Pages: 1 ... 3 4 [5]   Go Down

Author Topic: 5D3 vs. D800 - great real world head-to-head tests  (Read 57729 times)

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
Re: 5D3 vs. D800 - great real world head-to-head tests
« Reply #80 on: May 25, 2012, 08:58:47 am »

The real thing that separates a good shot from blurry tripe is a tripod. Nothing new here.

But that also depends on factors such as lighting conditions and subject movement. When the subject is moving, a tripod doesn't help, nor does IS or VR.

When lighting conditions allow for a fast shutter speed at the desired aperture and ISO, a tripod is not needed, except in specific circumstances when one needs a slow shutter speed whatever the lighting conditions, as when photographing a waterfall using an ND filter, and perhaps when using a real monster of a lens that's too heavy to hold.
Logged

David Watson

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 480
    • David Watson
Re: 5D3 vs. D800 - great real world head-to-head tests
« Reply #81 on: May 25, 2012, 03:45:11 pm »

But that also depends on factors such as lighting conditions and subject movement. When the subject is moving, a tripod doesn't help, nor does IS or VR.

When lighting conditions allow for a fast shutter speed at the desired aperture and ISO, a tripod is not needed, except in specific circumstances when one needs a slow shutter speed whatever the lighting conditions, as when photographing a waterfall using an ND filter, and perhaps when using a real monster of a lens that's too heavy to hold.

Couldn't agree more.  In addition there are many situations where a tripod is a positive hindrance to getting the shot.  IMO the D800(E0 has moved the goalposts a little in this respect.  The dynamic range, level of detail and lack of noise up to 1600 at least make this a very versatile and usable camera.  I shot a wedding last weekend (I was doing the "reportage" and another photographer was doing the set pieces).  He was struggling around the church with his 5D and a monopod trying to get the shots with a 70-200.  I was using an 85 1.4 handheld mostly at f2 and 1600 ASA.  No problems at all.  This camera excels on a tripod but it is also a class act hand held.
Logged
David Watson ARPS

Lightsmith

  • Full Member
  • ***
  • Offline Offline
  • Posts: 197
Re: 5D3 vs. D800 - great real world head-to-head tests
« Reply #82 on: May 26, 2012, 04:07:00 pm »

The photographer with the 5D was most likely needing the tripod by virtue of using either the non-IS or the f4 version of the 70-200mm Canon zoom. The 5D is perfectly capable of ISO 3200 with very usable images for wedding photography.

Having shot many thousands of images with Nikon and Canon cameras as well as doing the post editing on thousands taken by second photographers under identical lighting situations it becomes very clear what the strengths and weaknesses are of each system. The strength of the Canon cameras is with the in-camera processing to produce more film like images that are ready to print while the weakness has been with autofocus accuracy and fill flash exposure accuracy. I can take a Nikon RAW file further with post processing than I can most Canon cameras' files but I need to add my time as an additional cost.

Prior to the D3 the movement was from Nikon to Canon to get higher ISO capabilities but in the past few years it has been from Canon to Nikon and the most often cited reason was a desire for autofocus accuracy. Canon users have for years been sending cameras and lenses back to Canon for "recalibration" and noted wedding photographer Jeff Ascough would do this at the end of every season sending all his cameras and lenses in to Canon.

A real world ISO test needs to reflect subjects that cover the color spectrum and light sources that cover only a portion of the visible spectrum and so need different amounts of signal amplification for the RGB segments of the scene or subject. I like to use people, both light and dark skinned, shot under fluorescent and tungsten light sources (and not 6000K flash or sunlight) and then look at the tonal range provided. NR will compress the tonal range and the result is the plastic looking skin that is often mentioned with regard to certain cameras, like the 5D at ISO 3200.

I did a series of test shots of people under tungsten and fluorescent light sources this past weekend with my D800E. The D800 provided usable images (moderate post processing of noise a matter of choice) at ISO 3200 with all subjects and up to ISO 6400 with subjects with light skin. This was comparable to what I have seen with the D3 and is remarkable considering the smaller photosites and greater signal amplification of the D800.

