Whilst on the topic of....optimal subexposure length

[markckram]

and...before I get a barrage of but "hold on, that's not right, you see...its not that simple, in fact its a lot more complicated than that....blah, blah"

I found this interesting and succint article on determining your best sub length based on some basic info like your equipment, sky flux, etc > Signal to Noise and Subexposure Calculations

For those who want to cut to the chase...here is a spreadsheet which calculates it for you ( assuming you have the necessary and accurate data like sensor read noise, sky flux, etc ) > http://www.starrywonders.com/subexposurecalculator_V3.xls

My questions is...does anyone have/know where to get any of the data required for a Canon DSLR to pop into the spreadsheet...unfortunately, only certain ccd cameras are catered for

[MartinB]

This is exactly what you should do to determine optimum sub exposure length. Difficult to get the data for DSLRs and I gues the gain is dependent on ISO value.

Here is some info on read noise for some cameras but not sure about the gain data.

Clarkvision: Digital Camera Sensor Performance Summary

[omneferuss]

I'm just being a contrary butt-in here, but where is the spirit of pioneering adventure in all this "is it fixated for me on a spreadsheet somewhere" approach to things ??

As you were.. carry-on.. 'paint-by-numbers' was never my thing.......

[MartinB]

I'm just being a contrary butt-in here, but where is the spirit of pioneering adventure in all this "is it fixated for me on a spreadsheet somewhere" approach to things ??

As you were.. carry-on.. 'paint-by-numbers' was never my thing.......

Each to their own I suppose!

[FraserClarke]

You could try and measure it yourself. You can determine it from a series of flat fields with different light levels. Here is a description of how to do it;

CCD Gain

[markckram]

Thanks Martin and TeaDwarf,

I have a bit of reading to do now, as well as on the Canon DSLR digital astro forum Canon_DSLR_Digital_Astro : Canon DSLR Digital Astro I recenty signed up at.

Quite a lot of interesting discussing there too....

[markckram]

Earlier, I had a brief look through the article you posted Martin, it certainly is the most comprehensive and detailed one on the subject of Canon DSLR sensor characteristics on the net that is still fairly easily digestible by a lay person i've come cross.

I'll certainly need a day or 3 to read and let all that try and sink in

Once again thanks for the link!

[markckram]

You could try and measure it yourself. You can determine it from a series of flat fields with different light levels. Here is a description of how to do it;

CCD Gain

Just read the article, seems easy enough on paper, but what software do I need to calculate the pixel values in my images?

The article describes theory but there is no link to software I need to do the calcs.

[FraserClarke]

Dunno I'm afraid . I don't know much about DSLRs. What is the data format?? TIFF?? You need something that can give you the statistics on an image region. I'd use IRAF, but that's not appropriate here. You could do it by hand in photoshop (use the cursor and the 'info' panel to measure the value of individual pixels) -- but it would be pretty painful. There may be a photoshop plug-in that would do it, but I don't know if it exists.

I don't have any experience with the other common packages I'm afraid, but maybe someone else knows? You just need to get the mean and the standard-deviation of an image area. It's not hard to calculate in software, so I'm sure some package must do it!?!?

[dph1nm]

I suspect it is very tricky to do these calculations correctly with a DSLR. Once you have converted from the raw bayer matrix back into real pixels (which you have to do to run any of these software packages for measuring noise) your pixel values are going to be correlated, thus artificially reducing the noise.

NigelM