Lenses and sharpening

On 9/18/2014 2:57 AM, Floyd L. Davidson wrote:
Eric Stevens wrote:
On Thu, 18 Sep 2014 00:01:24 -0400, nospam
wrote:
floyd cannot acknowledge that there are other completely valid meanings.

If you want to argue with what he said then you have to use the same
meaning that he did.

I would point out that the average 10 year old tries to
use that mechanism just about 2 times, but usually
understands before a third attempt.

It's dishonest, lacks integrity and results in loss of
credibility.


But noaspam is not your average ten year old.

--
PeterN

On 2014-09-18 17:40:03 +0000, PeterN said:

On 9/17/2014 7:34 PM, Floyd L. Davidson wrote:
PeterN wrote:
On 9/17/2014 4:02 AM, Floyd L. Davidson wrote:

snip

The topic was sharpening, and the differences in ways to
do that. Abobe's programs are not even close to the
only way to sharpen. In fact *most* users that actually
get into the more sophisticated aspects of sharpening
cease using anything that Abobe provides for that
purpose, and shift to better tools.

Generic atributes of sharpen tools can and should be
discussed absent references to specific implementations.
When specific attributes are discussed it doesn't make a
great deal of sense to look at low end products designed
to appeal to the lowest common denominator, as might
well be discussed in your "Abobe Tools for Dummies"
manual.


OK, I am an admitted oversharpener.

What tools would you recommend, and why.
I request that the why be in simplistic terms.

Given what you like, it's really easy!

A good high pass sharpen tool and almost any USM tool.

Where everyone else has to look is into "smart sharpen"
tools. Usually these use an edge detection scheme to
mask off everything else, and then sharpen only the
edges. That tends to reduce the amount of noise that
gets sharpened.

People taking pictures of stars like to use
Richardson-Lucy Deconvolutional sharpen, and in general
wavelet sharpen does something similar. For general
photography they amount to a lot of work for very little
benefit. In particular they don't benefit your style at
all!

I'd expect you would be relatively happy with just
a high pass filter style of sharpen. The harshness
that can be produced with a little effort can be
easily controlled if the tool has adjustable parameters
for radious, sigma, and amount. Unfortunately many
sharpen tools only give you control of the amount.
Most UnSharpMask tool have more parameters, and it
will pretty much to the same thing.


I have found that using high pass on the luminiscence layer in LAB
tends to minimize halos.

Actually it is a good idea to do any/all/most sharpening on a
luminosity layer, LAB or not.

BTW: Using LAB doesn’t seem to be working for you.

When I am in the mood, I will use a light application of surface blur,
with a mask, to eliminate noise. Having said that, noise doesn't bother
me as much as it seems to bother others in this group.

Why? You have the NIK Suite, use Define, either applying a total
profile, or selectively with a brush.
You wallow in over-sharpening artifacts, high ISO noise, JPEG
artifacts, & dirty sensors, and over cooking post processing.

…but to each their own, if you enjoy noise so much most of your work
has it in spades. With a D800 no less.


--
Regards,

Savageduck

In article , PeterN
wrote:

to put it another way, i can change the amount of unsharp mask on an
image i processed a year ago, without having to redo *anything* else i
did. all of the retouching, white balance, etc. remain the same (unless
i choose to adjust those too).

But you can't do that once the image has been exported.

nonsense, of course i can, and have.

But not to the exported image,

so what?

the fact is i can change anything without needing to redo the entire
process.

floyd said the entire process would need to be redone. he's wrong.

not with the original sidecar file.

there is no sidecar file.

A documented demonstration would support your position.

any book on lightroom documents exactly what i've been saying.

In article 2014091811153644303-savageduck1@REMOVESPAMmecom,
Savageduck wrote:

I have found that using high pass on the luminiscence layer in LAB
tends to minimize halos.

Actually it is a good idea to do any/all/most sharpening on a
luminosity layer, LAB or not.

not always, since the conversion to lab and back is not lossless.

On 2014-09-18 18:17:42 +0000, nospam said:

In article 2014091811153644303-savageduck1@REMOVESPAMmecom,
Savageduck wrote:

I have found that using high pass on the luminiscence layer in LAB
tends to minimize halos.

Actually it is a good idea to do any/all/most sharpening on a
luminosity layer, LAB or not.

not always, since the conversion to lab and back is not lossless.

There I was referring to sharpening in general on a luminosity layer,
not necessarily LAB, personally I hardly ever use LAB.

--
Regards,

Savageduck

In article 2014091812115170933-savageduck1@REMOVESPAMmecom,
Savageduck wrote:

I have found that using high pass on the luminiscence layer in LAB
tends to minimize halos.

Actually it is a good idea to do any/all/most sharpening on a
luminosity layer, LAB or not.

not always, since the conversion to lab and back is not lossless.

