a7S high ISO claim - - -over hyped or not ?

Started Jul 3, 2014 | Discussions thread
Rishi Sanyal
Rishi Sanyal dpreview Admin • Posts: 803
Re: a7S high ISO claim - - -over hyped or not ?

andrewD2 wrote:

Two words were in caps for clarity.
Their data for the 6D and 5DIII is presented as 13.1 stops, 12.7 stops so the accuracy of that data should be +/-0.05 of a stop. Since my comparative testing puts them between 2.5 to 3 stops apart for real world useable DR then the DXO numbers are highly misleading and quoted to an unrealistic degree of accuracy.

I do wish DxO put error bars on their measurements. I understand though that they may have considered it and figured it complicated the visual presentation too much.

The comparative tests are simple to perform, I did mine side by side but its possible to standardise the test so you could compare them with a set of bracketed photographs when you don't have the camera to hand any more.

Exactly my point. This is why we only occasionally do side-by-sides... b/c while valuable, they are of limited utility down the road when you can't compare future cameras b/c the scene has changed. And you don't have the old camera with you, nor do you have the time to test every possible combination. That said, in the absence of a better, standardized test, I fully support & encourage side-by-side tests, which is why I did the high ISO comparison against the A7S, A7R, and 5DIII (and 1Dx, D4s, GH4, though these cameras have not been added to the widget yet).

A tool like your excellent ISO comparison tool shows real world results, you can see camera A at ISO800 matches camera B at ISO3200 and can conclude the number of stops apart they are. A tool like that is what is required for seeing what real world DR differences look like.


Thank you, and yes a tool like that is required to see the real world differences. I agree. In fact, you'll be very happy, then, with the real-world DR article I'm working on now pitching the A7S against the A7R.

That said, this sort of real-world test is not the way to do it if you can have a lab setup with a scene that tests the same thing. Admittedly, this is difficult, but we're working on it.

Furthermore, your comment about ISO X on camera A = ISO Y on camera B is something that DxO's SNR data can address (and note, as I point out on page 2 of my High ISO Compared article, that such talk of 'ISO Equivalence' must also specify the particular tone in question, as ISO performance will vary for dark vs. bright tones). So your ostensible distaste for 'read noise this / SNR that' is misguided - it's this very data that can tell you exactly what ISO on camera A looks like what ISO on camera B for what tone (by looking at the points where SNR curves intersect). This data can then inform real-world tests that verify what the measurements show/predict, but are not particularly necessary other than for visualization purposes (which, I agree, are important in the end - since as photographers, we care about the image).


-- hide signature --

“It is a capital mistake to theorize before one has data. Insensibly one begins to twist facts to suit theories, instead of theories to suit facts.” -- Sherlock Holmes

 Rishi Sanyal's gear list:Rishi Sanyal's gear list
Sony RX100 IV Nikon D810 Canon MP-E 65mm f/2.5 1-5x Macro Nikon 85mm F1.8G Sigma 35mm F1.4 DG HSM Art +3 more
Post (hide subjects) Posted by
(unknown member)
Keyboard shortcuts:
FForum PPrevious NNext WNext unread UUpvote SSubscribe RReply QQuote BBookmark MMy threads
Color scheme? Blue / Yellow