Hi quality jpg vs tiff?

Apparently if any modifications are made to JPG files after first saving and the modification saved there is a reduction in the quality of the image, while with TIFF there is not. There more times the image is re-saved the greater the deterioration. I have not tested this and I don’t know how long before the deterioration is very noticeable. I guess if the originals are stored on CD or DVD and any re-saving would be with a copy then the JPG would probably be fine.
John McGerr
"Jan Kohl" wrote in message

I’ve been scanning and working with graphics for about 8 years now, and

always kept my documents in uncompressed files for

storage.

However, I’ve recently volunteered to do a possibly VERY large order of

photos/negatives, and in the interests of keeping costs

(and storage costs) down, I’m looking at possibly using high quality jpgs.

I’ve got Photoshop 6, and if I save the jpg at max

quality (single pass), I see more than 1/2 reduction in the size over a

tiff, with a negligable difference in the image. In

fact, in a slideshow (or closeup inspection) of the images, I really

cannot see much difference at all. Is there something here

that I’m missing when considering using high-quality jpgs or tiffs?
Cheers!

"Jan Kohl" wrote in message

I’ve been scanning and working with graphics for about 8 years now, and

always kept my documents in uncompressed files for

storage.

However, I’ve recently volunteered to do a possibly VERY large order of

photos/negatives, and in the interests of keeping costs

(and storage costs) down, I’m looking at possibly using high quality jpgs.

I’ve got Photoshop 6, and if I save the jpg at max

quality (single pass), I see more than 1/2 reduction in the size over a

tiff, with a negligable difference in the image. In

fact, in a slideshow (or closeup inspection) of the images, I really

cannot see much difference at all. Is there something here

that I’m missing when considering using high-quality jpgs or tiffs?
Cheers!

"John McGerr" wrote in
news:wv0ob.3125$:

Apparently if any modifications are made to JPG files after first saving and the modification saved there is a reduction in the quality of the image, while with TIFF there is not. There more times the image is re-saved the greater the deterioration. I have not tested this and I don’t know how long before the deterioration is very noticeable. I guess if the originals are stored on CD or DVD and any re-saving would be with a copy then the JPG would probably be fine.

Don’t get me wrong, I’m not faulting you for repeating an extremely common warning, but I don’t think it’s true, at least not with Photoshop.

Jpegs simply take info shortcuts. A tiff defines each pixel with color info; a jpeg may say "All pixels between [X1/Y1] and [X4/Y4] are this shade of blue". The more compression you go for, the lower the threshold between contrasting colors, resulting in the ‘pixelated’ look because the smooth transition between colors has been chopped. Pretty much the same effect, to varying degrees, as a 256-color gif.

But when Photoshop opens the file again, it translates this info back into raw data, essentially the same as a tiff. If the info was lost during the first save (high compression), you’re not getting it back. But subsequent saving *at the same compression* simply sets the same parameters as the first time. You’re not ‘compressing the compression’, you’re just repeating the shortcuts.

Like Bob Hatch, I’ve tested this. Repeated saves at a compression of ‘5’ (Photoshop 5.0) looked exactly the same as the original, even when viewed at 500%.

– Al.

—
To reply, insert dash in address to separate G and I in the domain

"Jan Kohl" wrote in
news:W20ob.25582$:

I’ve been scanning and working with graphics for about 8 years now, and always kept my documents in uncompressed files for storage.
However, I’ve recently volunteered to do a possibly VERY large order of photos/negatives, and in the interests of keeping costs (and storage costs) down, I’m looking at possibly using high quality jpgs. I’ve got Photoshop 6, and if I save the jpg at max quality (single pass), I see more than 1/2 reduction in the size over a tiff, with a negligable difference in the image. In fact, in a slideshow (or closeup inspection) of the images, I really cannot see much difference at all. Is there something here that I’m missing when considering using high-quality jpgs or tiffs?

Chances are, no. But it’s hard to say on an overall basis, since it needs to be judged individually.

Jpegs *are* a shortcut, and they reduce the info. But at the lowest compression (highest quality), whether this makes any visible difference to the image is questionable. It will all depend on the detail of the original image, how much the scan captured, and potentially what color-bit-depth you’re using (I do all my work in 24-bit, so I’m hesitant to say how well you’d do at higher).

The greater the resolution, the less chance you’d see any difference at all. But never take anyone’s word for it. Do a couple of tests on different images, most especially on a) lots of fine detail, and b) large areas with very gradual color transitions (sunset skies, increasing shadows, etc.). Save them, and re-open them to examine at 200 to 400%. If you can’t see anything significant at that size, you have nothing to worry about.