There I was referring to sharpening in general on a luminosity layer,
not necessarily LAB, personally I hardly ever use LAB.

do you mean a luminosity blend mode? that can be useful.

On 2014-09-18 19:12:39 +0000, nospam said:

In article 2014091812115170933-savageduck1@REMOVESPAMmecom,
Savageduck wrote:

I have found that using high pass on the luminiscence layer in LAB
tends to minimize halos.

Actually it is a good idea to do any/all/most sharpening on a
luminosity layer, LAB or not.

not always, since the conversion to lab and back is not lossless.

There I was referring to sharpening in general on a luminosity layer,
not necessarily LAB, personally I hardly ever use LAB.

do you mean a luminosity blend mode? that can be useful.

Yes.

--
Regards,

Savageduck

On 2014-09-18 17:59:38 +0000, PeterN said:


A documented demonstration would support your position. Since you don't
want to use what is allegedly your image, everybody here has my
prmission to play with the crappy image I posted. Indeed I am placing
it in th epublic domain solely for purposes of expirimentation.
Now either prove your point, or STFU.

I have just emailed a link to a fully non-destructive working version
of your Central Park image. Opened as a Smart Object and all other
adjustments done with NIK stuff & one return to ACR (by double clicking
on the adjustment layer thumbnail) Not merged or flattened. To reopen
the filter dialogs double click on the effect name. To change the
blending mode click on the icon to the right of the name. I didn't crop
this time.
It is an 800+MB PSD

I saved a version for web first.
http://adobe.ly/1Dm0nII

--
Regards,

Savageduck

On 18 Sep 2014 15:07:50 GMT, Sandman wrote:

In article , Eric Stevens wrote:

Eric Stevens:
On this occasion, at least, Floyd most certainly knows what he
is talking about.

Sandman:
Haha, no.

Eric Stevens:
A problem seems to be that very few other people seem to.

Sandman:
Well, we all know that YOU rarely have the first clue about what
you're talking about, so I have no problem understanding why
you're here supporting ignorant Floyd.

Floyd's usage is strictly in accordance with
http://en.wikipedia.org/wiki/Reversi...rmodynamics%29

Great input, if the topic had been about thermodynamics.

The article is about thermodynamics. The concepts and the mathematics
and the concept of entropy apply to a number of topics including
information processing which, in turn, includes image processing.

as it applies information theory. If you think there is no room for
reversible processes in information theory see
http://tinyurl.com/otp5pug

Why can't you read?

1. Floyd thinks that HPS + JPG compression can be reversed. It can not.

He never claimed that the JPG can be reversed. As I have already
written, he said that the original image can be recovered after
sharpening by HPS even after the image has been saved as a JPG.

2. Floyd thinks that a non-destructive workflow is not a reversible
process, it is.

It's not a reversible process in the way that he used the term.

You previously quoted from a dictionary. Here is what the New Shorter
Oxford English Dictionary says of this particular usage of
'reversible':

"2 Physics. Of a change or process: that is capable of complete
and detailed reversal; spec designating or undergoing an ideal
change in which a system is thermodynamic equilibrium at all
times."

As nospam has so often told us, Lightroom (and other software using
side car files) do not actually change the file being edited until it
is in the process of being exported. In most case, all you see on the
screen is a simplified simulacrum of what the edited file will look
like, when the editing instructions are executed.

Once you export the file - that's it. You cannot reverse the changes.
All you can do is edit the original all over again but this time
slightly differently.

Now it's interesting that Lightroom does incorporate something a
little bit like the reversible process that Floyd was talking about
but neither nospam or Savageduck seem to realise the fact.
See http://tinyurl.com/p5sus42 From blur to sharpness on the one
slider. But this is not actually a reversible process: it's a change
in the instruction to the final edit which will only be executed when
the image is exported.

I do not know the type of sharpening used by Lightroom or whether or
not it is truly reversible. If it uses HPS then it is possible to take
a sharpened exported image and then reprocess it using the function
above to remove the sharpening and return it to it's original state.
However, it will not be possible to do this if the sharpening uses
USM: there will always be artifacts of the original sharpening
remaining.

Just the plain facts.

I'm afraid they are not as plain or as straightforward as you would
like to have them.
--

Regards,

Eric Stevens

On Thu, 18 Sep 2014 11:51:54 -0400, nospam
wrote:

In article , Eric Stevens
wrote:

Floyd's usage is strictly in accordance with
http://en.wikipedia.org/wiki/Reversi...rmodynamics%29
as it applies information theory. If you think there is no room for
reversible processes in information theory see
http://tinyurl.com/otp5pug

since when does editing digital images on a computer become a
thermodynamics problem?

As soon as you apply information processing theory.
http://en.wikipedia.org/wiki/Entropy...tion_theory%29
--

Regards,

Eric Stevens