About picture re-sizing....

[Carmen Gauvin-O'Donnell]

Hey folks! This is just out of curiosity...

I understand why we'd want to *resize* pictures for sending, but the reason
I hesitate to do it when I'm asked is because I wonder how it will affect
print quality should the person at the other end wish to print them, in the
case of family pictures or something like that.

How does it affect print quality? And if it does, what would be a good
resize, then, in the case of a picture that's about 1 meg in size originally
(which is what I set my camera at since I have a card with lots of room on
it) if I'm assuming they'd want to print it out at 4 x 6 inches?

Any ideas?

Thanks in advance!

C.

 

[Ron]

I would not resize a 1 MB file if I KNEW the person the other end wanted
to print it. However - in my experience relatively few people bother to
print e-mailed pictures or even their own. Why bother when they look so
much better on screen?

Ron

 

[J.S. Jackson]

The minimum dpi for crisp, clear printing of a photograph is 300. To print
a 4 x 6 photo at 300 dpi, it will need to have a resolution of 1200 x 1800
pixels. Hope that helps.

 

[Carmen Gauvin-O'Donnell]

Thanks!
C.

The minimum dpi for crisp, clear printing of a photograph is 300. To
print a 4 x 6 photo at 300 dpi, it will need to have a resolution of 1200
x 1800 pixels. Hope that helps.

 

[Michael J. Mahon]

The minimum dpi for crisp, clear printing of a photograph is 300. To
print a 4 x 6 photo at 300 dpi, it will need to have a resolution of
1200 x 1800 pixels. Hope that helps.

Actually, after encountering a host of confusing and mutually
contradictory pronouncements on printer resolution requirements,
I did some experiments on printer resolution which I recommend
to others who are curious about this.

I took a high resolution photo containing fine detail and downsampled
it in Photoshop to create a set of files with DPIs differing by about
20%, covering the range of 300DPI down to 100DPI in seven or eight
files.

I then printed each of the files at the same size, so that the actual
DPI of each print varied as designed. I wrote the DPI value on the
back of each print in a coded form.

Then I shuffled the pack of prints and gave them to some critical
viewers and asked them to sort them by resolution.

After several trials with several viewers, it was clear that the
sorts were in DPI order up to a certain value, after which they
were essentially in random order. The highest dependably ranked
DPI then represents the *actual* DPI capability of the printer
and paper being used.

I did this test a few years ago with a good photo printer and
found that the maximum discernable DPI was about 180-200DPI.

I recommend that you try this with your printer and see what
its practical DPI limit is.

BTW, I have never seen any publication of printer resolution tests
using a method as objective and accurate as this one.

This method could also be used to settle all kinds of questions
about the retention of image quality.

The principle is simple: if you can't reliably see a difference,
then there isn't any.

-michael

"The wastebasket is our most important design
tool--and it's seriously underused."