Nikon D800 tested at DxOMark, gets the #1 spot

[Shadowblade]

Originally Posted by ejenner

I don't take these scores seriously. There are so many inconsistencies and purely from a physics point of view I don't believe the DR of the D800. Compare for instance the DR and ISO scores of the 5DII and S100 - it makes absolutely no physical sense.

Totally what I expected though. 5DIII will come in just above the 5DII, DR no more than 12.5. In reality both will have a useable DR of about 7-8 stops.

Downloading raw files and processing them yourself is the only way to really get a handle on the differences, I certainly wouldn't base a buying decision of DxO scores.

In that case, do you believe the DR of the iQ180 (and other recent Phase One backs)?

Having seen them in action, I certainly do - you can push their shadows a long way to extract detail. You can push D7000 shots just as hard. Try doing the same with a 7D, 5D2 or even 1Ds3, and all you'll get is noise.

[tgamron]

Originally Posted by Colorblinded

Anyone who bases all their decisions off what DxO has to say is usually a bit silly.

It's not about vision, it's about the score.

[jwcdds]

Now if you can only convince Nikon to pump out some quality TS lenses to help out the landscape crowd. If they ever do that, then I think the landscapers of the world can finally rejoice.

[tgamron]

Originally Posted by rayman102

Holy Batman! I just canceled my 5DIII and just ordered the D800e. It is going to be long wait but a score of over 14 on DR in Landscape! WOW

I'll give you $400 for the 17-40.

[tgamron]

Originally Posted by Shadowblade

In that case, do you believe the DR of the iQ180 (and other recent Phase One backs)?

The iQ180 is nice, but costly.

[AngryCorgi]

Originally Posted by tkbslc

I've always considered DxO severely biased ever since they ranked a Nikon D5100 ahead of Canon 5Dmk2 and D3100 ahead of 7D.

Even though they are using the same criteria as before? To be fair, they actually rate the 5D2's DR HIGHER than most people.

http://www.sensorgen.info/CanonEOS_5D_MkII.html

These guys give it only 11.2EV of DR and give the D7k (no D5100 in their list) 14EV of DR. Are they getting paid by Nikon too? You may not LIKE or AGREE with the applicability of the scores, but they are standardized and use a controlled process to achieve their data.

As far as the D3100 vs 7D bit, they basically say they are equal, not that the D3100 is appreciably better. But they are measuring sensors. The 7D has a TON of capabilities the D3100 lacks, due to their pricing points/placement.

[bomzai]

What an earth shattering report!! Sony sensors scoring 2 stops higher than Canon! Wow... just like in any other report for last two years.

*Yawn*, moving on.

[AngryCorgi]

Originally Posted by bomzai

What an earth shattering report!! Sony sensors scoring 2 stops higher than Canon! Wow... just like in any other report for last two years.

*Yawn*, moving on.

True. Anyone who didn't see this score coming had to have been asleep for the past 5 years.

[woos]

I'm a bit curious how they got the DR to over 14 ... anyone have a good explanation for this? lol

[Möntgomery]

Originally Posted by woos

I'm a bit curious how they got the DR to over 14 ... anyone have a good explanation for this? lol

The same reason they were able to get the D5100 & D7000 to 13.9
Sony just has better tech...

[woos]

Yeah, 13.9 makes sense, but over 14 seems odd cuz it's only 14-bit raw. :P

[Shadowblade]

Originally Posted by woos

Yeah, 13.9 makes sense, but over 14 seems odd cuz it's only 14-bit raw. :P

It makes sense. Dynamic range and bit depth have nothing to do with each other.

Dynamic range is the difference in brightness between the lowest level that can be discerned from background noise, and the highest level the sensor can record. 14.4 bits of DR would mean that the brightest signal recordable is 21619 times brighter than the darkest signal discernable against background noise.

14-bit raw just means that this range is divided up into 16384 discrete luminance levels. You could have a sensor with 18 stops of DR and only 14-bit output - it just means that the 18 stops of DR will be divided up in to 14 bits of discrete luminance values.

[rayman102]

Darn Shadowblade, you sure knows your landscape Algebra.

[Ricku]

Well, this report certainly seals the deal for me.

I would stick with Canon if I knew they had a similar sensor up their sleeves, but I think it is safe to say that they don't. At least not within the next 3 years.

[Jahled]

Hahaha this thread is mental