How many of you print large with APS-C? I mean LARGE!!!

[Pontoneer]

I routinely print to super A3 on our Epson SP4900 and have a few times sent work out to be printed to 6x4 ( that’s six feet by four feet ) for wall displays .

My images are also displayed on LCD and plasma screens ranging up to 80” and projected onto screens up to 20ft wide .

APSC works pretty well for me , hence I haven’t jumped to 35mm .

 

[Chris Malcolm]

With a 14MP APS-C DSLR I struggled to do a good sharp A3 print, although with care and good luck I did sometimes manage it, whereas it was quite easy to do a good sharp A4 print. I wondered about upgrading to a full frame camera. Then I got a 24MP APS-C camera. With the same lenses I was now often able to produce very sharp A3+ prints. I was sufficiently amazed and delighted that I actually bought an A3+ printer. A3+ is quite good enough for me. I've entirely stopped hankering after a full frame model.

 

[Tourlou]

Which model was the 14 Mpix camera? 14 MPix on a 400mm length gives you 5.72 line pairs per mm. Which is perfect if you look at the image from a 350mm distance and more. 24 mpix gives you 7.5 line pairs per mm, which is of from 275mm or more. No big difference. Was your previous camera a little soft by nature like the SONY SLT? Nikon used a 14mpix sensor on the d3100 I think and it wasn't exactly the best one on the market...

 

[Lobbamobba]

Ive printed about 2m high full color prints.

I used an APS-C Sigma and a Fuji XT2.

The prints from the Sigma looked great, even up close.
But it was stretching the Fuji X-T2s capabilities. Even if the Fuji had a much better lens.
It just didn't have the color depth to match the Foveon.

 

[Trebor1]

We're back to the 6 line pairs per degree of the subjective quality scale. It was true when Grainger came out with that in the 50s I beleive, and it's still true today! The eye perception doesn't change because technology can cram more pixels per surface unit on a wafer of silicone... Back to bacics... Contrast and color. Luminance first and then chrominance...

They're now making cell phones with 25MPix cameras... for looking at images on a 3 MDots screen the size of a card deck. I don't get it. I think someone big in the camera sensor industry has shares in a memory chip factory.

The 6 line pairs per degree was only achieved by increasing the viewing distance, from the normal standard (for close inspection of detail) of 25cm, to a value of 34 cm. In order to achieve consistency the central 6 line pairs per degree of the Atkins SQF curve should be multiplied by 35/24 = 8.16.

MTF - Modulation transfer Function Explained - Bob Atkins Photography

An explantaion of MTF (Modulation Transfer Fundtion) and SQF (Subjective Quality Facror). Measurements of the optical performance of a lens.

www.bobatkins.com

Since the original 1972 work by Granger and Cupery, there have been some slight revisions to the Contrast Sensitivity Function curve used and the peak value has risen from 6 to just below 8.

Acutance and SQF (Subjective Quality Factor) | Imatest

Another article on the Imatest website now appears somewhat out of date, written when 6.3 MP cameras were the norm and 12 MP was state of the art!

Sharpness

Sharpness is arguably the most important single image quality factor: it determines the amount of detail an image can convey. The image on the upper right illustrates the effects of reduced sharpness (from running Image Processing with one of the Gaussian filters set to 0.7 sigma). Device or...

www.imatest.com

Viewing distances are again greater than the old standard so a correction is also required in this instance to maintain consistency.

There are also recommendations from Zeiss, Norman Koren and others that differ from Atkins. Zeiss suggest a maximum resolution of 8 lp/mm, on the print in their 2008 How to Read MTF Curves by H.H Nasse and return to the standard 25 cm viewing distance for an A4 print (page 11). This corresponds to 1600 line pairs per Picture Height as the limit and this was also the standard used by Roger Hicks as mentioned in several of his books.

Loading…

www.lagunabeachbikini.com

There is obviously some 'wriggle room', between relatively ancient and more modern standards. :-D The Norman Koren results are again relatively old but in 2004 he was recommending 240 PPI on an Epson 1270 printer for best quality and 300 PPI when he moved to an Epson 2200. Today, the magic number may well be 360 PPI, with 240 PPI representing very good print quality? Over on the printing forum there are some who are even using 720 PPI!

