Pages: 1 2 [3] 4 5   Go Down

Author Topic: D800 and E... poor for landscape use?  (Read 35869 times)

jeremypayne

  • Guest
Re: D800 and E... poor for landscape use?
« Reply #40 on: June 30, 2012, 06:58:17 pm »

I think this deserves a reference ...

http://luminous-landscape.com/tutorials/resolution.shtml

"You have all the data at hand, but take the green-yellow light and f/8-f/11 aperture values as a reference. It represents a realistic, not too demanding case. Consider a 35mm system with a lens at f/11. At best, the maximum resolution you will get is equivalent to 16 MP, even if your camera has 22 or 25 MP. In the case of an APS-C based system the limit goes to 7 MP, and 4 MP considering a Four Thirds format. Stopping down to f/22 the limit of the effective resolution of the 35mm based system goes to 4 MP!"

Logged

texshooter

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 575
Re: D800 and E... poor for landscape use?
« Reply #41 on: June 30, 2012, 09:22:54 pm »

according to this experiment, the loss of sharpness on a D800 due to diffraction at f22 can be recovered with post process sharpenning. its really not that bad.
http://fstoppers.com/what-is-lens-diffraction-on-dslr-camera

all this fuss over diffraction is so " princess and the pea"
Logged

Bart_van_der_Wolf

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 8913
Re: D800 and E... poor for landscape use?
« Reply #42 on: June 30, 2012, 09:36:33 pm »

all this fuss over diffraction is so " princess and the pea"

Not really. At f/22 on a 4.88 micron sensel pitch your fully diffraction limited resolution for the 564nm wavelength (close to the weighted average of luminance) will be at 78.3585 cycles/mm, which is 23.5% below Nyquist (at (102.4590 cycles/mm) for the D800(E). No fairytales, just plain physics.

But by all means feel free to throw away almost 1/4th of the peak resolution to achieve equal unsharpness across a deeper DOF.

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

texshooter

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 575
Re: D800 and E... poor for landscape use?
« Reply #43 on: June 30, 2012, 09:51:37 pm »

if diffraction is such a big deal, then wouldnt the D800E be a better choice than the D800 because the D800E has no anti alliacing filter--thats one less surface to add to diffraction. put a better way,

what is worse, 1) the added diffraction of the D800 or 2) the added Moire of the D800E? No more equations please. real world results are needed here.
Logged

Bart_van_der_Wolf

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 8913
Re: D800 and E... poor for landscape use?
« Reply #44 on: June 30, 2012, 10:12:19 pm »

if diffraction is such a big deal, then wouldnt the D800E be a better choice than the D800 because the D800E has no anti alliacing filter--thats one less surface to add to diffraction. put a better way, ...

Diffraction limits are solely determined by aperture number and wavelength of light. AA-filtering has nothing to do with it, unless the image is not fully diffraction limited like it is in the earlier f/22 example.

The difference with barely diffraction limited signals at f/5.6 is apparently less than 1% on the D800 versus the D800E.

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

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
Re: D800 and E... poor for landscape use?
« Reply #45 on: July 01, 2012, 12:00:54 am »

There are two broad but distinct aspects of this issue which I think it helps to be clear about. One is the theoretical, diffration limited resolution of a lens in terms of the sensor pixel count that comprises the image, ie. does the additional pixel count of the D800 serve any useful resolution purpose when the lens is used at F22, for example?

And the other aspect is the practical significance of such loss of resolution due to diffraction, at particular print sizes.

The first theoretical aspect is a guide or hint as to what to expect. The second aspect you really should dertmine for yourselves through making your own comparisons. Experiment a bit. The LiveView feature makes this process so much easier than it used to be because you can now be certain that your focus is exact.

I'll always remember the surprise I experienced when I bought a Canon 5D and compared resolution of the same scene shot at both F8 and F16 using the Canon 24-105/F4 zoom at the same focal length. The difference in sharpness in the plane of focus was very marginal at 100% on the monitor, representative of a really huge print.  However, the differences in sharpness away from the plane of focus was very much greater in the F16 shot.

In other words, in order to achieve that significant increase in DoF that F16 provides, compared with F8, the sacrifice in sharpness at the plane of focus was relatively insignificant. I made other comparisons ranging from F4 to F22, and found that F22 seemed to be the aperture at which there was a clearly noticeable fall-off in resolution, compared with F8, F11 and F16. I try to avoid using F22.

