XT vs. XTI in very low light level conditions

Started Sep 5, 2006 | Discussions thread
ForumParentFirstPrevious
Flat view
RicksAstro
RicksAstro Veteran Member • Posts: 3,879
XT vs. XTI in very low light level conditions

I have a keen interest in the low light performance of the XTi. I have used the 10D, 300D and XT for astrophotography applications, and was intrigued about the higher resolution, and supposedly better microlens and sensitive pixel area of the XTi.

I went to CC today, and they had one! The price was $899 with kit lens, but I mentioned their website had it for $850 on Sat. The guy checked, and he saw it was removed. He believed me, and promptly took $50 off!!!

Anyway, I ran a series of tests comparing the XT and XTi. I used my 70-210 f/4 lens at f/32 in a darkened room this evening under controlled lighting conditions. I used TC80N3 timer remote with plug modified for the Rebels and too 2 minute exposures are ISO800. For low light photography, I’ve found very little Signal/Noise difference between the various ISOs, so I just use them to select dynamic range based on the situation. My normal is ISO800.

With each camera, I took one in very dim light (only my monitor on with dark background) and a lighter one (turned on other monitor with white background and an incandescent lamp)…again all at f/32 for 2 minutes at ISO800…all in dual Raw/JPG.

I used Canon’s RAW converter, with Auto color and no sharpening. For the dim shots, I pushed the exposure up 2 stops using the raw converter. For the lighter shots, I did not compensate. I converted to 16 bit TIFs and, in PSCS2 converted those to JPG using quality 9. No individual treatment was done to the exposures other than the workflow described above (no sharpening, no noise reduction, no levels or curves, no brightening or contrast.)

I’ve posted the full resolution JPGs to my website. They can be downloaded here:

From Rebel XT:
Light conditions:

Dim conditions:

From Rebel XTI:
Light conditions:

Dim conditions:

I need to go over these (really the 16 bit TIFs) to draw a conclusion. Overall, they are fairly similar, with perhaps an edge going to the XTi, particularly when the image is rescaled to match the XT’s pixel scale.

On the issue of the XTi being underexposed compared to the XT…horray! I always left my XT with a ½ stop down compensation to not blow out highlights, so the XTi should be perfect for me. Note that this has nothing to do with inherent noise levels whatsoever, just dynamic range control. My tests show that for these longer exposures, there really is very little difference (certainly not close to ½ stop).

Remember, these are both 2 minute exposures with no noise reduction at f/32 in dim light conditions...both are simply incredible when you think about it!!!

ForumParentFirstPrevious
Flat view
ForumParentFirstPrevious
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow