1. I have never said that the DxO measurement is wrong as far as being one possible measure of a camera's sensitivity, since after all, it is just a slight variation of the ISO saturation-based measure of sensitivity, Ssat. What I have said is that is wrong for anyone to declare that this is the unique ISO sensitivity measurement, and that camera makers are therefore wrong when they report other ISO measures of sensitivity such as ISO Standard Output Sensitivity [SOS] or ISO Recommended Exposure Index [REI].
BJL, This is what you wrote a while back. "You would then learn that if indeed DxO claims that "most manufacturers exaggerate their ISO sensitivity", then DxO is utterly wrong: instead, most manufacturers are following the current ISO standard, using measurements according to either "Standard Output Sensitivity" or "Recommended Exposure Index", as is in fact required of Japanese camera manufacturers. They are simply not using DxO's preferred choice of equating "more than the standard-specified _minimum_ amount of highlight headroom in raw files" with "overstating the sensitivity".
Now, I have never used the word 'wrong' in relation to any manufacturer's methodology of setting their ISO standard. I've used terms such as exaggerate and overstate, but of course I understand that such overstatement can only be described as an overstatement in relation to a different standard, or different methodology.
The essential requirement for me, and I suspect many others, is that whatever method of measuring and describing ISO sensitivity is used, it should be a consistent
method so that meaningful comparisons can be made.
If it is true that DXOMark's methodology is consistent, then it's clear that most other manufacturer's methodology of nominating ISO is not consistent
, and therefore not reliable.
For example, one manufacturer overstates their ISOs by 1/3rd of a stop, on average, and another manufacturer, such as Olympus, overstates the ISOs of a particular model, the OM-D E-M5, by a full stop, on average, in relation the consistent ISO standard of DXO
Each manufacturer will have its own reasons and justification for allowing a particular degree of highlight headroom, and that degree of headroom would seem to vary from brand to brand.
My understanding is that most people who use DSLRs shoot in jpeg mode. This is my experience when chatting with other people using DSLRs, during my frequent travels. It's therefore quite understandable to me that manufacturers, also realising that most of their customers shoot in jpeg mode, will employ all sorts of tricks to help them avoid blowing out highlights.
One such trick that Canon used was termed, "Highlight Tone Priority", and it worked.
The reason why it worked so well is because the real ISO setting the camera used when one took a shot in 'Highlight Tone Priority' mode was automatically set one stop lower than that idicated on the dial. This is why 'Highlight Tone Priority' was not available at base ISO, because there was no lower ISO available for the camera to use.
It seems to me that Olympus have used a similar trick to reduce the risk of blowing highlights. They've simply renamed all their ISO settings one full stop higher. What is actually ISO 100, from the perspective of a RAW shooter who is concerned about ETTR, Olympus describe as ISO 200 on the E-M5, and what is actually ISO 800 (or 782 to be precise), Olympus describe as ISO 1600.
This is just as effective as Canon's Highlight Tone Priority. In fact, in a sense it's even more effective because it can be used at base ISO.
Now let me say, I have no objection to Olympus catering to the needs of its jpeg shooters. In the distant past, on the few occasion I used jpeg because I was running out of memory, in the days when a 1GB Compact Flash card was considered enormous, I recall being annoyed with myself for not underexposing all shots by one full stop, because I realised too late that that's what would have been required to avoid blowing out highlight detail in sky and waterfall in jpeg mode.
I understand perfectly the reason why Olympus would overstate all its ISO settings by one full stop, in relation to the 'sensor saturation' methodology used by DXO which is of course mainly relevant to those who shoot in RAW mode and who attempt to achieve an ETTR exposure.
In fact, if I were in charge of designing cameras for some manufacturer, and the market research had indicated that the vast majority of customers shot mainly in jpeg mode, I would probably recommend such overstatement of ISO sensitivity, realising that only a few of us are able to achieve, as a matter of normal practice, the full DR potential of our cameras without blowing essential highlights.