Thread: CCD vs dSLR
View Single Post
  #41  
Old 15-06-2014, 05:55 PM
Shiraz's Avatar
Shiraz (Ray)
Registered User

Shiraz is offline
 
Join Date: Apr 2010
Location: ardrossan south australia
Posts: 4,918
Quote:
Originally Posted by Terry B View Post
Interesting idea but not quite true. It does depend on the well depth and the gain used in the CCD. My ST10XME is 16 bit and I use it to do photometry. If the number of levels of grey was less I would not be able to measure the difference on brightness of 2 stars anywhere near as accurately. It is not just stretched data. Big CCD pixels have greater well depth so can be broken into many more individual levels.
I don't know about the noise level at very low signal but this is not where I measure the stars.
your ST10 has one of the expensive chips that does provide a bit better than 12 bits of real data (but still not 16 bits). The read noise level at high levels is exactly the same as that at low levels and you cannot measure to better accuracy than the noise will allow. Your chip has a gain of about 1 electron per ADU (from memory), but a noise level of about 8 electrons RMS, so you cannot make use of the lower 3 bits - your measurements will have an error of about 8ADU RMS because of the noise. ie, you have about 13 bits of usable data.

However, I agree that my earlier statement could have been better worded and it now is.

Last edited by Shiraz; 17-06-2014 at 01:52 PM.
Reply With Quote