Pages: 1 [2] 3 4   Go Down

Author Topic: Is the D800E expected to have noticeable better IQ than the D800?  (Read 23124 times)

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6600
Re: Oversample the signal enough and then apply LPF still seems best overall
« Reply #20 on: February 13, 2012, 02:05:50 pm »

BJL:  In audio D/A conversion ... You can use a classic sinc filter ... You can choose other filters ...
The choice of low-pass vs no lo-pass for the D800 vs D800E is not that different.
I see a big difference of degree: "no LPF at all" is not an option I have ever heard seriously proposed in digital audio, or in any other area of signal processing. The options in audio are more akin to the varying strengths of OLPF's used in various cameras.
Logged

hjulenissen

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2051
Re: Oversample the signal enough and then apply LPF still seems best overall
« Reply #21 on: February 13, 2012, 03:30:36 pm »

P.S. has anyone in the digital audio field advocated removing the LPF in A/D conversion, complaining about he loss of high frequency response cause by the LPF? Maybe that would happen if audio had the equivalent of 100% pixel peeping, which would be playing music back at half speed or less, and then complaining about the muddy highs caused by the LPF.
Not the filter in the A/D to my knowledge, but the equivalent filter in the D/A:
http://www.diyaudio.com/forums/digital-source/108227-nos-adc-discussion-3.html
"It’s unfortunate the term non-oversampled (NOS) usually implies no reconstruction filter (NRF), as well."

It seems that usually the lowpass chatacteristics of loudspeakers are relied upon.

-h
Logged

hjulenissen

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2051
Re: Oversample the signal enough and then apply LPF still seems best overall
« Reply #22 on: February 13, 2012, 03:37:24 pm »

You can use a classic sinc filter for reconstruction of the analog signal, but then you get ringing and other artifacts.
I dont think that a sinc filter can actually be implemented in a physical circuit. What can be done is a windowed approximation, and then there is the question of what window function is applied.
Quote
You can choose other filters, which minimize ringing, but then you'll have lowered frequency response and some aliasing.  Some D/A converters even offer user-selectable filter choices for this reason.  There's no perfect solution -- just tradeoffs and preferences.
It is hard to make the case that ringing in frequency ranges that no-one seems to be able to hear matters all that much. Off course, it _may_ drive some HF loudspeaker units into nonlinear operation, but that can certainly not be claimed to be a proven, general phenomenon that anyone have hear with any certainty.

I do believe that AD and DA functions from the major, established manufacturers tends to suppress aliasing down to levels where it is generally accepted hard to prove audibility in a listening test. But then, audible differences between properly designed AD and DAs are really hard anyways.

The possibility to use simpler analog filters is often asserted as a benefit of high-sampling-rate delivery formats, but I find this hard to believe - the same advantage can (and is nearly always) be offered through oversampling internally in the AD or DA.

-h
« Last Edit: February 13, 2012, 03:40:50 pm by hjulenissen »
Logged

BJL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 6600
Re: Oversample the signal enough and then apply LPF still seems best overall
« Reply #23 on: February 13, 2012, 04:20:41 pm »

It seems that usually the lowpass chatacteristics of loudspeakers are relied upon.
I think I recognize that one: high enough PPI compared to ink dot smearing size.
Logged

MarkL

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 475
Re: Is the D800E expected to have noticeable better IQ than the D800?
« Reply #24 on: February 13, 2012, 05:16:26 pm »

I'm interested to see some more examples (from raw files). The D800 files may be able to be sharpened far more making the difference very close and not worth the moire/aliasing risks.
Logged

EverPhoto

  • Newbie
  • *
  • Offline Offline
  • Posts: 12
A good D800 to D800E resolution comparison shot
« Reply #25 on: February 13, 2012, 09:00:38 pm »

http://www.ephotozine.com/article/nikon-d800-d800e-digital-slr-hands-on-review-18420

Scroll about 1/3 of the way down the page to the shots of the snowy landscape.

