Back to Scanners and Comparisons.

rafe b wrote:
. . . . . .

I keep tabs on eBay auctions of drum scanners, hoping to find one selling at
a decent price and within driving distance. For some odd reason I can't
bring myself to buy an Epson 4990. The Microtek (for which I paid $1K) just
wasn't up to the job.



A good source for used drum scanners and service:

http://genesis-equipment.com

They have a few Howtek models at reasonable prices. The downside of many
older drum scanners is a lack of modern software. That can mean a need
to run a dedicated older computer and OS just for scanning.

Ciao!

Gordon Moat
A G Studio
http://www.allgstudio.com

On Wed, 07 Dec 2005 13:10:07 -0800, Gordon Moat
wrote:


Perhaps if you read a bit closer to what I stated; I don't print from
JPEGs, and my concern with image files is their printing quality.
Obviously, storing TIFF files would take up a huge amount of space, and
is not practical for websites. The JPEG algorithm also functions with a
sharpening effect, even at the least compressed settings.

I stand by what I stated: "a JPEG is a poor way to judge how a printed
item will turn out". That has nothing to do with "fear", and I highly
disagree in your assertion that my "concerns" are "unfounded". If you
want to believe otherwise, you are entitled to your opinion.


Hmm.. Your words we "I have been mostly opposed to these
internet challenges mostly due to the degradation of images
by posting and viewing JPEGs." That's the statement I was
refering to.

Also, you're dead wrong about sharpening being implicit in
"the JPG algorithm."

My contention is that, in a fair test, you would not be able
to distinguish between the same image as TIF vs. high-
quality (low-compression) JPG. Whether you view it on
screen or print it -- makes no difference.


rafe b
www.terrapinphoto.com

rafe b wrote:
On Wed, 07 Dec 2005 13:10:07 -0800, Gordon Moat
wrote:


Perhaps if you read a bit closer to what I stated; I don't print from
JPEGs, and my concern with image files is their printing quality.
Obviously, storing TIFF files would take up a huge amount of space, and
is not practical for websites. The JPEG algorithm also functions with a
sharpening effect, even at the least compressed settings.

I stand by what I stated: "a JPEG is a poor way to judge how a printed
item will turn out". That has nothing to do with "fear", and I highly
disagree in your assertion that my "concerns" are "unfounded". If you
want to believe otherwise, you are entitled to your opinion.


Hmm.. Your words we "I have been mostly opposed to these
internet challenges mostly due to the degradation of images
by posting and viewing JPEGs." That's the statement I was
refering to.

Also, you're dead wrong about sharpening being implicit in
"the JPG algorithm."

My contention is that, in a fair test, you would not be able
to distinguish between the same image as TIF vs. high-
quality (low-compression) JPG. Whether you view it on
screen or print it -- makes no difference.


rafe b
www.terrapinphoto.com

rafe is right and here are the images to back it up.
These are down sampled to get a good clean image.
http://www.sewcon.com/temp/DSC02676%20sized.tif
http://www.sewcon.com/temp/DSC02676%20sized.jpg

jpeg does no shapening.

Scott

In article .com,
"Scott W" wrote:

rafe is right and here are the images to back it up.
These are down sampled to get a good clean image.
http://www.sewcon.com/temp/DSC02676%20sized.tif
http://www.sewcon.com/temp/DSC02676%20sized.jpg

jpeg does no shapening.

Scott

What you and perhaps Rafe are failing to see is that Gordon works with
offset printers, and clients that send cheap low res jpegs they pull
out of their yin yang to save actually hiring someone to re shoot and
supply the printer with a respectable size file. The problem with jpegs
are that you can't manipulate them to a great extent and so therefore
once in the hands of a print shop where they may need to be adjusted for
the requirements of the specific device, they become a problem, do to
that nature. ....that is the more they are opened the quicker they
degrade for printing purposes, yes they may look fine but in print they
will suck. You can perhaps get away with giving people clients lower res
jpegs or and its all well and good until they need a bigger file like a
tif.

So you can look at jpegs all you like on screen and state there's no
issue but i believe your misinformed.
--
"To announce that there must be no criticism of the President,
or that we are to stand by the President, right or wrong,
is not only unpatriotic and servile, but is morally treasonable
to the American public."--Theodore Roosevelt, May 7, 1918

www.gregblankphoto(dot)com

On Wed, 07 Dec 2005 22:39:56 -0500, Gregory Blank
wrote:



What you and perhaps Rafe are failing to see is that Gordon works with
offset printers, and clients that send cheap low res jpegs they pull
out of their yin yang to save actually hiring someone to re shoot and
supply the printer with a respectable size file. The problem with jpegs
are that you can't manipulate them to a great extent and so therefore
once in the hands of a print shop where they may need to be adjusted for
the requirements of the specific device, they become a problem, do to
that nature. ....that is the more they are opened the quicker they
degrade for printing purposes, yes they may look fine but in print they
will suck. You can perhaps get away with giving people clients lower res
jpegs or and its all well and good until they need a bigger file like a
tif.

