Lenses and sharpening

In article , Eric Stevens
wrote:

All image effects in Photoshop are 100% reversible.

image effects in photoshop *can* be reversible, but they are not
always because photoshop is at its core, a pixel editor. you have to
take additional steps for something to be non-destructive.

True - but the undo function in modern software is the perfect
definition
of a reversible process. It can reverse any effect done by anything in
Photoshop.

undo only exists while you're using the app and it's within its undo
history. at some point, it won't be undoable, generally when the file
is saved but sometimes well before that, depending on the app. that's
the whole problem with a traditional workflow.

with a non-destructive workflow, anything can be reversed at any time
because the original data remains unaltered, including long after the
app has been quit and even archived and later worked on with a
different computer.

the difference is very important, and one which floyd and eric do not
fully understand (or at all).

Of course I understand it!

based on what you've written in this thread and in others, no you do
not.

You should stop and think a little about what you are actually doing.
You are NOT editing the image until you export it in some way.

proof that you do not fully understand it.

Then you explain what happens.

i have many times. go revisit the thread from a month ago about files.

or peruse adobe's documentation and/or books/videos on lightroom.

the information is out there.

In article , Eric Stevens
wrote:

This discussion started when in response to Kevin McMurtrie Floyd
wrote in Message-ID:

"The digital form of unsharp mask is the inverse of a blur.
There's both a frequency (diameter) and an intensity.

Not the case. It is the high pass sharpen tool that is the inverse
of blur. They can use the exact same algorithm with different
parameters. Using one and then the other virtually reverses the
results.

UnSharpMask is not reversible."

You completely failed to understand what Floyd was talking about and
have added your inestimable contributions ever since.

once again, in a non-destructive workflow, unsharp mask along with
everything else *is* reversible. this is a fact no matter how much you
and floyd argue otherwise.

It's not a reversible process as it is conventionally defined.

yes it is.

someone can unsharp mask today and remove it tomorrow and put it back
the day after that.

the following week, that same someone can remove all colour (convert to
b/w) and the week after that, can reverse that, exactly how it was in
the original image, because it *is* the original image.

that's what just about everyone would call reversible.

In article , Eric Stevens
wrote:

as i said before, in a non-destructive workflow, a user can do any
adjustment they want, including usm, and then at some later point,
reverse it.

that's reversible, no matter how much you try to argue it isn't.

Try reversing it after you have exported the image.

you obviously don't understand how it works.

You explain it then.

i have, many times.

In article , Eric Stevens
wrote:

some time before that, maybe two years ago, you said there could never
be an internal wifi antenna on an slr because the camera body is metal.
that too is wrong. there are over a billion devices with a metal body,
including laptops, tablets and smartphones, and work just fine with
wifi. all it takes is an antenna aperture or putting the antenna on the
outside rather than the inside (and still not visible). it's trivial to
do.

Gee! An internal antennae works fine as long as it's on the outside.

learn to read before you say more stupid stuff.

i never said an internal antenna is on the outside.

" ... putting the antenna on the outside rather than the inside ... "

that is *after* the word 'or'.

do you not understand what 'or' means?

do you see the word 'or' in there?? what do you think 'or' means?

It means you should have thought a little before writing that. :-)

what i wrote is perfectly clear.

nevertheless, since you're having trouble, here it is again:

all it takes is
1) an antenna aperture
or
2) putting the antenna on the outside rather than the inside
it's trivial to do

What is the 'it' that "is all it takes'?

designing the product.

I know: " ... getting an internal antennae to work ... "

1 & 2 are two separate solutions. thus the word 'or'.

Even solution 1 is a form of getting it to the outside.

no it isn't. the antenna is internal in 1 and external in 2.

exhibit a: ipad, macbook pro
exhibit b: iphone 4s

On 2014-09-19 02:30:16 +0000, Eric Stevens said:

On Thu, 18 Sep 2014 16:23:48 -0700, Savageduck
wrote:

On 2014-09-18 21:58:25 +0000, Eric Stevens said:

On Thu, 18 Sep 2014 07:52:56 -0700, Savageduck
wrote:

On 2014-09-18 09:03:35 +0000, Eric Stevens said:

On Wed, 17 Sep 2014 21:32:17 -0700, Savageduck
wrote:
On 2014-09-18 04:17:59 +0000, (Floyd L. Davidson) said:
Savageduck wrote:
On 2014-09-18 01:00:46 +0000, Eric Stevens said:
On Wed, 17 Sep 2014 06:05:14 -0700, Savageduck
wrote:
On 2014-09-17 09:22:00 +0000, Eric Stevens said:
On Tue, 16 Sep 2014 22:27:43 -0700, Savageduck
wrote:
On 2014-09-17 04:08:19 +0000, Eric Stevens said:
On Tue, 16 Sep 2014 07:53:15 -0700, Savageduck
wrote:
On 2014-09-16 10:36:29 +0000, (Floyd L. Davidson) said:

--- snip ---

The reverse process performed on a lossy, compressed JPEG is not going
to reverse the HPF to return to the original state. That was lost once
the save was executed.
That's why I never included a conversion to JPG in
my example of a
reversible process.
Â...but that genius Floyd did.

