Reading mode: Light Dark

DxO Mark and MF x Nikons

Go to last post
R

rhlpetrus

Forum Pro
Messages
27,478
Solutions
3
Reaction score
5,418
Location
Campinas, BR
Thanks for posting Renato. I didn't know they had this data yet. Amazing camera. Yep, this should start a "healthy" discussion now. Here is the S/N data for deep shadow also.

From the Full SNR log plots:

S/N @ 0.1% Gray (ISO)
Phase One P45_ 11.3dB (50)
Nikon D3X 15.4dB (100)

If this is backed up by real world image quality, then it really just leaves the increase in resolution which should be noticeable but not huge. I think you have seen these shots by Paul Richman, but for those who haven't, check them out here:

http://forums.dpreview.com/forums/readflat.asp?forum=1021&thread=30863275

He does use some PP with Nik software. The images are breathtaking even so.
--
David
 
sidelight said:
Thanks for posting Renato. I didn't know they had this data yet.
Amazing camera. Yep, this should start a "healthy" discussion now.
Here is the S/N data for deep shadow also.

From the Full SNR log plots:

S/N @ 0.1% Gray (ISO)
Phase One P45_ 11.3dB (50)
Nikon D3X 15.4dB (100)

If this is backed up by real world image quality, then it really just
leaves the increase in resolution which should be noticeable but not
huge. I think you have seen these shots by Paul Richman, but for
those who haven't, check them out here:

http://forums.dpreview.com/forums/readflat.asp?forum=1021&thread=30863275

He does use some PP with Nik software. The images are breathtaking
even so.
--
David
Click to expand...
They look a little overprocessed for my taste.
--
http://dslr-video.com/blogmag/
 
Iliah Borg said:
http://www.luminous-landscape.com/essays/eyes-vs-numbers.shtml

--
http://www.libraw.org/
Click to expand...
Completely agree with the points made in the article. Thanks for posting. I would just add that, while some may only be interested in the numbers, I think it is clear that any experienced photographer always wants to see the images, as for most of us this is what we love. There are increasing numbers of high quality D3X photos being posted and I think most can appreciate the quality the camera is capable of. I believe many of us would love to see comparison shots from MF cameras in order to assess the differences which I am sure are there. It is not a contest, but rather a topic which is interesting because I think most want to see how well the best 35mm cameras can compare to the far more expensive and more cumbersome MF cameras. If 35mm does not surpass MF digital (and it obviously cannot resolution wise), I will not be surprised, but if it gets close, I will be pleasantly surprised and happy about it.

I will also say that Reichmann's discussion of the subtleties of high end audio I don't believe quite applies to photography. Audiophiles discovered that harmonic distortion is actually desirable in many cases, as electric guitarists have known forever. So some high end makers went back to tubes to add "good" imperfections back in. In digital photography, I think the analogous qualities would be S/N, DR, and color depth. While photographers might want to vary these qualities to achieve certain looks, I don't think we can say that less performance in these areas is desirable.

I believe DxOMark is a reasonable starting point for gear comparisons, but in the end, I do agree that the proof is in the print.
--
David
 
--

David, Im not so sure that the audiophile analogy is off base. Remember, many still think that digital looks "sterile" and actually add noise to their digital prints. Interesting times ahead.
 
Iliah Borg said:
http://www.luminous-landscape.com/essays/eyes-vs-numbers.shtml
Click to expand...
I would disagree that the choice is eyes vs numbers. Photography is a visual medium and what pleases the eye (or, really, the brain) is the most important thing. It's interesting that MR talks about high end audio. It might well be that consumers don't worry about the numbers, but you can be sure that the designers of the gear do. The only ones who don't are the crooks who've realised they can charge huge amounts for cr4p, based on the 'magic ears' of a few trusted reviewers. That is the risk when subjective evaluation becomes dominant.

The trick in quantitative evaluation is to produce numbers which actually correlate well with peoples' subjective choices. It's much easier to choose things that are easy to measure, and that's what tends to happen, as in DxO. Alternatively, you can choose to measure things that fit your own ideology, as does DPR.

I did a small test some time ago, using some DPR people as guinea pigs. One of the interesting outcomes is that rendering of 'detail' seems to be much more important for most peoples' appreciation of 'image quality' that noise, despite the obsession with the latter here.

I would hazard a guess that it is the superior rendering of detail, caused by higher pixel counts and lower magnification, that causes MF images to be subjectively 'better' than 35mm ones, even if it is now clear that noise and DR may not be as good.

--
Bob

 
headofdestiny said:
--
David, Im not so sure that the audiophile analogy is off base.
Remember, many still think that digital looks "sterile" and actually
add noise to their digital prints. Interesting times ahead.
Click to expand...
Yes, which is why I said that people may want to vary the image qualities, depending on what they are after. I don't think anyone except possibly an old B&W film purist (and I love B&W) wants low quality captures by default. It is not inherently better unless you love grain, noise, low DR, and poor color separation, whereas a good case can be made that some harmonic distortion sounds better. Yes, very interesting times, camera-wise at least.
--
David
 
--

True. I guess Im referring more to the "quality" of noise, as some cameras may have a bit more noise than others, but their noise (grain?) is more appealing.
 
But as sidelight states:
"but in the end, I do agree that the proof is in the print. "
This is in fact the only key that matters. What we, clients, see on paper.

The other not often mentioned reality to the 35mm vs MF comparisons is the "significantly" higher price point because of the back needing the body and lens line. Now throw in useability in the field vs a 35mm dslr with a vast array of lens capability and very quickly the little DR and resolution difference "as seen on paper", quickly take a back seat IMO for which system gives the overall best capability. The anti Nikon and "ticked off price point" people for now will keep up the pressure, but the bottom line is that the D3X certainly appears to be the best 35mm dslr to date.