Loading…

www.normankoren.com

Sharpness

Sharpness is arguably the most important single image quality factor: it determines the amount of detail an image can convey. The image on the upper right illustrates the effects of reduced sharpness (from running Image Processing with one of the Gaussian filters set to 0.7 sigma). Device or...

www.imatest.com

 

[The Davinator]

We're back to the 6 line pairs per degree of the subjective quality scale. It was true when Grainger came out with that in the 50s I beleive, and it's still true today! The eye perception doesn't change because technology can cram more pixels per surface unit on a wafer of silicone... Back to bacics... Contrast and color. Luminance first and then chrominance...

They're now making cell phones with 25MPix cameras... for looking at images on a 3 MDots screen the size of a card deck. I don't get it. I think someone big in the camera sensor industry has shares in a memory chip factory.

The 6 line pairs per degree was only achieved by increasing the viewing distance, from the normal standard (for close inspection of detail) of 25cm, to a value of 34 cm. In order to achieve consistency the central 6 line pairs per degree of the Atkins SQF curve should be multiplied by 35/24 = 8.16.

http://www.bobatkins.com/photography/technical/mtf/mtf4.html

Since the original 1972 work by Granger and Cupery, there have been some slight revisions to the Contrast Sensitivity Function curve used and the peak value has risen from 6 to just below 8.

http://www.imatest.com/docs/sqf/

Another article on the Imatest website now appears somewhat out of date, written when 6.3 MP cameras were the norm and 12 MP was state of the art!

http://www.imatest.com/docs/sharpness/

Viewing distances are again greater than the old standard so a correction is also required in this instance to maintain consistency.

There are also recommendations from Zeiss, Norman Koren and others that differ from Atkins. Zeiss suggest a maximum resolution of 8 lp/mm, on the print in their 2008 How to Read MTF Curves by H.H Nasse and return to the standard 25 cm viewing distance for an A4 print (page 11). This corresponds to 1600 line pairs per Picture Height as the limit and this was also the standard used by Roger Hicks as mentioned in several of his books.

http://www.lagunabeachbikini.com/docs/papers/CLN_MTF_Kurven_EN.pdf

There is obviously some 'wriggle room', between relatively ancient and more modern standards. :-D The Norman Koren results are again relatively old but in 2004 he was recommending 240 PPI on an Epson 1270 printer for best quality and 300 PPI when he moved to an Epson 2200. Today, the magic number may well be 360 PPI, with 240 PPI representing very good print quality? Over on the printing forum there are some who are even using 720 PPI!

http://www.normankoren.com/Tutorials/MTF3.html

http://www.imatest.com/docs/sharpness/

For large prints...say over 30", 240ppi is more than enough. That said, when I am working with b&w scans from 4x5 Delta 100, for large prints, I have 16bit scans done at 4000ppi on an Aztek. I then output to 24"x30" at 720ppi using the carbon piezographic system on an Epson.

 

[Gato Amarillo]

In the past I often printed 16x20 and larger from 4/3 chip cameras. I have printed as large as 40 inches from small chip cameras.

Currently I own and use 1-inch, m4/3 and full frame (though I rarely use the FF). I print more for gallery exhibit now and stick to 12x16 because that's the largest pigment printer I can afford. I like the control of printing at home.

I have seen my prints in galleries alongside prints from full frame and probably medium format. No one has ever said anything about the difference -- except photographer friends who shoot larger sensors and know what I use. The want to know how I do it.

(Contrary to what some want to believe, viewers in galleries do move in close to check detail in large prints. It's a wonder some of them don't leave nose prints on the glass.)

As I have seen people move in closer to prints in galleries around the world from the top landscape photographers, don't proclaim it doesn't happen. I see it all the time. No one is talking about nose against glass. Don't be obtuse. They most certainly get within a couple feet as opposed to standing wait back like you for some reason wish to claim.

In my opinion pixel count is at least as important as sensor size when striving for detail in prints. At least from my experience, m4/3 at 20MP to FF at 24MP is not a huge difference, in part I think because of the greater depth of field with the smaller sensor. But 20MP 4/3 to 36MP FF can be significant -- if you can afford the best lenses and have good technique in both shooting and processing.

