E-5 Raw noise performance = E-30/ ballpark E-3's

Started Dec 18, 2010 | Discussions thread
bobn2 Forum Pro • Posts: 61,149
Re: Look at this

Hi Gusti

mcabato wrote:

i just now read your postings here and congrats.

sorry, not to you, but to me, beacuse 2 weeks ago i commented the reason why the dxo results of the e-5 are so different to what other tests show and so different to the dxo-results of other cams by explaining, similarily as you did here, as being the iso-determination of dxo as basis of their testing, where they do not consider where the cam designer has chosen to put the exposure to a given iso-setting in respect to the point of highlight clipping. maybe from my wording not as precise as yours, but the same explanation.

Well, you understood what was going on, that's more than many. There's a discussion going on about just that here:

so, i need to revers - congrats to you. i had not been thinking that you would actually give the right reasoning instead of emphasizing "slow" g sensor size and dxo results supporting the sensor argument.

you've been listening to the wrong people as to what I'm about.

so, i am happy to see your fair argumentation coming from you.

Really, I try not to say things that aren't fair. I think what I say is generally fair and neutral - it's just that fair and neutral is a threat to some people - it's always been that way. people who's agenda is propaganda after some goal always see the truth as a threat.

actually, the iso comparison of dxo between the e-3 and the e-5 is measured correctly as I tested with both e-3 and e-5 by myself - 1/3 stop difference in exposure between e-3 and e-5. but this is just a minor issue. the real issue is, as you explained, the selection of dxo of the point of correct exposure in their opinion which differs from that which oly has planned in the cam design.

DPR understood this too. See this from their review of the E-PL1:

'Dynamic range has traditionally been one of the shortcomings of the Four Thirds system but this has been addressed on recent cameras by changing what part of the sensor's dynamic range is used to represent middle gray (for more details, see here). As a result, you get significantly greater dynamic range at ISO 200 and above than you do at ISO 100'

dxo seems to just stick to a certain raw-headroom they are used from nikons and apply that for any canon or oly as basis for their measurement - thus canons show worse than nikons in the average at dxo. and olys of course even more ...

What they do is find out how much light it takes to saturate the sensor, and then apply the ISO formula for EI determination from 100% - but that formula doesn't really apply to camera sensors or raw files. ISO really only describes the relationship between metered scene brightness and output image brightness, and there is no output image brightness until you process a raw file - rally ISO is all about processing.

they should adjust their measurements in relation of how the cam exposes jpg ... and that would change the results.

But DxO is all about raw. They should stop calling it 'ISO'. In my data files for Sensorgen I call it 'DxOSO'. I used to call it 'Index of Sensor Operation'.

boah, i did not expect you doing well fair g

Well, thanks for the post.

so, today i am not bulling.

No-one accused you of it. They accused me of it.

thanks. gusti.

and all the best to you.

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