One little pointer from doing much the same thing: I used to tell masters from web-sized images apart because of file extension. The tiffs were always masters. Now that I’ve stopped doing this, I find it easier to sort through folders by appending the filename with something, such as "M- Photo.jpg" to designate a master copy. Setting up specific folders, or dumping to CD as soon as you have a full disc, can help too. You’re less likely to mix them up or accidentally overwrite because your mind says absently "Jpeg, this is a copy" ;-).

– Al.

—
To reply, insert dash in address to separate G and I in the domain

But editing an image introduces changes that will potentially/probably force the "shortcuts" to be rewritten. This may introduce additional degradation, and repetition can only make the situation worse.

The rationale, I think, for this drumbeat is to disrupt the high-compression, repeat-edit cycle that the unaware can fall into.

On Thu, 30 Oct 2003 05:50:23 GMT, Al Denelsbeck
wrote:

"John McGerr" wrote in
news:wv0ob.3125$:

But when Photoshop opens the file again, it translates this info back into raw data, essentially the same as a tiff. If the info was lost during the first save (high compression), you’re not getting it back. But subsequent saving *at the same compression* simply sets the same parameters as the first time. You’re not ‘compressing the compression’, you’re just repeating the shortcuts.

Like Bob Hatch, I’ve tested this. Repeated saves at a compression of ‘5’ (Photoshop 5.0) looked exactly the same as the original, even when viewed at 500%.

Pardon me for butting in. A rule of thumb that was passed on to me during a PS training class was to always convert JPG captures into PSD or TIF for manipulation and storage, and to only store corrected files in JPG if there was a storage space issue. I’ve done the same as was suggested here, that is multiple test opening & closing of JPGs and have never really seen what I’d call image deterioration, esp. with larger JPG files. However small, web intended files are often fairly skimpy on data to begin with and it’s therefore probably a good idea not to deteriorate the file via compression / "decompression", etc.

"Al Denelsbeck" wrote in message

"John McGerr" wrote in
news:wv0ob.3125$:

Apparently if any modifications are made to JPG files after first saving and the modification saved there is a reduction in the quality of the image, while with TIFF there is not. There more times the image is re-saved the greater the deterioration. I have not tested this and I don’t know how long before the deterioration is very noticeable. I guess if the originals are stored on CD or DVD and any re-saving would be with a copy then the JPG would probably be fine.

Don’t get me wrong, I’m not faulting you for repeating an extremely common warning, but I don’t think it’s true, at least not with Photoshop.
Jpegs simply take info shortcuts. A tiff defines each pixel with color info; a jpeg may say "All pixels between [X1/Y1] and [X4/Y4] are

this

shade of blue". The more compression you go for, the lower the threshold between contrasting colors, resulting in the ‘pixelated’ look because the smooth transition between colors has been chopped. Pretty much the same effect, to varying degrees, as a 256-color gif.

But when Photoshop opens the file again, it translates this info back into raw data, essentially the same as a tiff. If the info was lost during the first save (high compression), you’re not getting it back. But subsequent saving *at the same compression* simply sets the same

parameters

as the first time. You’re not ‘compressing the compression’, you’re just repeating the shortcuts.

Like Bob Hatch, I’ve tested this. Repeated saves at a compression of ‘5’ (Photoshop 5.0) looked exactly the same as the original, even when viewed at 500%.

– Al.

—
To reply, insert dash in address to separate G and I in the domain

On Thu, 30 Oct 2003 04:01:58 GMT, "Jan Kohl" wrote:

I’ve been scanning and working with graphics for about 8 years now, and always kept my documents in uncompressed files for
storage.

However, I’ve recently volunteered to do a possibly VERY large order of photos/negatives, and in the interests of keeping costs
(and storage costs) down, I’m looking at possibly using high quality jpgs. I’ve got Photoshop 6, and if I save the jpg at max
quality (single pass), I see more than 1/2 reduction in the size over a tiff, with a negligable difference in the image. In
fact, in a slideshow (or closeup inspection) of the images, I really cannot see much difference at all. Is there something here
that I’m missing when considering using high-quality jpgs or tiffs?

i’ve used hi rez jpgs for offset printing from time to time, with good results. Most stock photo CDs we buy, the images are in jpg form to start with. If you dont keep re-opening and re-saving your jpgs (which might degrade the image), you’ll probly be fine. Some of this depends on how discerning your client is.

Had you looked into saving TIFFs with lossless encoding? Both Zip and LZW are lossless.
You’ll probably get better compression of scanned images with zip and of computer generated line art with LZW (based on my likely-flawed understanding of how the two kinds of compression actually work). You are not giving up any kind of image quality (or introducing any generational issues) by using lossless encoding. The downside is that you may find that some of your image browsers cannot read compressed TIFFs.
Oh – you might also think about getting a DVD writer – you can put a suprisingly large number of (even big) images on a 4.2G disc. Your clients most probably have DVD readers in their PCs/Macs already. Bob

On Thu, 30 Oct 2003 04:01:58 GMT, "Jan Kohl" wrote:

I’ve been scanning and working with graphics for about 8 years now, and always kept my documents in uncompressed files for
storage.

