Daaaaaaaamn! Check out the D5200 sensor!

Started Jan 24, 2013 | Discussions thread
bobn2
bobn2 Forum Pro • Posts: 57,446
Re: Aptina?

Jack Hogan wrote:

dethis2 wrote:

@ Bobn2 and/or Iliah

I had the impression that PRNU should be the same at all ISOs ... but trying to fit DxO's data I calculate about 0.50% for ISO100 and about 0.60% for ISO 200.

I get 0,52% 0,59% 0,61% 0,64% 0,59% 0,00% 0,00% ISO 100-6400, by assuming that all of the reduced SNR at 100% compared to the electon shot noise tangent is entirely due to PRNU (1/SNRtot^2-1/SNRshot^2). These figures are in line with the D800's, but worse than the D7k/D600/D4.

My calculations from DxO SNR gaphs

I think other than ISO 100 the difference can be explained by DXO's raw data massaging and fitting. I still do not quite understand ISO 100 iin the D5200. The sensor/ADC system saturates 1 stop higher at ISO 100 than at ISO200 (i.e. twice the photons arrive), but the data at ISO100 is 'noisier' than expected... Could it be something like Aptina's DR extender kicking in? If so ISO 100 could indeed have very different characteristics than the rest of the ISO scale.

Jack

I find that the figures at base ISO are often anomalous, for many cameras. Enough so that I exclude them from the QE calculation. I think the reason is that the top end is driven into the sensor's non-linear region, the shoulder of the pixel output transistor's response. The gain calculations are made on the assumption of a generally linear response, when it ceases to be, the gain calculation goes a little awry.

-- hide signature --

Bob

Post (hide subjects) Posted by
cpw
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow