largest print out for your full frame

[EinstStein]

What is the largest print out that has diminishing return in image quality?
Which camera is it from? Is this from 35mm full frame or medium format?
Oh, forget about stitched or composite pictures such as works through gigapan.

[Christoph C. Feldhaim]

It depends a lot on your viewing distance.
It is said 300 ppi viewed from 25 cm is very high quality and 200 ppi still acceptable.
This would mean for a 24 MP image you could do a

13.3 x 20 inch print at 300 ppi or

20 x 30 inch print at 200 ppi

and still watch these from 25 cm.

If you double the viewing distance you need only a quarter of the MP for the same impression of quality.

Do the math or better - see for yourself ...

Cheers
~Chris

[ErikKaffehr]

Hi,

I made an experiment recently, making two cropped prints from an 24 MP DSLR and 39 MP MFD. The prints corresponded to 86 cm height. Very little difference could be seen, but with a slight advantage to the MFD. The difference was hard to see, unless you have identified certain details. I would also add that we were viewing at 30-40 cm. I guess that 70x100 cm is no problem, going larger is OK probably, but I have no experience.

I could add that I have 70x100 print made from an 10 MP DSLR on my wall. It is absolutely OK if you view it from 80 cm distance, but viewed close it falls apart.

Finally, it also depends on your sharpening and your subject.

Best regards
Erik

What is the largest print out that has diminishing return in image quality?
Which camera is it from? Is this from 35mm full frame or medium format?
Oh, forget about stitched or composite pictures such as works through gigapan.

[langier]

Since going to digital, I've learned to ignore the math on large prints. In my studio is a real nice 45x30 inch print from probably my D300 (12mp). My partner has a photo in the Museum of Monterey that is a similar size but shot with a Nikon D200 (10mp). I've got a show in the National Steinbeck Center with beautiful prints from cameras as low-end as in iPhone to the upper end taken with the D800 and 5DIII.

Last year at the National Steinbeck Center Museum my colleague had a lovely photo from his iPad. That made a nice 24x36 photo and we had to go larger than that to be able to crop to 2:3 and there were many other examples of cameras many would not touch because they weren't sexy nor new enough nor had enough MP.

In our current show there, we have quite a few shots taken with an iPhone 4s in addition to photos from the 5DII, 5DIII, D700, D300, D800, Konica-Minolta 7000, probably D100 and D200 and a whole bunch of other cameras we've used along the way. Oh, and some were taken at ISO 3200, ISO 6400 and even some of those were cropped to fit the show.

I've seen too many fuzzy photos that lack concept from very expensive cameras and jewels that nailed the photographer's vision from very low end devices. I prefer the latter.

Bottom line isn't the MP, full-frame, camera, lens, etc., it's those tools in capable hands crafting the well-visualized photograph that matters here.

After our opening the other night, we took along the executive director and curator of the museum along. We asked either if they could determine which photo was taken by which camera. Neither could. We made our point and nobody will be the wiser.

[Christoph C. Feldhaim]

Bottom line isn't the MP, full-frame, camera, lens, etc., it's those tools in capable hands crafting the well-visualized photograph that matters here.

Well ... this is the answer for the grand scheme of things and maybe the best statement one can get out of this thread.
I considered the OP question to be on a much narrower scale.
But maybe its often better to stand a step back and see from there.

Cheers
~Chris

[ErikKaffehr]

Hi,

I won't argue with that!

I have an A2 size print from Iceland on my wall, it was shot with a 6MP DSLR. It lacks resolution, but it is a great image. That image also needs a viewing distance, however.

I have "graduated" from 6MP APS-C to 39MP MFD (Phase One P45+), but I don't feel I made my best images with the P45+ or the 24 MP DSLRs I own, 12 MP was just fine.

Best regards
Erik

Since going to digital, I've learned to ignore the math on large prints. In my studio is a real nice 45x30 inch print from probably my D300 (12mp). My partner has a photo in the Museum of Monterey that is a similar size but shot with a Nikon D200 (10mp). I've got a show in the National Steinbeck Center with beautiful prints from cameras as low-end as in iPhone to the upper end taken with the D800 and 5DIII.

Last year at the National Steinbeck Center Museum my colleague had a lovely photo from his iPad. That made a nice 24x36 photo and we had to go larger than that to be able to crop to 2:3 and there were many other examples of cameras many would not touch because they weren't sexy nor new enough nor had enough MP.

In our current show there, we have quite a few shots taken with an iPhone 4s in addition to photos from the 5DII, 5DIII, D700, D300, D800, Konica-Minolta 7000, probably D100 and D200 and a whole bunch of other cameras we've used along the way. Oh, and some were taken at ISO 3200, ISO 6400 and even some of those were cropped to fit the show.

I've seen too many fuzzy photos that lack concept from very expensive cameras and jewels that nailed the photographer's vision from very low end devices. I prefer the latter.

Bottom line isn't the MP, full-frame, camera, lens, etc., it's those tools in capable hands crafting the well-visualized photograph that matters here.

After our opening the other night, we took along the executive director and curator of the museum along. We asked either if they could determine which photo was taken by which camera. Neither could. We made our point and nobody will be the wiser.

[Ellis Vener]

So far the largest high quality print I've had made (too big for my Canon iPF6300) from a D800, without going through heavy  interpolation heroics  is 30 x 45 inche. There is enough real world detail along with a lack of pixelization and interpolation artifacts that can stand very close up scrutiny and the detail still look continuously smooth.

It was carefully shot however: with a great lens(Nikon AF-S 105mm f/2.8G VR Micro-Nikkor at an aperture in the lens' optimum range)  with the camera in live view mode, remotely triggered, mounted on a heavy tripod, and the subject lit with flash.

[Telecaster]

The only full frame electronic format I currently use is Micro Four-Thirds. And it is indeed "full frame" in that the lenses are matched to the sensor dimensions.     My largest print sizes normally run 20" in the longest dimension, though up to 45" or so for stitched panos on roll paper. Don't do much stitching these days.

The largest sensor I use is the 33x44mm chip in my Pentax 645D. Same print dimensions as above. The largest exposure medium I use is 4x5" sheet film...it's been awhile, though, since I loaded up the Wista.

-Dave-

[BernardLanguillier]

The largest single frame print I have made is about 1m high, it is from a D3x file printed on Lyve canvas in B&W and is arguably very nice.

As far as panos go, I have a few 2m wide, detail is staggering (files being 300mp and up). These are some of the most immersive photographs I have seen, the print generates such a gravitational effect towards it that I normally forbid pregnant woman from getting too close.

Cheers,
Bernard

[tim wolcott]

Well in the old days we always used the standard of apparent viewing distance.  That basically means the closest your eyes can focus.  Thats how all should be looked at.  I have made 210 inch prints with that in mind but they are shot with the Phase One system and 9 images stitched together at 40 megapixel each.  I believe its 1.2 gigabytes.

I'm making a 168 inch pano on a hand made cherry folding screen.  Its going to be awesome. 

Just shot some pano stitches in colorado that are 2.4 gigs.  should be 300x 40 but how can you move something that big.  My limits is how big I can get one great piece of wood.  Unless its a folding screen.  Tim

[Bart_van_der_Wolf]

What is the largest print out that has diminishing return in image quality?

Hi,

Depends on your definition of 'quality' and the level you find acceptable.

Creative content aside, quality is usually expressed as resolution. Resolution is a quantifiable metric, but contrast is also part of the experience.

When resolution of the human eye is the guiding principle, in concert with required capture and output resolution, then you can use the first section of my DOF output quality planning tool.

Which camera is it from? Is this from 35mm full frame or medium format?

That's relatively unimportant, as long as it meets the optical and sensel resolution requirements for output that you adopt. Of course some platforms are better suited for some subjects than others (I would not necessarily shoot sports with a MFDB).

Cheers,
Bart

[wolfnowl]

This might be of interest: http://theonlinephotographer.typepad.com/the_online_photographer/2013/11/in-search-of-a-vocabulary-for-image-quality.html

Mike.

[EinstStein]

I'm asking what's the largest print out for YOUR full frame.
What's the largest print out YOU think has diminishing return?
Which YOUR camera is it from?
Whats' YOUR opinion?
What's the definition of "quality"? use YOUR definition.

[Bart_van_der_Wolf]

I'm asking what's the largest print out for YOUR full frame.

Hi,

Same question, same answer. It depends. I've witnessed people stopping in their pace, walking through the aisles of a super-yacht exhibit, when confronted with a booth wall (sized 2.5 x 3.5 metres, ) action image of a super-yacht crossing a narrow sea passage with recognizable people on board, shot from a distance with a 500mm lens on a 36x24mm sensor (same as my Canon EOS-1Ds Mark III). Although the viewing distance helped, I must admit that even from within the booth, sitting at 1 metre from the image, it was still pleasant to look at the details (which obviously were not pin sharp anymore, but not objectably so).

