 14.0 - without any dark subtraction.

Terry

[Gama]

Stars will always show down to low magnitude, its the large surface areas of a galaxy or nebulae that has problems.
The 20D has a 6.4 micon pixels, the 350 has the same 6.4 micron as for the D200, its got 6 micron pixels.
While a little smaller than i'd like, still isnt a huge difference.
Regardless, the modified 300D is a great Astro camera, my end results with the Nikon are heading towards the same mods, that is remove the IR filter..
'

[CometGuy]

Just something that completely slipped my mind regarding 2 x 2 binning.

I forgot that random noise sources add in 'quadrature' so that even in software if you bin 2 x 2 you will - at least for an extended object - improve your signal 4 fold, but the noise will only increase square root of 4 or 2x. Therefore the net result should be doubling the signal to noise in an extended object (at the expense of resolution). I see where your going with this now - a 2.5 MPixel 12 micron pitch camera.

There is plenty of software that will bin your images. Actually, essentially a standard resize (i.e bilinear resample) should achieve a similiar result I think.

Terry

[Gama]

Im just hanging out to test it.
It should be here tomorow (Wednesday), but i need a T mount..
I just remembered..... Now look for a Nikon T mount to connect the bleedin thing to the scope..
The biggest killer is the filters for this model. Its pretty bad in terms of killing UV and IR. Hopefully not so much.. Cause in the end the filter is gettin ripped out !.