--
Mel
http://www.mellockhartphotography.zenfolio.com
http://www.mellockhartphotography.net
 
He mixes right and wrong arguments.

1) He mentions that MF cameras don't use AA filters. Ok, mostly correct. So what? If that implies images have more detail, which is also true, and has some disagreable side-effects as well, it's not related to the tests, which are about sensors, not final image detail. Thus, if one is concerned with that, should factor it in.

2) Then he says: how come a P45 is rated below A900 and 5DII? Well, very simple and easy, because of low-light ability and how DxOMark measures that (ISO where 18% SNR reaches 30db, granted at least 9stops of DR). Of course the MFs are not capable of that, they use CCD sensors which are of old design and don't have in-sensor NR. If you're not interested in shooting above ISO50 (or 100), then the rest of the ratings are relevant.

3) He claims that dslrs' sensors use NR circuitry, berfore RAW, etc. Correct, CMOS actually needs that, since it has, per se, less SNR than CCD. Does that actually affect IQ? That's an important point, and just saying it does is not a good answer.

4) The only reasonable, controlled, with similar quality lens, test I've seen between similar pixel count dslrs and MF is one done by someone at ... LL:

http://luminous-landscape.com/forum/index.php?showtopic=29821

P25, 1DsIII and 5DII. To me the detail is a tie, maybe an edge to the 1DsIII. DR and colors, maybe an edge to P25.

So, detail the holy grail of MF, well, not such a big difference after all. I've seen the D3x is at least a tie in detail, maybe better, than the 1DsIII. DR and color, all say it's better. Summing up, maybe the D3x is about equivalent to the P25, which is the back it should be compared to, after all.

Reichman's argument is, well, not well-founded. Just saying MFs are better, or claiming most pros think so, maybe is living in the past world of 12MP dlsrs.

5) One big issue: lenses. All the MF are used with topnotch lenses, the Hassies use Zeiss lenses, or Voigtlander, etc. So, Nikon, ok, the zooms are very nice, but maybe they should invest a bit into some really pro-grade primes, not only PC ones, but for more general use. Let's see how the legendary 21mm Zeiss Distagon, in F-mount, will perform. Some say it's the only lens that beats the 14-24 zoom.

One probably needs a full range of such lenses to use the D3x to its full capabilities. Which, may well prove to be about same as MF in IQ terms (ISO100) and much better re everything else.

--
Renato.
http://www.flickr.com/photos/rhlpedrosa/
OnExposure member
http://www.onexposure.net/

Good shooting and good luck
(after Ed Murrow)
 
DxO Mark is a silly number. It doesn't take resolution into account. All it is doing is comparing the raw output photo site to photo site which a bias toward high iso performance.

Yes, the photo sites on a D3x are more advanced than those on a +P45, but that means about as much as saying a 400 asa film emulsion is faster than one rated at 50 asa.

To their credit, though, they freely admit the limitations of their "Mark." Read this:

http://www.dxomark.com/index.php/eng/Technologies/Medium-format-ranking
 
A major shortcoming of DxO's rating system is that it doesn't include the camera's resolution. For some photographers, resolution is obviously important, and for others, less so. But for DxO to come up with an absolute ranking without taking resolution into account is a bit of a joke.

That said, DSLRs are clearly making inroads into MF territory in the areas of DR and color fidelity. And of course, there's the smaller size, lighter weight, much longer lasting batteries, and wider range of lenses on the DSLR side to consider.

--
I Reject Your Reality And Substitute My Own

Web Site - http://www.hgiersberg.com/
 
Agreeing! lens quality will always be the most important, far more so then cameras, they come and go, lenses stays forever.

Yes, top-notch MF Zeiss lenses are hard to beat, maybe only by LF lenses from Schneider and Rodenstock. The Schneider and the Zeiss. 38 mil Biogon are so far the sharpest lenses Ive ever used or seen.

Theres however an old stigma that MF MUST be better then top say Nikon lenses, maybe so in the past but not today. Some of the new top of line Nikon Primes, even the new 24-70 and 14-24 shows in fact same values and curves. Leica, Leitz optics are another proof of absoloute top notch 35-mil optics.
 
I disagree about the resolution aspect. To me, resolution is obvious with digital. One camera has more or less MP than another and we can calculate the linear resolution potential difference. What should DxOMark show us? It is clear that 15-20% increase in MP is slightly noticeable and that doubling MP count only gives us about a 40% increase in linear resolution. DxOMark clearly states what their ranking criteria are and also say how to interpret the ranking numbers and how much difference is even noticeable. I don't care about the rankings as much as the data they present in their plots. One can draw their own conclusions from the data as to how one camera stacks up to another and then decide if the cost differences are justified. Now, people may argue about their test protocols as I know some do, and that is certainly fair. But to their credit, they publish these as well for all to review. Their rankings are like any rankings - just sort of a rough guide. You need to dig deeper into the results to find what matters to you.
--
David
 
You must log in or register to reply here.

Keyboard shortcuts

f
Forum
m
My threads
Mobile site
About
Editorial content
Cameras & Lenses
Community
All content, design, and layout are Copyright © 1998–2025 Digital Photography Review All Rights Reserved.
Reproduction in whole or part in any form or medium without specific written permission is prohibited.
When you use DPReview links to buy products, the site may earn a commission.
©GPS Media - Guides, Products, Services.
Back
Top