However, I’ve recently volunteered to do a possibly VERY large order of photos/negatives, and in the interests of keeping costs
(and storage costs) down, I’m looking at possibly using high quality jpgs. I’ve got Photoshop 6, and if I save the jpg at max
quality (single pass), I see more than 1/2 reduction in the size over a tiff, with a negligable difference in the image. In
fact, in a slideshow (or closeup inspection) of the images, I really cannot see much difference at all. Is there something here
that I’m missing when considering using high-quality jpgs or tiffs?
Cheers!

»» Jan Kohl

::: computer security consultant :::
the pits – http://www.theuspits.com
castle graphics – http://www.castlegraphics.com

Bob Hatch wrote:

"Jan Kohl" wrote in message

I’ve been scanning and working with graphics for about 8 years now, and

always kept my documents in uncompressed files for

storage.

However, I’ve recently volunteered to do a possibly VERY large order of

photos/negatives, and in the interests of keeping costs

(and storage costs) down, I’m looking at possibly using high quality jpgs.

I’ve got Photoshop 6, and if I save the jpg at max

quality (single pass), I see more than 1/2 reduction in the size over a

tiff, with a negligable difference in the image. In

fact, in a slideshow (or closeup inspection) of the images, I really

cannot see much difference at all. Is there something here

that I’m missing when considering using high-quality jpgs or tiffs?
Cheers!

?? Jan Kohl

No, you’re not missing much of anything. The loss by saving 1 or 2 times at highest jpg is really not noticeable. I recently did a test with an image from my Canon D60. Took an image from the camera, high jpg. Did color correction etc and saved as tiff. Took the same image opened it and rotated 2 or more times, changed color or density then restored the image to the original settings and saved as high jpg, then closed the image. Reopened and repeated the process 5 times. Made a copy of the image and did the same thing 5 more times.

In the end I had 3 images. The tiff, a jpg that had been opened and saved 5 times and one that had been opened and saved 10 times. Set them side by side on a 24" wide page and printed the page out on my Epson 7600. There is no real difference between the tiff and the 10 time saved jpg.

I know I’m gonna catch hell for this but so be it. Now, as a matter of practice I capture the images hi jpg, any thing we modify is saved and sent to the lab as a tiff.
YMMV.

Two things. Photoshop may be clever enough to realize that in the end you’ve done nothing to the image. So it possibly did not recompress the image.

One clue is to check the file sizes. If they are different, then there was more compression, even if you did not see it.

—- Paul J. Gans

"Paul J Gans" wrote in message

Two things. Photoshop may be clever enough to realize that in the end you’ve done nothing to the image. So it possibly did not recompress the image.

Files were save using the Save As command, then yes to replace.

One clue is to check the file sizes. If they are different, then there was more compression, even if you did not see it.

The 5 save file and the 10 saved file are different sizes. 5 saved is 3,333. 10 saved is 3,367. Tiff is 14,663. Both jpgs when opened in PS are 14.3 meg. If saved again as tiff the save at 14,663.
—
"Your money does not cause my poverty. Refusal to believe this is at the bottom of most bad economic thinking." –P. J. O’Rourke http://www.bobhatch.com

"Gerald G. McGeorge" wrote in message

Pardon me for butting in. A rule of thumb that was passed on to me during

a

PS training class was to always convert JPG captures into PSD or TIF for manipulation and storage, and to only store corrected files in JPG if

there

was a storage space issue. I’ve done the same as was suggested here, that

is

multiple test opening & closing of JPGs and have never really seen what

I’d

call image deterioration, esp. with larger JPG files. However small, web intended files are often fairly skimpy on data to begin with and it’s therefore probably a good idea not to deteriorate the file via compression

/

"decompression", etc.

You could spend the entire next week opening and closing the same JPEG without seeing any deterioration!
Opening and closing a JPEG has never ever damaged it.
You lose quality ONLY if you modify your image and then recompress (close) it.
Modifying like in adjusting levels, curves, resizing, etc etc. Now how much of the loss is visible I have no idea but some people tested it and probably published their results.
Personally, I don’t really care because I can afford hard drives and CDs so I save as TIFF.
To me JPEGs are for Web work and email.

Stephan

JPEG is a ‘lossy’ image data compression algorithm.
It is optimized for human vision which is pretty forgiving and easily tricked.

TIFF is a image data storage scheme. It was designed as a lossless way to record and move image data via files from one device to another; optical imager to PC, PC to PC, etc.

A bad analogy can be made by comparing JPEG to a photo print and TIFF to a negative. Depending on the photo printer, the image can be astounding, but it is just a representation of the original image capture. Any further processing of that representational image starts without all of the original captured data. The TIFF ‘negative’ is an absolute record of the image data. As long as you have the TIFF file, you can always go back to the original captured data.

For most applications, JPEG with it’s human vision optimization is perfectly reasonable.
TIFF is only an absolute necessity when the image data is going to be multi-purposed or archived for future processing.

Your course of action depends on the source material and intended use. Example:
Early photo and negatives – the source material may be fragile and only available for one scanning session.
Are the scanned images for academic use?
How are the scans going to be cataloged/stored/distributed/viewed/printed? Will the images be published? paper or electronic?
Which image format is expected by the publisher/end user/owner? Will the scanned images be subjected to specialized image processing at a later date?

It could be that big picture concerns may temper storage size issues.

HTH

Mr3

"Bob Kour

On Thu, 30 Oct 2003 05:50:23 GMT, Al Denelsbeck
wrote:

Like Bob Hatch, I’ve tested this. Repeated saves at a compression of ‘5’ (Photoshop 5.0) looked exactly the same as the original, even when viewed at 500%.

Sorry Al, but I have to disagree. I’ve seen repeated saves degrade detail on images. After about 10 saves there is a noticeable blurring of fine detail in the image. And that’s viewed at twice normal size.

—

Hecate

veni, vidi, relinqui

"Hecate" wrote in message

On Thu, 30 Oct 2003 05:50:23 GMT, Al Denelsbeck
wrote:

Like Bob Hatch, I’ve tested this. Repeated saves at a compression of ‘5’ (Photoshop 5.0) looked exactly the same as the original, even when viewed at 500%.

Sorry Al, but I have to disagree. I’ve seen repeated saves degrade detail on images. After about 10 saves there is a noticeable blurring of fine detail in the image. And that’s viewed at twice normal size.
—

My tests do not show the same thing.
—
"Your money does not cause my poverty. Refusal to believe this is at the bottom of most bad economic thinking." –P. J. O’Rourke http://www.bobhatch.com

"Peavey_Hermann" wrote in message

On Thu, 30 Oct 2003 04:01:58 GMT, "Jan Kohl" wrote:

I’ve been scanning and working with graphics for about 8 years now, and

always kept my documents in uncompressed files for

storage.

However, I’ve recently volunteered to do a possibly VERY large order of

photos/negatives, and in the interests of keeping costs

(and storage costs) down, I’m looking at possibly using high quality

jpgs. I’ve got Photoshop 6, and if I save the jpg at max

quality (single pass), I see more than 1/2 reduction in the size over a

tiff, with a negligable difference in the image. In

fact, in a slideshow (or closeup inspection) of the images, I really

cannot see much difference at all. Is there something here

that I’m missing when considering using high-quality jpgs or tiffs?

i’ve used hi rez jpgs for offset printing from time to time, with good results. Most stock photo CDs we buy, the images are in jpg form to start with. If you dont keep re-opening and re-saving your jpgs (which might degrade the image), you’ll probly be fine. Some of this depends on how discerning your client is.

As I said in this same thread, opening and closing a JPEG does not degrade it

Stephan

"Stephan" wrote in message

As I said in this same thread, opening and closing a JPEG does not degrade it

Of course not. Opening and closing (as oppposed to opening and saving) does not invoke any further re-compression.

Every book, manual, and class I’ve ever taken has said that each and every JPG *save* results in more compression. Furthermore, all the "experts" I’ve follow agree that visible degradation results after X number of saves, but debate is open to the general value of X since it seems to be image dependent. Fair enough.

Personally, a JPG is an end-product for me so re-saves are never an issue.

—–= Posted via Newsfeeds.Com, Uncensored Usenet News =—– http://www.newsfeeds.com – The #1 Newsgroup Service in the World! —–== Over 100,000 Newsgroups – 19 Different Servers! =—–

Bob Hatch wrote:

"Paul J Gans" wrote in message

Two things. Photoshop may be clever enough to realize that in the end you’ve done nothing to the image. So it possibly did not recompress the image.

Files were save using the Save As command, then yes to replace.

One clue is to check the file sizes. If they are different, then there was more compression, even if you did not see it.

The 5 save file and the 10 saved file are different sizes. 5 saved is 3,333. 10 saved is 3,367. Tiff is 14,663. Both jpgs when opened in PS are 14.3 meg. If saved again as tiff the save at 14,663.
—
"Your money does not cause my poverty. Refusal to believe this is at the bottom of most bad economic thinking." –P. J. O’Rourke http://www.bobhatch.com

The tiffs should not change size as that’s a lossless format. I’d guess that the jpegs were not further compressed.

—- Paul J. Gans

Hecate wrote:

On Thu, 30 Oct 2003 05:50:23 GMT, Al Denelsbeck
wrote:

Like Bob Hatch, I’ve tested this. Repeated saves at a compression of ‘5’ (Photoshop 5.0) looked exactly the same as the original, even when viewed at 500%.

Sorry Al, but I have to disagree. I’ve seen repeated saves degrade detail on images. After about 10 saves there is a noticeable blurring of fine detail in the image. And that’s viewed at twice normal size.

Fine detail is exactly were you’d expect to see it. JPEG compression works by, in effect, eliminating very fine detail and then compressing. It is why one can get excellent compression for the web by doing a bit of Gaussian blur before saving as a JPEG. The advantage is that you get to determine where the fine detail will be lost.

—- Paul J. Gans

On Thu, 30 Oct 2003 18:07:38 -0800, "Bob Hatch" wrote:

"Hecate" wrote in message

On Thu, 30 Oct 2003 05:50:23 GMT, Al Denelsbeck
wrote:

Like Bob Hatch, I’ve tested this. Repeated saves at a compression of ‘5’ (Photoshop 5.0) looked exactly the same as the original, even when viewed at 500%.

Sorry Al, but I have to disagree. I’ve seen repeated saves degrade detail on images. After about 10 saves there is a noticeable blurring of fine detail in the image. And that’s viewed at twice normal size.
—

My tests do not show the same thing.

<shrug> I can only go by what I see.

—

Hecate

veni, vidi, relinqui

On Fri, 31 Oct 2003 21:03:55 +0000 (UTC), Paul J Gans
wrote:

Hecate wrote:

On Thu, 30 Oct 2003 05:50:23 GMT, Al Denelsbeck
wrote:

Like Bob Hatch, I’ve tested this. Repeated saves at a compression of ‘5’ (Photoshop 5.0) looked exactly the same as the original, even when viewed at 500%.

Sorry Al, but I have to disagree. I’ve seen repeated saves degrade detail on images. After about 10 saves there is a noticeable blurring of fine detail in the image. And that’s viewed at twice normal size.

Fine detail is exactly were you’d expect to see it. JPEG compression works by, in effect, eliminating very fine detail and then compressing. It is why one can get excellent compression for the web by doing a bit of Gaussian blur before saving as a JPEG. The advantage is that you get to determine where the fine detail will be lost.

Yes. I have a feeling that the people who say it doesn’t happen are just opening and closing, rather than opening and resaving.

—

Hecate

veni, vidi, relinqui

"Paul J Gans" wrote in message

The tiffs should not change size as that’s a lossless format. I’d guess that the jpegs were not further compressed.

And you would guess wrong.
But that’s OK.
—
"Just as the Left was anti-anticommunist,
so too then are they anti-antiterrorist." –Robert Spencer http://www.bobhatch.com
http://www.boycottcbs.com

"Hecate" wrote in message

—

My tests do not show the same thing.

<shrug> I can only go by what I see.

<shrug> Me too. 🙂
—
"Just as the Left was anti-anticommunist,
so too then are they anti-antiterrorist." –Robert Spencer http://www.bobhatch.com
http://www.boycottcbs.com

On Fri, 31 Oct 2003 16:51:46 -0800, "Bob Hatch" wrote:

"Paul J Gans" wrote in message

The tiffs should not change size as that’s a lossless format. I’d guess that the jpegs were not further compressed.

And you would guess wrong.
But that’s OK.

I scanned some B&W negs. Some weren’t orientated correctly but for speed I scanned as laid out on the scanner bed. The resultant tiffs were about 5mb each. After rotating as required they increased to about 22mb.

MJ

"Hecate" wrote in message

Yes. I have a feeling that the people who say it doesn’t happen are just opening and closing, rather than opening and resaving.

Might be true in some cases but in my test I did not just open and close or open and save. I did an open modify in some way then Save As. I knew better than just open and close.
—
"Just as the Left was anti-anticommunist,
so too then are they anti-antiterrorist." –Robert Spencer http://www.bobhatch.com
http://www.boycottcbs.com

Hecate wrote in
news::

On Thu, 30 Oct 2003 05:50:23 GMT, Al Denelsbeck
wrote:

Like Bob Hatch, I’ve tested this. Repeated saves at a
compression of
‘5’ (Photoshop 5.0) looked exactly the same as the original, even when viewed at 500%.

Sorry Al, but I have to disagree. I’ve seen repeated saves degrade detail on images. After about 10 saves there is a noticeable blurring of fine detail in the image. And that’s viewed at twice normal size.

