November 28, 2014, 07:28:16 AM

Author Topic: D800 just took DXO Mark top spot...  (Read 9101 times)

well_dunno

  • Canon 6D
  • *****
  • Posts: 356
    • View Profile
Re: Nikon D800 tested at DxOMark, gets the #1 spot
« Reply #30 on: March 23, 2012, 01:32:15 PM »
indeed!

canon rumors FORUM

Re: Nikon D800 tested at DxOMark, gets the #1 spot
« Reply #30 on: March 23, 2012, 01:32:15 PM »

unfocused

  • Canon EF 300mm f/2.8L IS II
  • *******
  • Posts: 2205
    • View Profile
    • Unfocused: A photo website
Re: Nikon D800 tested at DxOMark, gets the #1 spot
« Reply #31 on: March 23, 2012, 01:34:38 PM »
Quote
But in the end, i think image quality on all these upper tier cameras are getting so good across the board that we are really splitting hairs with these numbers

True that. In fact, I'd say that's the case with almost all the DSLRs, even the Rebels. Make a 16x20 print from any Canon or Nikon and for 99% of shooting conditions, no one will be able to tell the difference between the $700 and the $6,000 camera. 

The debates on this forum remind me of Sayre's Law: "Academic politics is the most vicious and bitter form of politics, because the stakes are so low."
pictures sharp. life not so much. www.unfocusedmg.com

preppyak

  • 1D Mark IV
  • ******
  • Posts: 802
    • View Profile
Re: Nikon D800 tested at DxOMark, gets the #1 spot
« Reply #32 on: March 23, 2012, 01:36:32 PM »
That is the interesting thing. According to DXO the 7D sensor rates very poorly but I'm not sure I could tell the difference between it and the D7000 sensor in an actual image.
The true test will be when they start running lens tests on the camera.
That's what makes the sensor rating sort of irrelevant as a whole. The sensor is only as good as the glass you put in front of it. So, if anything, DxO is saying that Nikon lenses are generally inferior, if people loved the 5dMII more than its competitors, yet the competitors sensors were better.

As for the ISO #'s, DxO does say to take them with a grain of salt.
Quote
    As with all DxOMark scores, we take into account only image quality. It does not address such other important criteria as image signal processing, mechanical robustness, ease of use, flexibility. These scores are obtained after normalization see “Detailed computation of DxOMark Sensor normalization” for more information.

EYEONE

  • Canon 7D MK II
  • *****
  • Posts: 623
    • View Profile
Re: Nikon D800 tested at DxOMark, gets the #1 spot
« Reply #33 on: March 23, 2012, 01:37:39 PM »
Quote
But in the end, i think image quality on all these upper tier cameras are getting so good across the board that we are really splitting hairs with these numbers

True that. In fact, I'd say that's the case with almost all the DSLRs, even the Rebels. Make a 16x20 print from any Canon or Nikon and for 99% of shooting conditions, no one will be able to tell the difference between the $700 and the $6,000 camera. 

The debates on this forum remind me of Sayre's Law: "Academic politics is the most vicious and bitter form of politics, because the stakes are so low."

Very true!

I'd applaud you if I could.
Canon 5D Mark III w/BG-E11, Canon 7D w/BG-E7: EF 24-70mm f.2.8L, EF 70-200mm f2.8L IS II, EF 40mm f2.8 Pancake STM, Speedlite 430EXII + 430EXI, Canon EOS 3

WilliamG

  • Guest
Re: Nikon D800 tested at DxOMark, gets the #1 spot
« Reply #34 on: March 23, 2012, 01:38:42 PM »

In the past they wrote an article on how the S95's sensor performed better than that of the D3s. Does that sound like bias towards Nikon to you?

And does that sound like a remotely credible conclusion to you?

I don't think DxO is "biased" per se but I do believe that their methodology is inherently skewed against Canon sensor technology, and not necessarily in a fair - or significant-to-the-end-result - way.

Someone else has suggested he'd expect that in Real World use, there'd be very little between the 7D and D7000: I can say with absolute rock-solid certainty that he's right, because I've torture-tested umpteen D7000 files in comparison to my 7D's output (I shoot shoulder-to-shoulder with someone who regularly uses a D7000) and in actual use there's absolutely sweet FA to choose between them - even when hammering the shadows in PP - once you're beyond 100 ISO.

Yet that tiny little advantage - the extra usable DR at base ISO - gets the D7000 a hugely "superior" DxO rating, which is utterly irrelevant in the Real World. 

