I hope everyone realises that sensor manufacturing is very capital-intensive, really very with fabs costing usd 1 billion and more
as a result, a relatively small company like Nikon, just cannot enter that field - it's a much more logical decision to leave it to someone else to make the sensors (and earn a low economic return doing that), and earn a higher economic return by adding value on top of the sensor, typically (1) via image processing, and (2) by wrapping the sensor in a very competent camera body.
- until recently anyone that took a Nikon DSLR apart, saw "Sony" printed on the sensor
- more recently, some sensors have shown up with "Nikon" printed on them. Which simply means that Nikon as a customer, asked Sony to please print "Nikon" on the Sony chips
The m43 sensors tend to earn lower markss with DxO because, among other things:
- their IQ at higher ISOs while good, is still behind best practice, and so they get penalised
- the black box used by DxO seems to award a fairly high weight to the number of pixels, but m43 sensors still have fairly low pixel counts
At the end of the day, who cares? DxO is just a black box tool to grade sensors, and history have shown the limits of that black box (the first time they graded mid-frame sensors those came out very low, so every serious photographer made fun of DxO, so they modified the black box...).
What matters is not the DxO score: it's whether the photographer likes the IQ from a given camera and its lens or lenses; and whether the photographer likes the overall package including size, weight, user-friendliness. And you also have to factor in how much it costs.
It is dark in that DXO black box and we cannot see clearly, but it's not only the DXO black box that is dark. The way they behave is even darker, if you consider the timing of the release of the scores for E-M5. This is the most significant and interesting camera that numerous people have wanted to get scored by DXO and openly and repeatedly requested that. Yet DXO only released the scores months later after measurements were taken and known, after all the new cameras have been announced in Photokina, and after the scores for much later released cameras were published. And most intriguing, if not frustrating, is the fact that those results show nothing really unexpected or unacceptable to warrant such a unreasonable delay!!!
Could it be that DXO had little care for consumers who have come to trust and rely on DXO for confirmation of quality, little attention and priority to their own blog members who repeatedly asked for the E-M5 rating, and even no concern for their own growing reputation as a trusted provider of sensor quality information? I think not. There is obviously considerable procrastination and consideration of what their action in this case is doing. Something is very wrong, particularly as no explanation was given. Even when they decided to take a month off and announced that, they would still not bother to release the scores already in their hands.
It is not unreasonable to think that something very secretive has been going on. I think they fully realise they could be influencing buying decisions, which can have very significant financial consequences for Olympus as well as its competitors. I also think that it is not just a co-incidence that those scores are so reasonable as published. Could the scores be very different when measured, since it makes no sense for the to be delayed for so long otherwise? It is unthinkable but then it is not impossible, given what transpired, that there might have been adjustments to the black box, i.e. these scores were cooked. If there is any doubt, then can we trust DXO any more
if scores can be cooked, or measuring methods changed without explanation or notification ? If not
cooked , then there is an even more serious possbility/effect, i.e. the delay could be helping some other brands whether done intentionally or unintentionally. Without a proper explanation, is it unreasonable for us to doubt the integrity of this company and its credibility?
