Lenses perfomance on D800E v. D810

[papo]

<p>Hello,<br>

According to www.dxomark.com, various performance indicators (DoXMark Score, Sharpness and Transmission) are always slightly better for each lens when used with the D800E than with the most recent D810 (these indicators are quite higher than those obtainable using the D800).<br>

Does anyone have an idea why?</p>

<p>Thanks for your replies.</p>

<p>Olivio Agrenti FRPS</p>

[Andrew Garrard]

<p>Hmm. No, no idea, unless they tested everything at ISO 200 or something (middle ISOs are a little better on the D800 than D810, and they may have changed the filters a bit). The D810 should be very slightly better than the D800e in that it doesn't split and then re-join light paths - I'm assuming Nikon didn't actually change the sensor stack thickness on the D810 when they took out the low pass filters, because that would actively be detrimental. I've no complaints about the image quality from the D810, although I'd take a little better performance at higher ISO. I used to have a D800E and I've certainly not noticed any step back - and the number of other minor improvements made the upgrade significant to me.</p>

[ilkka_nissila]

Test variability (it is very difficult to get identical results at high precision every time something is measured) and variability

in the sensor samples (the components do operate in the physical world and there are small differences between

components in each camera sample.

[Hector Javkin]

<p>The differences in scores between the two bodies are small. The graphs almost look like two exemplars of the same camera. I don't know if DXOMark publishes their estimate of the margin of error of their scores, but measurement error, or sample to sample variation, could account for the differences.</p>

[joseph_smith3]

I have and use both bodies with my Nikon lenses. I cannot detect any differences in sharpness.

 

Joe

[rodeo_joe1]

There's a period of, what,

four years(?) between the

D800E and the D810. If DXO are still using the same sample of lens, a lot can happen to a lens in 4 years. Wear can introduce a small amount of decentring for example, or contamination from outgassing grease could reduce contrast.

 

OTOH, if it's not the exact same lens; well, there's the explanation for different results.

[Two23]

<p>I've shot both cameras; own two D800E. In reality there is no discernible difference in image quality between the cameras. So, I saved myself about $2K and kept using the D800E bodies.</p>

<p>Kent in SD</p>

[ShunCheung]

<p>Nikon introduced the D800/D800E in February 2012: http://www.photo.net/nikon-camera-forum/00ZyTv<br>

And the D810 as, at least IMO, a minor update in June 2014: http://www.photo.net/nikon-camera-forum/00cfUw</p>

<p>In other words, even the D810 is over 2.5 years old and an update is overdue.</p>

[Andrew Garrard]

<p>I would hesitate to make the D810 sound too much out of date, or too minor an update over the D800e. Each individual improvement was minor compared with the D800, but in combination the update was, in my experience, substantial - although one could say the same about the D4s vs the D4. The D810 remains pretty competitive against the competition. It's at the low end of the resolution curve, but I'm one of the most vocal requesters of more pixels, I think, and even I'm not complaining all that loudly; I'd really like a moderate resolution boost for crop shooting and to make 8K time lapse an option (I have an eye on the forthcoming Dell UP3218K monitor) - and give integer sampling for any 4K mode that's in there - but the 5Ds doesn't do 4K either. I'd like slightly better high ISO, as the A7R2 achieves, but it's no worse than the 5Ds. The D810 beats both the A7R2 and 5Ds at base ISO (which matters to me, because I do a lot of shadow pushing) - I'll take more if offered, and that might mean 16-bit raw files. The AF is decent - more reliable, if a little less low-light capable, than the D800, covering a wider area than the D750, and only passed by the most recent D5/D500 module, which I'm sure any successor will get (bonus marks for on-sensor phase detect AF for video). It'll hit 6fps if you drop to a 1.2x crop at 25MP, which is fast enough for me. There are lots of minor tweaks I'd like, and other cameras do have their advantages too, but I don't think Nikon are lagging the market here in the way that, say, Pentax are (sorry Pentax) - the D810 is still a very good camera.<br />

<br />

But I do expect an update soon too! I just hope it works this time...</p>

[dan_brown4]

<p>The lenses don't change, rather the DXOMark calculations change. This is a big reason why I put very little faith in DXOMark scores.</p>

[ShunCheung]

<p>I go even farther from Dan Brown, I pay not attention at all to DXO Mark scores. I evaluate cameras and lenses by using them and observe how they perform in actual shooting conditions. One needs to experience that in various shooting environments and different subject matters. Of course, only those conditions that are relevant to you matter. However, some simple score is more convenient for those who prefer something like a score of 98 is "better" than 96.</p>

[Andrew Garrard]

<p>It would be lovely if we could all try out every piece of kit we're considering and spend a few weeks doing exhaustive tests. Sadly, that's simply not practical, at least for anyone not working in a camera hire shop. Even if that were an option, there needs to be some input to what we even consider testing. There has to be a merit in on-line tests, along with the opinions of trustworthy reviewers.</p>

<p>We certainly shouldn't look at a single review and consider it to be all-encompassing. There's variation between samples, no review tests everything, and different reviewers have different biases. Nonetheless it's hard to get comparisons just from looking at images, so numerical information also has its place. Images too, obviously - there are many lenses that test well, but whose actual image quality concerns me. The Nikkor 85 f/1.4 is an example - tests well, many people love it, but I see enough LoCA that bothers me that I've never wanted one. Had I seen more sample images before I got my Nikkor 85mm f/1.8 I may have reached the same conclusion - I saw enough to see that the bokeh improved over the AF-D version (which also tested well but whose bokeh was horrible), but I'm not enjoying the green backgrounds. Can I live with the mechanical vignetting of the Sigma? Still TBD.</p>

<p>I absolutely look at DxO - and it has an easy way to bring up comparisons. I happen to use their software, so I'm happy to point people at their site. What I don't do is look at a single score - I check the graphs for sensor behaviour and the interactive field maps for lens behaviour. I see how the item behaves at the settings I care about. And I also look on senscore/lenscore, check for reviews at Photozone, google for other reviews, usually check Thom Hogan's site and sometimes others I trust, see what lensrentals have to say about sample variation and scores (and repairability), and I'll even check Hypnoken to see what he has to say (with a large grain of salt). They all might point out something important that the rest have missed. As far as I know, nobody really complained about field curvature in the original gushing reviews of the 14-24; there was very little discussion of the LoCA of the 135 f/2 DC, and every early review of the 150-500 Sigma claimed it was amazingly sharp (later reviews changed their mind). Many reviews are old - I believe an 80-200 f/2.8 is sharp throughout its range on 35mm film, but "sharp" has a different definition on a 36+MP sensor.</p>

<p>Summary: More information is better than none. DxO aren't particularly biased or incompetent as far as I can tell, but they're just another test. That they have different interpretations of the most recent pair of 400 f/2.8 Nikkors from some other sites is interesting information, and you can read things into that. If you're buying anything, do your research, and that's why we discuss things here - if we have personal experience, we can share it, and those of use who obsess over reviews can point out any potential oddities when an item comes up for discussion. It's about the pictures, but for lenses it's really hard to find a site that does a good and realistic comparison that's consistent, and even if there was they'd probably have a limited sample count. Pay attention to everything, but pay too much attention to nothing!</p>

[papo]

<p>Many thanks to all of you for taking the time to add your points of view and suggestions, in your replies to my question. Much appreciated.</p>