The GetDPI Photography Forum

Great to see you here. Join our insightful photographic forum today and start tapping into a huge wealth of photographic knowledge. Completing our simple registration process will allow you to gain access to exclusive content, add your own topics and posts, share your work and connect with other members through your own private inbox! And don’t forget to say hi!

Photons to Photos analysis of IQ4 150 MP sensor

Shashin

Well-known member
Well, "Photographic Dynamic Range" is not dynamic range. It is a rather complex number that is trying to define the perception on a print. I am not really sure what it actually means.
 

SrMphoto

Well-known member
"This edges out the IQ3 100MP for the highest PDR at PhotonsToPhotos (but not by a statistically relevant margin)."

There is not that much difference between H6D-100c, IQ3 100MP, and IQ4 150MP:

http://www.photonstophotos.net/Charts/PDR.htm#Hasselblad%20H6D-100c,Phase%20One%20IQ3%20100MP,Phase%20One%20IQ4%20150MP

It seems that Phase One shifts their ISO quite a bit.

Bill C.: "From limited DxOMark Measured ISO on other Hasselblad cameras I'd guess the Hasselblad is closer to the "truth" and the Phase One is shifted to the right."
 

narikin

New member
The unusually small drop between 200 and 400 is notable.
Anyone care to explain why this step down is so much smaller than other steps?
Dual gain? they don't advertise it as having this if so.

Does this mean we could use ISO400 for no/miniscule cost over ISO200, or am I missing something here?

Interesting how the IQ3-100 has better DR at 200asa, just, than IQ4. But this is reversed at all other sensitivities.
(select Phase IQ3-100 in the side bar to compare both)
 

Oren Grad

Active member
Well, "Photographic Dynamic Range" is not dynamic range. It is a rather complex number that is trying to define the perception on a print.
Not quite. Bill explains here where his numbers come from:

http://photonstophotos.net/GeneralTopics/Sensors_&_Raw/Sensor_Analysis_Primer/Engineering_and_Photographic_Dynamic_Range.htm

The absolute DR values determined via this approach don't translate directly to a perceptual meaning for most readers. But given the consistent noise threshold applied for all cameras in the calculation, the values are useful for comparing performance.
 

Shashin

Well-known member
Not quite. Bill explains here where his numbers come from:

http://photonstophotos.net/GeneralTopics/Sensors_&_Raw/Sensor_Analysis_Primer/Engineering_and_Photographic_Dynamic_Range.htm

The absolute DR values determined via this approach don't translate directly to a perceptual meaning for most readers. But given the consistent noise threshold applied for all cameras in the calculation, the values are useful for comparing performance.
Actually, Bill does state it is about perception (if that is what your "not quite" reference is about): http://photonstophotos.net/GeneralT...Primer/Photographic_Dynamic_Range_Summary.htm
 

Oren Grad

Active member
Actually, Bill does state it is about perception (if that is what your "not quite" reference is about): Sensor Analysis Primer ?Photographic Dynamic Range Summary
We may actually be having an agreement rather than a disagreement. The "consumer" of the data can't translate those numbers into what they mean for subjective perception of dark tone noise character in long-scale subjects without calibrating them against the experience of processing files from the respective cameras. Even then, cameras with the same numbers may have different noise patterns with different subjective impact.

I think best use of these numbers is to calibrate general expectations about the subject brightness range that's reasonably within the grasp of cameras that are under consideration, and to shed light - so to speak - on design objectives and implementation details for each camera. For example, some are optimized for base ISO performance, some for high ISO performance; some scale smoothly at intermediate ISO values while some do not; and some switch to software tags rather than in-camera amplification beyond certain ISO thresholds. With all that used to help narrow the field, final judgments of suitability for purpose are then best made by processing sample files to taste.
 

Oren Grad

Active member
The unusually small drop between 200 and 400 is notable.
Anyone care to explain why this step down is so much smaller than other steps?
Dual gain? they don't advertise it as having this if so.
Here's what Bill Claff said about it in the DPR thread:

Notice the slight dip at ISO 400. This behaves like dual conversion gain although I see nothing about it in the IMX411 specs.
It's clearer when you look at Input-referred Read Noise.
 
Top