Ad
Ad
Ad
Pages: « 1 [2] 3 4 »   Bottom of Page
Print
Author Topic: Does the D800E reach MF quality ?  (Read 19431 times)
BartvanderWolf
Sr. Member
****
Offline Offline

Posts: 3650


« Reply #20 on: May 09, 2012, 02:13:35 AM »
ReplyReply

[..]
Therefore I think the dxomark DR measurement is a bit one-dimensional. It would be interesting if one could measure in things like "skin-tone tonal range in well-exposed pictures" etc.

Hi,

The DR test is just that. It gives a good impression about how fast the shadows will block up. What you are looking for is the SNR 18% test (medium brightness levels), because there the photon shot noise will dominate. Larger sensels generally have a benefit since they can collect more photons before they saturate, and more photons usually lead to smoother gradations.

Cheers,
Bart
Logged
marcmccalmont
Sr. Member
****
Offline Offline

Posts: 1724



« Reply #21 on: May 09, 2012, 03:51:09 AM »
ReplyReply

Bernard, Erik, et al I have my D800E/Leica R lenses and my IQ180/Rodenstock HR lenses in Tokyo so any tests or comparisons you want let me know. First day with the D800 just getting to know it and figuring out where things are on a Nikon but initial shots from my apartment balcony looked promising. Not sure if pixel for pixel one could tell the 2 apart? More pixels on the IQ but longer lenses for the D800E, certainly better pixels than on my 5DII.
Marc
Logged

Marc McCalmont
torger
Sr. Member
****
Offline Offline

Posts: 1503


« Reply #22 on: May 09, 2012, 03:56:03 AM »
ReplyReply

I tried it. Put  lenses into the equation, it's not M9 quality yet.

Yes me too think lenses is an important factor. For people photographers and other short DoF photography the bokeh and other aspects of lens look will be important, but also very much a subjective thing. One may prefer leica look or maybe nikon look, or maybe one thinks it doesn't matter because they look similar enough, or one sees it as just two different looks and both are ok.

For large DoF landscape photography I personally think the technical cameras with Rodenstock/Schneider lenses is ahead in sharpness and flexibility when combined with a MF sensor, but it would be interesting with a bit more testing on this. It is also a matter of taste and photography style if you consider for example zoom lenses to be more important flexibility factor than the ability to freely shift and tilt (for me it is the latter).
Logged
BJL
Sr. Member
****
Offline Offline

Posts: 5129


« Reply #23 on: May 09, 2012, 08:51:07 AM »
ReplyReply

Torger raises a point that I have thought about theoretically, so I am glad to hear of it being put to the test, and look forward to image comparisons.

I can imagine that in a comparison between
 - a larger CCD with larger full well capacity, and
- a smaller active pixel CMOS sensor with dark noise lower by enough to have a higher engineering dynamic range,
that the comparisons of noise levels in images could vary over four zomes, defined in terms of exposure level counted down from full well capacity:

Zone 3: "bright"
So much light that noise is irrelevant. Hopefully true down to somewhere below the midtones in exposures as base ISO speed.

Zone 2: "light shadows"
Enough light that noise is dominated by photon shot noise, but little enough that some noise is visible. Then the larger sensor, which counts more photons at each lighting level, has better SNR, and could look better. Inexpect that this is true down till four or more stops below midtones, beyond which a print of normal contrast is rendering everything as black anyway.

Zone 1: "deep shadows"
So little light that dark noise from the sensor and other electronics is significant, at least with the CCD, and the CMOS sensor has a practical advantage in SNR, useful for pulling these deep shadows up in scenes of High Subject Brightness Range.

Zone 0: "darkess"
So little light that the SNR from either sensor is unacceptably low, and you want not want to print or display thisnas anything other than black. Having a SNR of 4:1 instead of 2:1 is nothing but a paper victory for artistic photography, even though it can be a big deal for astronomy or surveillance.
Logged
BJL
Sr. Member
****
Offline Offline

Posts: 5129


« Reply #24 on: May 09, 2012, 09:02:30 AM »
ReplyReply

The DR test is just that. It gives a good impression about how fast the shadows will block up. What you are looking for is the SNR 18% test (medium brightness levels), because there the photon shot noise will dominate.
Thanks; I should have mentioned that SNR 18% as a useful measure at the midtones.