You can disagree with my results all you like. You might do well to see them first, though 😉

Started with a mildly compressed jpeg, opened, added text, and saved at compression of 5, ten times, new filename. Compared Save One, especially the text, but also the fine detail and the broad area colors, with Save Ten. Like I said, at 500%. Not a damn bit of difference. Want ’em? I’ll e- mail them to you. The date-time markers on the file, as well as the additional text, should make it clear no tricks are being played.

I did *not* make any radical changes, playing with the color register or such like that. And I have yet to see any image where I ever have to do that more than once, so I’m not too concerned.

You will also note, elsewhere in the thread, where I recommended several tests on different images. If you prefer, I’d add in maintaining a master image and not screwing with it (beyond initial color correction and dust spotting), with is my routine anyway. Good habits will stand you in better stead than web folklore. If you have a tendency to take your master image and change it umpteen times and resave it, then you’re shooting yourself in the foot anyway. But I’d put heavy money down that you’ll trash more images from a bad jpeg marker or a bad disc sector that corrupts the file, than from ‘repetitive compression degradation’.

– Al.

—
To reply, insert dash in address to separate G and I in the domain

On Sat, 01 Nov 2003 04:41:39 GMT, Al Denelsbeck
wrote:

Hecate wrote in
news::

On Thu, 30 Oct 2003 05:50:23 GMT, Al Denelsbeck
wrote:

Like Bob Hatch, I’ve tested this. Repeated saves at a
compression of
‘5’ (Photoshop 5.0) looked exactly the same as the original, even when viewed at 500%.

Sorry Al, but I have to disagree. I’ve seen repeated saves degrade detail on images. After about 10 saves there is a noticeable blurring of fine detail in the image. And that’s viewed at twice normal size.

You can disagree with my results all you like. You might do well to see them first, though 😉

I don’t disbelieve you, it’s just not what I found. I used a picture of a tree against a blue sky. At 10 recompressions there was a just noticeable difference. At 20 there was a definitely noticeable difference.

All I did was open, did a quick small clone at the bottom of the image (not where I was comparing) and resaved using Save As. (I was using the branches/twigs/leaves at the top of the picture as a comparison.).

And the fine detail in the branches was definitely fuzzier.

Having said that, it’s really moot anyway. I always save as Tiff and work on a copy. I never save as jpg unless I’m putting the image on the web. 🙂

—

Hecate

veni, vidi, relinqui

Bob Hatch wrote:

"Paul J Gans" wrote in message

The tiffs should not change size as that’s a lossless format. I’d guess that the jpegs were not further compressed.

And you would guess wrong.
But that’s OK.

I don’t think so. As I said in an earlier post, resaving a jpg will, in general, cause it to go through the compression process again with subsequent loss of detail.

However, some programs are smart enough to know that the image has not been changed. They don’t bother doing any further compression.

I understand that Photoshop will do this under some circumstances. I might be wrong about that though.

So we agree. Going through the compression sequence twice will cause loss. But not going through it the second time will not cause further loss.

—- Paul J. Gans

The format theory speaks for itself. Lossy means something is lost, whether it is perceiveable or not. As for repeated saves—if there is ANY reduction in file size, there is a reduction in the information left.. again, whether it is preceivable or not, there is a loss. This loss is cumulative, again, whether it is perceiveable or not. Some images will reveal the losses more obviously than others.

Two artists may paint a solid yellow canvas, one chooses a two-inch brush, the other chooses a one-eighth-inch brush. When the latter is asked why he would use such a small brush to paint the entire canvas a single color he replies, "can’t you tell the difference?" It depends on exactly what you are looking at.. a colored canvas, or the paint on the canvas.

"Paul J Gans" wrote in message

Bob Hatch wrote:

"Paul J Gans" wrote in message

The tiffs should not change size as that’s a lossless format. I’d guess that the jpegs were not further compressed.

And you would guess wrong.
But that’s OK.

I don’t think so. As I said in an earlier post, resaving a jpg will, in general, cause it to go through the compression process again with subsequent loss of detail.

However, some programs are smart enough to know that the image has not been changed. They don’t bother doing any further compression.

I understand that Photoshop will do this under some circumstances. I might be wrong about that though.

So we agree. Going through the compression sequence twice will cause loss. But not going through it the second time will not cause further loss.

—- Paul J. Gans

On Fri, 31 Oct 2003 17:09:45 -0800
in alt.graphics.photoshop
"Bob Hatch" muttered something like this:

"Hecate" wrote in message

—

My tests do not show the same thing.

<shrug> I can only go by what I see.

