Come across this "Diary" via the DPR Forum where the poster introduces his diary, named the "645Diary" himself.
I have had a look at the 645Diary and have some good impressions on it. So, I would like to recommend it here, at my Blog.
Other than those photography stories shared by the author, there are also a few sample photos in small size mostly and occasionally in larger size. In particular, this ISO 800 sample from the 645D (RAW converted) has good noise performance and tonality, which is worth to take a special look.
Wednesday, October 13, 2010
Logging you in...
Comments by IntenseDebate
Subscribe to:
Post Comments (Atom)
Anu · 757 weeks ago
Anyhow, only raw-files should be used for serious noise analysis, and there only a base ISO image is proper to use (vis-a-vis cameras where ISOs beyond the base ISO are software generated).
RiceHigh 110p · 757 weeks ago
Anu · 757 weeks ago
Different DSLRs have different points from which on one should not up the ISO when shooting in raw - dxo-camera database is a good source: one should look at the dynamic-range-tab and when a stop higher ISO loses a stop of dynamic range, the ISO should be avoided. Interestingly, my K20D seems to have a software-created ISO 200, but photosite amplified signal from ISO 400 upwards. But this is slightly off-topic.
RiceHigh 110p · 757 weeks ago
Unless you can provide me the datasheet of the 645D imager (which is a Kodak one btw) that proves the abnormality you claim is true, I just can't believe it, as technically it just doesn't make any sense at all.
Anu · 756 weeks ago
Anu · 757 weeks ago
And yes, I am aware that the 645 uses a Kodak sensor (http://www.kodak.com/global/plugins/acrobat/en/business/ISS/datasheet/fullframe/KAF-40000LongSpec.pdf).
Sample CCD from Kodak - Leica M9: http://www.dxomark.com/index.php/en/Camera-Sensor... - check the DR page. As you see, for each increase in ISO one loses a stop of DR - ie. no PGA.
Medium format http://www.dxomark.com/index.php/en/Camera-Sensor... - don't know if it is a Kodak sensor, but again, for each stop (apart from 50 -> 100, or actually 42->81 is a little tiny bit less, almost certainly not due to PGA or such, though) one loses a stop of DR.
Pentax K10D - a CCD SLR, again, one stop lost for each doubling of sensititivy: http://www.dxomark.com/index.php/en/Camera-Sensor...
Nikon D70 - CCD and the same thing: http://www.dxomark.com/index.php/en/Camera-Sensor...
Phase One, CCD and... http://www.dxomark.com/index.php/en/Camera-Sensor...
Mamiya, CCD, same story: http://www.dxomark.com/index.php/en/Camera-Sensor...
Now, it is actually GOOD if this happens to DR with increased ISO, as it indicates that with the lowest ISO (excessive amount of) signal is not lost due to electric noise.
There is absolutely no reason to believe that Pentax 645 is somehow different.
mrgreycard · 756 weeks ago
quoting one of my trusted "people" .......Many medium format digitals just use a single gain before digitizing the sensor charges, and ISO is just logistics for exposure guidance, and interpreting the RAW file..................... http://forums.dpreview.com/forums/read.asp?forum=...
The 645 differs a bit in it's iso range then expected:Auto (ISO 200-1000)
• Boost (100,1600)
so in all honesty I can't say either of you is right for the 645D........though the lack of a larger normal iso range may imply no PGA....
One more quote re: Pentax k10 from the same referenced thread.......
If one of these high bit depth ADC's were used, there wouldn't be a need to analog amplify the signal before ADC, which is pretty much how the Pentax K10D worked with its 22 bit ADC. However, as this high bit depth would only be of benefit for ISO's less than 400 and only for very very dark levels of less than eight or nine stops below the maximum clipping level, it isn't a necessity given that we can accomplish close enough to the same ends using amplification............... http://forums.dpreview.com/forums/read.asp?forum=...
Anu · 756 weeks ago
Auto-ISO as a mode is irrelevant vis-a-vis this framework of discussion, as is the boost. For maximum image quality they should be avoided, but they're there for comfort and completeness and tell nothing about the underlaying technology.
If it had adjustable gain, it would be kind of revolutionary for a medium format camera - Pentax would make lots of noise about it if that were the case.
(btw, nice to notice you read Sheehy et al on DRP - if only many more did the world would be a better place :) .)