Frankly, I couldn't give a toss about what DxO has to say about either the D800 or the 5D Mk III: my eyes tell me what I need to know, and everything I've seen so far suggests that while each will have its strengths, neither will be significantly better than the other at the image level.

And nothing else matters.

I could not agree more with this statement. These numbers are all well and good, but seriously....  who CARES?! If you want to take side-by-side photos from two different bodies and compare them, perfect! But these numbers games are just silly, especially with no lenses on the cameras.

lol judging what camera is the best with no lens on just boggles my mind.

lol

  • Canon 7D MK II
  • *****
  • Posts: 514
    • View Profile
    • My dA
Re: Nikon D800 tested at DxOMark, gets the #1 spot
« Reply #35 on: March 23, 2012, 01:56:39 PM »
It might help understand the DxO tests and what they mean. Which basically means you should ignore the overall score and look at the components contributing to it.

The "colour depth" score I believe is influenced in large part by the strength of the colour filters. Stronger colour filters lead to better colour separation, at the cost of more light lost. Canon have generally used weaker filters, especially the red one, and can give noticeably colour differences. Unless Canon have changed there, this isn't a strong spot.

The "dynamic range" score appears to be derived from the maximum at low ISO, where Nikon/Sony/Pentax have in recent times been doing something to get more range at low ISO. Canon so far has not. So this is another weak area for Canon.

Finally, there is the "low light ISO" score. I believe this is a simple measurement of signal to noise level. Where it may deviate from practice is I do not believe it has consideration for the character of noise. That is, some types of noise patterns may be more objectionable even if it is of the same nominal level.

Overall... regardless of how the sensor performs, I wouldn't expect the 5D3 to score well unless Canon are doing something radically different in the sensor from the past, which as far as can be gathered so far, doesn't appear to be the case.
Canon 1D, 300D IR, 450D full spectrum, 600D, 5D2, 7D, EF 35/2, 50/1.8, 85/1.8, 135/2+SF, 70-300L, 100-400L
EF-S 15-85, TS-E 24, MP-E 65, Zeiss 50/2 macro, Sigma 150 macro, 120-300/2.8 OS, Samyang 8mm fisheye

Marsu42

  • Canon EF 400mm f/2.8L IS II
  • *******
  • Posts: 5124
  • ML-66d / 100L / 70-300L / 17-40L / 600rts
    • View Profile
    • 6D positive spec list
Re: Nikon D800 tested at DxOMark, gets the #1 spot
« Reply #36 on: March 23, 2012, 01:58:55 PM »
I could not agree more with this statement. These numbers are all well and good, but seriously....  who CARES?! If you want to take side-by-side photos from two different bodies and compare them, perfect! But these numbers games are just silly, especially with no lenses on the cameras.

I'm eagerly awaiting the moment when Canon releases a high mp body at last - my prediction is that numbers will matter again when our brand is in the leading position... if that should ever happen again.

True that. In fact, I'd say that's the case with almost all the DSLRs, even the Rebels. Make a 16x20 print from any Canon or Nikon and for 99% of shooting conditions, no one will be able to tell the difference between the $700 and the $6,000 camera. 

... as long as you don't crop too much! Which is more necessary when using a prime lens, but imho is one of the main advancements in digital photography in comparison to the good ol' days. But usually I'm the one stating that you're just fine with standard rebel gear and don't need a ff body to get a good picture :-p

The debates on this forum remind me of Sayre's Law: "Academic politics is the most vicious and bitter form of politics, because the stakes are so low."

... "low stakes" is relative: Spending a good part of your savings for your recreation is a serious thing to some (including me), and for professional photography it's worth money and concerns your future to have an edge.

canon rumors FORUM

Re: Nikon D800 tested at DxOMark, gets the #1 spot
« Reply #36 on: March 23, 2012, 01:58:55 PM »

awinphoto

  • Canon EF 300mm f/2.8L IS II
  • *******
  • Posts: 2010
    • View Profile
    • AW Photography
Re: Nikon D800 tested at DxOMark, gets the #1 spot
« Reply #37 on: March 23, 2012, 02:10:49 PM »
The debates on this forum remind me of Sayre's Law: "Academic politics is the most vicious and bitter form of politics, because the stakes are so low."

... "low stakes" is relative: Spending a good part of your savings for your recreation is a serious thing to some (including me), and for professional photography it's worth money and concerns your future to have an edge.

It is but we're getting to the point where differences are getting so razor thin, such as sensor test scores, 51 vs 61 AF points, etc...  You give a professional photographer a 5d3 and top of the line lens and a D800 with equivalent lens or even a 1dx and D4 or D3s for that matter and lens, print all at 16x20, I doubt you will be able to tell one from another.  Heck i'm up for that... Anyone want to mail me all those cameras and lenses for testing purposes, i'm willing to test and get back to you all with the results =) 
Canon 5d III, Canon 24-105L, Canon 17-40L, Canon 70-200 F4L, Canon 100L 2.8, Canon 85 1.8, 430EX 2's and a lot of bumps along the road to get to where I am.

EYEONE

  • Canon 7D MK II
  • *****
  • Posts: 623
    • View Profile
Re: Nikon D800 tested at DxOMark, gets the #1 spot
« Reply #38 on: March 23, 2012, 02:12:10 PM »


The debates on this forum remind me of Sayre's Law: "Academic politics is the most vicious and bitter form of politics, because the stakes are so low."

... "low stakes" is relative: Spending a good part of your savings for your recreation is a serious thing to some (including me), and for professional photography it's worth money and concerns your future to have an edge.

That was not the point. I believe he's referring to the stakes of forum arguments.
Canon 5D Mark III w/BG-E11, Canon 7D w/BG-E7: EF 24-70mm f.2.8L, EF 70-200mm f2.8L IS II, EF 40mm f2.8 Pancake STM, Speedlite 430EXII + 430EXI, Canon EOS 3

psolberg

  • Canon 7D MK II
  • *****
  • Posts: 483
    • View Profile
Re: Nikon D800 tested at DxOMark, gets the #1 spot
« Reply #39 on: March 23, 2012, 02:25:09 PM »
I fully expect these results to be discredited in a board based on canon gear, but regardless, the nikon D800 is indeed a home run hit and at a price point that should be praised. The dynamic range alone shows just how nikon/sony know what they are doing sensorwise and its low light score puts to rest a lot of the fears about 36MP. It is an amazing acomplishment even if it is not #1 at everything.

let's not filter things with a canon lens (no pun intended). this is a great body and deserves recognition as one.

dilbert, they also significantly improved the AF on the 5D3.

I feel the tests DxO runs is heavily biased towards Nikon anyways.  It's hard to believe not a single Canon camera is in the top 10, and the 7D isn't even in the top 50.

That is the interesting thing. According to DXO the 7D sensor rates very poorly but I'm not sure I could tell the difference between it and the D7000 sensor in an actual image.

And I'm really confused by the DXO rating on ISO. The real world images do NOT reflect what those numbers are saying. Maybe it has to do with upscaling or whatever (I don't know much about that however). Maybe someone can explain better.

they explain it themselves. don't look at the overall score for answers on that. look at all the numbers. dynamic range, color depth, etc. not just at the final number. Like ALL synthetic tests, you may find the differences in the actual images you make to not be as great.

Quote
I'm eagerly awaiting the moment when Canon releases a high mp body at last - my prediction is that numbers will matter again when our brand is in the leading position... if that should ever happen again.

well said. what's the point of trying to pretend things like these don't matter when they don't read favorably to our choices and then switch opinions when they justify our decisions. why can't people just enjoy what they have instead of pointlessly comparing it to make them feel good.

Quote
The thing everyone should keep in mind is that DXO isn't testing with Lenses yet... It's testing JUST THE SENSOR...

While it may be the best sensor available...
the consensus on every review that I've seen is that it is TOO GOOD...
How is it that even possible?
Well, the lenses can't hold up to the quality of the sensor...
one brief example is Chromatic aberration...
If the 85 1.4 was producing a 1.5 px wide magenta fringe on the D700, it'll now produce a 3px wide aberration. 
Sensor size didn't change... but the number of recording pixels increased.. increasing resolution.. but  simultaneously increasing artifacts caused by lens flaws.

The true test will be when they start running lens tests on the camera.

Until then...I wouldn't even sweat it...

OH... FYI... a few years back canon halted development on the 1DsIV because the sensor they were developing was better than their glass...

that is off course NOT correct. pixel pitch is what is the main factor. the D800 has a more forgiving pixel pitch than a 7D and the same as a D7000, neither which stresses lenses beyond what they can deliver confortably. 36 sounds like a lot but they are spread out. it is harder on a lens to resolve a single 7D image than a D800 image. So far I think few complain their lenses can't handle the 7D or D7000. I don't see why the D800 changes anything in that regard.

