What quality level on the 0-100 scale?

jfriend · April 3, 2007

I'm experimenting with Adobe Lightroom. When exporting to JPEGs, it uses a 0-100 scale for JPEG quality. I'm used to the 1-12 scale in CS2 and have been happily using quality level 10 for my Smugmug uploads.

Does anyone know what the equivalent of quality level 10 (on the 1-12 scale) is for the 0-100 scale. Is 80 roughly equivalent?

andymillson · April 3, 2007

jfriend wrote:

I'm experimenting with Adobe Lightroom. When exporting to JPEGs, it uses a 0-100 scale for JPEG quality. I'm used to the 1-12 scale in CS2 and have been happily using quality level 10 for my Smugmug uploads.

Does anyone know what the equivalent of quality level 10 (on the 1-12 scale) is for the 0-100 scale. Is 80 roughly equivalent?

Assuming that the scales in PS and lightroom are linear, then

100/12=8.33
8.33x10 = 83.3

So yeah your 80% is going to be pretty close. Id go to 85% personally, but then I like to be different 8-)

Make sense?

Andy

jfriend · April 3, 2007

andymillson wrote:

Assuming that the scales in PS and lightroom are linear, then

100/12=8.33
8.33x10 = 83.3

So yeah your 80% is going to be pretty close. Id go to 85% personally, but then I like to be different 8-)

Make sense?

Andy

I wasn't sure the scales were actually the same so I didn't want to rely on just the math. I was hoping someone who's done a lot of printing at Smugmug and who uses the 0-100 scale (I'm seeing this in Adobe Lightroom) could tell me what they use.

pat.kane · April 3, 2007

John, I want to know this as well. I was also using 10 for Photoshop. I've been using Lightroom for just under a month and I've been saving with 85. I haven't printed anything yet though.

Duffy Pratt · April 4, 2007

I'd give it an 86. It reminds me of Bandstand -- has a nice tune and you can dance to it.

Seriously, does anyone know if either the 1-12 progression or the 1-100 progression is linear? If not, then the conversion may not be any good. If I were you, I would take a couple of sample pictures and save them in lightroom at intervals of 10, then see if I could tell the difference with the output devices I was using.

Duffy

jfriend · April 4, 2007

Duffy Pratt wrote:

I'd give it an 86. It reminds me of Bandstand -- has a nice tune and you can dance to it.

Seriously, does anyone know if either the 1-12 progression or the 1-100 progression is linear? If not, then the conversion may not be any good. If I were you, I would take a couple of sample pictures and save them in lightroom at intervals of 10, then see if I could tell the difference with the output devices I was using.

Duffy

Yeah, if I can find the time, I guess I just need to run an experiment. Assuming there's a similarly decent JPEG compression engine in both CS2 and Lightroom (probably the same code in both), I can probably just figure out what 1-100 setting produces the same size JPEG as quality level 10 in CS2 starting from the same source image. If it's the same or comparable JPEG compression engine and you truly have the same source image, you can use file size as a proxy for compression level/artifacts/etc... Just need to find the time to run the comparison.

Edit: Now that I think about it more, the experiment isn't that easy. If you have one setting (like sharpening) set differently in the two programs, the file size can be impacted greatly and would no longer be a good proxy for JPEG compression level. I'll have to think about this some more.

colourbox · April 4, 2007

jfriend wrote:

Edit: Now that I think about it more, the experiment isn't that easy. If you have one setting (like sharpening) set differently in the two programs...

This may be easier (sort of) than you think. Who needs two programs? Check this. If you Save As a JPEG in Photoshop, you get the 1-12 scale. But, still in Photoshop, if you Save for Web and pick JPEG, you get...the 1-100 scale. I've never had time to reconcile them myself.

The challenge may be making sure all of the other options in the dialog boxes are equivalent because they're very different. You would probably want to start from a image with no EXIF or other metadata, because that can cause a file size difference since Save for Web strips those. But the point is you do have two scales in the same program, assuming they actually do go through the same engine.