But might it also be useful to look at SNR in the moderate shadows, say at one, two, and three stops below the midtones? I take three stops below the midtones as the bottom of the main range for scenes of average Subject Brightness Range.

Can this information be gleaned roughly by looking at the SNR 18% figures at exposure indices (ISO speeds) of  one, two, and three stops higher?
Logged
BJL
Sr. Member
****
Offline Offline

Posts: 5129


« Reply #25 on: May 09, 2012, 09:05:56 AM »
ReplyReply

I tried it. Put  lenses into the equation, it's not M9 quality yet.
Which lenses did you use on the D800E in your comparisons to the M9?
Logged
BartvanderWolf
Sr. Member
****
Offline Offline

Posts: 3650


« Reply #26 on: May 09, 2012, 11:11:53 AM »
ReplyReply

But might it also be useful to look at SNR in the moderate shadows, say at one, two, and three stops below the midtones? I take three stops below the midtones as the bottom of the main range for scenes of average Subject Brightness Range.

Can this information be gleaned roughly by looking at the SNR 18% figures at exposure indices (ISO speeds) of  one, two, and three stops higher?

Hi,

Well the SNR 18% chart sort of allows to do that, but there can be (sometimes significant) differences in how the cameras use amplification to boost the underexposed signals at higher ISOs. Some cameras and/or Raw converters also change their default processing before writing the Raw data or the demosaiced image based on the ISO setting. So then we're adding amplification and signal processing in the mix, which of course is interesting as well for the resulting image unless one strictly wants to evaluate e.g. low ISO performance at different brightness levels.

Cheers,
Bart
Logged
Stefan.Steib
Sr. Member
****
Offline Offline

Posts: 419



WWW
« Reply #27 on: May 09, 2012, 05:00:51 PM »
ReplyReply

I also did a full shift up 10mm and one down 10mm of the overview, but I thought this is not of relevance as I only wanted to produce as much and as defined moiree as possible.If anyone is interested I can also put these 2 up.

Regards
Stefan
Logged

Because Photography is more than Technology and "as we have done this all the time"
www.hartblei.de     www.hcam.de    www.spectralize.com
torger
Sr. Member
****
Offline Offline

Posts: 1503


« Reply #28 on: May 10, 2012, 03:21:47 AM »
ReplyReply

Here's a few comparisons between H4D-40 and D800:

http://www.buschphoto.com/blog/

What I find most remarkable is the D800 color error in the portrait example:
http://www.buschphoto.com/storage/port_compare1.jpg?__SQUARESPACE_CACHEVERSION=1334260989094
(left hasselblad, right D800 with yellowish tone)

but I suppose that can be because Nikon don't care much to help out Adobe to reproduce good color in Lightroom, perhaps nx2 would yield better result, don't know. Neither Canon or Nikon has been very helpful in letting third party software process their raw files, which I think is a big mistake.
Logged
Ti29er
Newbie
*
Offline Offline

Posts: 9


« Reply #29 on: May 10, 2012, 08:31:43 AM »
ReplyReply

Can we return to the question posed: namely, does the 800E reach MF quality?
I am still shooting film on the MF kit BTW.

I got a little lost in the cross-banter and was wondering, like many others, if the 800 or the 800e is going to be best for me, replacing the old stalwarts, the D700.

I get the impression from one well known site that the filter in the non-E could not be replicated in any 3rd party software, but in this post that is diputed with the LR 4.1 software, so if you do suffer some moire, LR will right the wrong?

That being the case, is there any real case to take the 800 over the slightly more expensive 800e?

Lastly, & since most of the lenses I have are circa 2000, albeit f2.8 on the 3x zooms, is this something of Fools' Gold since the older lenses are not good enough for the 800e to make its self known beyond the 800?

So, aside from my own questions: does this 800e reach levels that might render some MF kit somewhat redundant?

