Digital BW, The Print

Reading an old thread on levels and grain I had some questions...

Regarding the difference between levels and curves... aren't they doing exactly the 
same thing, just with different interfaces?  Additionally, curves gives you more 
flexibility than levels (although levels has some features curves doesn't)... but, the end 
result as far as the manipulation that is done to the image is identical.  So, one should 
not be "better" than the other WRT image degradation and grain.

So if I was to put a quantitative number to the amount of contrast I need to add to an 
image... lets say 20 points of contrast (whatever that may be).  If I add 20 points of 
contrast on an image using levels and 20 points of contrast in curves on a copy of the 
same image, one image isn't going to be less degraded or less grainy because I used 
curves instead of levels.

Am I correct on this?  This is why I'm not sure I understand why people say levels is 
more destructive and causes more grain than curves.

Shouldn't the main point be to have as much contrast or do as much manipulation in 
levels or curves, making sure not to clip information unnecessarily?

Jeff
http://www.jeffsingerphotography.com

> From: jsinger986 [mailto:jsphotos@...]
>
> Regarding the difference between levels and curves... aren't they
> doing exactly the
> same thing, just with different interfaces?  Additionally, curves
> gives you more
> flexibility than levels (although levels has some features curves
> doesn't)... but, the end
> result as far as the manipulation that is done to the image is
> identical.  So, one should
> not be "better" than the other WRT image degradation and grain.

I think you're correct. As far as I know, all Levels does is put a prettier
user interface on the underlying Curves engine. This makes it easier to do
certain very useful things, like adjust the color balance with an
eyedropper, but it can't do other things, like boosting contrast with an S
curve.

--

Ciao,               Paul D. DeRocco
Paul                mailto:pderocco@...

Levels is more intuitive for doing something like expanding your 
tonality, but you are right, if you only do one operation Levels 
should't be any more destructive than any other curve adjustment, BUT 
you will usually follow up a levels adjust with a s-curve for 
contrast which adds a second curve adjust so the sum of your 
destruction is going to be greater.  Better to do everything with one 
curve if possible.  I personally don't like using the "gamma slider" 
for contrast because it is only modifying the middle point on a 3 
point curve while a simple s-curve gives you much more control and 
flexiblity WRT the tonal region where you are actually modifying the 
gamma.  There really aren't really "20 pts of contrast", just the 
local slope of your curve adjustment.  

mark

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, "jsinger986" 
<jsphotos@t...> wrote:
> Reading an old thread on levels and grain I had some questions...
> 
> Regarding the difference between levels and curves... aren't they 
doing exactly the 
> same thing, just with different interfaces?  Additionally, curves 
gives you more 
> flexibility than levels (although levels has some features curves 
doesn't)... but, the end 
> result as far as the manipulation that is done to the image is 
identical.  So, one should 
> not be "better" than the other WRT image degradation and grain.
> 
> So if I was to put a quantitative number to the amount of contrast 
I need to add to an 
> image... lets say 20 points of contrast (whatever that may be).  If 
I add 20 points of 
> contrast on an image using levels and 20 points of contrast in 
curves on a copy of the 
> same image, one image isn't going to be less degraded or less 
grainy because I used 
> curves instead of levels.
> 
> Am I correct on this?  This is why I'm not sure I understand why 
people say levels is 
> more destructive and causes more grain than curves.
> 
> Shouldn't the main point be to have as much contrast or do as much 
manipulation in 

> levels or curves, making sure not to clip information unnecessarily?
> 
> Jeff
> http://www.jeffsingerphotography.com

While it's possible to get very similar results from using the curves 
dialogue and the levels dialogue, they are not the same and do not 
function the same.

When you pull in the white point slider in the levels dialogue, you 
are clipping all the data above the slider.

If you do not move the end points in the curves dialogue and increase 
the contrast,say, with a traditional "s" curve you are reassigning 
values in the
image but not clipping any.

There is a difference and curves is much less destructive than levels.