Orion

  • Guest
Re: Nikon D800 tested at DxOMark, gets the #1 spot
« Reply #40 on: March 23, 2012, 02:31:23 PM »
talking about biased DxO mark scores when the D800 is scored @ 95, compared to the Phase One Q180, which stands @ 91?????? That must be a whole lot of bias there!  ::)

I also don't really care about pixel/atomic level views of a camera, which is why I stay on the subject that if a camera costs so much more money than a competitor that offers the same AND more, then I need to stop in my tracks and ask questions and ponder things . . . . especially since Canon sells more cameras than practically Nikon and Sony combined (*?), or close to it.

5DmkIII is a great camera, as is the D800 . . for sure! There is another side to it though, and it has nothing to do with pixel peeping. . .

RichATL

  • PowerShot G1 X II
  • ***
  • Posts: 65
    • View Profile
Re: Nikon D800 tested at DxOMark, gets the #1 spot
« Reply #41 on: March 23, 2012, 02:39:40 PM »


that is off course NOT correct. pixel pitch is what is the main factor. the D800 has a more forgiving pixel pitch than a 7D and the same as a D7000, neither which stresses lenses beyond what they can deliver confortably. 36 sounds like a lot but they are spread out. it is harder on a lens to resolve a single 7D image than a D800 image. So far I think few complain their lenses can't handle the 7D or D7000. I don't see why the D800 changes anything in that regard.

Actually its a simple matter of physics...
The sensor size didn't change... but the number of pixes tripled (between the D700 and the D800)
No matter HOW you look at it... the higher the resolving power of the sensor, the better... but on the same token... the flaws of the lenses are also increased.

a 1px wide CA on D700 witha particular lens WILL equate to a 3px CA on the D800...
Need a calculator?

awinphoto

  • Canon EF 300mm f/2.8L IS II
  • *******
  • Posts: 2010
    • View Profile
    • AW Photography
Re: Nikon D800 tested at DxOMark, gets the #1 spot
« Reply #42 on: March 23, 2012, 02:40:25 PM »
I fully expect these results to be discredited in a board based on canon gear, but regardless, the nikon D800 is indeed a home run hit and at a price point that should be praised. The dynamic range alone shows just how nikon/sony know what they are doing sensorwise and its low light score puts to rest a lot of the fears about 36MP. It is an amazing acomplishment even if it is not #1 at everything.

let's not filter things with a canon lens (no pun intended). this is a great body and deserves recognition as one.

dilbert, they also significantly improved the AF on the 5D3.

I feel the tests DxO runs is heavily biased towards Nikon anyways.  It's hard to believe not a single Canon camera is in the top 10, and the 7D isn't even in the top 50.

That is the interesting thing. According to DXO the 7D sensor rates very poorly but I'm not sure I could tell the difference between it and the D7000 sensor in an actual image.

And I'm really confused by the DXO rating on ISO. The real world images do NOT reflect what those numbers are saying. Maybe it has to do with upscaling or whatever (I don't know much about that however). Maybe someone can explain better.

they explain it themselves. don't look at the overall score for answers on that. look at all the numbers. dynamic range, color depth, etc. not just at the final number. Like ALL synthetic tests, you may find the differences in the actual images you make to not be as great.

Quote
I'm eagerly awaiting the moment when Canon releases a high mp body at last - my prediction is that numbers will matter again when our brand is in the leading position... if that should ever happen again.

well said. what's the point of trying to pretend things like these don't matter when they don't read favorably to our choices and then switch opinions when they justify our decisions. why can't people just enjoy what they have instead of pointlessly comparing it to make them feel good.

Quote
The thing everyone should keep in mind is that DXO isn't testing with Lenses yet... It's testing JUST THE SENSOR...

While it may be the best sensor available...
the consensus on every review that I've seen is that it is TOO GOOD...
How is it that even possible?
Well, the lenses can't hold up to the quality of the sensor...
one brief example is Chromatic aberration...
If the 85 1.4 was producing a 1.5 px wide magenta fringe on the D700, it'll now produce a 3px wide aberration. 
Sensor size didn't change... but the number of recording pixels increased.. increasing resolution.. but  simultaneously increasing artifacts caused by lens flaws.

The true test will be when they start running lens tests on the camera.

Until then...I wouldn't even sweat it...