--- snip ---

No one who understood what we were trying to talk about would claim
that a JPG conversion is a reversible process.
Â...but that genius Floyd did.
I've had a look and I cant see where. Could you refer
me to the
message?

With pleasure.

That wasnâEUR(Tm)t too tough to find:
Posted: Mon, 15 Sep 2014 13:44:18 -0500
Message ID:

Wherein Floyd stated the following:

âEURoeA non-destructive workflow means you can *undo* and then *redo*.

That is not a reversible function.

For example, you can add sharpening with a high pass
sharpen tool to an image, save it as a JPEG, send it to
someone else, and they can use a blur tool to reverse
the sharpen.

If the sharpening is done with UnsharpMask that cannot
be done. USM is not reversible.âEUR?

Note, the words, âEURoesave it as a JPEG,âEUR?.

As I said, that genius Floyd did.

So we now we know you can't read.

What I said was that *high pass sharpen is reversible*.
It is, even if a few people are unable to either understand
or accept that it is.

From a lossy, compressed JPEG?
You did say ?save it as a JPEG? didn?t you?
Have you also developed the mathematics to reverse the compression and
loss in that High-Pass sharpened file, so that you can return it to its
original state?

HPS might well be reversible, but returning the file to its truly
original state after being saved as a JPEG is improbable.

It's not the file that is being returned to the original state: it's
the sharpening.

Holy spluttering wake up call!!
Eric, just take a look at what you have written and tell me you aren?t
grasping at straws?

I'm talking about the very subject that was being discussed at the
start of this now entirely wrecked thread.

If you actually start to explore the potential of Photoshop, it will be
well worth all you have suffered through.

I've been using Photshop for more than 6 months. I'm not suffering
from the experience.

The experience I inferred you were suffering through was this "wrecked
thread" not anything you have got from using PS for the last 6 months.

However, I have the tools to do that, with HSP, and even USM regardless
of how much you stamp your feet and say I can?t.

But you are not reversing it: you are reverting and replacing it. Not
the same thing.

Actually, I am not replacing anything, and in an odd way I am not
?reversing? the effect of any of those sharpening methods. I am
readjusting the parameters, and in making that readjustment I can end
up in any state, including the original so I can produce another
version. In fact I am not even going to go through the pointless
esoteric exercise of working on that saved JPEG, it will remain as a
snapshot of the state of the working image file when it was saved.

If you are still talking about Lightroom, the edited image is only
created when it is exported.

I have only been talking about Photoshop, because you and Peter have
yet to include Lightroom in your workflow. Have you not been reading
what I have been saying?

I don't have to be familar with all the ins and outs of Lightroom to
understand the principles behind it's working. It's far from being the
only software with the ability to edit instructions via an external
file.

Agreed. However, I have been talking about Photoshop methods, because
the two of you, who are CC subscriber have yet to employ Lightroom as
part of your workflow paired with PS. I am also aware that there is
other software for image editing out there.

You can adjust what you like before that but all you are changing is a
simplified simulacrum of what the final
image will be. When you export it, whatever you get is fixed. You can
of course treat it as a new image and re-edit it, but that comes back
to Floyd's original point: if in the first edit you sharpened the
image with USM you will not be able to restore the exact sharpness of
the original by editing the exported image. USM is not fully
reversible.

However, Floyd claims the following which is a theoretical exercise
destined to failu
"“For example, you can add sharpening with a high pass
sharpen tool to an image, save it as a JPEG, send it to
someone else, and they can use a blur tool to reverse
the sharpen.”

He can restore the sharpen even if he can't fully restore the JPG.

That's nice.

...and in PHOTOSHOP I can reverse either the HPS or the USM for real.

I revisited Peter's Central Park shot and did a 100% non-destructive &
reversible/readjustable rendition of that using two Smart object
layers, several NIK tools and one revisit to ACR when in the adjustment
layer.
https://dl.dropbox.com/u/1295663/FileChute/screenshot_908.jpg

I can use the layer mask which is part of the *Smart Filter* and
selectively remove or reduce the filter?s effect as I please. You still
have something to learn regarding just how you deal with *Smart
Objects* & *Smart Filters* in Photoshop. All I have to do is double
click on the *Smart Filter* layer and the filter dialog will reopen and
I can re adjust away, and if I don?t want that particular sharpening
and want to return to an original state without the issues presented by
a lossy, compressed JPEG, I can do one of three things; return
parameters to their original, pre-effect state, turn off the visibility
of the adjustment layer (you can do that you know), or just delete the
layer.

This type of legerdemain is not what is meant by fully reversible in
the context that Floyd was using it.

Because Floyd doesn't use any Adobe software, and has no idea what any
of us who use it the way it was intended to be used are talking about.
So he just repeats his long winded arcane verbiage. That "type of
legerdemain" (I see you had your thesaurus handy) is exactly what makes
for a truly non-destructive workflow, where each element in that
workflow can be adjusted, and dare I say it, reversed.

All of which is fine, but the idea of a reversible process predates
all of us. Even if you can reverse edits in Lightroom you do not do it
via a reversible process.

Did you not notice that I haven't been talking about Lightroom and its
particular variety of non-destructive editing/adjusting?
I have been sticking to PHOTOSHOP methods.

Whatever I do, I have reversed what was presented, reverting to the
original uncompressed state, without any compression artifact issues.
As I have said before, that JPEG was nothing but aversion, or a
snapshot of the current state of the working image file. I can go on to
have many versions all of the same quality, none showing generational
degradation.

Unlike Floyd, you actually have the opportunity to see for yourself
that what I am saying is true, Just open PS and experiment. Learn
something about the capability of the software you have installed and
are allegedly using.

I know what you are saying and I understand what you are saying, but
you are missing the point.

Then open PS and start learning.

How many times do I have to tell you that it's got nothing to do with
that?

It's got everything to do with that. For you at least.

I suggest that you actually start thinking about what is happening
when you do these things in the context of what Floyd originally said.

Actually I don't need to do anything in the context of Floyd says at any time.


--
Regards,

Savageduck

On Thu, 18 Sep 2014 19:14:58 -0400, nospam
wrote:

In article , Eric Stevens
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.

Not strictly correct:

it is completely correct.

we went through this about six months ago, and apparently will again.

https://www.ledet.com/margulis/ACT_p...LAB-damage.htm

" I have always thought that moving from either CMYK or RGB to Lab
and back was a damage free process, that is, you would end up with
the same color co-ordinates when you arrived back from Lab mode.

"RGBLABRGB is damage free, but CMYKLABCMYK is not. The damage
isn't all that great, so in many images it pays to come out of CMYK
so as to take advantage of LAB's strengths; sharpening, however, is
not one of these cases.
....
Dan Margulis"

you clearly don't understand what you're reading, since that link
agrees with what i said!

as the other posts in your link clearly show, dan margulis is wrong (as
he is about a lot of things).

read the *very* next post, from chris murphy,

Converting to and from Lab has never been a damage free process.

He's talking about quantization errors. You get those with most
operations. It's not a problem unique to colour space conversions.

and the one after that,

RGBLABRGB is damage free, but CMYKLABCMYK is not.
I disagree. If you start out with all of the same spaces for RGB and
CMYK, and use only those spaces - then convert to and from Lab, you
will get some quantization errors with both.

... again

and andrew rodney's post:
RGBLABRGB is damage free
You1re not serious are you Dan?
Take an RGB file. Duplicate it. Do an RGB to LAB to RGB conversion
and subtract the two. You can turn on or off the 8 bit dither. When
you subtract the two and create a new document and look at the
Histogram in Levels, you will see there certainly is data loss and a
change. Move the sliders of the Levels Histogram over and you1ll see
the effects of what differences between the two files you produced.
Are you saying this isn1t data loss?

that test is trivial to do. try it yourself.

--

Regards,

Eric Stevens

In article , Eric Stevens
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.

Not strictly correct:

it is completely correct.

we went through this about six months ago, and apparently will again.


https://www.ledet.com/margulis/ACT_p...ACT-LAB-damage.
htm

" I have always thought that moving from either CMYK or RGB to Lab
and back was a damage free process, that is, you would end up with
the same color co-ordinates when you arrived back from Lab mode.

"RGBLABRGB is damage free, but CMYKLABCMYK is not. The damage
isn't all that great, so in many images it pays to come out of CMYK
so as to take advantage of LAB's strengths; sharpening, however, is
not one of these cases.
....
Dan Margulis"

you clearly don't understand what you're reading, since that link
agrees with what i said!

as the other posts in your link clearly show, dan margulis is wrong (as
he is about a lot of things).

read the *very* next post, from chris murphy,

Converting to and from Lab has never been a damage free process.

He's talking about quantization errors. You get those with most
operations. It's not a problem unique to colour space conversions.

in other words, switching to lab and back is *not* lossless.


and the one after that,

RGBLABRGB is damage free, but CMYKLABCMYK is not.
I disagree. If you start out with all of the same spaces for RGB and
CMYK, and use only those spaces - then convert to and from Lab, you
will get some quantization errors with both.