The D800E 100% crop is noticeably "sharper" but does it actually contain any additional information? Hard to tell from this shot. I took the D800 crop and applied some unsharp to it and I was able to get it to look almost exactly like the D800E crop. Of course, when I applied some unsharp to the D800E shot it went even sharper, but I started noticing artifacts more.
Logged

torger

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3267
Re: Is the D800E expected to have noticeable better IQ than the D800?
« Reply #26 on: February 14, 2012, 04:12:59 am »

I think we need RAW files to play around with. JPEGs, even at 100% quality setting, will soften the files somewhat.

I do suspect though that with typical landscape f-numbers (some diffraction onset) and after deconvolution sharpening there will be very small difference between the two.

I think AA filter is a good thing generally, and the advantages of not having it has been greatly exaggerated (by medium format, Sigma and Leica owners I suppose :-)), but still I would probably buy the D800E out of curiosity and I think that the resolution is high enough that I won't have much problems with aliasing or moiré. I don't do much short DOF photography at high res, and at large DOF I think the f/8-f/11 diffraction alone will work good enough as an AA filter, so blurring it further would be unnecessary.

Having overly sharp aliased pixels sucks for post-processing (lens corrections, rotation, upscaling etc), but as said I don't think it is a risk here as the resolution is sufficiently high.
Logged

Scott O.

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 315
    • Photography by Scott and Joyce
Re: Is the D800E expected to have noticeable better IQ than the D800?
« Reply #27 on: February 17, 2012, 07:32:30 pm »

Read the recently released Nikon Technical Guide ( http://www.nikonusa.com/en_US/o/Y6wrkA9OU_z04IreazIXl_22UII/PDF/D800_TechnicalGuide_En.pdf ).  The summary is that your technique better be spotless as everything is magnified.  Tripod, Live View or mirror lockup, perfect focusing and nothing smaller than f11!  Makes me think the 'e' model might be slightly less forgiving.  Neither camera will tolerate sloppy technique...  Nikon didn't pull any punches on this one.
« Last Edit: February 17, 2012, 08:01:59 pm by soberle »
Logged

BernardLanguillier

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 13983
    • http://www.flickr.com/photos/bernardlanguillier/sets/
Re: Is the D800E expected to have noticeable better IQ than the D800?
« Reply #28 on: February 17, 2012, 07:43:39 pm »

Read the recently released Nikon Technical Guide ( http://www.nikonusa.com/en_US/o/Y6wrkA9OU_z04IreazIXl_22UII/PDF/D800_TechnicalGuide_En.pdf ).  The summary is that your technique better be spotless as everything is magnified.  Tripod, Live View or mirror lockup, perfect focusing and nothing smaller than f11!  Makes me think the 'e' model might be slightly less forgiving.  Neither camera will tolerate sloppy technique...

I just love the very open and down to the earth content of this white paper.

Cheers,
Bernard

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: Is the D800E expected to have noticeable better IQ than the D800?
« Reply #29 on: February 17, 2012, 09:04:49 pm »

Hi,

All that applies to the D7000, too. The test below were shot with a Sony SLT 55, a 16 MP camera. Also, solid tripod, non moving mirror, cable release, live view focus and flash.

Check out: http://echophoto.dnsalias.net/ekr/index.php/photoarticles/49-dof-in-digital-pictures?start=1

The top right image would be considered sharp by most DoF tables. The left column shows effects of diffraction, and in my view it's quite obvious that we start loosing sharpness already at f/8.

That said sharpening plays an important role: http://echophoto.dnsalias.net/ekr/index.php/photoarticles/49-dof-in-digital-pictures?start=2

Also we need to keep in mind that differences may be much smaller in print than at 1:1 view.