Tests using flash or sunlight might as well be of a black cat in a coal mine for all the "real world" value they provide. In a studio a high ISO setting is not needed as there is enough light for ISO 200 at f11 with any set of strobes. Outdoors in the bright sun it is much the same. I want a situation where the ambient light and or the color reflected from the subject has a bias that requires extra boost of one part of the spectrum as that is what generates the most visible chroma noise. A black person with a dark red or purple dress in a room dimly lit with a combination of tungsten and fluorescent lamps is my favorite test of how high I can safely set the ISO of the camera and get an image that can be printed without NR needed or with some NR or where the NR is going to be so great that detail is going to be visibly absent in the print.
Logged

Fine_Art

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 1172
Re: 5D3 vs. D800 - great real world head-to-head tests
« Reply #83 on: May 27, 2012, 03:32:47 am »

But that also depends on factors such as lighting conditions and subject movement. When the subject is moving, a tripod doesn't help, nor does IS or VR.

When lighting conditions allow for a fast shutter speed at the desired aperture and ISO, a tripod is not needed, except in specific circumstances when one needs a slow shutter speed whatever the lighting conditions, as when photographing a waterfall using an ND filter, and perhaps when using a real monster of a lens that's too heavy to hold.

Certainly there are types of photography where a tripod is a hindrance. That doesn't change the fact that when people are talking about mush detail with these cameras it's probably not from lenses that can't keep up to 5 micron pixels. It's probably from other factors. Who is going to buy a camera like that to put poor lenses on it? I mean besides a reviewer making a point. No, people are going to match it up with some decent glass. Those are both good cameras.
Logged

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
Re: 5D3 vs. D800 - great real world head-to-head tests
« Reply #84 on: May 27, 2012, 05:01:06 am »

Certainly there are types of photography where a tripod is a hindrance. That doesn't change the fact that when people are talking about mush detail with these cameras it's probably not from lenses that can't keep up to 5 micron pixels. It's probably from other factors. Who is going to buy a camera like that to put poor lenses on it? I mean besides a reviewer making a point. No, people are going to match it up with some decent glass. Those are both good cameras.

It used be said that all lenses are equal at F8. If that's a slight exaggeration, then it's probably truer that all lenses are equal at F11. When Photodo used to carry out MTF tests purely on the lenses without camera body, which now seems to be a process which is too expensive, they were often asked why they didn't test at F11. The answer was, because all lenses are equally bad at F11.

If this is true (and I admit it's not quite true), any lens at F11 will deliver more resolution when used with a sensor with more pixels, up to a point. There's obviously a law of diminishing returns at play here.

I've confirmed for myself that a Canon 50/1.4 at F11 delivers more resolution with my 15mp 50D (38mp full-frame pixel-density equivalent), than it does with my 10mp 40D (26mp FF pixel-density equivalent).

In my experience, blurry images generally result from a shutter speed which is too slow, or sometimes due to misfocussing. For reliably sharp images with a slow shutter speed, a tripod is necessary, with MLU enabled. To escape misfocussing issues, LiveView is a great help.

A modern IS or VR lens can sometimes produce remarkably sharp results at a slow shutter speed, with a static subject, but I don't believe this method is as reliable as use of a tripod.

One can test this for oneself by taking several hand-held shots of the same target at a slow shutter speed, with an IS or VR lens. Most shots might be slightly blurry. However, one or two will likely be tack sharp. I think the tripod is more reliable in such situations.
Logged

BernardLanguillier

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 13983
    • http://www.flickr.com/photos/bernardlanguillier/sets/
Re: On Ansel Adams, stitching and very large prints
« Reply #85 on: May 27, 2012, 06:59:25 pm »

"A technically perfect photograph can be the world’s most boring picture."
in "Total Photography"

‎"The fact that a (in the traditional sense) technically deficient photograph can have greater emotional impact than a technically flawness picture probably comes as a shock to those who are naive enough to believe that technical exellence alone is a measure of a value of a photograph."
in "The Color Photobook"

Could it be that the creator of Instagram read the same book?  ;)

Besides, on the point, aren't we overlooking the possibility that some photographers may enjoy creating technically perfect photographs? I am sure nobody does create boring image on purpose - although some images sure look like there is deliberate intend related to the willingness to create a new and unique "style" - but I am sure that some photographers set high standards for themselves in terms of technical perfection, and sometimes prioritize this a bit too much, which may lead them to look down at some subjects simply because they are not confident they can capture them with the right level of technical perfection.

Oops... would stitchers be likely representatives of this category?  ???

If that were the case, then higher resolution cameras may in fact act as creativity liberators?

Cheers,
Bernard
« Last Edit: May 27, 2012, 11:40:13 pm by BernardLanguillier »
Logged

Ray

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 10365
Re: On Ansel Adams, stitching and very large prints
« Reply #86 on: May 27, 2012, 08:53:43 pm »

30 megapixels?

Before I read AAs books I read Andreas Feinigers books. Two quotes of him which guid me trough my photographic career are:

"A technically perfect photograph can be the world’s most boring picture."
in "Total Photography"

‎"The fact that a (in the traditional sense) technically deficient photograph can have greater emotional impact than a technically flawness picture probably comes as a shock to those who are naive enough to believe that technical exellence alone is a measure of a value of a photograph."
in "The Color Photobook"

Best,
Johannes

This is one of those adages under the guise of wisdom which is really total twaddle.

First, every photograph that has ever been taken is boring at least to someone, no matter how great the photo is considered by others to be. Interests vary enrmously amongst any population. What one person considers sublime, another may consider total crap.

Secondly, if it were possible to determine what could be the world's most boring photo, through a world-wide competition perhaps, technical perfection would definitely exclude any photo from winning the prize, because so far, technical perfection does not exist.

If someone were to produce a technically perfect photograph, its perfection would be a wonder to behold no matter how boring the subject may be in the eyes of some. The perfection itself would become the focus of interest. How is it possible, for example, to achieve such 3-dimensional realism, perfectly accurate colors and such sharply defined detail which simply increases in quantity the closer one gets to the photo?

The prize for the world's most boring photo would have to be awarded to a boring subject in conjunction with totally lousy technical quality.

Of course, I'm making certain assumptions about human rationality here. I'm assuming, for example, that whenever a person takes a photograph, it is of something that at least interested the photographer at the time, that motivated him to raise the camera and take the shot. When such shots, after processing, sometimes fail to satisfy, it must be due to technical limitations of some sort. Perhaps the dark ominous clouds are no longer ominous because the sky was blown due to incorrect exposure. The brilliant green bush in the foreground, drenched in the light of the setting sun, has lost its impact due to misfocussing and other factors such as inappropriate choice of F stop, or the color sensitivity and dynamic range of the camera was too limited, or the color gamut of the printer was not wide enough, and/or the Photoshop skills of the operator were not up to the job, etc. etc.

However, don't think that I've missed the underlying point. It's pretty obvious that an interesting composition that is technically inadequate may be more valuable than an uninteresting subject which is nevertheless technically impressive. Look at any family photo album for such examples. They are usually full of technically inadequate shots which are of immense value to the family members.
Logged

Tony Jay

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 2965
Re: 5D3 vs. D800 - great real world head-to-head tests
« Reply #87 on: May 28, 2012, 08:29:07 pm »

In photography there is no doubt a tension exists between intent and result, creativity and technical limitations.
Moreover, in each image the dynamic will change.
I totally agree that in an absolute sense a technically perfect photograph is still a pipedream.
The art and science of photography optimally employed is to give the illusion of technical expertise - hopefully with an aesthetically pleasing result.

I also agree that what constitutes a successful photograph is entirely in the eyes of the beholder. However, those images destined for public viewing are judged much in the manner of democratic elections - ultimately, successful images will be judged favourably by the majority even if this takes time, up to decades in certain situations.

Bernard makes the point that many photographers probably avoid certain types of scenes when contemplating stitching a large scene. He is absolutely correct but no more than the bird photographer who realizes that shooting at 1/25th second shutter speed with a 500 mm lens and a 1.4 teleconverter in gloomy light just won't work (belive me I have been there and have the images to prove it - good for a laugh but not much else).

Nonetheless, I believe that technical expertise in photography is important, and vital, but only as a means to an end (I am trying not to be too obvious here). I have images that, according to public opinion anyway, really have a wow factor that were technically very simple to execute. Other images that also excite a lot of interest were much much more difficult to do such as multiple shot HDR panoramas. Some of these have taken years to accomplish - mainly because previous results were not quite up to scratch. Unfortunately, not many people were at all interested in how the image was made but nonetheless really enjoyed the noiseless detail and wonderful tonal relationships.

The bottom line, of course, is that technical expertise can never substitute for lack of artistic motive and intent and that will be reflected in the result. Combat photography often is lacking technical merit yet some of those images have - rightfully - won worldwide acclaim for the memories, emotions, and messages that they evoke.

My few thoughts

Tony Jay
« Last Edit: May 28, 2012, 08:32:07 pm by Tony Jay »
Logged

ziocan

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 426
Re: 5D3 vs. D800 - great real world head-to-head tests
« Reply #88 on: May 30, 2012, 02:30:50 am »

Still arguing over these things....
all these cameras are plenty good!
Logged

wofsy

  • Newbie
  • *
  • Offline Offline
  • Posts: 38
Re: 5D3 vs. D800 - great real world head-to-head tests
« Reply #89 on: May 30, 2012, 09:23:42 am »


I watched these videos and thought that they were glossing over the clear win of Nikon's 33 megapixels over Canon's 23.  This gave me pause on pressing the Buy Now button for the 5d Mark 3. Until I see the cameras I will not know for sure, but it seems that Canon will have to catch up if it wants to stay in the game. I wonder if a 5d Mark4 is not far off, a thirty megapixel, low noise at high ISO, super auto focus, perfectly quiet shutter, answer to the D800.

I remember years ago photographer's telling me that cramming all of those extra pixels onto a sensor did not mean sharper images and  they also argued that no one made a lens with sufficient optics to capture the extra detail even it if was theoretically possible. But these objections were wrong and the 5d Mark 2's  with its 22 megapixel sensor became the standard for 35mm image quality. This makes me worry that Canon has fallen behind or has made a marketing mistake.

That said, for my own photography, the Mark 3 improves over the Mark 2 for three reasons: First, the shutter is nearly silent. This makes it possible to shoot meditation and other silent activities: Second, the autofous enhances the ability to shoot candid photos of ecstatic dance: Third, high quality high ISO files allow shooting yoga in low lit rooms and especially in early morning.

So I am torn about the Mark 3.
Logged

kers

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4386
    • Pieter Kers
Re: Fred Miranda just put both through their paces...
« Reply #90 on: May 30, 2012, 12:03:48 pm »

Live view focusing wide-open should be easily fixable with a firmware update - there's no hardware-specific reason it can't be done.

I very much like the fact you can use liveview at the desired F-stop - Some lenses need to be focussed on the F-stop you use.- one good example is the 24mmPCE lens.
(Maybe that is why people think it is a dud...)

I have heard that you cannot use liveview in the corners with the 5DmkIII - is that true?
Logged
Pieter Kers
www.beeld.nu/la

MrSmith

  • Guest
Re: 5D3 vs. D800 - great real world head-to-head tests
« Reply #91 on: May 30, 2012, 03:10:17 pm »

No.
Logged

torger

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 3267
Re: Fred Miranda just put both through their paces...
« Reply #92 on: June 07, 2012, 03:40:00 am »

I have heard that you cannot use liveview in the corners with the 5DmkIII - is that true?

If it is like on the 5Dmk2 you cannot move to the corners if you have autofocus enabled, but if you disable it you can move to the corners.

The reason for not being able to move into corners with autofocus enabled is probably that the contrast-detecting AF-algorithm may not work that well in lens corners wide open.
Logged

kers

  • Sr. Member
  • ****
  • Offline Offline
  • Posts: 4386
    • Pieter Kers
Re: 5D3 vs. D800 - great real world head-to-head tests
« Reply #93 on: June 28, 2012, 05:09:36 pm »

Thank you very much Torger,

This is more informative than just a ...NO...   

Logged
Pieter Kers
www.beeld.nu/la
Pages: 1 ... 3 4 [5]   Go Up