That said, the notion that a combed histogram is bad is much too much 
overstated, and generally speaking the problems attributed to them 
arise from
other issues.

J.Z.

> From: Jon Zax [mailto:lotus@...]
>
> While it's possible to get very similar results from using the curves
> dialogue and the levels dialogue, they are not the same and do not
> function the same.
>
> When you pull in the white point slider in the levels dialogue, you
> are clipping all the data above the slider.
>
> If you do not move the end points in the curves dialogue and increase
> the contrast,say, with a traditional "s" curve you are reassigning
> values in the
> image but not clipping any.
>
> There is a difference and curves is much less destructive than levels.

That's true. That's one of the things that you can do with Curves that you
can't do with Levels. Indeed, it's useful to bring up Levels, move the
endpoints inward to the edges of the histogram, note the numeric values,
cancel it, and then use those numbers as the starting point for the Curves
dialog. (Now that PSCS has full 16-bit support, there's no point.)

But if you do the same thing with Curves that you do with Levels, i.e., move
the endpoints inward, then you do indeed get the same results. I think there
was some speculation that the underlying Curves engine would somehow result
in less posterization or noise than Levels.

> That said, the notion that a combed histogram is bad is much too much
> overstated, and generally speaking the problems attributed to them
> arise from other issues.

The combing may manifest itself as posterization, in a quiet image. But in a
noisy image, the noise will drown out any posterization.

--

Ciao,               Paul D. DeRocco
Paul                mailto:pderocco@...

Jeff,

While I cannot add anything to this discussion, I would just like to 
say that the woman whose face graces your home page is absolutely 
stunning! Just my $0.02

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, "jsinger986" 
<jsphotos@t...> wrote:
> Reading an old thread on levels and grain I had some questions...
> 
> Regarding the difference between levels and curves... aren't they 
doing exactly the 
> same thing, just with different interfaces?  Additionally, curves 
gives you more 
> flexibility than levels (although levels has some features curves 
doesn't)... but, the end 
> result as far as the manipulation that is done to the image is 
identical.  So, one should 
> not be "better" than the other WRT image degradation and grain.
> 
> So if I was to put a quantitative number to the amount of contrast 
I need to add to an 
> image... lets say 20 points of contrast (whatever that may be).  If 
I add 20 points of 
> contrast on an image using levels and 20 points of contrast in 
curves on a copy of the 
> same image, one image isn't going to be less degraded or less 
grainy because I used 
> curves instead of levels.
> 
> Am I correct on this?  This is why I'm not sure I understand why 
people say levels is 
> more destructive and causes more grain than curves.
> 
> Shouldn't the main point be to have as much contrast or do as much 
manipulation in 

> levels or curves, making sure not to clip information unnecessarily?
> 
> Jeff
> http://www.jeffsingerphotography.com

With all due respect, you are wrong here.  A two point linear curve 
is exactly the same as a Levels adjust with no gamma slider tweak.  
Someone speculated that PS uses a different curve subroutine for 
Levels than Curves, which doesn't make sense, but is of course 
possible, but that would be the only reason one would be more 
destructive than the other.  That a Curve adjust is preferable does 
have merit, especially since you can usually apply just one 
adjustment curve instead of two (after Levels a curve adjust is 
usually required).  The reason to use Levels is generally to expand 
your data to fill the entire tonal range, not to clip your data, and 
it is easier to do this in levels since you see the entire tonal 
range in the histogram and can easily set your black and white points 
accordingly.

Just out of curiousity, how do you think Levels works in PS?  
Mathematically you shift your data to allign your black point and 
then scale the data so your white point comes out right (or visa-
vera).  The simplest way to do this is with a linear curve adjust.  

