Back to Ganoksin | FAQ | Contact

Digital photo question - JPEG settings


#1

When I adjust my jewelry photos, then save them in JPEG, the window
asks me if I want “Progressive”, if I want to “Optimize Huffman
codes”, if I want “Smoothing”, and how much; and then there is a
"Component Sampling" setting. It also asks if I want “Best
Compression” or “Best Quality”. I’ve been going for "Best Quality"
there, but don’t have a clue what the rest of these settings do.
Help please!

Thanks!
Beth


#2

Beth; Hi; Don’t know for sure what some of the choices mean, but I
might hazard a guess. There should be somewhere a help file in the
software, online help, or a book that will answer these questions.
What software is it?

The best compression best quality is easiest.

If you are going over the web - go with compression. The amount
will depend on the detail in your photo. It might even convert your
pics into TIF or JIF files. It will throw out a lot of redundant
(maybe even lose some detail in the process). Will
probable bring the resolution down to 72 dpi (the image has 72 dots
or pixels in an inch long line). It might even reduce the color
choices to 256 if there is serious compression.

Progressive might be for optimizing the picture for video display
over a computer monitor as opposed to a NTSC or video monitor.
Computer monitors and video (TV) monitors read different. Computer
or RGB monitors read the picture line by line - going 1, 2, 3, 4, 5 .
. . . etc. until the end. NTSC monitors read two fields that are
different. Each field reads alternate lines. The first field is
composed of the odd number liens i.e. 1, 3, 5, 7, 9 . . . etc until
the lost one. The other field is 2, 4, 6, 8, . . . etc. until the
end. Each of these fields is displayed 30 times a second, giving 60
different pictures a second to our eyes. Because of the way our
eyes work each pair of frames is seen as one picture (a phenomena
called persistence of vision) Progressive scan also reads 60 times a
second.

The optimize Hoffman codes might be choices for converting the image
to a video friendly image. It could pull up a list of choices about
which field first (odd or even). Also NTSC, PAL, SECAM. These are
types of video. In the US and, I believe Japan, we use NTSC. Most
of Europe uses PAL (which is a progressive scan 50 times a second).
SECAM is used by some other countries. There is one or two others
that are used also but not much. They are not interchangeable. You
can’t bring a PAL video home from Europe and play it on your VCR.

Won’t happen - at least that anyone will want to watch. They can be
converted but there are artifacts in the conversion version (nice
words)and it is an expensive process. It’s not something you get
done on every street corner and the people who do it like to have you
walking into their store backwards and bent over already. It makes
getting at your wallet so much easier. You’ll generally get a coffee
cup though.

Smoothing might be an anti-aliasing working. It will lessen the
stair step details in a diagonal line and at the edges of things.

These are just mights and maybes. Rereading your post, I see that
they are JPEG choices. So maybe throw out the TIF and JIF
conversions. But I’ve seen some programs that do the conversions
when it is told you are optimizing for web. But progressive is
possibly true.

Eric


#3

Beth, The real question is what are you using the photos for. If you
are going to put them on the Web then you want a medium quality with
various options. If you are going to use them for printing on a
color printer then you want best quality. If you want to save them
for some future use then save them as best quality of if using
PhotoShop save them as a PhotoShop file.

The details. Progressive - Used for the Web only, allows the image to
be loaded in sections that show a fuzzy image and there a clearer
one as the images finishes loading.

Optimize Huffman Codes - I am not familiar with this but I assume it
is a type of compression.

Smoothing - Makes a judgment about how the image should look and
smooths the jagged edges.

Component Sampling - Again I am not familiar with this one but it
has to do with compression the image.

Best compression - Makes the smallest possible file by throwing away
data.

Best Quality - Creates the largest file with the most image

My suggestion is to take the original image with the highest
resolution that your camera can produce and archive it on the
computer. Then copy it and make all changes to the copy not the
original. This will allow you to take the photo once and not retake
it if you need a larger image later. Once you do any kind of
compression on the image data is lost. Lost data means a fuzzier
file and you can never get the back into that file.

Hope this helps.

Larrie Easterly
Easterly Creative Services
http://www.larrie.com