... again

and andrew rodney's post:
RGBLABRGB is damage free
You1re not serious are you Dan?
Take an RGB file. Duplicate it. Do an RGB to LAB to RGB conversion
and subtract the two. You can turn on or off the 8 bit dither. When
you subtract the two and create a new document and look at the
Histogram in Levels, you will see there certainly is data loss and a
change. Move the sliders of the Levels Histogram over and you1ll see
the effects of what differences between the two files you produced.
Are you saying this isn1t data loss?

that test is trivial to do. try it yourself.

have you done the test he describes?

it's rather revealing and the difference can be more than just
quantization errors, depending on the image.

On Thu, 18 Sep 2014 22:55:27 -0400, nospam
wrote:

In article , Eric Stevens
wrote:

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.

which means the changes are reversible.

You are not reversing the changes: you are substituting for them.
Surely even you can see that?

you're overanalyzing things again.

the user makes a change to an image and quits the app. the next day

All they have done up to this point is create a list of edits and view
a simulacrum of their effect in screen. They haven't actually edited
the image.

they resume working on the image and decide to reverse what they did
the day before.

So they change the list of edits and againview the changed simulacrum
which results. They haven't actually created any image to be changed
at either point. That oonly happens once they execute the list of
edits by exporting the image.

with a non-destructive workflow, they can do that.
without a non-destructive workflow, the changes cannot be reversed.

Once they have executed the list of edits by exporting a file the
changes can't generally be reversed either.



the key here is the workflow.

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.

not only do i realize it but that's what i've been saying all along.

you are *so* confused.

And I have pointed out that you cannot reverse a change which has not
actually been made. Even if it is reversible, you can't reverse
something before you have done it.

the change *has* been made, just not to the pixels themselves.

And to what has the change been made?

as i said, you're confused.

I'm confused?


--

Regards,

Eric Stevens

On Thu, 18 Sep 2014 22:55:28 -0400, nospam
wrote:

In article , Eric Stevens
wrote:

All image effects in Photoshop are 100% reversible.

image effects in photoshop *can* be reversible, but they are not
always because photoshop is at its core, a pixel editor. you have to
take additional steps for something to be non-destructive.

True - but the undo function in modern software is the perfect
definition
of a reversible process. It can reverse any effect done by anything in
Photoshop.

undo only exists while you're using the app and it's within its undo
history. at some point, it won't be undoable, generally when the file
is saved but sometimes well before that, depending on the app. that's
the whole problem with a traditional workflow.

with a non-destructive workflow, anything can be reversed at any time
because the original data remains unaltered, including long after the
app has been quit and even archived and later worked on with a
different computer.

the difference is very important, and one which floyd and eric do not
fully understand (or at all).

Of course I understand it!

based on what you've written in this thread and in others, no you do
not.

You should stop and think a little about what you are actually doing.
You are NOT editing the image until you export it in some way.

proof that you do not fully understand it.

Then you explain what happens.

i have many times. go revisit the thread from a month ago about files.

or peruse adobe's documentation and/or books/videos on lightroom.

the information is out there.

I knew you would say that.

The truth is that you can't explain what happens because you do not
actually understand it. All you know is what buttons to push to get a
particular result. This is evident from the way that what is going on
in the background never seems to concern you.

You have said that I do not fully understand what is going on. You now
need to show that you do, by explaining it.
--

Regards,

Eric Stevens

On 2014-09-19 04:48:46 +0000, Eric Stevens said:

On Thu, 18 Sep 2014 22:55:27 -0400, nospam
wrote:

In article , Eric Stevens
wrote:

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.

which means the changes are reversible.

You are not reversing the changes: you are substituting for them.
Surely even you can see that?

you're overanalyzing things again.

the user makes a change to an image and quits the app. the next day

All they have done up to this point is create a list of edits and view
a simulacrum of their effect in screen. They haven't actually edited
the image.

they resume working on the image and decide to reverse what they did
the day before.

So they change the list of edits and againview the changed simulacrum
which results. They haven't actually created any image to be changed
at either point. That oonly happens once they execute the list of
edits by exporting the image.

with a non-destructive workflow, they can do that.
without a non-destructive workflow, the changes cannot be reversed.

Once they have executed the list of edits by exporting a file the
changes can't generally be reversed either.



the key here is the workflow.

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.

not only do i realize it but that's what i've been saying all along.

you are *so* confused.

And I have pointed out that you cannot reverse a change which has not
actually been made. Even if it is reversible, you can't reverse
something before you have done it.

the change *has* been made, just not to the pixels themselves.

And to what has the change been made?

as i said, you're confused.

I'm confused?

What are you guys smoking down there in New Zealand?
....it can't possibly be the same **** found in Alaska, but if it is
that might explain something.

--
Regards,

Savageduck