It's always bugged me that Photoshop had those two JPEG scales in there. This problem did not start with Lightroom.

jfriend · April 4, 2007

Some test results
OK, I ran a test. I took a JPEG-fine image out of a D70s and brought that JPEG into Lightroom. I then saved a copy of the JPEG at several different quality levels. Then, I opened the same image in CS2 and saved it at quality level 10. This is what I see in the file size:

Original out-of-camera: 2.6MB

In Lightroom:
Quality level 100: 3.2MB
Quality level 95: 3.2MB
Quality level 90: 1.8MB
Quality level 85: 1.8MB
Quality level 80: 1.1MB

In CS2:
Quality level 12: 3.1MB
Quality level 11: 1.8MB
Quality level 10: 1.1MB

So, it appears to me that:
level 80 ~= level 10
level 85-90 ~= level 11
level 100 ~= level 12

So, if you assume there are comparable JPEG compression engines behind these two apps, then it appears that level 80 on the 0-100 scale should be about equal to level 10 on the 0-12 scale. And, it appears that the scale is not linear, but quantized some. There is basically no difference between 85 and 90 or between 95 and 100.

Andy · April 4, 2007

jfriend wrote:

OK, I ran a test. I took a JPEG-fine image out of a D70s and brought that JPEG into Lightroom. I then saved a copy of the JPEG at several different quality levels. Then, I opened the same image in CS2 and saved it at quality level 10. This is what I see in the file size:

Original out-of-camera: 2.6MB

In Lightroom:
Quality level 100: 3.2MB
Quality level 95: 3.2MB
Quality level 90: 1.8MB
Quality level 85: 1.8MB
Quality level 80: 1.1MB

In CS2:
Quality level 12: 3.1MB
Quality level 11: 1.8MB
Quality level 10: 1.1MB

So, it appears to me that:
level 80 ~= level 10
level 85-90 ~= level 11
level 100 ~= level 12

So, if you assume there are comparable JPEG compression engines behind these two apps, then it appears that level 80 on the 0-100 scale should be about equal to level 10 on the 0-12 scale. And, it appears that the scale is not linear, but quantized some. There is basically no difference between 85 and 90 or between 95 and 100.

Hi John, make a test order, and email me the order number, I'll void the charges. Make sure you know which images are which, and let's make a nice test.

Make say, 4 8x10s. Differing quality settings. I'll ship a copy of the images to myself, too, and we'll compare notes. OK?

jfriend · April 4, 2007

Andy wrote:

Hi John, make a test order, and email me the order number, I'll void the charges. Make sure you know which images are which, and let's make a nice test.

Make say, 4 8x10s. Differing quality settings. I'll ship a copy of the images to myself, too, and we'll compare notes. OK?

Will do, but I'll have to see how soon I can find time. I've got to get 1000 images up from a big school event this last Sunday. I ended up with 500 JPEGs from two other photographers who were helping me that I decided to process in Lightroom for the first time (since I wanted efficient JPEG processing) which is what got all this started (normally I use ACR for my RAWs). When I get that event done (their 500 plus my 500), I'll help run the test.

Boy, if you want to really learn about the strengths and weakensses of your own workflow, try processing an event of 1000 images in highly variable stage lighting (lots of exposure and white balance tweaks required).

RogersDA · April 5, 2007

Andy wrote:

Hi John, make a test order, and email me the order number, I'll void the charges. Make sure you know which images are which, and let's make a nice test.

Make say, 4 8x10s. Differing quality settings. I'll ship a copy of the images to myself, too, and we'll compare notes. OK?

Yet another reason to love Smugmug!!

wellman · April 19, 2007

jfriend wrote:

OK, I ran a test. I took a JPEG-fine image out of a D70s and brought that JPEG into Lightroom. I then saved a copy of the JPEG at several different quality levels. Then, I opened the same image in CS2 and saved it at quality level 10. This is what I see in the file size:

Original out-of-camera: 2.6MB

In Lightroom:
Quality level 100: 3.2MB
Quality level 95: 3.2MB
Quality level 90: 1.8MB
Quality level 85: 1.8MB
Quality level 80: 1.1MB

In CS2:
Quality level 12: 3.1MB
Quality level 11: 1.8MB
Quality level 10: 1.1MB

So, it appears to me that:
level 80 ~= level 10
level 85-90 ~= level 11
level 100 ~= level 12

So, if you assume there are comparable JPEG compression engines behind these two apps, then it appears that level 80 on the 0-100 scale should be about equal to level 10 on the 0-12 scale. And, it appears that the scale is not linear, but quantized some. There is basically no difference between 85 and 90 or between 95 and 100.

Thanks for posting these results. This is extremely valuable info.
-Greg

jfriend · April 19, 2007

wellman wrote:

Thanks for posting these results. This is extremely valuable info.
-Greg

Ahhh, the activity in this thread reminds me. Now that I'm back from vacation, I should take Andy up on the offer to do some test prints at different 0-100 compression levels. I'll see if I can set that up this weekend so we can get some data from real prints, not just file sizes.