Thanks
Tim
Logged
ndevlin
Sr. Member
****
Offline Offline

Posts: 515



WWW
« Reply #30 on: May 10, 2012, 10:02:42 AM »
ReplyReply

Tim,

If you are used to shooting MF film (and scanning, I presume) the resolution of the D800 will astound you, as will the usable depth of shadows.  Yesterday I processed a frame (in LR4) where a pedestrian was in total silhouette shadow in front of a day-lit building.  Just using the shadows slider I was able to open the highlights so fully as to read his watch. Even my wife, a knowledgable non-photographer, was blown away.

Suffice to say, DR is phenomenal. 

Resolution far exceeds what any MF kit will produce, and I say that as a long-time shooter of everything from 645 to 69.  As with all digital, the 'look' is different than analog printed to analog. But in terms of resolution, it well and truly is a step above.

As for 800E/800, the "E" is better if you care about a slight increase in accutance and prefere the look of non-AA'd chips. I do, so it was a no-brainer.  Get an 800 if you shoot a lot of fabrics or architectual work. The difference is not huge, but perceptible.

As regards lenses, don't get too worried. All of Nikon's better lenses will produce good results on the 800s, but the newer and better lenses will get more out of it. In particular, the 35mm f1.4 and 85mm f1.8 have proven astonishing in my early testing. The 16-35, however, is also really far better than it has any right to be. I'd suggest you try what you have and, if you see sub-opptimal performance, replace your most used lens with something state-of-the-art. But don't be fooled into thinking you need brand new primes to see the 800's advantages. That is simply not true.

Lastly, I personally find this notion that software can undo the effects of an AA filter to be kind of funny. You can safely ignore that. While talking about it appears to make some theoreticians happy, in terms of actual photography, forget it.
« Last Edit: May 10, 2012, 10:04:41 AM by ndevlin » Logged

Nick Devlin   @onelittlecamera
yaya
Sr. Member
****
Offline Offline

Posts: 1130



WWW
« Reply #31 on: May 10, 2012, 10:59:02 AM »
ReplyReply

Lastly, & since most of the lenses I have are circa 2000, albeit f2.8 on the 3x zooms, is this something of Fools' Gold since the older lenses are not good enough for the 800e to make its self known beyond the 800?

Tim are you sure they are that old? Or are they film lenses? (am asking because right until the D3 came out there were no FX lenses available)

Yair

PS does you username mean you ride a Ti 29er? Which one?
Logged

Yair Shahar | Product Manager | Mamiya Leaf |
e: ysh@leaf-photography.com | m: +44(0)77 8992 8199 | www.mamiyaleaf.com | yaya's blog
ErikKaffehr
Sr. Member
****
Online Online

Posts: 7424


WWW
« Reply #32 on: May 10, 2012, 11:35:50 AM »
ReplyReply

Hi,

I essentially agree with Nick, will just add a couple of points.

1) I have done a lot of tests comparing my Pentax 67 using Velvia and Ektar 100 with my Sony Alpha 900 and found the digital camera to be superior. You can check my tests here:

http://echophoto.dnsalias.net/ekr/index.php/photoarticles/59-sony-alpha-900-vs-67-analogue-round-2

http://echophoto.dnsalias.net/ekr/index.php/photoarticles/16-pentax67velvia-vs-sony-alpha-900

If you check out my tests, keep in mind that you need to click on the images once or twice to get full size.

Another point I would make is that the sensor in your D7000 has the same pixel pitch as the one in the D800, so if your lenses work well with the D7000 they will also work with the D800, unless you are pixel peeping in the corners, where weakness normally shows up. Lenses from 2000 may be just fine.

Regarding the OLP filtering, the way I see it I would make the following points:

- It seems that landscape photographers seldom have issue with moiré even with moiré prone cameras.
- Many subjects are moiré prone, like textile
- The effect of the AA-filter can be compensated for by improved sharpening, but that may also increase noise in the image
- Moiré can be reduced by a special filter brush in Lightroom

So you can go either way and be happy. If you shoot moiré prone subjects you would probably choose the D800.

http://www.luminous-landscape.com/forum/index.php?topic=66893.msg528667

Best regards
Erik



Tim,

If you are used to shooting MF film (and scanning, I presume) the resolution of the D800 will astound you, as will the usable depth of shadows.  Yesterday I processed a frame (in LR4) where a pedestrian was in total silhouette shadow in front of a day-lit building.  Just using the shadows slider I was able to open the highlights so fully as to read his watch. Even my wife, a knowledgable non-photographer, was blown away.