Read the recently released Nikon Technical Guide ( http://www.nikonusa.com/en_US/o/Y6wrkA9OU_z04IreazIXl_22UII/PDF/D800_TechnicalGuide_En.pdf ).  The summary is that your technique better be spotless as everything is magnified.  Tripod, Live View or mirror lockup, perfect focusing and nothing smaller than f11!  Makes me think the 'e' model might be slightly less forgiving.  Neither camera will tolerate sloppy technique...  Nikon didn't pull any punches on this one.
Logged
Erik Kaffehr
 

Doug Peterson

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4210
    • http://www.doug-peterson.com
Re: Is the D800E expected to have noticeable better IQ than the D800?
« Reply #30 on: February 17, 2012, 11:49:14 pm »

It's just a 22% improvement in resolution. I don't think it will change the world.

Erik, I find it odd that in two pages of this thread I find the most useful answer so far in your reply despite it being the shortest!

I'd add from hands-on experience with digital backs I'd say 15-20%. If you're lenses aren't very sharp, your focus is off slightly, or the aperture selected has moderate or high diffraction, then the difference is much less. There also seems to be a slight change in the "feel" of the sharpness that I don't pretend to be able to qualify mathematically, nor would I claim is night and day.

I don't know how closely that experience will correlate to this Nikon given that the D800E is not an AA-free filter in the sense of just removing the AA filter.

It seems to me that resolution and IQ seem are often used as interchangeable terms. I suppose it's somewhat subjective but I'd argue very strongly that resolution is one facet of IQ and that pure resolution is often 3rd or 4th on our client's lists of facets of image quality.

Doug Peterson (e-mail Me)
__________________

Head of Technical Services, Capture Integration
Phase One Partner of the Year
Mamiya Leaf, Leica, Arca Swiss, Cambo, Profoto, LaCie, Canon, TTI, Broncolor & More

National: 877.217.9870  |  Cell: 740.707.2183
Newsletter | RSS Feed
Buy Capture One 6 at 10% off
« Last Edit: February 19, 2012, 11:07:48 am by dougpetersonci »
Logged

MrSmith

  • Guest
Re: Is the D800E expected to have noticeable better IQ than the D800?
« Reply #31 on: February 18, 2012, 04:32:11 am »

still a significant jump, a lot of people would have been delighted with 22% improvement. especially in a 22% reduction in cost or 22% bigger screen on a digital back  ::)
Logged

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: Is the D800E expected to have noticeable better IQ than the D800?
« Reply #32 on: February 18, 2012, 05:18:29 am »

Hi,

Lest not forget, the 22% increase in resolution comes from a 60% reduction in cost for Nikon owners.

Best regards
Erik

still a significant jump, a lot of people would have been delighted with 22% improvement. especially in a 22% reduction in cost or 22% bigger screen on a digital back  ::)
Logged
Erik Kaffehr
 

BernardLanguillier

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 13983
    • http://www.flickr.com/photos/bernardlanguillier/sets/
Re: Is the D800E expected to have noticeable better IQ than the D800?
« Reply #33 on: February 18, 2012, 06:27:32 am »

Hi,

Lest not forget, the 22% increase in resolution comes from a 60% reduction in cost for Nikon owners.

Besides, the removal of the AA filter should be worth a few bonus extra %, no? :-)

Cheers,
Bernard

PhotoEcosse

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 712
Re: Is the D800E expected to have noticeable better IQ than the D800?
« Reply #34 on: February 18, 2012, 10:03:28 am »

If you download the Technical Guide for the D800/D800E, Nikon seem to be saying the the E version will lead to enhancements with some type of image but detract from others.

Reading between the lines, just as several professional photographers have said that they will need to use both the D800 and the D4 to cover all situations, it seems like Nikon are saying they will require both a D800 and a D800E (as well as the D4, of course).
Logged
************************************
"Reality is an illusion caused by lack of alcohol."
Alternatively, "Life begins at the far end of your comfort zone."

DaveCurtis

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 508
    • http://www.magiclight.co.nz
Re: Oversample the signal enough and then apply LPF still seems best overall
« Reply #35 on: February 18, 2012, 02:51:50 pm »


The choice of low-pass vs no lo-pass for the D800 vs D800E is not that different.  They are really two discrete options on a continuous scale of possible choices (imagine a user-tunable filter!).  Someone else posted recently in this forums that you can either start with a lower frequency response (give you a cleaner starting point) and then try to get back some of the highest frequencies with deconvolution / sharpening ... or you can start with a higher frequency response then deal with aliasing artifacts when/if they arise.  I think that's a very sensible way of looking at it.

