Yahoo Groups archive

Digital BW, The Print

Index last updated: 2026-04-28 22:56 UTC

Message

Re: [Digital BW] Sharpening vs. Print size [was: Image Density vs. Print Size]

2004-04-13 by Steve Kale

Broadly consistent with the Photokit guys.  Capture sharpening, Creative
sharpening (if desired) and then Output Sharpening.


From: hogarth <hogarth@...>
Reply-To: DigitalBlackandWhiteThePrint@yahoogroups.com
Date: Tue, 13 Apr 2004 13:19:56 -0400
To: "digitalblackandwhitetheprint@yahoogroups.com"
<digitalblackandwhitetheprint@yahoogroups.com>
Subject: [Digital BW] Sharpening vs. Print size [was: Image Density vs.
Print Size]

Ummm... why not? Worst that can happen is I get flamed as a heretic.
Again. This theory is not supported by house painters, however. 8-(

Sharpening should be done in several stages. All scanning softens the
image by definition; scanning lays a deterministic sampling grid over a
stochastic spread of grain clumps. First, you do what I call a grain
sharpening, to restore the sharpness of the image after scanning.

Digital output also softens an image. Printing, for instance, converts
square pixels into round-ish blobs of ink. Resampling algorithms either
sort through data and throw some of it away, or sort through data and
manufacture more based on what the algorithms see. This of course
softens the image. How much softening depends on how much you change,
and how much you can see.

For example, take an image (at 360dpi) and print it at printer
resolution 1440, and at 2880 (or any other two printer resolutions).
When you compare the images, the 2880 often looks sharper than the 1440.
It's not because the data in the file you sent to the printer was
sharper; that data didn't change. It's because in one case the printer
used 4 ink dots per pixel, and the other case it used 8 ink dots per
pixel. The 8 ink dots can more accurately reproduce the pixel, and thus
the print appears a bit sharper.

To deal with this, here's what I do. Of course, YMMV. First, I do a
light grain sharpening just after scanning (let the cries of "heresy!"
begin). Then, just before output, I do a heavier sharpening (after I
resize for output). If you are doing severe downsampling for web
publishing, you'll need to really sharpen the image to get it to be
representative of the image when viewed on a monitor. If I'm outputting
to a printer and printing a smallish image (8x10 say) the sharpening is
lighter. As the size goes up, so does the output sharpening. All of this
just to get the image to "look consistent" across sizes. To my eye.

Unfortunately, I haven't come up with a rule of thumb for sharpening on
either end. On the scanning end, it's going to depend on your scanner
and your enlargement factor, your film, your processing, etc... On the
output end, it's going to depend on your output device, your enlargement
factor, the detail in your image, etc....

There are those that say that sharpening should be a three step thing,
with a local area sharpening done as part of image manipulation. I've
never seen the need for that with my images. Might be useful for some
though.

So.... While I wish that sharpening were a one size fits all print sizes
thing, it doesn't seem to work out that way. I'll say it again: YMMV.





[Non-text portions of this message have been removed]

Attachments

Move to quarantaine

This moves the raw source file on disk only. The archive index is not changed automatically, so you still need to run a manual refresh afterward.