[TonyW]

Left is pixellated, right is soft.

Yes, but what I was looking for was opinion on what you would prefer to see close up to a print. Perhaps more information to put this into some context.

The image is a scan of two real prints made by me at the same time with the same paper etc. side by side from two different applications.

The area you are looking at is a crop from the original image.

I wanted to really test the algorithms of the two applications and set to resize to 11,667 x 14125 pixels, a pretty big enlargement yielding a print of 38.9" x 47"

Viewing at a 'normal' distance considering the print size I cannot see any difference. Getting closer differences start to become a little clearer and at photographers viewing distance what you observe here is what I see. I do actually have a preference as I get a feeling for more information in one over another at close range but I was seeking others views.

---------- Post added 11-03-17 at 04:15 PM ----------

Are the "jaggies" present on the left-side print or are they scan artifact?

Steve

Not scan artefacts. Jaggies present on the left side print and viewable at 12", pretty much irrelevant after 24"

[stevebrot]

Yes, but what I was looking for was opinion on what you would prefer to see close up to a print. Perhaps more information to put this into some context.

The image is a scan of two real prints made by me at the same time with the same paper etc. side by side from two different applications.

The area you are looking at is a crop from the original image.

I wanted to really test the algorithms of the two applications and set to resize to 11,667 x 14125 pixels, a pretty big enlargement yielding a print of 38.9" x 47"

Viewing at a 'normal' distance considering the print size I cannot see any difference. Getting closer differences start to become a little clearer and at photographers viewing distance what you observe here is what I see. I do actually have a preference as I get a feeling for more information in one over another at close range but I was seeking others views.

---------- Post added 11-03-17 at 04:15 PM ----------

Not scan artefacts. Jaggies present on the left side print and viewable at 12", pretty much irrelevant after 24"

My sense of aesthetic considers those to be bad processing artifact and unacceptable. They are present on the right hand example too, just not as bad.


Steve

[TonyW]

My sense of aesthetic considers those to be bad processing artifact and unacceptable. They are present on the right hand example too, just not as bad.


Steve

The question has nothing to do with aesthetic considerations and solely to be taken as a comparison of a massive enlargement via two different applications.

The full image consisted of only 1398 x 1745 pixels (original downscaled by a factor of 4 I believe before I got hold of it). This image would only yield a print of 4.66" x 5.82" at 300 ppi and be very acceptable.

What you are seeing in the presented crop is an image area measuring a mere 5" x 8" from a print that would measure 38" x 47". A 'correct' viewing distance for this size would be 7' 5" so really little wonder that you are seeing artefacts.

[RGlasel]

IrfanView is a free image utility program that offers six different resampling filters for enlarging. Hopefully the OP exports the developed RAW image as a TIFF file, which can be resampled by IrfanView, in batch mode if desired.

[biz-engineer]

Left is pixellated, right is soft.

When I use the K3, I liked the idea of upsizing 24Mp images to 51Mpixels because it instantly felt like I was shooting 645z without having spent any money.
But the 645z feeling vanished fairly quickly when I looked at the images and, when I saw how much more size files were taking in memory and on hard drive.

[Bob 256]

Well, that's an interesting opinion.



I guess you didn't read this part of the question.

Sorry - my bad. I should have used the word "answered" rather than "related" and that was an observation - not an opinion. I didn't mean to imply there wasn't good info provided in the following posts - just that the OPs question about the Pentax software hadn't been addressed (which it has in subsequent posts).
All the other information is very pertinent to the OP's issue of up sizing.

[TonyW]

Why would you want to look "close up to a print"? I've seen people do that with paintings in the Musee d' Orsay but I'm pretty sure that misses the point.

ALL photographers do, it is a perfectly natural thing for them. The distance limited by either the size of their belly or length of their nose :-)

A ‘correct’ viewing distance is fine and the theory holds good in practice. However if you cannot physically limit the viewing distance or you put a very large print with lots of detail in a small room the shortfall in IQ will be noticeable. You only need to look at the image posted here. Sit at screen viewing distance the images look poor (naturally) move away from the screen a couple of feet further not too bad - bear in mind that a correct viewing distance for this image 7 foot.

[beholder3]

I suggest reading this thread here including the links given:
resizing algorithms - XnView Software
The whole topic is way more complex than most people think.

Lanczos seems the best option. XnView offers:

• Lanczos (3)
• next neighbor
• bilinear
• hermite
• Gaussian
• Bell
• Bspline
• Mitchell
• Hannig
• Cubic

Feel free to find the best for your specific case.

Just from Lanczos there are five subversions.

[TonyW]

...The whole topic is way more complex than most people think.....

, Yes 100%

I would suggest that for anyone with a little doubt about differences just read one example and move the mouse over the cat images to see what result can be expected from various algorithms -
Optimizing Digital Photo Enlargment

---------- Post added 11-04-17 at 06:29 AM ----------

Weird, I don't stick my nose on my Kindle, TV, magazine covers or books with photographs. My wife did want a 60"x40" canvas of a photograph I made of the Eiffel Tower. Printer only wanted 150 PPI. Hangs over a stairwell. Viewing distance is about 16 ft. Visitors will stand on the steps near a corner and remark on the detail. Compared to a 300 PPI paper print, its coarse but the point is, it puts you there, in Paris where most haven't been or makes the memory of those who have, vivid. Which is why my wife wanted it. Conversely, I was underwhelmed at an Ansel Adams exhibit. With few exceptions, the prints had much less impact than expected.

Well, why would you stick your nose to kindle,tv or magazine or photo books - they have already been designed for best view at an average viewing distance for someone with 20/20 vision so handholding maybe somewhere around 12" requiring only around 286 PPI to give the impression of continuous tone

Viewing a print at 16 ft only requires around 20 PPI native file resolution to give the appearance of contone anything above prior to interpolation of benefit. Viewing your 60x40 from 3 ft only requires 96 ppi, but...

Canvas prints can get away with a native resolution much less than 150 PPI - due to the structure of the base they are just not able to resolve fine detail period compared to photo paper 100 PPI native resolution is just fine however as previously stated:

Regardless of what your printer requested your image would be resampled by the print driver to the printers require ppi which would be most likely 300 or 360 ppi depending on the actual printer, (unlikely to make much impact at all relating to canvas)

Ansel Adams never made digital prints and has not made analogue prints for over 35 years since his death in 1982. If the prints did not impress you then maybe you were not looking at original prints made by Adams or just poorly executed stuff or simply you did not like the images?


---------- Post added 11-04-17 at 06:29 AM ----------

[TonyW]

Non forsit.