> I will assume your post was not a joke.

Thank you.   :)

Ya, no not a joke.  Maybe I sould qualify myself a bit at the risk
of seeming bold.  After my CS degree and a few government contracts
I wrote games (all aspects music, gfx, skews, etc.) on the Commodore64
and Vic20.  The Amiga was released and I wrote a game on it.  Discovered
the money and fun available in doing video graphics and related software
was better so I got into that along with building up what later became
a huge Amiga distribution network for central Japan.  The Newtek Toaster
came along and I got very much into broadcast video post production,
editing, and camera work.  Many TV commercials and a few TV Shows later
I was headhunted by the university system here in Japan where I have
been teaching VFX, CG Techniques, and the specific operation of Maya
LightWave, Digital Fusion, and etc. for the last 5 or 6 years.  I still
do as much CG on the side as time permitts.


> Both of these cameras output jpg files.  If you want to compare the
> two then that is all there is.  How do you propose to compare image
> quality with out using pictures produced by the cameras?

Well,  I don't really know.  I think if you are stuck with a camera
that can only do JPEG then you can't.  At least not on a hardware level.
The subject of "noise" or "grain" becomes obserdly moot.  In a different
place in this thread someone asked "what do you think of the sharpness
of the text on the sighboard" (paraphrased).  I went there to look and
the first thing that was obvious /to me/ was that it had been smooshed
by jepeg error.  I wanted to jump in and say so but all I got from
my initial post were smerks so I held off.

But there are other aspects of "Image Quality" besides noise and grain
that are still perfeclty compareable and discussable.


> You may have heard in the past that resaving jpgs can reduce the
> quality.  Several years ago I saw a post by someone who heard this and
> thought that every time he viewed a jpg id would lose quality and
> eventualy "wear out".
>
> You need have no fear about the files wearing out.  When one copies
> the file to a website and downloads it there is no added loss in
> quality.   The file one downloads is the same as the one produced by
> the camera, no loss in quality.

Yes when I made the comment about "online jpegs" I wasn't referring
to generational loss caused by software that decompress to a raster
port of some type and then recompresses durring a save operation.
You're right iExplorer /mostly/ "copies" an image from it's server
location to you local drive with no recompression (that I know of)
ocurring in the process.  What I was considering was the quality
of the decompression system used.  To exemplify this define your
desk top as big as it'll go still maintaining 32bit color. Hopefully
you have at _least_ 1600x1200 preferbly higher.  Now open Photoshop,
your web browser, and something like ACDSEE (image viewer) and display
one of those OnLine example photos in all three applications.  Tile
the display so that you can compare in a strip-test like across your
monitor.  You /should/ be floored by the differences.  Especailly in
terms of how much noise and grain is "aparrent" in the three renderings.
Photoshop /should/ be the best by far.  Now maximise Photoshop, zoom
the image to about 200% or 300% and select "save as". Select the file
type for Jpeg and click "OK".  You should be presented with an option
box with quality settings from 1 to 10 and a preview checkbox.  With
the preview checkbox checked, slide the quality settings from 1 to 10
and wait each time for the preview to update in the window.  Every image
will be a little different but in most cases you'll notice that /extra/
grain and noise won't disappear until the quality setting is at 8 or 9.
Try the same thing with an uncompressed photo if you can get access to
one and you'll be surprised at how much /more/ of a difference there is
as you slide the quality setting gadget across.  Make sure photoshop
is not set to display using "Difusion Dither" for these tests.

From examining most jpeged images from <$600msrp cameras it /seems/ to me
by the appearence and the "compression ratio" of the image, that they
are applying about the same as a photoshop quality setting of 5 or maybe
6 on some better models.  Well, golly!  That introduces so much noise
and grain that there is really no way you could tell the difference
between two cameras given the same CCD size, about the same lens design,
and similar setting in thier respective firmware.  Well, as far as noise
and grain are conserned anyway.  I do believe the original poster
specified "noise" and "grain" as issues of comparrison right?

My comment was and still is:  Hmmm, no, I don't think you can...  Not
with JPeg images anyway.

Besides all that there are the issues of focus, camera shake, DOF, and
any preprocessing done by the camera itself.  This brings comparring
cameras via online images to an ironic level of obserdity IMO.  As I
pointed out in my last post I believe this is why reviewers spend most
of thier time in a camera review discussing the hardware/firmware
specifications followed by handeling, /cool/ features and a little
impressionistic feeling.


I could be wrong though...  What do you think?  Does that sound about
right to you?