A four point "s-curve" will not be the same as a Levels end point 
adjust, but a linear Curve adjust with appropriate  endpoints will be 
identical. 

mark


--- In DigitalBlackandWhiteThePrint@yahoogroups.com, Jon Zax 
<lotus@i...> wrote:
> While it's possible to get very similar results from using the 
curves 
> dialogue and the levels dialogue, they are not the same and do not 
> function the same.
> 
> When you pull in the white point slider in the levels dialogue, you 
> are clipping all the data above the slider.
> 
> If you do not move the end points in the curves dialogue and 
increase 
> the contrast,say, with a traditional "s" curve you are reassigning 
> values in the
> image but not clipping any.
> 
> There is a difference and curves is much less destructive than 
levels.
> 
> That said, the notion that a combed histogram is bad is much too 
much 

> overstated, and generally speaking the problems attributed to them 
> arise from
> other issues.
> 
> J.Z.

I don't know why this question comes up all the time or why it is so 
difficult to grasp, but just to put it to rest...

1) open image file (p1)
2) duplicate image (p2)
3) apply Levels adjustment to endpoints to p1 (input levels)
4) apply curve adjust to p2 with same input levels at the end points
5) Calculate -> Subtract p1 from p2
6) Check histogram and note that you have a mean of zero and a std. 
deviation of zero
7) Correctly conclude that Levels and Curve adjustments are identical 
in outcome.

Since they both have the same algorithm options it should have been 
obvious that they do the same thing.

now I think that should settle it...

mark

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, "Mark Hahn" 
<markhahn2000@y...> wrote:
> With all due respect, you are wrong here.  A two point linear curve 
> is exactly the same as a Levels adjust with no gamma slider tweak.
...

it does indeed.

-----Original Message-----
From: Mark Hahn [mailto:markhahn2000@...] 
Sent: Thursday, November 06, 2003 10:43 PM
To: DigitalBlackandWhiteThePrint@yahoogroups.com
Subject: [Digital BW] The Proof - Re: Levels and grain... again...


I don't know why this question comes up all the time or why it is so 
difficult to grasp, but just to put it to rest...

1) open image file (p1)
2) duplicate image (p2)
3) apply Levels adjustment to endpoints to p1 (input levels)
4) apply curve adjust to p2 with same input levels at the end points
5) Calculate -> Subtract p1 from p2
6) Check histogram and note that you have a mean of zero and a std. 
deviation of zero
7) Correctly conclude that Levels and Curve adjustments are identical 
in outcome.

Since they both have the same algorithm options it should have been 
obvious that they do the same thing.

now I think that should settle it...

mark

--- In DigitalBlackandWhiteThePrint@yahoogroups.com, "Mark Hahn" 
<markhahn2000@y...> wrote:
> With all due respect, you are wrong here.  A two point linear curve 
> is exactly the same as a Levels adjust with no gamma slider tweak.
...



Yahoo! Groups Sponsor	

ADVERTISEMENT
 
<http://rd.yahoo.com/M=267637.4116730.5333196.1261774/D=egroupweb/S=1705
019182:HM/A=1853619/R=0/SIG=11tijtneq/*http://www.netflix.com/Default?mq
so=60178356&partid=4116730> click here	
 
<http://us.adserver.yahoo.com/l?M=267637.4116730.5333196.1261774/D=egrou
pmail/S=:HM/A=1853619/rand=585792023> 	

Please visit the Group Homepage to check the Files, Bookmarks, Polls and
other resources as they are often being updated. The page is at:

http://groups.yahoo.com/group/DigitalBlackandWhiteThePrint

If you wish to receive no emails or just a daily digest, or you wish to
unsubscribe, please edit your Membership preferences by visiting this
same page.

Please follow these basic guidelines:
- Include your full name with your message.
- Include the address of your website, if you have one.
- As threads develop, trim off excess portions of earlier messages to
keep them short.
- As the topic of a thread changes remember to change the subject
header.
- Good manners are required at all time. No personal attacks or flames
- Complete your Yahoo profile.
- Before posting a question, search the message archives and the various
resources on the homepage. 




Your use of Yahoo! Groups is subject to the Yahoo! Terms of
<http://docs.yahoo.com/info/terms/> Service. 




[Non-text portions of this message have been removed]