A lot depends on raw-conversion and post-processing and output sharpening quality, assuming the Capture technique + Capture sharpening is as should be.

Cheers,
Bart

[EinstStein]

In another thread, someone posted the link to your tool for calculating the print size. It's an interesting tool.


I think you have a very basic mistake. You assume every pixel is equivalent from camera to camera?
You tool shows Leica C can produce larger print out than Leica M8.2. Really? Where did you get that idea?   

[Bart_van_der_Wolf]

In another thread, someone posted the link to your tool for calculating the print size. It's an interesting tool.


I think you have a very basic mistake. You assume every pixel is equivalent from camera to camera?
You tool shows Leica C can produce larger print out than Leica M8.2. Really? Where did you get that idea?   

Hi,

The tool uses human visual acuity as the limit of resolvable detail. Pixels are not the basis. Given that (angular resolution) limit, because it is also influenced by contrast and illumination level. Those are not added to the equation, I assume optimal viewing conditions. Under worse conditions the resolution will be even lower and we could magnify more (as long as viewing conditions cannot become better).

How we sample the original scene, i.e. which lens and sensor we use is relatively unimportant for the detail that we cannot resolve by eye anyway. When detail is smaller than we can resolve by eye, its quality is not relevant. However, for the detail that we can resolve by eye, the pixel quality does matter. But also output sharpening matters, in fact the MTF curve of the entire optical chain (lens, sensor, Raw conversion, post-processing, output modality, viewing conditions, eye) is what will determine the visual impression.

The choice of sensor (amongst others) will therefore indeed make a difference, but only for the detail that we can ultimately resolve by eye.

The sensor dimensions and sampling density also do not factor in lens quality, but they are simply used to set a physical limit to the COC that can be used. A COC smaller than the sensel will not change the DOF, the sensel size dictates (and assumes a large fill-factor). The same applies to diffraction, when diffraction becomes an absolute limit for high spatial frequency detail (=low contrast) the COC will be adjusted even if the sensor could resolve more detail (detail is missing in the sensor's input signal).

So there are many more factors that come into play, but once we can no longer resolve details by eye they become less relevant. 'Sensel quality' also equates to a different Capture sharpening requirement (as outlined in this thread), which will even out some of the differences because Capture sharpening should restore the original input signal (in the original scene) as much as possible.

Cheers,
Bart

[xpatUSA]

 . . . then you can use the first section of my DOF output quality planning tool . . .

Hello Bart,

OT, but I can't open the link in FireFox nor Safari even after resetting the Hughes satellite modem. Permanent 'DNS lookup error'. Any problems on your site today (18 Nov 2013)?

Ted

[Bart_van_der_Wolf]

OT, but I can't open the link in FireFox nor Safari even after resetting the Hughes satellite modem. Permanent 'DNS lookup error'. Any problems on your site today (18 Nov 2013)?

Hi Ted,

Just tried it with Firefox 25.0 and 25.0.1, no problem. MS Internet Explorer 11, and Google Chrome (version 31.0.1650.57) also open it without issues. My service provider reports no issues.

Cheers,
Bart

[jjj]

It depends a lot on your viewing distance.
It is said 300 ppi viewed from 25 cm is very high quality and 200 ppi still acceptable.
This would mean for a 24 MP image you could do a

13.3 x 20 inch print at 300 ppi or

20 x 30 inch print at 200 ppi

and still watch these from 25 cm.

Why on earth would you view a big print from that distance?

I considered the OP question to be on a much narrower scale.
But maybe its often better to stand a step back and see from there.

Now that's better. 

Big advertising posters are often 'abysmal quality' when viewed close, but yet when viewed at their intended distance look great. i.e. a lot further away than a mere 25cm. It's medium size prints or double page spreads in magazines that actually need to be the highest quality. Usually as they are the one ones that fill the widest field of ones view in reality.

[Christoph C. Feldhaim]

Why on earth would you view a big print from that distance?
Now that's better. 

Big advertising posters are often 'abysmal quality' when viewed close, but yet when viewed at their intended distance look great. i.e. a lot further away than a mere 25cm. It's medium size prints or double page spreads in magazines that actually need to be the highest quality. Usually as they are the one ones that fill the widest field of ones view in reality.

And now what?