OH... FYI... a few years back canon halted development on the 1DsIV because the sensor they were developing was better than their glass...

that is off course NOT correct. pixel pitch is what is the main factor. the D800 has a more forgiving pixel pitch than a 7D and the same as a D7000, neither which stresses lenses beyond what they can deliver confortably. 36 sounds like a lot but they are spread out. it is harder on a lens to resolve a single 7D image than a D800 image. So far I think few complain their lenses can't handle the 7D or D7000. I don't see why the D800 changes anything in that regard.

Indeed the D800 is a fine camera... no one is disputing this or even saying the canon 5d3 is superior to that of the D800... it is what it is... but then again, everything in consideration, it is what it is, a tool.  For what it's worth as well, the D7000 was a 16MP apc and the 7d was 18Mp... The 7D if made into a full frame, would be approx 45-46MP which would put the D700 closer to 40Mp, more dense than the D800, but that's neither here nor there... The 36MP is closer to the 12-13MP APS-C sensor.  Everyone seems to be losing the fact and purpose in looking at these tests... just because the D800 may have a #1 sensor on one websites test based on their conclusions, valid or not, does not mean that the 5d3 is crap.  It doesn't diminish the 5d3 in it's role or place or purpose.  Both are very fine cameras...  It isn't one or nothing... Best or not... all this debate is silly.  It's like being pissed that you got honor roll in school but not Valedictorian...  Lets just settle down, enjoy our cameras, and if the 5d3 or any canon camera doesn't tickle your fancy, dont get it or find some other camera that does do it for you.  It doesn't make your decision or our decisions any more right or wrong, it is what it is. 
Canon 5d III, Canon 24-105L, Canon 17-40L, Canon 70-200 F4L, Canon 100L 2.8, Canon 85 1.8, 430EX 2's and a lot of bumps along the road to get to where I am.

canon rumors FORUM

Re: Nikon D800 tested at DxOMark, gets the #1 spot
« Reply #42 on: March 23, 2012, 02:40:25 PM »

CNfuzzy

  • Guest
Re: Nikon D800 tested at DxOMark, gets the #1 spot
« Reply #43 on: March 23, 2012, 02:47:59 PM »
I thought the new Canon and Nikon were actually 14bits - max theoretical DR would be 14 stops. DxOMark DR score for the D800 is 14.4 Evs. Am I missing anything?

psolberg

  • Canon 7D MK II
  • *****
  • Posts: 483
    • View Profile
Re: Nikon D800 tested at DxOMark, gets the #1 spot
« Reply #44 on: March 23, 2012, 02:58:07 PM »


that is off course NOT correct. pixel pitch is what is the main factor. the D800 has a more forgiving pixel pitch than a 7D and the same as a D7000, neither which stresses lenses beyond what they can deliver confortably. 36 sounds like a lot but they are spread out. it is harder on a lens to resolve a single 7D image than a D800 image. So far I think few complain their lenses can't handle the 7D or D7000. I don't see why the D800 changes anything in that regard.


Actually its a simple matter of physics...
The sensor size didn't change... but the number of pixes tripled (between the D700 and the D800)
No matter HOW you look at it... the higher the resolving power of the sensor, the better... but on the same token... the flaws of the lenses are also increased.

a 1px wide CA on D700 witha particular lens WILL equate to a 3px CA on the D800...
Need a calculator?
you're wrong, peirod.

it is a matter of physics and the physics of surface area tells us that 36MP full frame is more spread out than 18MP of an APSC sensor. It is basic geometry. therefore the same CA fringe of say 1mm recorded on the surface of a denser sensor will look LARGER than on the less dense sensor regardless of their pixel count when viewed at 100% cropping becaus the denser sensor has an inherited LARGER magnification when viewed on your screen at 100% crops, and thus it will MAGNIFY CA and other flaws. need a calculator? pixel pitch is what matters, NOT pixel count because the areas are NOT THE SAME!

the bottom line is that it is a myth that the d800 is more difficult to shoot or more demanding than the D7000 which has less pixel density than the 7D. So in this regard the 7D is by far more problematic. not hearing much complains from 7D's so D800 shooters are A-ok. period.

several nikon "experts' have already came out and said it
http://bythom.com/
« Last Edit: March 23, 2012, 03:01:08 PM by psolberg »

canon rumors FORUM

Re: Nikon D800 tested at DxOMark, gets the #1 spot
« Reply #44 on: March 23, 2012, 02:58:07 PM »