MyNameIsNeo · August 10, 2007

Results?
John & Andy,

I'd sure like to hear the results of the test printing. Many thanks!

Andy · August 10, 2007

MyNameIsNeo wrote:

John & Andy,

I'd sure like to hear the results of the test printing. Many thanks!

Waiting on Mr. Friend

cabbey · August 14, 2007

JPEG FAQ has some usefull information
From an old usenet faq: http://www.faqs.org/faqs/jpeg-faq/, which might be a bit 'dated' but it's still pretty accurate.

Subject: [4] How well does JPEG compress images?

Very well indeed, when working with its intended type of image (photographs
and suchlike). For full-color images, the uncompressed data is normally 24
bits/pixel. The best known lossless compression methods can compress such
data about 2:1 on average. JPEG can typically achieve 10:1 to 20:1
compression without visible loss, bringing the effective storage requirement
down to 1 to 2 bits/pixel. 30:1 to 50:1 compression is possible with small
to moderate defects, while for very-low-quality purposes such as previews or
archive indexes, 100:1 compression is quite feasible. An image compressed
100:1 with JPEG takes up the same space as a full-color one-tenth-scale
thumbnail image, yet it retains much more detail than such a thumbnail.

For comparison, a GIF version of the same image would start out by
sacrificing most of the color information to reduce the image to 256 colors
(8 bits/pixel). This provides 3:1 compression. GIF has additional "LZW"
compression built in, but LZW doesn't work very well on typical photographic
data; at most you may get 5:1 compression overall, and it's not at all
uncommon for LZW to be a net loss (i.e., less than 3:1 overall compression).
LZW *does* work well on simpler images such as line drawings, which is why
GIF handles that sort of image so well. When a JPEG file is made from
full-color photographic data, using a quality setting just high enough to
prevent visible loss, the JPEG will typically be a factor of four or five
smaller than a GIF file made from the same data.

Gray-scale images do not compress by such large factors. Because the human
eye is much more sensitive to brightness variations than to hue variations,
JPEG can compress hue data more heavily than brightness (gray-scale) data.
A gray-scale JPEG file is generally only about 10%-25% smaller than a
full-color JPEG file of similar visual quality. But the uncompressed
gray-scale data is only 8 bits/pixel, or one-third the size of the color
data, so the calculated compression ratio is much lower. The threshold of
visible loss is often around 5:1 compression for gray-scale images.

The exact threshold at which errors become visible depends on your viewing
conditions. The smaller an individual pixel, the harder it is to see an
error; so errors are more visible on a computer screen (at 70 or so
dots/inch) than on a high-quality color printout (300 or more dots/inch).
Thus a higher-resolution image can tolerate more compression ... which is
fortunate considering it's much bigger to start with. The compression
ratios quoted above are typical for screen viewing. Also note that the
threshold of visible error varies considerably across images.

Subject: [5] What are good "quality" settings for JPEG?

Most JPEG compressors let you pick a file size vs. image quality tradeoff by
selecting a quality setting. There seems to be widespread confusion about
the meaning of these settings. "Quality 95" does NOT mean "keep 95% of the
information", as some have claimed. The quality scale is purely arbitrary;
it's not a percentage of anything.

In fact, quality scales aren't even standardized across JPEG programs.
The quality settings discussed in this article apply to the free IJG JPEG
software (see part 2, item 15), and to many programs based on it. Some
other JPEG implementations use completely different quality scales.
For example:
* Apple used to use a scale running from 0 to 4, not 0 to 100.
* Recent Apple software uses an 0-100 scale that has nothing to do with
the IJG scale (their Q 50 is about the same as Q 80 on the IJG scale).
* Paint Shop Pro's scale is the exact opposite of the IJG scale, PSP
setting N = IJG 100-N; thus lower numbers are higher quality in PSP.
* Adobe Photoshop doesn't use a numeric scale at all, it just gives you
"high"/"medium"/"low" choices. (But I hear this is changing in 4.0.)
Fortunately, this confusion doesn't prevent different implementations from
exchanging JPEG files. But you do need to keep in mind that quality scales
vary considerably from one JPEG-creating program to another, and that just
saying "I saved this at Q 75" doesn't mean a thing if you don't say which
program you used.

In most cases the user's goal is to pick the lowest quality setting, or
smallest file size, that decompresses into an image indistinguishable from
the original. This setting will vary from one image to another and from one
observer to another, but here are some rules of thumb.

For good-quality, full-color source images, the default IJG quality setting
(Q 75) is very often the best choice. This setting is about the lowest you
can go without expecting to see defects in a typical image. Try Q 75 first;
if you see defects, then go up.

