"Real ISO" again and DXO charts (not again!)

Started May 1, 2012 | Discussions thread
ForumParentFirstPreviousNextNext unread
Flat view
Peter 13
Veteran MemberPosts: 8,301
Like?
"Real ISO" again and DXO charts (not again!)
May 1, 2012

When comparing different bodies for low light performance, the only thing that matters is the noise at the same f-stop, same shutter speed, same lens and scene. What ISO is reported is irrelevant. I think that we all agree on that.

This makes me wonder if the way DXO reports their data corresponds to that. They test the body A, at, say ISO 1600. If they find out that this sensor is very far from saturation level, than, say, body B, they shift the graph to the left. In other words, they say – not bad but this is really ISO 800, and for ISO 800, it is not so great. Well, to saturate the sensor so much, you have to expose twice as long. But this contradicts our definition of low light performance.

So this brings us to the question – is that shift justified? Assuming that the tests were done at the same shutters speed and aperture – shouldn’t the data be reported at the manufacturer’s ISO? This can change a lot the way we compare cameras.

I can see the value of what DXO does at ISO 100. If you expose to the right, even if that means a lower shutter speed, you can have a photo with better IQ. But I fail to see that for higher ISO.

I guess the answer to that question also depends how they actually test - do they adjust the lighting to reflect the "real ISO", for example, etc.

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