<shrug> Me too. 🙂

http://0utpatient.home.comcast.net/noartifacts/

—
OuTpaTienT / outpatient°AT°rocketmail°DOT°com
http://www.0utpatient.com
http://www.oeyec.com

On Sat, 01 Nov 2003 04:41:39 GMT
in alt.graphics.photoshop
Al Denelsbeck muttered something like this:

Hecate wrote in
news::

On Thu, 30 Oct 2003 05:50:23 GMT, Al Denelsbeck
wrote:

Like Bob Hatch, I’ve tested this. Repeated saves at a
compression of
‘5’ (Photoshop 5.0) looked exactly the same as the original, even when viewed at 500%.

Sorry Al, but I have to disagree. I’ve seen repeated saves degrade detail on images. After about 10 saves there is a noticeable blurring of fine detail in the image. And that’s viewed at twice normal size.

You can disagree with my results all you like. You might do well to see them first, though 😉

Started with a mildly compressed jpeg, opened, added text, and saved at compression of 5, ten times, new filename. Compared Save One, especially the text, but also the fine detail and the broad area colors, with Save Ten. Like I said, at 500%. Not a damn bit of difference. Want ’em? I’ll e- mail them to you. The date-time markers on the file, as well as the additional text, should make it clear no tricks are being played.
I did *not* make any radical changes, playing with the color register or such like that. And I have yet to see any image where I ever have to do that more than once, so I’m not too concerned.

You will also note, elsewhere in the thread, where I recommended several tests on different images. If you prefer, I’d add in maintaining a master image and not screwing with it (beyond initial color correction and dust spotting), with is my routine anyway. Good habits will stand you in better stead than web folklore. If you have a tendency to take your master image and change it umpteen times and resave it, then you’re shooting yourself in the foot anyway. But I’d put heavy money down that you’ll trash more images from a bad jpeg marker or a bad disc sector that corrupts the file, than from ‘repetitive compression degradation’.

– Al.

A case against jpeg: http://0utpatient.home.comcast.net/noartifacts/

—
OuTpaTienT / outpatient°AT°rocketmail°DOT°com
http://www.0utpatient.com
http://www.oeyec.com

Excellent page. Especially the cool filesize graph bars comparing the different formats!! lol

"OuTpaTienT" wrote in message

On Fri, 31 Oct 2003 17:09:45 -0800
in alt.graphics.photoshop
"Bob Hatch" muttered something like this:

"Hecate" wrote in message

—

My tests do not show the same thing.

<shrug> I can only go by what I see.

<shrug> Me too. 🙂

http://0utpatient.home.comcast.net/noartifacts/

—
OuTpaTienT / outpatient

OuTpaTienT wrote:

A case against jpeg: http://0utpatient.home.comcast.net/noartifacts/

I fully agree with you. Further; no photographer in his or her right mind will sacrifize quality for convenience by scanning or capturing original (master) files in .jpeg.

If a file has once been saved in the lossy .jpeg format, the information lost can never be retrieved, and the quality can never be what it could have been if there had been a lossless master copy.

It is of course fully acceptable to make jpeg’s when that format fits the purpose, but one must always keep a master which has never been lossily compressed.

Per Inge Oestmoen, Norway
http://www.coldsiberia.org/

Peavey_Hermann wrote:

If you dont keep re-opening and re-saving your jpgs (which might degrade the image), you’ll probly be fine. Some of this depends on how discerning your client is.

Please understand that this is imprecise at best.

JPEG compression does reduce the quality of the image. The fact that 95% of all people seem to only look at a picture’s composition and originality with a relative disregard for technical quality, does not alter this sad fact.

It is a true tragedy that so much of the images we leave to posterity are made and saved in .jpeg. This represent a throwaway of much of the pictorial quality that would otherwise have been available to future connoisseurs of photographic excellence.

Per Inge Oestmoen, Norway
http://www.coldsiberia.org/

Stephan wrote:

As I said in this same thread, opening and closing a JPEG does not degrade it

Very true, but that statement bypasses the most important point.

The main thing is that a JPEG-file means that much information, and thereby quality and editability, has already been irretrievably lost.

Per Inge Oestmoen, Norway
http://www.coldsiberia.org/

Mr3 wrote:

For most applications, JPEG with it’s human vision optimization is perfectly reasonable.

No. It is only for non-critical applications like testing, and display on low-quality computer screens, where the use of JPEG can be justified. Of course, printing in magazines and newspapers must be considered non-critical when it comes to quality.

If we create our original files in lossy JPEG instead of a non-lossy format like TIFF we make a major mistake. We should always scan or capture into uncompressed file formats and keep that master intact forever.

Per Inge Oestmoen, Norway
http://www.coldsiberia.org/

On Mon, 03 Nov 2003 12:07:48 GMT, OuTpaTienT
wrote:

A case against jpeg: http://0utpatient.home.comcast.net/noartifacts/

And a very good case, apart from the conclusion. I agree, it would be nice to use png. But, until MS fix their problem with png (i.e. never) they’re unusable because the vast majority of people use IE.