So you can look at jpegs all you like on screen and state there's no
issue but i believe your misinformed.



You are asking us to argue with a strawman.

Gordon does not talk about multiple cycles of
JPG encode/decode. Separate issue.

Gordon's argument is that even *one* cycle
of JPG encode/decode produces a result
visibly inferior to TIF. And that is only the case
when high levels of JPG compression are used.

Furthermore, the context of our discussion is
(to use Gordon's phrase) "these internet challenges"
presumably, full-res film scan or digicam output,
and the like (like the stuff on my scan snippets
site, or Leigh Perry's site, which is the subject
of the thread...)

In all such cases, high-quality (low-compression)
JPG settings are used.

It's a silly thing to argue about, really, because it
costs nothing but a bit of time to verify all this for
one's self.


rafe b
www.terrapinphoto.com

Gregory Blank wrote:
In article .com,
"Scott W" wrote:

rafe is right and here are the images to back it up.
These are down sampled to get a good clean image.
http://www.sewcon.com/temp/DSC02676%20sized.tif
http://www.sewcon.com/temp/DSC02676%20sized.jpg

jpeg does no shapening.

Scott

What you and perhaps Rafe are failing to see is that Gordon works with
offset printers, and clients that send cheap low res jpegs they pull
out of their yin yang to save actually hiring someone to re shoot and
supply the printer with a respectable size file. The problem with jpegs
are that you can't manipulate them to a great extent and so therefore
once in the hands of a print shop where they may need to be adjusted for
the requirements of the specific device, they become a problem, do to
that nature. ....that is the more they are opened the quicker they
degrade for printing purposes, yes they may look fine but in print they
will suck. You can perhaps get away with giving people clients lower res
jpegs or and its all well and good until they need a bigger file like a
tif.

So you can look at jpegs all you like on screen and state there's no
issue but i believe your misinformed.
--


So I have posted a tiff and jpeg, try to do any edit that you might do
before printing that will show a difference between them. In fact they
are are same to within one level, in other words in the noise.

Scott

Scott W wrote:
Gregory Blank wrote:
In article .com,
"Scott W" wrote:

rafe is right and here are the images to back it up.
These are down sampled to get a good clean image.
http://www.sewcon.com/temp/DSC02676%20sized.tif
http://www.sewcon.com/temp/DSC02676%20sized.jpg

jpeg does no shapening.

Scott

What you and perhaps Rafe are failing to see is that Gordon works with
offset printers, and clients that send cheap low res jpegs they pull
out of their yin yang to save actually hiring someone to re shoot and
supply the printer with a respectable size file. The problem with jpegs
are that you can't manipulate them to a great extent and so therefore
once in the hands of a print shop where they may need to be adjusted for
the requirements of the specific device, they become a problem, do to
that nature. ....that is the more they are opened the quicker they
degrade for printing purposes, yes they may look fine but in print they
will suck. You can perhaps get away with giving people clients lower res
jpegs or and its all well and good until they need a bigger file like a
tif.

So you can look at jpegs all you like on screen and state there's no
issue but i believe your misinformed.
--


So I have posted a tiff and jpeg, try to do any edit that you might do
before printing that will show a difference between them. In fact they
are are same to within one level, in other words in the noise.

I've checked carefully and I do not see any difference in the detail.

Scott W wrote:
rafe b wrote:

On Wed, 07 Dec 2005 13:10:07 -0800, Gordon Moat
wrote:



Perhaps if you read a bit closer to what I stated; I don't print from
JPEGs, and my concern with image files is their printing quality.
Obviously, storing TIFF files would take up a huge amount of space, and
is not practical for websites. The JPEG algorithm also functions with a
sharpening effect, even at the least compressed settings.

I stand by what I stated: "a JPEG is a poor way to judge how a printed
item will turn out". That has nothing to do with "fear", and I highly
disagree in your assertion that my "concerns" are "unfounded". If you
want to believe otherwise, you are entitled to your opinion.


Hmm.. Your words we "I have been mostly opposed to these
internet challenges mostly due to the degradation of images
by posting and viewing JPEGs." That's the statement I was
refering to.

Also, you're dead wrong about sharpening being implicit in
"the JPG algorithm."

My contention is that, in a fair test, you would not be able
to distinguish between the same image as TIF vs. high-
quality (low-compression) JPG. Whether you view it on
screen or print it -- makes no difference.


rafe b
www.terrapinphoto.com


rafe is right and here are the images to back it up.
These are down sampled to get a good clean image.
http://www.sewcon.com/temp/DSC02676%20sized.tif
http://www.sewcon.com/temp/DSC02676%20sized.jpg

jpeg does no shapening.

