All this said, what this does show is that the competition is making great strides in dynamic range at low ISOs and I hope Canon follows suit in their future bodies.
Amen to that. By the way, where is this 1DsMark 4?
thedigitalbean wrote:
I think I've said this when DXO comes up and I'll say it again. I can appreciate the value of the graphs they generate. I am uncomfortable with the way in which they come up with a single number so summarize each graph but can deal with it. However the way in which they then come up a single number to represent the quality of a sensor is (without using four letter expletives) retarded.
...
This.
Well known Nikon blogger Thom Hogan (see 'DxO Hysteria') says as much.
I've come to find more value in looking at the comparison graphs...
I don't care about DxO... I do care about the D7000. That cam looks pretty sweet. Well, apart from it being a cropper. I am seriously considering one, instead of the 7D, for studio work. Kudos to Nikon! Also, it makes me wonder how the D700 upgrade will be...
Daan B wrote:
I don't care about DxO... I do care about the D7000. That cam looks pretty sweet. Well, apart from it being a cropper. I am seriously considering one, instead of the 7D, for studio work. Kudos to Nikon! Also, it makes me wonder how the D700 upgrade will be...
Yea the D7000 should be a knock out for studio work. The shadows look as clean as the D3x which of all 35mm cameras had the cleanest shadows. Should be a great tool.
DxoMark summary scores are definitely not very useful. However the detailed DxoMark data is actually very useful. Its kind of like reading a camera review on dpreview or a lens review on photozone. Do you go by the final score? At any review site you have to read the detailed analysis and ignore the useless attempts to sum up the whole analysis with a single number.
Once you dig into the data, there is nothing too surprising in the D7000 versus 5DII analysis. 5DII has close to 1 stop advantage in terms of SNR which is about what you would expect (D7000 does slightly better in terms of "performance per unit area" which is not surprising given its a newer sensor). In color sensitivity and tonal range, 5DII is appreciably better at anything over ISO 200.
So that leaves the DR - first note that above ISO400, 5DII is better in terms of dynamic range than D7000. Its only at lower ISO's that the D7000 has higher dynamic range. Second you have to keep in mind that the way DxoMark defines DR is not field relevant. It is defined in terms of the point where SNR = 1 (0db) i.e. where signal is equal to noise. If you choose a more reasonable lower SNR limit then the results will be different. Say if I define my lowest acceptable SNR to be 20DB (signal is 10 times the noise) then DR at ISO100 will be about 8 stops for D7000 and 7.5 stops for 5DII. Thats not too surprising specially given that canon has often been limited in the DR due to higher read noise.
So in short 5DII is easily the superior sensor overall which is a given considering its larger size. However D7000 sensor performs pretty damn good for an APS-C sensor and is better than the 5DII in terms of "performance per unit area"
jeremy_clay wrote:
If the 1Ds4 doesn't deliver, I'm swapping systems.
Just in case you're serious about this, what would you do for a 50/1.2L or an 85/1.2L? (I know you've shot with at least the first one.) Honest question.
wickerprints wrote:
Just in case you're serious about this, what would you do for a 50/1.2L or an 85/1.2L? (I know you've shot with at least the first one.) Honest question.
For what i do, especially the fashion work, more DR would be more important to me then the 1.2's.
Why can't we all accept that sensor technology is improving and sensors are getting better and better? Meaning, why is it such a surprise that the D7000 sensor is almost as good as the D700 and 5D II? After all, the 5D II's sensor is now an older-generation sensor. I cannot wait for Canon to the release of 5D III .... hopefully next year.
The individual DXO tests seem pretty valuable to me. A person should read how they are tested and know what the test results mean.
As far as the sensor rankiong, they do not claim that it relates to image IQ or anything like that, I really do not see any use for that overall number.
However, its good to know the ISO rating, the DR rating, etc so I can come to my own conclusion as to how the sensor might perform for my usage.