If the image was less than perfect quality to begin with, you might be able
to drop down to Q 50 without objectionable degradation. On the other hand,
you might need to go to a *higher* quality setting to avoid further loss.
This is often necessary if the image contains dithering or moire patterns
(see "[9] What are some rules of thumb for converting GIF images to JPEG?").

Except for experimental purposes, never go above about Q 95; using Q 100
will produce a file two or three times as large as Q 95, but of hardly any
better quality. Q 100 is a mathematical limit rather than a useful setting.
If you see a file made with Q 100, it's a pretty sure sign that the maker
didn't know what he/she was doing.

If you want a very small file (say for preview or indexing purposes) and are
prepared to tolerate large defects, a Q setting in the range of 5 to 10 is
about right. Q 2 or so may be amusing as "op art". (It's worth mentioning
that the current IJG software is not optimized for such low quality factors.
Future versions may achieve better image quality for the same file size at
low quality settings.)

If your image contains sharp colored edges, you may notice slight fuzziness
or jagginess around such edges no matter how high you make the quality
setting. This can be suppressed, at a price in file size, by turning off
chroma downsampling in the compressor. The IJG encoder regards downsampling
as a separate option which you can turn on or off independently of the Q
setting. With the "cjpeg" program, the command line switch "-sample 1x1"
turns off downsampling; other programs based on the IJG library may have
checkboxes or other controls for downsampling. Other JPEG implementations
may or may not provide user control of downsampling. Adobe Photoshop, for
example, automatically switches off downsampling at its higher quality
settings. On most photographic images, we recommend leaving downsampling
on, because it saves a significant amount of space at little or no visual
penalty.

For images being used on the World Wide Web, it's often a good idea to
give up a small amount of image quality in order to reduce download time.
Quality settings around 50 are often perfectly acceptable on the Web.
In fact, a user viewing such an image on a browser with a 256-color display
is unlikely to be able to see any difference from a higher quality setting,
because the browser's color quantization artifacts will swamp any
imperfections in the JPEG image itself. It's also worth knowing that
current progressive-JPEG-making programs use default progression sequences
that are tuned for quality settings around 50-75: much below 50, the early
scans will look really bad, while much above 75, the later scans won't
contribute anything noticeable to the picture.

wannabe1979 · August 29, 2007

Here is a question for everyone. I'm used to the 1-12 scale in PS too and see where most labs recommend level 10. Why wouldn't you want to save photos at the best level 12? I had been saving at level 12 until I had read my labs' policies on uploads, but they never say why 10? Afterall, isn't 12 better?

Andy · August 29, 2007

wannabe1979 wrote:

Here is a question for everyone. I'm used to the 1-12 scale in PS too and see where most labs recommend level 10. Why wouldn't you want to save photos at the best level 12? I had been saving at level 12 until I had read my labs' policies on uploads, but they never say why 10? Afterall, isn't 12 better?

In theory, sure 12 is better. But I don't know a single soul who can tell the difference between 10 and 12 in a print.

People choose 10 for easier uploads, mostly. Today's cameras have bigger chips, and produce bigger files. 12 --> 10 is a huge difference in file size. If you are uploading one or two, no big deal, but to some, uploading 500 files, it's a giant difference.

jfriend · August 29, 2007

wannabe1979 wrote:

Here is a question for everyone. I'm used to the 1-12 scale in PS too and see where most labs recommend level 10. Why wouldn't you want to save photos at the best level 12? I had been saving at level 12 until I had read my labs' policies on uploads, but they never say why 10? Afterall, isn't 12 better?

Level 12 has an additional cost over level 10. It costs you a little more time to save it. It takes about 3x the disk space. It takes 3x the time to upload. It takes longer to load into a viewer or editor. It takes longer to copy. It takes longer to backup or archive.

If none of these additional costs affect you in any way, then feel free to use level 12. But, most people find that, with no visible difference in print quality, they'd rather have the efficiency advantages of level 10 files.

wannabe1979 · August 30, 2007

Thanks Andy,

That clears that question up.

Andy wrote:

In theory, sure 12 is better. But I don't know a single soul who can tell the difference between 10 and 12 in a print.

People choose 10 for easier uploads, mostly. Today's cameras have bigger chips, and produce bigger files. 12 --> 10 is a huge difference in file size. If you are uploading one or two, no big deal, but to some, uploading 500 files, it's a giant difference.

What quality level on the 0-100 scale?

Comments