Years later when I bought a Canon 50D, which is effectively a cropped version of a 38mp full-frame DSLR, I compared resolution at various F/stops with my 10mp 40D, equivalent to a cropped 25.6mp full-frame. I wanted to know if F16 would provide any resolution increase with the 50D.

Surprisingly it did. Resolution at F16 with a good quality prime lens, the Canon 50/1.4, was approximately equal to resolution on the 40D at F11, at the plane of focus. Of course, away from the plane of focus, the 50D at F16 provided significantly sharper results than the 40D at F11; in other words the 50D provided greater DoF without any sacrifice in sharpness at the plane of focus. That was a worthwhile improvement of the 50D sensor which many uninformed people on the internet claimed would serve no purpose at apertures beyond F8 and certainly F11. However, at F22, it was clear that the 50D provided no additional benefit of a practical nature. That aperture of F22 was effectively the cut-off point.

My advice to you guys is, get out your cameras, set up your tripods and take some test images for comparison purposes so that you can get a feel for the magnitude of any differences observed at various F stops.
Logged

Scott O.

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 315
    • Photography by Scott and Joyce
Re: D800 and E... poor for landscape use?
« Reply #46 on: July 01, 2012, 01:19:32 am »

"I'll reserve judgement until I have a bit more experience, but it seems rather ironic that a camera that at first seems tailor made for use in landscape photography is irrevocably limited by diffraction softening."


I like the more experience comment, but it seems that you have already reached an opinion...

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: D800 and E... poor for landscape use?
« Reply #47 on: July 01, 2012, 02:04:22 am »

Hi,

AA filter does not add diffraction. Diffraction is solely depending on light and aperture. The AA filter does reduce contrast on small details. The effect of the aliasing filter is most noticeable at optimal aperture and it would have little effect in a diffraction limited image.

The simple truth is that photography is often diffraction limited, but the diffraction limit is more obvious if you raise the ribbon by using more pixels, printing larger and looking at the same distance and so on.

Diffraction works very well with sharpening, if radius is correctly chosen. So most of the impression of sharpness can be regained with adequate sharpening. If the lens is stopped down to f/22 or f/32 MTF will probably drop to zero at the pixel pitch, in which case resolution would be irrevocably lost.

The other point to see it, any good lens will be diffraction limited at f/11 or so, so if the lenses are used stopped down there is little reason to use the most expensive lenses. The same also applies to sensors, but we normally don't pay a premium price for high resolution sensors.

Going to larger formats does not help. All circumstances being equal you would need to stop down more on a MF or large format camera the diffraction eating up any format related advantage.

This is recommended reading: http://www.photodo.com/topic_138.html

The author makes comparison on film, 135 (at f/5.6) 120 (at f/11) and 4x5" (at f/22) using Tri X and T-MAX 100 film.

Here is one of the observations:
"And here is the big surprise: a 35 mm negative taken with the T-Max 100 is sharper than the large format negative taken with the Tri-X! Who would have thought it? It is important to add here that the small format picture was taken with a super sharp lens, the camera mounted on a tripod, and using a cable release; not exactly a free hand shot. But the large format picture was taken using the exact same technique. A 35 mm negative handled in the right way can be very clear, even when compared to large format."

Best regards
Erik


if diffraction is such a big deal, then wouldnt the D800E be a better choice than the D800 because the D800E has no anti alliacing filter--thats one less surface to add to diffraction. put a better way,

what is worse, 1) the added diffraction of the D800 or 2) the added Moire of the D800E? No more equations please. real world results are needed here.
Logged
Erik Kaffehr
 

marcmccalmont

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1780
Re: D800 and E... poor for landscape use?
« Reply #48 on: July 01, 2012, 02:08:37 am »

Just for argument sake you have 2 choices for landscape photography, 1. a gigapixel camera where every pixel is soft at all apertures but the overall scene is quite nicely represented or 2. a 4 pixel camera where all four pixels are very sharp at all apertures but the overall scene is poorly represented. A leaf looks just like a house, 4 squares but those squares are tack sharp! Which would you use?

Another way of looking at it is at f5.6 my D800E is a D800E at f8.0 it is a D800 at F16 it is a 5DIII at f22 it is a 5D. However my 5D is a 5D at f22, f16, f8 and f5.6 it will never be a D800E at any aperture

Leaves in my 5D files look like they were painted, leaves in my 5DII files look like they were photographed, leaves in my D800E files look real.

