Thread: Bits of the LMC
View Single Post
  #7  
Old 01-01-2012, 08:56 AM
avandonk's Avatar
avandonk
avandonk

avandonk is offline
 
Join Date: Aug 2005
Location: Melbourne
Posts: 4,786
Quote:
Originally Posted by midnight View Post
That's a mighty fine image there Bert.

I hope you don't mind me asking how do you find using ISO1600 as opposed to say ISO800? Would your cooler fridge help in this or is it more so optimising ISO to the particular camera model?

Darrin...

Darrin I use 1600 ISO as it records the very dim stuff over more bits in the final raw image. This makes the faint data easier to realistically separate from the noise.
This ISO is camera dependant.

Cooling the camera to -12C really helps to keep thermal noise down. Without cooling the faint stuff is actually at or less than the thermal noise.
The read noise is the same at 1600 ISO as at 200 ISO. At 1600 ISO the dynamic range is about three stops less than at 200 ISO but my HDR method overcomes this.

There are a lot of rather technical reasons for my methods. This is only a guide as trial and error will sort out the real world performance. Dithering and stacking is the best way to minimise any noise.

Two very useful sources of information about sensors from two very smart and knowledgeable blokes.

http://www.clarkvision.com/articles/...html#low_light


http://www.brayebrookobservatory.org...CD_SENSOR.html

Note the Canon 5D is one of the best performing low light DSLRs. It can only be beaten by the Canon 5D MII IMHO.

Some camera manufacturers cheat to improve signal to noise by simply cutting out the data in the lowest four bits where all the dim stuff is. This has the effect of having very clean backgrounds devoid of faint stars and faint nebulousity!

Bert

Last edited by avandonk; 01-01-2012 at 09:07 AM.
Reply With Quote