—

Hecate

veni, vidi, relinqui

On Tue, 04 Nov 2003 00:50:17 +0100, "pioe[rmv]" <"pioe[rmv]"@coldsiberia.org> wrote:

Stephan wrote:

As I said in this same thread, opening and closing a JPEG does not degrade it

Very true, but that statement bypasses the most important point.
The main thing is that a JPEG-file means that much information, and thereby quality and editability, has already been irretrievably lost.

true, however, there are instances where it’s sort of irrelevant to the given task at hand, and, a hi-rez jpg would be fine. Depends on the project.

On Tue, 04 Nov 2003 00:46:45 +0100, "pioe[rmv]" <"pioe[rmv]"@coldsiberia.org> wrote:

Peavey_Hermann wrote:

If you dont keep re-opening and re-saving your jpgs (which might degrade the image), you’ll probly be fine. Some of this depends on how discerning your client is.

Please understand that this is imprecise at best.

JPEG compression does reduce the quality of the image. The fact that 95% of all people seem to only look at a picture’s composition and originality with a relative disregard for technical quality, does not alter this sad fact.

true enough, i see your point, however, you also have to consider the parameters of the project. If the client isnt all that discerning, and file size is a real, viable issue for the particular project, then, hi rez jpg is probably a good choice, for that project.

On Tue, 04 Nov 2003 00:56:34 +0100, "pioe[rmv]" <"pioe[rmv]"@coldsiberia.org> wrote:

Mr3 wrote:

For most applications, JPEG with it’s human vision optimization is perfectly reasonable.

No. It is only for non-critical applications like testing, and display on low-quality computer screens, where the use of JPEG can be justified. Of course, printing in magazines and newspapers must be considered non-critical when it comes to quality.

display on low-quality computer screens, printing in magazines and newspapers, thats pretty much my entire business. And I’ve used hi rez jpgs often.

I’m not a photographer. Maybe you are?

Do I assume from that last statement that IE doesn’t support png even though it is a MS product?
"Hecate" wrote in message

On Mon, 03 Nov 2003 12:07:48 GMT, OuTpaTienT
wrote:

A case against jpeg: http://0utpatient.home.comcast.net/noartifacts/

And a very good case, apart from the conclusion. I agree, it would be nice to use png. But, until MS fix their problem with png (i.e. never) they’re unusable because the vast majority of people use IE.
—

Hecate

veni, vidi, relinqui

On Tue, 4 Nov 2003 18:40:22 -0500, "Birddog"
wrote:

Do I assume from that last statement that IE doesn’t support png even though it is a MS product?

I don’t know where you got the idea that it’s an MS product. For more info see

http://www.w3.org/Graphics/PNG/

And no, IE doesn’t support it. And I would suspect, given the ever closer relations between Adobe and MS, that they won’t.

—

Hecate

veni, vidi, relinqui

Hecate. thanks for responding. I assumed IE meant Internet Explorer . What does IE mean?
"Hecate" wrote in message

On Tue, 4 Nov 2003 18:40:22 -0500, "Birddog"
wrote:

Do I assume from that last statement that IE doesn’t support png even

though

it is a MS product?

I don’t know where you got the idea that it’s an MS product. For more info see

http://www.w3.org/Graphics/PNG/

And no, IE doesn’t support it. And I would suspect, given the ever closer relations between Adobe and MS, that they won’t.

—

Hecate

veni, vidi, relinqui

IE is Internet Explorer and, obviously, is an MS product; it’s PNG that’s not an MS "product" and is not well supported by IE, etc.

On Wed, 5 Nov 2003 09:21:35 -0500, "Birddog"
wrote:

Hecate. thanks for responding. I assumed IE meant Internet Explorer . What does IE mean?
"Hecate" wrote in message

On Tue, 4 Nov 2003 18:40:22 -0500, "Birddog"
wrote:

Do I assume from that last statement that IE doesn’t support png even

though

it is a MS product?

I don’t know where you got the idea that it’s an MS product. For more info see

http://www.w3.org/Graphics/PNG/

And no, IE doesn’t support it. And I would suspect, given the ever closer relations between Adobe and MS, that they won’t.

—

Hecate

veni, vidi, relinqui

On Wed, 5 Nov 2003 09:21:35 -0500, "Birddog"
wrote:

Hecate. thanks for responding. I assumed IE meant Internet Explorer . What does IE mean?

Yes, IE does mean Internet Explorer and that is an MS product. However, you seemed to be saying that PNG was an MS product… 🙂

—

Hecate

veni, vidi, relinqui