Suffice to say, DR is phenomenal. 

Resolution far exceeds what any MF kit will produce, and I say that as a long-time shooter of everything from 645 to 69.  As with all digital, the 'look' is different than analog printed to analog. But in terms of resolution, it well and truly is a step above.

As for 800E/800, the "E" is better if you care about a slight increase in accutance and prefere the look of non-AA'd chips. I do, so it was a no-brainer.  Get an 800 if you shoot a lot of fabrics or architectual work. The difference is not huge, but perceptible.

As regards lenses, don't get too worried. All of Nikon's better lenses will produce good results on the 800s, but the newer and better lenses will get more out of it. In particular, the 35mm f1.4 and 85mm f1.8 have proven astonishing in my early testing. The 16-35, however, is also really far better than it has any right to be. I'd suggest you try what you have and, if you see sub-opptimal performance, replace your most used lens with something state-of-the-art. But don't be fooled into thinking you need brand new primes to see the 800's advantages. That is simply not true.

Lastly, I personally find this notion that software can undo the effects of an AA filter to be kind of funny. You can safely ignore that. While talking about it appears to make some theoreticians happy, in terms of actual photography, forget it.

Logged

Ti29er
Newbie
*
Offline Offline

Posts: 9


« Reply #33 on: May 14, 2012, 05:48:05 PM »
ReplyReply

Hi.
Shooting the MF film on an old Mamiya 7.
Scanning done by a pro lab in town (££!) although I was looking at buying an Epson V600 maybe.
A few issues aside it (not focucing correctly and the markers for the various lenses doesn't work!) works well enough and shooting 160 film, but shifting to Reala for a change.
The Ti29er was a Lynsky bike made by them to an On-One design. Stopped riding 2yrs ago after a problem with my L leg, still unsure what happened but it was on a training ride a few weeks after the UK 24hr solo championships. All my pals are racing well and climbing the race leaderboard getting podiums. I'm not. They're all on carbon 29ers now, Scott, Fisher and Spec' bikes.
Next appt with a dr / specialist is actually in a few hrs to review the MRI scans as they showed nothing out of the ordinary.

Cheers
Tim
Logged
ndevlin
Sr. Member
****
Offline Offline

Posts: 515



WWW
« Reply #34 on: May 14, 2012, 09:08:24 PM »
ReplyReply


Is just me, or do the Nikon exposure in the Nikon v. 'Blad test linked above consistently seems to have greater exposure? Especially the kitchen still-life.

- N.
Logged

Nick Devlin   @onelittlecamera
Frank Doorhof
Sr. Member
****
Offline Offline

Posts: 1521


WWW
« Reply #35 on: May 15, 2012, 02:11:02 AM »
ReplyReply

It's a bit like comparing apples to oranges in my opinion.

First of all :
Sensor size.
You can not compare a DSLR sensor to a sensor from a MF camera.
Due to the larger sensor the LOOK of the images is different, already mentioned before the DOF is totally different but also the angle of view. Of course one could choose a 50mm on a DSLR and a 80mm on a MF but still the look is different.

Then we have :
1 fps for MF which is a great tempo to work with for for example fashion, this could be solved by firmware on a DSLR I know, but it isn't yet.

With leaf shutter lenses you can go up to 1/1600 with strobes or higher.
I know that with SOME strobes and Pocketwizard you can get there also using Hypersync but for example in my setup (all A heads on Rangers/Quadras) it simply doesn't go higher than 1/350, I can go higher with a D-lite but those I don't use outside and that's were it counts.

Diffraction, with the 33MP Leaf AptusII I can shoot without any problem on F16-F22, with my 5DMKIII I see a degradation in the image above F11 on some lenses.

But maybe more importantly, when shooting for example fashion MF makes it possible to shoot from the SAME distance with longer lens and higher shutter speed with strobes, meaning we can get shallow DOF while fighting the sun (using strobes on full power).


