Canon EOS 1000D v 450D Dark Noise @ ISO800

[Psychobilly]

I left the 1000D and 450D cameras capturing Dark Sequences in the OBS last night under the control of Yoddha's APT Software...

20 frames were captured at 300, 600 , 900 and 1200 Seconds with 30 seconds left between each frame at ISO800

In camera Long Exposure and Hi ISO Noise reduction were both turned off...

I shot Raw + L Jpeg and have assembled the middle Jpeg from each set of exposures into a Comparison Chart that can be downloaded from here...

Its a 6MB Download and needs to be viewed zoomed to compare the images...

Theres no point in posting a thumbnail as all the noise would be lost...

Peter...

[jgs001]

An interesting comparison Peter. Looks like the 450d is better controlled...

[Psychobilly]

It does indeed... but then again I would have hoped so...

The newer 500D is even better....

Think i'll fill inthe gaps tonight with a set of 150,450,750,1050 Darks...

Bias noise is also lower...

Peter...

[jgs001]

Aye very true, but it's nice to see it in "print" as it were, and the 550d is better yet...

[Psychobilly]

The smaller pixels on the 550D might be offset with better gapless microlensing as well...

Its nice to have 2 instances of a single app controlling both cameras from the same computer as well

Yoddha is doing a might fine job with APT can't go wrong for the money really...

We did find something out with it if your using Antivibration Delay with Digic III based Cameras - I guess he may add more info... and there's a workround for it...

Peter...

[iamjulian]

Interesting comparison, thanks.

Photography and astronomy hold equal sway for me, and thankfully they are nicely compatible. To that end I have just upgraded to the 550D The ISO comparisons I have seen for it looks like quite a nice step up.

[jgs001]

Now that sounds interesting Peter... Hopefully he will... I've not had a chance to try it yet... I suspect I don't actually really need to worry about using mirror lockup with my current setup...

[Psychobilly]

Its the fact that you can get away with a very short inter sub gap (I leave 30s which is plenty of time to make sure the camera saved the data to the card and PC ) with both cameras that makes them so useful and was the main reason ffrom moving on from the 350D ...

Need to get an 62mm R72 filter for the 350D now as its likely just to be used for IR work from now on...

John,

Me niether... the software implements Mirror lock in a few different ways and its better to use either the 2s or 10s self timer trick which it supports and automatically adds the tiem to the exposure duration... I was rather worried though when i first tried using it using the arbitary length mode...

Peter...

[Deneb]

Im sure when I bought the 450D, I was making the right choice .., a very good comparision Peter.

Nadeem.

[Psychobilly]

I still think the 1000D is excellent VFM... and has decent noise performance when compared to the earlier cams... and was about to buy another when your Modified 450D came up

Peter...

[dph1nm]

One has to be rather careful with these comparisons. There is a similar test here, where the 1000D looks noisier to the eye than the 450D. But if you look at the graphs lower down you see that the 450D, although it has a lower mean dark signal, has a higher standard deviation than the 1000D (and is the standard deviation which matters).

NigelM

[Psychobilly]

I dont do Maffs..

If CS3's analysis tools are to be believed...

This is what it reckons for the last "dark" at each exposure and is for the Raw files...

It's all Geek to me....

Peter...