It would be interesting to compare the same image taken with both cameras and deconvolution sharpening applied to the D800. How close would it get to the D800E.
Logged

JeffKohn

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1668
    • http://jeffk-photo.typepad.com
Re: Oversample the signal enough and then apply LPF still seems best overall
« Reply #36 on: February 18, 2012, 03:53:53 pm »

It would be interesting to compare the same image taken with both cameras and deconvolution sharpening applied to the D800. How close would it get to the D800E.
I agree, this is what I want to see before making my purchase decision; hopefully I'll be able to get a hold of some optimally-shot RAW images that I can sharpen myself to compare. The D800/D800e comparison in that Nikon article is interesting, but I can't help thinking the D800 image isn't optimally sharpened.

To be honest I'm not so much worried about moire as a landscape shooter, it's more the false color and to a lesser extent edge aliasing that concern me. Some people don't seem to mind these artifacts, or even like the fact that they can create an initial impression of extra sharpness.  I think they make images look 'digital', and they only get worse as you sharpen (even if you use mild or no capture sharpening, output sharpening can be problematic).

So to me, if a D800 image with deconvolution sharpening gets me almost the same detail/sharpness but without the artifacts, that would be my preference.  I'm just not sure how close they two will be after careful processing, we'll have to wait till more samples are available to judge.

Logged
Jeff Kohn
[url=http://ww

Bart_van_der_Wolf

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 8913
Re: Oversample the signal enough and then apply LPF still seems best overall
« Reply #37 on: February 18, 2012, 06:16:49 pm »

I agree, this is what I want to see before making my purchase decision; hopefully I'll be able to get a hold of some optimally-shot RAW images that I can sharpen myself to compare. The D800/D800e comparison in that Nikon article is interesting, but I can't help thinking the D800 image isn't optimally sharpened.

Hi Jeff,

Postponing one's decision until actual useful comparison material is available makes a lot of sense to me. The D800 vs D800E resolution differences will probably be small when proper (deconvolution) sharpening is applied to the 'non-E' files. Sharpening 'E' files is probably bad idea because it will enhance the (depending on image content) unavoidable artifacts such as false color artifacts and jaggies. Paying a bonus for a higher chance on artifacts (which will also hamper further enlargement capabilities) seems like a bad decision ...

Cheers,
Bart
Logged
== If you do what you did, you'll get what you got. ==

BernardLanguillier

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 13983
    • http://www.flickr.com/photos/bernardlanguillier/sets/
Re: Oversample the signal enough and then apply LPF still seems best overall
« Reply #38 on: February 18, 2012, 08:43:10 pm »

Postponing one's decision until actual useful comparison material is available makes a lot of sense to me. The D800 vs D800E resolution differences will probably be small when proper (deconvolution) sharpening is applied to the 'non-E' files. Sharpening 'E' files is probably bad idea because it will enhance the (depending on image content) unavoidable artifacts such as false color artifacts and jaggies. Paying a bonus for a higher chance on artifacts (which will also hamper further enlargement capabilities) seems like a bad decision ...

That's my current view also.

So I don't see the value of buying a camera today that makes me spend time look for moire for those non landscape images I know I would be taking with it... for a few % of additional sharpness.

Cheers,
Bernard

Mike Guilbault

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1040
    • Mike Guilbault Photography
Re: Is the D800E expected to have noticeable better IQ than the D800?
« Reply #39 on: February 19, 2012, 11:04:35 am »

I find it interesting that most descriptions of the D800E say that it's targeted to those requiring hi-resolution images such as Landscape and FASHION photographers.  Wouldn't the moire problem of the E be worse for Fashion photography since it's all about the clothing where moire is more prominent?
Logged
Mike Guilbault
Pages: 1 [2] 3 4   Go Up