Marc
Logged
Marc McCalmont

free1000

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 465
    • http://www.foliobook.mobi
Re: D800 and E... poor for landscape use?
« Reply #49 on: July 01, 2012, 02:33:43 am »

Just for argument sake you have 2 choices for landscape photography, 1. a gigapixel camera where every pixel is soft at all apertures but the overall scene is quite nicely represented or 2. a 4 pixel camera where all four pixels are very sharp at all apertures but the overall scene is poorly represented. A leaf looks just like a house, 4 squares but those squares are tack sharp! Which would you use?

Another way of looking at it is at f5.6 my D800E is a D800E at f8.0 it is a D800 at F16 it is a 5DIII at f22 it is a 5D. However my 5D is a 5D at f22, f16, f8 and f5.6 it will never be a D800E at any aperture

Leaves in my 5D files look like they were painted, leaves in my 5DII files look like they were photographed, leaves in my D800E files look real.

Marc

I think this puts it very well.  The D800E is worth getting (rather than the D800) because at optimal use its better than anything else in its class. Its worth it because it makes that quality an option. 

My expectations are high because of what has been written about this camera. Generalities aside, what I need is not something better than the best other DSLR though, but something from which 50x40" images can compare reasonably in a exhibition with images from my Aptus 75 on Mamiya with the 550-110 zoom. This is because I would like to use this camera to add to an existing project and I'd like things to look consistent.  This is a big ask, but in an exhibition, people (not photographers) don't pixel peep and when prints are 50x40 they are standing well back.

My initial test against my Aptus shows that the D800E with 50 at f8 compares very well at f8 against the Aptus on a Mamiya with 80 f2.8. This is my reference point, I'm not really expecting the D800E to be at the level of the MF equipment, but it does very well. Well enough to provide me with something approaching that detail in the image in conditions that would defeat the Mamiya/Aptus combination, hand held, low light, long exposures etc. etc.  My task now is to determine what lenses at my preferred focal lengths will get the best out of this camera.  The DOF is greater of course with the D800E and this is useful. 
 
Initial tests suggest to me that the A75 (at ISO 50) remains superior in DR and colour to the D800E. I need to do a bit more 'peeping' to confirm whether this is the case as the light was changing rapidly when I did that test.

So far I conclude the D800E is really a great camera.  I'll shoot on the Aptus when practical but use the D800E when it isn't.
 
 



 
Logged
@foliobook
Foliobook professional photo

ErikKaffehr

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 11311
    • Echophoto
Re: D800 and E... poor for landscape use?
« Reply #50 on: July 01, 2012, 02:59:35 am »

Hi,

My guess is that your Nikon D800E with an excellent lens shot at f/5.6 would be about equal in resolution with your Aptus shot at f/11.
Because the pixel pitch is larger on the A75 the corresponding detail will have higher MTF (edge contrast) on the the Aptus. Using the Aptus at f/11 would reduce MTF enough so the two system may be equal. They would also give the same DoF.

Shooting the Aptus at f/5.6 should give better sharpness.

From the data and samples I have seen I would really expect that the Nikon D800E would have smoother shadows if both images were correctly exposed to the right (that is near clipping in raw data).

Best regards
Erik

I think this puts it very well.  The D800E is worth getting (rather than the D800) because at optimal use its better than anything else in its class. Its worth it because it makes that quality an option. 

My expectations are high because of what has been written about this camera. Generalities aside, what I need is not something better than the best other DSLR though, but something from which 50x40" images can compare reasonably in a exhibition with images from my Aptus 75 on Mamiya with the 550-110 zoom. This is because I would like to use this camera to add to an existing project and I'd like things to look consistent.  This is a big ask, but in an exhibition, people (not photographers) don't pixel peep and when prints are 50x40 they are standing well back.

My initial test against my Aptus shows that the D800E with 50 at f8 compares very well at f8 against the Aptus on a Mamiya with 80 f2.8. This is my reference point, I'm not really expecting the D800E to be at the level of the MF equipment, but it does very well. Well enough to provide me with something approaching that detail in the image in conditions that would defeat the Mamiya/Aptus combination, hand held, low light, long exposures etc. etc.  My task now is to determine what lenses at my preferred focal lengths will get the best out of this camera.  The DOF is greater of course with the D800E and this is useful. 
 
Initial tests suggest to me that the A75 (at ISO 50) remains superior in DR and colour to the D800E. I need to do a bit more 'peeping' to confirm whether this is the case as the light was changing rapidly when I did that test.

So far I conclude the D800E is really a great camera.  I'll shoot on the Aptus when practical but use the D800E when it isn't.
 
 



 
Logged
Erik Kaffehr
 

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
Re: D800 and E... poor for landscape use?
« Reply #51 on: July 01, 2012, 04:25:46 am »

Don't guess. Do the f***ing test.
Logged

MrSmith

  • Guest
Re: D800 and E... poor for landscape use?
« Reply #52 on: July 01, 2012, 05:30:16 am »

Just for argument sake you have 2 choices for landscape photography, 1. a gigapixel camera where every pixel is soft at all apertures but the overall scene is quite nicely represented or 2. a 4 pixel camera where all four pixels are very sharp at all apertures but the overall scene is poorly represented. A leaf looks just like a house, 4 squares but those squares are tack sharp! Which would you use?

i would use the 4 pixel camera with a long lens.
and do a bit of stitching..
Logged

jeremypayne

  • Guest
Re: D800 and E... poor for landscape use?
« Reply #53 on: July 01, 2012, 08:17:17 am »

Initial tests suggest to me that the A75 (at ISO 50) remains superior in DR and colour to the D800E.

Please share these results when you can ... given that other testers have found about a 2 stop advantage for the D800, I'd love to know what convinced you otherwise.

How are you measuring dynamic range?
Logged

Bart_van_der_Wolf

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 8913
Re: D800 and E... poor for landscape use?
« Reply #54 on: July 01, 2012, 09:42:02 am »

Another way of looking at it is at f5.6 my D800E is a D800E at f8.0 it is a D800 at F16 it is a 5DIII at f22 it is a 5D. However my 5D is a 5D at f22, f16, f8 and f5.6 it will never be a D800E at any aperture

Leaves in my 5D files look like they were painted, leaves in my 5DII files look like they were photographed, leaves in my D800E files look real.

Hi Marc,

It indeed all depends on the aperture used, and at a very narrow aperture like f/22 these cameras will produce virtually identical absolute output resolution (due to the influence of the absolute (MTF=0) diffraction limit). So the D800 is no better or worse of a camera for landscape than the others when DOF is maximized this extreme.

Things will start looking better for the higher sensel density cameras when the aperture is chosen to be a bit wider.

To illustrate, I've attached 2 charts that show the Optical Transfer Function (combination of diffraction and defocus) on an x-axis of absolute resolution in cycles/mm, for a D800(E) and a 5D2 (or 1Ds3) at f/22.

I've also attached 2 charts that show that at f/16 the D800 already starts to gain actual resolution over the 5D2, which would become apparent on printed output of 10x magnification (24x36cm or 9.4 x14.2 inch) or more. I've only calculated the response up to the Nyquist frequency of both cameras, 102.5 cy/mm for the D800(E) and 78.4 cy/mm for the 5D2 .

A different comparison is when a physically larger sensor size is used because, while it will hit the same physical f/22 limitation, it requires less output magnification to reach the same output size. Less magnification will retain higher absolute resolution. However, the requirement for longer focal lengths will reduce some of the benefit.

My conclusion is that the D800(E) is a great camera for landscape, especially when it is used at wider apertures than f/18. Narrower apertures take away any resolution benefit it might otherwise have (assuming comparable optics).

Cheers,
Bart


P.S. The Circle of Confusion (CoC) diameters I used in the charts are for 1.5, 2, 3, and 4x the sensel pitch.
« Last Edit: July 01, 2012, 06:30:14 pm by BartvanderWolf »
Logged
== If you do what you did, you'll get what you got. ==

Wim van Velzen

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 372
    • http://www.fotografiewimvanvelzen.nl
Re: D800 and E... poor for landscape use?
« Reply #55 on: July 01, 2012, 12:04:52 pm »

A technique I sometimes use with my gear (Rollei 6008i and older 22MP digital back) is take two shots from the same tripod position, one say f8 @ 1/60th and one f22 @ 1/8th. Post production is the same; I then put the f8 tiff as a layer on top of the f22. I then erase the top f8 layer on those areas I'd like more DoF (for example heather in the foreground of a moor vista).

It works far easier than focus stacking with the same f stop, as the field of view stays exactly the same.

Another option would of course be to do the opposite and just have the f8 layer visible on the parts where sharpness is paramount.
Logged
I don't have a signature.

uaiomex

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1211
    • http://www.eduardocervantes.com
Re: D800 and E... poor for landscape use?
« Reply #56 on: July 01, 2012, 02:38:27 pm »

Thanks for sharing this technique. It seems it leads too to interesting and some time unexpected results.
Eduardo

A technique I sometimes use with my gear (Rollei 6008i and older 22MP digital back) is take two shots from the same tripod position, one say f8 @ 1/60th and one f22 @ 1/8th. Post production is the same; I then put the f8 tiff as a layer on top of the f22. I then erase the top f8 layer on those areas I'd like more DoF (for example heather in the foreground of a moor vista).

It works far easier than focus stacking with the same f stop, as the field of view stays exactly the same.

Another option would of course be to do the opposite and just have the f8 layer visible on the parts where sharpness is paramount.
Logged

Guillermo Luijk

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2005
    • http://www.guillermoluijk.com
Re: D800 and E... poor for landscape use?
« Reply #57 on: July 01, 2012, 06:18:33 pm »

Your diffraction limit is the same as any 35mm format camera. Use f/16 and enjoy your camera.

Yes and no. While it is true that a high resolution sensor such as the one in the D800 will not suffer from diffraction in a print copy of a given size more than any other lower resolution sensor, it is also true that the higher the resolution the more difficult is to actually enjoy it in large DOF applications such as landscape or arquitecture.

So using f/16 will allow you to enjoy the D800 like any other FF camera, but will not allow you to enjoy D800's 36Mpx.

If a certain very high resolution in Mpx is needed for a given large DOF, three solutions are possible:
  • Focus stacking a series of shots with wider aperture set on the lens
  • Tilting to achieve large DOF at wide apertures (needs specialized gear and it is not applicable to all scenes)
  • Stitching several shots of different areas of the scene so that DOF requirements on each of them get severely reduced (you need to individually focus each of the shots)

As long as you don't apply any of the techniques above, using very large resolution sensors (let's say >24Mpx) at reduced apertures (e.g. f/16) means you will never get 24Mpx of effective information, so in practical terms you are throwing away those Mpx you paid for. Your files will be 36Mpx in size, but won't hold 36Mpx of information.

Regards
« Last Edit: July 01, 2012, 06:25:11 pm by Guillermo Luijk »
Logged

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
Re: D800 and E... poor for landscape use?
« Reply #58 on: July 01, 2012, 09:35:50 pm »

So using f/16 will allow you to enjoy the D800 like any other FF camera, but will not allow you to enjoy D800's 36Mpx.

Not necessarily, Guillermo. My own tests comparing the 10mp Canon 40D with the 15mp 50D, indicate that the 50D at F16 produces approximately the same level of detail, at the plane of focus, as the 40D at F11.

Extrapolating the pixel densities of these cameras to full-frame, the comparison becomes 25.6mp versus 38.4mp (full-frame). Such differences are very close to the difference between the 24mp D3X and the 36.3mp D800.

I would predict that the D800 used at F16 will give approximately the same level of detail as the D3X at F11, in the plane of focus.

Someone please do the test to prove me wrong. I'm so rarely proved wrong nowadays.  ;D
Logged

Bart_van_der_Wolf

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 8913
Re: D800 and E... poor for landscape use?
« Reply #59 on: July 02, 2012, 11:21:39 am »

I would predict that the D800 used at F16 will give approximately the same level of detail as the D3X at F11, in the plane of focus.

Someone please do the test to prove me wrong. I'm so rarely proved wrong nowadays.  ;D

Hi Ray,

Not sure where the D3x came from, but attached is the comparison between the D3x and the D800 (once expressed in cy/mm and once in cy/pixel resolution), from an optical point of view. Obviously the D800 has more resolution, but the optical response at the same aperture is of course identical. Only if one were to crop the D800 to the size of the D3x (no idea why one would do that, other than for argument's sake), then the per pixel response would change, but in favor of the larger sensel pitch camera (less diffraction per pixel due to lower resolution).

These are OTFs (diffraction+defocus, CoC blur assumed to be 1.5x sensel pitch), so a perfect camera is assumed. The actual MTF curves could never exceed these, but will be somewhat lower, depending on things like AA-filter and sensel aperture.

This hopefully also demonstrates where the confusion comes from that some people experience, because they fail to translate to output quality related cy/mm resolution.

Cheers,
Bart
« Last Edit: July 02, 2012, 11:23:32 am by BartvanderWolf »
Logged
== If you do what you did, you'll get what you got. ==
Pages: 1 2 [3] 4 5   Go Up