I really wonder if we had the same conversation if the D800 would have been released with 24MPs. It seems that as soon as people see a camera with the same amount of MPs that MF has the discussion starts up again.

By the way I will not say that the D800 or 5DMKIII is a bad camera, but I base my decision which one to buy/use on the "horses for courses" idea. Meaning with AF nothing beats a DSLR, High ISO DSLR again, recovery from shadow areas seems the D800(E), however when I want to shoot a studio fashion series where I control the light there is no need for lighting up the shadows, but here the 1fps is ideal, the extra shutter speeds with strobes and of course the totally different DOF, meaning for those sessions nothing at the moment beats a MF camera.....


Now fast forward a few years (maybe shorter)
The first REAL mirror less camera is released, no more limitations in shutter speeds with strobes, we get a 24MP foveon type sensor (real 24MP) I think now is the time I will sell my MF gear and forget about the different look, working with one camera for all sounds very tempting, especially because we travel around a lot, and dragging two complete sets can be very cumbersome (and expensive). But I will NEVER make that decision based on MP's alone.

I do wish to add that the D800(e) is a very fine camera, especially the Sony sensor looks very promising, but when I look at 100% crops we shot during a workshop in our studio I do find that fine details in hairs look a bit "chaotic" compared to the razor-sharp MF files, also even with the use of color checkers I can't get the colors in the neighborhood of the Leaf Aptus back, and skin tones are without a doubt the most important thing for me, so even when I would forget all the rest, for skin tones alone I would not think about switching.

But the future looks promising let's hope the technology also drips through to MF camera, I would love to have a system I can use up to ISO6400 (heck even 3200 would be great) with the same quality my 5DMKIII gives me, add some killer AF and I'm more than happy, but up until than I will choose the camera for each assignment, and for studio/strobe work at THE moment nothing beats my MF system.

Now one final thought.
Can I deliver all my commercial work with a DSLR.... ?
yep.
No problem at all, it will look different, I can use ND filters to "help" with the strobes outside, 22MP is more than enough for almost any publication.
So why invest in a MF system?
Simply put... because I want the best quality I can get, it's a personal opinion, if you just want to make money get a good pro level DSLR with a 24-75 and 70-200 lens, some good strobes and you can work forever with that, never needing to upgrade, they did with far worse in the 70's and 80's and the magazines haven't gotten bigger Cheesy

If you however are driven by passion to deliver the absolute best out there and LOVE working the MF way, I think MF is the way to go for now.

In the end it's actually all a bit silly that we look at technique and forget about the thing that really counts.... getting the shot. And me personally I would rather get the shot on the very unprofessional "P" setting with Auto ISO and bracketing than not getting the shot Cheesy (not to say that I do, but I think you get the general idea), maybe we are too obsessed by the numbers and forget the essence of photography... getting the shot.
Logged
BernardLanguillier
Sr. Member
****
Offline Offline

Posts: 7996



WWW
« Reply #36 on: May 15, 2012, 02:28:25 AM »
ReplyReply

1 fps for MF which is a great tempo to work with for for example fashion, this could be solved by firmware on a DSLR I know, but it isn't yet.

Not sure whether it helps, but Setting d2 enables you to select the maximum continuous shooting fps between 1 and 5 (see page 293 of .pdf manual).

Cheers,
Bernard
Logged

A few images online here!
Frank Doorhof
Sr. Member
****
Offline Offline

Posts: 1521


WWW
« Reply #37 on: May 15, 2012, 02:34:36 AM »
ReplyReply

Is that option also on the new Nikons?
Would help some of my students, I know it's not in the Canon.
Logged
BernardLanguillier
Sr. Member
****
Offline Offline

Posts: 7996



WWW
« Reply #38 on: May 15, 2012, 04:45:20 AM »
ReplyReply

Is that option also on the new Nikons?
Would help some of my students, I know it's not in the Canon.

It has been there for a few years on the Nikon bodies.

Cheers,
Bernard
Logged

A few images online here!
Frank Doorhof
Sr. Member
****
Offline Offline

Posts: 1521


WWW
« Reply #39 on: May 15, 2012, 05:11:25 AM »
ReplyReply

Great Will them that.
Logged
Pages: « 1 [2] 3 4 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad