Do pixels matter? What 20" x 30" prints reveal.

Started Jun 14, 2013 | Discussions
LaszloBencze
OP LaszloBencze Contributing Member • Posts: 788
Re: Do pixels matter? What 20" x 30" prints reveal.
1

Thank you!

It's interesting that your experience is so similar to mine. I truly did expect to see a difference and was surprised I didn't.

I'm sure there are pictures which would reveal these large differences in pixel count. Pictures filled with fine detail—a wall filled with posters, a tree filled landscape, close ups of a butterfly's wings. But I take very few pictures like those. And I suspect most people also take pictures that don't easily reveal such differences in resolution.

I suppose the bottom line to all discussions about number of pixels is "Stop worrying about pixels! Refine your seeing. Work at developing yourself as a photographer. Let your pictures today be better than your pictures yesterday." That is the attitude I try to cultivate for myself.

LaszloBencze
OP LaszloBencze Contributing Member • Posts: 788
Re: Do pixels matter? What 20" x 30" prints reveal.
2

without posting any examples of your preposterous proposition I can only conclude this is a sneaky way of promoting you exhibition.

Actually it's a very straightforward way of promoting my exhibition. I'd like as many people to see it as possible. If I can entice a few users of this forum to the show, so much the better. They may come expecting to discern pixels and leave having enjoyed art.

LaszloBencze
OP LaszloBencze Contributing Member • Posts: 788
Re: Do pixels matter? What 20" x 30" prints reveal.
1

talico wrote:

I took the original post to mean good photos from good equipment are still good.

Yep. You got it. My post was meant to be reassuring to the many photographers who seem to be obsessed with technical specifications. Stop obsessing! Buying newer equipment with better specs is no guarantee of better pictures. Develop your skills with what equipment you now have. Enjoy the art of photography.

AnthonyL Senior Member • Posts: 2,931
Does it tie in with the theory?

I was coincidentally look at some resolution theory - see http://www.scss.com.au/family/andrew/camera/resolution/#background

How do your figures stack up with what is being said?

As an aside very few of my images are printed (and I don't really belong in this forum) but the only two which are framed and hanging in the house are 12" x 8" which my wife loves and were taken with a 3 megapixel Minolta Dimage 5.  None of my Canon 450d photos have yet made the grade

 AnthonyL's gear list:AnthonyL's gear list
Canon EOS 700D Canon EF-S 55-250mm f/4-5.6 IS Canon EF-S 10-22mm F3.5-4.5 USM Canon EF-S 18-55mm f/3.5-5.6 IS STM Canon EF 40mm f/2.8 STM +5 more
LaszloBencze
OP LaszloBencze Contributing Member • Posts: 788
Re: Does it tie in with the theory?
1

Anthony, I looked at that blog you linked to. It is filled with the sort of stuff I used to work with when I was a photo officer in the Air Force. Such figures are very important when you're trying to figure out the ultimate resolution of an aerial photograph. But in everyday life, none of that matters.

What matters is how you respond to a picture. Does it look "sharp" to you? Does it have distracting artifacts which do not look photographic? Does it's resolution support its subject matter. In short, does it look "right"? These are matters which cannot be easily quantified but which matter much more than the stuff that is quantifiable.

So there's no easy, numerical answer to the question you pose. There's only the artistic answer which must take into account your own standards, level of sophistication, and fussiness.

rwbaron Forum Pro • Posts: 14,031
Re: Do pixels matter? What 20" x 30" prints reveal.
1

LaszloBencze wrote:

Thank you!

It's interesting that your experience is so similar to mine. I truly did expect to see a difference and was surprised I didn't.

I'm sure there are pictures which would reveal these large differences in pixel count. Pictures filled with fine detail—a wall filled with posters, a tree filled landscape, close ups of a butterfly's wings. But I take very few pictures like those. And I suspect most people also take pictures that don't easily reveal such differences in resolution.

I suppose the bottom line to all discussions about number of pixels is "Stop worrying about pixels! Refine your seeing. Work at developing yourself as a photographer. Let your pictures today be better than your pictures yesterday." That is the attitude I try to cultivate for myself.

I'm 100% in agreement with all you've said and especially your last paragraph.

My experience is the accuracy of focus, the chosen aperture (both DoF and sharpest f stop) along with the placement of the subject in the frame, the quality of the optics and the technique at capture (minimizing vibration or shake) are all as important if not more so than the number of pixels and I believe DXO is now expressing similar views after testing camera/lens combinations.

Photography is a bit like golf in that when you press the shutter it's like beginning your swing. Like Jack Nicklaus said there are about 6 things he can think of when beginning his swing so you have to be certain they are the most important in acheiving the desired result

Bob

-- hide signature --
 rwbaron's gear list:rwbaron's gear list
Fujifilm X-T2 Fujifilm X-T20 Fujifilm XF 18-55mm F2.8-4 R LM OIS Fujifilm XF 55-200mm F3.5-4.8 R LM OIS Fujifilm XF 10-24mm F4 R OIS +8 more
rwbaron Forum Pro • Posts: 14,031
Re: Do pixels matter? What 20" x 30" prints reveal.
1

LaszloBencze wrote:

Thank you!

It's interesting that your experience is so similar to mine. I truly did expect to see a difference and was surprised I didn't.

I'm sure there are pictures which would reveal these large differences in pixel count. Pictures filled with fine detail—a wall filled with posters, a tree filled landscape, close ups of a butterfly's wings. But I take very few pictures like those. And I suspect most people also take pictures that don't easily reveal such differences in resolution.

I suppose the bottom line to all discussions about number of pixels is "Stop worrying about pixels! Refine your seeing. Work at developing yourself as a photographer. Let your pictures today be better than your pictures yesterday." That is the attitude I try to cultivate for myself.

Let me first say that I viewed your Mining, Transportation and Industry portfolio's on your website and they are excellent.  I really like your vision and style.

I think this further supports your point. Below is a photo from the last museum exhibition I paricipated in. Of the 4 images the juror selected, the museum purchased the photo second from the left for their permanent collection which was without doubt the image with the lowest technical quality of the 10 I submitted for review. They even used it on the cover of the program for the exhibition. It was taken with my 50D and 100-400L which had a decentering issue and was not even close to tack sharp at the captured FL besides being handheld. The image on the far right was captued with my G1X.

Bob

-- hide signature --
 rwbaron's gear list:rwbaron's gear list
Fujifilm X-T2 Fujifilm X-T20 Fujifilm XF 18-55mm F2.8-4 R LM OIS Fujifilm XF 55-200mm F3.5-4.8 R LM OIS Fujifilm XF 10-24mm F4 R OIS +8 more
AlbertTheLazy
AlbertTheLazy Veteran Member • Posts: 7,836
Re: Do pixels matter? What 20" x 30" prints reveal.

On the subject of big prints and exhibitions, I went to the annual photo festival at La Gacilly in Britanny last weekend. There were some seriously big prints, some of them in the 10 year retrospective, which means they would not be from current state of the art kit.

http://www.brittanytourism.com/things-to-do/events/brittany-s-main-events/peuples-nature-people-and-nature-photo-festival-la-gacilly-31-may-30-september-2013

As you can see from the people, the 'portrait' of an elephant is full length and about life size.

http://www.paysdelagacilly.com/envie-fete/le-festival-photo-peuple-et-nature/

Unfortunately they don't display the EXIF data 

-- hide signature --

Albert
Every photograph is an abstraction from reality.

 AlbertTheLazy's gear list:AlbertTheLazy's gear list
Canon PowerShot G15 Fujifilm X-T1 Sigma 70-300mm F4-5.6 APO DG Macro Fujifilm XF 18mm F2 R Fujifilm XF 35mm F1.4 R +4 more
LaszloBencze
OP LaszloBencze Contributing Member • Posts: 788
Re: Do pixels matter? What 20" x 30" prints reveal.
1

You have certainly proved our point. Everything is subsidiary to the impact of the image. Is it a compelling picture? A strong image? Something worth a second look? These are the factors that matter. An image of supreme technical competence that is boring is nothing but a boring picture.

Many years ago in Tucson, a young man who had just gotten involved in photography was running around town with an 8 x 10 camera. He went to a lot of trouble to lug this beast but every picture he took was a beginner's cliché. He would have been far better off to have used a 35mm camera and developed his seeing enough to justify the 8 x 10 camera. But he was seduced by equipment as are so many hobbyists.

XeroJay
XeroJay Senior Member • Posts: 2,015
My findings are...

After a decade and a half of shooting and printing digital images, megapixels is the final component in the chain of resolution and detail.

1. Movement during exposure is the first component to image detail. Mainly camera movement.

2. Lens characteristics and focus are the next most critical component.

3. Accutance of light sources is the third most likely to have affect on the image detail.

4. Image sensor size/total capture area. The larger the sensor; the less magnification relative to the final output.

5. Megapixels. This only noticeably comes into play after factors 1 through 4 are near optimum.

This isn't fact, and ymmv, but believe me, this comes from years of paying the mortgage by capturing and producing large prints. Megapixels are less of an issue today than ever before. If anyone thinks that they could ever tell the difference between my 1Dx and my 5D3 by spotting the extra 4mp from prints of any size, I would bet my pants they're mistaken.
--
If there's typos, you can be sure I used my iPhone.

 XeroJay's gear list:XeroJay's gear list
Canon EOS M6 Sony a7 III Canon EF 100mm F2.8L Macro IS USM Canon EF-M 22mm f/2 STM Sigma 20mm F1.4 DG HSM Art +4 more
LaszloBencze
OP LaszloBencze Contributing Member • Posts: 788
Re: My findings are...
1

I couldn't agree with you more. The other factors you list are far more important than pixel count. But perhaps the most crucial factor is simply whether the image is a good picture. The viewer tends to be far more forgiving of a good picture than of a mediocre one. A good picture takes your breath away. A mediocre one makes you want to count pixels for lack of anything better to do.

olliess Senior Member • Posts: 1,349
Re: My findings are...

LaszloBencze wrote:

A good picture takes your breath away. A mediocre one makes you want to count pixels for lack of anything better to do.

Harsh. But so true...

Great Bustard Forum Pro • Posts: 42,943
Excellent thread. I have a few thoughts about it.
1

LaszloBencze wrote:

Thank you for your congratulations.

Of course my point was to invite anyone who could attend to see for themselves and determine if THEY can tell the images from the two camera systems apart. Mind you, the point the of the exhibition is art, not pixel peeping. But for those who might wish to test their discernment in addition to appreciating the art, this presents an easy and public opportunity.

Indeed.  And, may I say, I thought it very cool of you to put your money where your mouth was with your previous offer -- I'm disappointed that DPR pulled the thread, ostensibly for that reason.

You do raise an interesting point that I've been discussing elsewhere. The claim was made that a 17 x 22 inch print from a 5 MP Olympus E1 was indistinguishable, without the use of a magnifying loupe, from the same size print from a 12 MP Olympus E30. Many came forward to support the claim:

http://www.dpreview.com/forums/post/51580883

and was continued in another thread:

http://www.dpreview.com/forums/post/51602273

Furthermore, there is a thread on this point in the Open Forum:

http://www.dpreview.com/forums/post/51635765

where I linked your original thread.

This is a very interesting subject to me. Let me restate some points I made in the second thread linked above:

This makes some sort of sense under certain conditions:

  • The lens is not sharp enough to make use of the additional pixels.
  • The aperture is so narrow that diffraction softening greatly diminishes the advantage of more pixels (e.g. if the f-ratio were f/16, or more narrow, the diffraction softening will certainly lessen the resolution advantage of 22 MP over 10 MP).
  • Motion in the scene or camera shake adds blur that greatly diminishes the advantages of more pixels.
  • The sensor with fewer pixels is more efficient, thus having less noise and/or greater DR.
  • 129 PPI (10 MP at 20 x 30 inches) with a "good enough" lens is beyond the visual threshold to distinguish higher resolutions still (e.g. 191 PPI from 22 MP) without going to "extraordinary efforts" to see the difference in resolution.

Another challenge that you could equally make is to see if people could tell the difference in noise between an ISO 800 and ISO 1600 photo printed at 20 x 30 inches, which, interestingly enough, will result in the same increase in noise as 22 MP will have over 10 MP.

dad_of_four Veteran Member • Posts: 8,467
20x30 of football player
1

A customer bought a 20x30 print of his son playing football. I cropped the photo per his request, and IIRC, the final DPI was about 76. Yet even upon examining the print up-close, you could see the individual breather-holes in his jersey, etc. Of course some of this could have been the Lab's Uprezzing algorithm, but the photo was crystal clear.

Oh, and this was with my 6MP (3000x2000) Nikon D70

 dad_of_four's gear list:dad_of_four's gear list
Nikon D70 Nikon D5000 Nikon D2Xs Nikon D700 Nikon AF-S DX Nikkor 18-70mm f/3.5-4.5G ED-IF +10 more
MASTERPPA Contributing Member • Posts: 867
Re: Do pixels matter? What 20" x 30" prints reveal.

Hello, the 1DII and 1DIII were very sharp for 8-10mp. I own all 5DIII now. BUT, something happened this week, that NEVER happened with the 1DII and 1DIII. A client said their massive image was TOO SHARP (and their was almost no sharpening applied.)  All my 16x20 prints, and the few 20x24, are A LOT cleaner, and sharper then my 1DII, 1DIII and ESP my old 30D.

At ISO 100, the 1DII and 1DIII were great. At 400, so so, at 800-1600, UGH (esp the 1DIII)

LaszloBencze wrote:

I currently have a show of 38 20" x 30" prints appearing at the Viewpoint Photographic Art Center in Sacramento. Most of the pictures were taken with either a 5D MKII or 5D MKIII. However, some of them come from the Canon 1D MKII and 1DMIII which are not full frame and have about half the number of pixels of the 5D cameras (8 & 10 megapixels vs over 20 for the 5Ds).

What I noticed is that I could not tell which pictures were taken with the lower pixel count 1D cameras. They do not stand out as obviously inferior. In fact all the pictures look good and are indistinguishable in terms of sharpness or resolution.

Now I'm sure that there are benefits of the 20 megapixel cameras. And I do own two of them. But my point is that real life photo situations with large areas of bland texture or out of focus areas do not reveal such differences. This is a bit of a surprise for me but a reassuring one.

If you happen to live in the Sacramento area, you're welcome to look for yourself. The gallery is at 2015 J street and is well identified with signage.

MASTERPPA Contributing Member • Posts: 867
Re: My findings are...

No one can on the same sized print.  (within reason on the print.

But on a 16x20, I can tell the differance between my 1DIII and my 5DIII very easy. And my old 20x24 on my 1DII looks horrible compared to the same type of print from my 5DIII.

If anyone thinks that they could ever tell the difference between my 1Dx and my 5D3 by spotting the extra 4mp from prints of any size, I would bet my pants they're mistaken.

evoprox
evoprox Senior Member • Posts: 1,469
Re: Do pixels matter? What 20" x 30" prints reveal.

Of course they matter but a lot depends on the subject. Coming from 30+ years of film, mostly B&W I'm usually more concerned about DR and tones and spend most of my PP time in that department. YMMV.

 evoprox's gear list:evoprox's gear list
Fujifilm X10 Sony Alpha NEX-7 Fujifilm X-Pro1 Sony a7 II
XeroJay
XeroJay Senior Member • Posts: 2,015
Re: My findings are...

No one can on the same sized print.  (within reason on the print.

But on a 16x20, I can tell the differance between my 1DIII and my 5DIII very easy. And my old 20x24 on my 1DII looks horrible compared to the same type of print from my 5DIII.

If anyone thinks that they could ever tell the difference between my 1Dx and my 5D3 by spotting the extra 4mp from prints of any size, I would bet my pants they're mistaken.

sure, but there's the crop factor there that's playing a bigger role. I'll bet if you cropped the 5d3 image to the same magnification as the 1d3, then you would be hard pressed to tell any difference, at any price to size.

-- hide signature --

If there's typos, you can be sure I used my iPhone.

 XeroJay's gear list:XeroJay's gear list
Canon EOS M6 Sony a7 III Canon EF 100mm F2.8L Macro IS USM Canon EF-M 22mm f/2 STM Sigma 20mm F1.4 DG HSM Art +4 more
MASTERPPA Contributing Member • Posts: 867
Re: My findings are...

Being that I owned both, and shot hundreds of thousand of images with them, cropping the 5D3 still looks better then the 1DIII. The 5DIII still has more resolution. At ISO 100 its harder to tell, but at ISO 800 and over, MAJOR.

I shot about 120K with my 1DIII, and over 200K with my 1DII. I am about at 150K between my 5DII(which I sold) and my 5DIIs.

On a 8x10 print. NO DIFFERENCE. On a 16x20, you see it. Anything bigger, you really see it..

sure, but there's the crop factor there that's playing a bigger role. I'll bet if you cropped the 5d3 image to the same magnification as the 1d3, then you would be hard pressed to tell any difference, at any price to size.

rwbaron Forum Pro • Posts: 14,031
Re: My findings are...

MASTERPPA wrote:

Being that I owned both, and shot hundreds of thousand of images with them, cropping the 5D3 still looks better then the 1DIII. The 5DIII still has more resolution. At ISO 100 its harder to tell, but at ISO 800 and over, MAJOR.

I shot about 120K with my 1DIII, and over 200K with my 1DII. I am about at 150K between my 5DII(which I sold) and my 5DIIs.

On a 8x10 print. NO DIFFERENCE. On a 16x20, you see it. Anything bigger, you really see it..

sure, but there's the crop factor there that's playing a bigger role. I'll bet if you cropped the 5d3 image to the same magnification as the 1d3, then you would be hard pressed to tell any difference, at any price to size.

Without knowing how you post process and print it's difficult to comment.  Are you trying to up res files from the 1D3 when printing large and if so by how much and by what method?

Bob

-- hide signature --
 rwbaron's gear list:rwbaron's gear list
Fujifilm X-T2 Fujifilm X-T20 Fujifilm XF 18-55mm F2.8-4 R LM OIS Fujifilm XF 55-200mm F3.5-4.8 R LM OIS Fujifilm XF 10-24mm F4 R OIS +8 more
Keyboard shortcuts:
FForum MMy threads