Scott


JPEG is a lossy compression algorithm. It functions by retaining more
information in higher resolution areas, then compressing lower
resolution areas. The effect is that contrast along high resolution
edges can change. Very difficult to see this in smaller images, though
it becomes more noticeable in larger images. This is different from
sharpening as most consider it, though if you recall I stated it has the
effect of sharpening. Follow this next paragraph for an example of this.

Simple test for your two images. Take the JPEG, copy it, then paste it
onto a layer above the JFax TIFF you provided. Then choose Difference as
the layer blending. If they were exactly the same, the result would be
completely black. If it looks that way, then zoom into the image to
enlarge the view, and spots will be apparent. Next flatten the image to
create one layer, which looks like the black spotted image. Then select
ImageAdjustThreshold and type in "2" as the value. This will give you
a field of white spots on a black background. If you view that at
exactly 100% magnification, and compare to the originals, you will see
the ghost image of the ship and rigging. Those spots represent the
changed edge contrast near the high resolution parts.

Okay, so lots of people here probably go "so what" or "who cares". If
you only print inkjet or some of the newer LightJet, Chromira or similar
digital/chemical processes, you should rarely see a problem, or much
difference; the dot gain will usually mask any small aberrations. This
also assumes you would only work with RGB files, letting the printing
software convert to CMYK (or CcMmYk, or other) instead of doing your own
CMYK files.

Gregory Blank got it exactly right. The difference becomes more
noticeable when handling files for offset printing. This is even more
true as the printed sizes grow larger. Do you have to look really close
to find it . . . sure, and therein might be the trouble . . . probably
many just won't look that close, or will accept any printed difference.
All of us have stated it many times, the technology, technique and
resolution will become less important when the image is of a compelling
enough nature.

Ciao!

Gordon Moat
A G Studio
http://www.allgstudio.com

Gordon Moat wrote:

JPEG is a lossy compression algorithm. It functions by retaining more
information in higher resolution areas, then compressing lower
resolution areas. The effect is that contrast along high resolution
edges can change. Very difficult to see this in smaller images, though
it becomes more noticeable in larger images.
First off this is not how jpeg does it compression, and second my
samples are 920 x 720, you trying to tell me that if these were larger
I could see a differance?

Simple test for your two images. Take the JPEG, copy it, then paste it
onto a layer above the JFax TIFF you provided. Then choose Difference as
the layer blending. If they were exactly the same, the result would be
completely black. If it looks that way, then zoom into the image to
enlarge the view, and spots will be apparent. Next flatten the image to
create one layer, which looks like the black spotted image. Then select
ImageAdjustThreshold and type in "2" as the value. This will give you
a field of white spots on a black background. If you view that at
exactly 100% magnification, and compare to the originals, you will see
the ghost image of the ship and rigging. Those spots represent the
changed edge contrast near the high resolution parts.

You are talking about a change of 2 levels, and if you look at the
differance you
will see that what you are looking at is noise, no edge detail.


Okay, so lots of people here probably go "so what" or "who cares". If
you only print inkjet or some of the newer LightJet, Chromira or similar
digital/chemical processes, you should rarely see a problem, or much
difference; the dot gain will usually mask any small aberrations. This
also assumes you would only work with RGB files, letting the printing
software convert to CMYK (or CcMmYk, or other) instead of doing your own
CMYK files.

Gregory Blank got it exactly right. The difference becomes more
noticeable when handling files for offset printing. This is even more
true as the printed sizes grow larger. Do you have to look really close
to find it . . . sure, and therein might be the trouble . . . probably
many just won't look that close, or will accept any printed difference.
All of us have stated it many times, the technology, technique and
resolution will become less important when the image is of a compelling
enough nature.

I would be willing to give heavy odds that if printed these two photos
would produce the same print to anybody's eye. But then your point
was that jpeg was so bad that you did not feel you could judge
differences in scanners if all you had to look at was the jpeg, this is
just not true. In fact put each into its own layer and blink between
them at 400% zoom, you can't tell one from the other.

Jpeg can be bad, very bad, if the compression is set too high. But for
a low compression jpeg images there is no visable change in the image,
even if it is a large image.

Scott

In article ,
rafe b rafebATspeakeasy.net wrote:

and the like (like the stuff on my scan snippets
site, or Leigh Perry's site, which is the subject
of the thread...)


Since I started this thread I am well aware of the subject.
& to correct you its not Leigh Perry's site its -

largeformatphotography.info is operated on a volunteer basis by Q.-Tuan
Luong and Tom Westbrook, using server space generously donated by Brian
Reid, refered to us by Tim Atherton.
--
"To announce that there must be no criticism of the President,
or that we are to stand by the President, right or wrong,
is not only unpatriotic and servile, but is morally treasonable
to the American public."--Theodore Roosevelt, May 7, 1918

www.gregblankphoto(dot)com