photo enlargement for large format

[bob]

...The enlargement will most likely disappoint and be poor in quality. I use 400% as the limit for enlargement of a bitmap as a rule of thumb before pixelization and visible defects become apparent. It will, of course, depend on the size of the image before any enlargement as well as the file's condition.

If you merely grab a corner and pull then 400% is probably a pretty good number. But if you use a spline fit algorithm, which are legion, you can expand to incredible sizes with no pixelization whatsoever. You will end up with some small pixel bloom and other annoying artifacts but in scale with the rest of the image they are small and not particularly noticeable.

I've been using PhotoZoom Pro and find it completely satisfactory for expanding reasonably sharp images to ridiculous sizes. There are many other proprietary products using the spline fit algorithm, I just happen to like this one. Apparently Corel likes it as well since they integrated the PhotoZoom engine into PhotoPaint X6 and I assume beyond..

 

[Signout]

Resize

We have used On1 software for resizing photos for years - the plug in used to be called genuine fractals, they now just call it resize. If I remember correctly they now can resize an image (original size) up to 1000% without any loss of detail.
https://www.on1.com/apps/on1photo10/

 

[rossmosh]

OK so I got pixel dimension and dpi mixed up.

I still go by megabytes. It is easy. A 1 meg files is not so good, and 15 meg file better, a 100 meg file better yet.

Except for the fact I can easily make a 100mb file that is stupid high resolution that would look like absolute trash if printed. People know to look at the quality of the image. They just don't consider it part of the evaluation process despite it being one of the most important parts.

We have used On1 software for resizing photos for years - the plug in used to be called genuine fractals, they now just call it resize. If I remember correctly they now can resize an image (original size) up to 1000% without any loss of detail.
https://www.on1.com/apps/on1photo10/

I have no doubts that that software does a very good job, even fantastic job, increasing the resolution on images. Anyone making claims that their software can take an image, res it up 1000%, and it will be the same quality, is telling a lie. Could it be 95% the same quality? Maybe. But saying no loss is just nonsense. It's like saying going from RAW to JPEG shows no loss. We all know it shows some loss. Is it good enough? More often than not, yes.

 

[player]

Except for the fact I can easily make a 100mb file that is stupid high resolution that would look like absolute trash if printed. People know to look at the quality of the image. They just don't consider it part of the evaluation process despite it being one of the most important parts.

I have no idea why you are arguing here. Any idiot knows I am not talking about interpolating a .5 meg file up to 100 megs. That's just so obvious.

 

[rossmosh]

I have no idea why you are arguing here. Any idiot knows I am not talking about interpolating a .5 meg file up to 100 megs. That's just so obvious.

Did you read this thread?

Also, quality is very subjective. What I think is good enough may not be by your standards. At this point, I've gotten enough oversized garbage files to know not to simply tell people to go based on file size.

 

[player]

Did you read this thread?

Also, quality is very subjective. What I think is good enough may not be by your standards. At this point, I've gotten enough oversized garbage files to know not to simply tell people to go based on file size.

Yes. You're right. I am usually creating. I was referring more to small files. If they are only a meg it's a sign it won't enlarge much.

 

[bob]

...Anyone making claims that their software can take an image, res it up 1000%, and it will be the same quality, is telling a lie...

Not quite. There are algorithms that do exactly that. The quality is the same plus or minus but the problem is that the no one has ever seen the image at the expanded size. There's not a whole hell of a lot of difference between expanding and image with one of these spline fit algorithms and looking at the original with an equivalent power magnifier. In other words, expand the original times 100 or look at the original through a 100 power lens. You'll see pretty much the same thing.

The image is the same it's just your view of it is new. And, of course, there's the matter of proper viewing distance. This being the diagonal of the work or greater. Anything closer than that is much like viewing the Mona Lisa from 2 inches away. All you'll see are brush strokes.