Pages: [1]   Go Down

Author Topic: D700 Read Noise Results from Optical Black Data Collection  (Read 2550 times)

bclaff

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 86
    • Photons to Photos
D700 Read Noise Results from Optical Black Data Collection
« on: December 16, 2011, 09:00:01 pm »

You may be aware that Optical Black photosites are captured in the raw data for most Nikon DSLRs.
Unlike values in the active area these are not zeroed and so they are a good source of read noise data.
Until recently I had assumed, erroneously, that all Optical Black photosites are created equal.
I came to question my assumption when I was contacted by someone gathering noise data with my program regarding inconsistent results.
This person, M.R. Blume, collaborated with me in resolving the issue and in collecting excellent D700 read noise data.

D700 raw data has four columns of Optical Back data, two to the left of the active area, and two to the right.
I first modified my program to collect Optical Black data by column.
We then discovered that Optical Black columns adjacent to the active area had values that increased when the image area was exposed to more light.
So I conclude that Optical Black photosites are not perfectly shielded from light.
To collect the best read noise data from Optical Black photosites we switched to "lens cap" shots at 1/8000s.
Effectively eliminating signal and minimizing thermal noise.

Even with this column data, there was an effect that is not surprising; the data was bimodal.
Because the even and the odd rows are going through different ADCs with slightly different offsets.
The program was modified again, to collect even and odd rows separately for each Optical Black column.
Now the data in each of the eight buckets is very clean.

The final twist, is that the numbers indicate a two-stage amplification that I have not previously seen in a Nikon DSLR.



Note that the open circles indicate analog gain and the close circles indicate "Lo" and "Hi" ISO equivalents.

Here's the raw data as well:

ISOnoise
1003.959
1253.988
1583.956
2003.985
2514.956
3176.292
4004.422
5035.615
6347.062
8005.896
10077.548
12699.210
160012.296
201515.075
253918.643
320023.765
403130.348
507937.813
640048.506
806330.205
1015938.100
1280048.646
2560048.948


:-)
Bill
« Last Edit: December 16, 2011, 09:19:08 pm by bclaff »
Logged
Visit me at Photons to Photos

Rob C

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 24074
Re: D700 Read Noise Results from Optical Black Data Collection
« Reply #1 on: December 17, 2011, 03:38:55 am »

Great stuff, Bill: do I sell my D700 or buy a second one?

Rob C

bjanes

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3387
Re: D700 Read Noise Results from Optical Black Data Collection
« Reply #2 on: December 17, 2011, 11:09:51 am »

Bill,

Very interesting data. IMHO, the graph of the data is more easily read by using a log base 2 scaling of the axes using the actual data rather than a linear graph of log 2 values.



Although the read noise expressed in ADU units (data numbers) is of interest in evaluation how the noise would appear in an image, the read noise in electrons is also of interest when one is choosing an ISO when shooting in low light where shutter speed and f/stop considerations limit exposure. Read noise in e- at various ISOs are given for the D700 in Sensogren, and I wanted to see how they compare with your data. I assumed that the sensor gain was the same as for the D3, where it is about 4.12 e-/DN and that the gain would halve for each doubling of ISO. The calculated read noise correlates well with the Sensogren data.



Regards,

Bill
Logged

bclaff

  • Jr. Member
  • **
  • Offline Offline
  • Posts: 86
    • Photons to Photos
Re: D700 Read Noise Results from Optical Black Data Collection
« Reply #3 on: December 17, 2011, 01:58:20 pm »

Bill,

Because of the initial length of the post I omitted the gain and electron data.
I have D700 gain information (actually a close fit to actual data) so here is my table:

ISOnoisegainnoise
ADUe/ADUe
1003.9593.85215.251
1253.9883.85215.365
1583.9563.85215.241
2003.9853.85215.351
2514.9563.05815.153
3176.2922.42715.269
4004.4221.9268.518
5035.6151.5298.584
6347.0621.2138.569
8005.8960.9635.679
10077.5480.7645.769
12699.2100.6075.588
160012.2960.4825.921
201515.0750.3825.762
253918.6430.3035.655
320023.7650.2415.722
403130.3480.1915.799
507937.8130.1525.735
640048.5060.1205.839
806330.2050.1203.636
1015938.1000.1204.587
1280048.6460.1205.856
2560048.9480.1205.892

These new values of 15.4, 8.5, 5.7, 5.9, 5.7, and 5.8 are still consistent with your initial comparison to
Sensogren values of 15.2, 9.1, 5.8, 5.3, 6.0, and 6.0

Regards,
Bill
Logged
Visit me at Photons to Photos

Riaan van Wyk

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 812
Re: D700 Read Noise Results from Optical Black Data Collection
« Reply #4 on: December 21, 2011, 03:13:10 pm »

Great stuff, Bill: do I sell my D700 or buy a second one?

Rob C

Rob, I usually try and wait till at least the eighth page ( with the required eight pages of graphs off course) with discussions like these before forming an opinion. Which doesn't happen anyway as I get more confused with each page and invariably stop reading on page two.   
Pages: [1]   Go Up