Gato

But like you said...no one in galleries gets close...despite me seeing otherwise.

What I said was people DO get close -- often very close.

 

[The Davinator]

Well...I certainly misread what you posted. My apologies.

 

[Trebor1]

There is obviously some 'wriggle room', between relatively ancient and more modern standards. :-D The Norman Koren results are again relatively old but in 2004 he was recommending 240 PPI on an Epson 1270 printer for best quality and 300 PPI when he moved to an Epson 2200. Today, the magic number may well be 360 PPI, with 240 PPI representing very good print quality? Over on the printing forum there are some who are even using 720 PPI!

http://www.normankoren.com/Tutorials/MTF3.html

http://www.imatest.com/docs/sharpness/

For large prints...say over 30", 240ppi is more than enough. That said, when I am working with b&w scans from 4x5 Delta 100, for large prints, I have 16bit scans done at 4000ppi on an Aztek. I then output to 24"x30" at 720ppi using the carbon piezographic system on an Epson.

That's 285.7 MP by my calculation, assuming an actual image area of 96 x 120mm. Enough to provoke severe Megapixel envy, although the cost per scanned shot - £25 + P&P here in the UK, for b&w (Tim Parkin), might prove to be a deterrent.

But can you charge at least twice as much for a print, compared to one captured with a Canon 5DS camera - only 241 PPI uninterpolated, at this image size, or 229 PPI for a Nikon D850?

 

[Don_D]

2) What is the largest print you have EVER made with a crop sensor camera and were you happy with the result?

Please do not include any data from a sensor larger than APS-C.

Well if you mean REALLY large, I once licensed an image made with a Fujifilm Fine Pix J150W camera...1/2.3 " sensor, 10mp and it was made into a roadside billboard about 10 x 20 feet.

It looked pretty good from a normal viewing distance.

I apologize that this doesn't address your crop sensor question directly.

--
Don

 

[Osvaldo Cristo]

If the original image was well done with no blurring, well exposed at base ISO (or closer) and you don't need to crop, usually I can get good print out up to 50cm by 30cm and with some intelligent postprocessing, I have had good lithographic printing using either Heidelberg or HP Indigo presses up to 70cm by 50 cm using my "low count pixel" Nikon D200 and D300S cameras (10 and 12Mpixel).

My own experience is the key for the quality is in the quality of the original image. Most very acceptable quality image when watched at a small size like a phone screen, are terrible when printed even at modest sizes like A4.

All the best.

 

[DigitalMaven]

I made a post on a similar topic on this site over 10 years ago which I believe
is still germane today. See:

http://forums.dpreview.com/forums/read.asp?forum=1018&message=19939667

I hope that helps with this discussion.

 

[Trebor1]

I made a post on a similar topic on this site over 10 years ago which I believe
is still germane today. See:

http://forums.dpreview.com/forums/read.asp?forum=1018&message=19939667

I hope that helps with this discussion.

 

[Gato Amarillo]

Well...I certainly misread what you posted. My apologies.

Apology accepted. No harm done.

 

[DigitalMaven]

The Diagonal of a image's size as rule of thumb for a image's typical viewing distance
predates digital photography, so I would hardly characterize it as bizarre.

The resolutions you cite, 240 - 400 PPI, are suitable for a small image, say 8"x10",
being viewed at a short distance (8"- 13") but not for banner sized images
(5' x 5' or larger) being viewed at a distance where the PPI could go as low
as 100.

--
When a hammer is your only tool, all problems begin to look like nails.

 

[Trebor1]

The Diagonal of a image's size as rule of thumb for a image's typical viewing distance
predates digital photography, so I would hardly characterize it as bizarre.

The resolutions you cite, 240 - 400 PPI, are suitable for a small image, say 8"x10",
being viewed at a short distance (8"- 13") but not for banner sized images
(5' x 5' or larger) being viewed at a distance where the PPI could go as low
as 100.

--
When a hammer is your only tool, all problems begin to look like nails.

Perhaps bizarre is too strong a term, it would be more accurate to ask: does it have a good scientific basis? The answer appears to be not really, or it depends.