Howdy...

Last week I posted a link to a layer masking tutorial I wrote for
graphics.com. Some had problems with the link, but they have it corrected now.
Just go to the main page:

http://www.graphics.com

The new tutorial is under 'The Cool Factor'.

One more thing...summer is about done, and I'm back into a heavy writing
schedule. I'm going to take down the discount on my website in 2 days, so if you
want to get the Photoshop downloads at a reduced price please do so today if
possible:  http://actionfx.com

The new book (Photoshop Most Wanted II) that I wrote with Colin Smith from the
Photoshop Cafe will be out in a couple weeks also...I'll keep you posted when
the link is updated on Amazon.

Take care,
Al , the NAPP Action Guy
Action Fx Photoshop Resources
http://actionfx.com

[in reference to several features in Photoshop, including putting the pen as an
option for the Shape tool, and several other complaints, someone said:]

> > In my opinion, Adobe needs to seriously rethink
> > some of their GUI policies simply because they
> > are outdated, inconvenient AND confusing. It's bad
> > for the end users which will ultimately effect
> > the usability of the product. It's pathetic.

I think not. While there are probably legitimate gripes about implementation of
some features, some gripes are born of failing to understand the implementation
(process and reasoning).

Making a program 100% customizable and perfect for everyone's desire is an
unlikely scenario. There will have to be defaults, at the very least, or else
you will be asking quite a lot of the new user as they will have to set up the
program before doing anything with it -- which becomes a catch 22. You can't
just approach GUI design from the standpoint of: What advanced users need, for
example, as then what happens to users that are not already advanced? If you
have ever done programming (and not just looked at different interfaces), it can
sometimes be tricky to accomplish a seamless implementation of tools that end up
LOOKING simple. When the interface appears to read your mind, this should be
considered a benefit, but regretfully those successes will be grandly overlooked
by users. People are only geared to complaining about problems. There will
rarely be praise for something that works like you expect. Expectations as to
how things should work vary from user to user, so almost any implementation can
lead to a complaint.

Sometimes it is best to be glad you can use it at all. Either that or glad you
can learn to use it as is. Generally, I vote for becoming talented with the
tools as they are.

Listing problems is not a solution. Saying: "XYZ tool should do more to read my
mind" doesn't point to the specifics of a problem that you want solved. Making
fun of an implementation does even less, I would suppose. Posing a solution
would probably do more good.

Presets on the options bar are really pretty handy -- if you use them. With a
few clicks I can change back and forth to settings that would have taken many
clicks to make otherwise. the only change I might make is allowing users to
assign default settings, PERHAPS, so they know what to expect and have something
useful when opening the program, instead of just relying on previous settings.
However, those who like previous settings would get jolted with this
change...The solution might be to allow either/or, and the solution swiftly
becomes a more complex one. Suggesting the tools and options will interchange is
silly [user suggested the tools would soon be found on the options bar and
options on the toolbar], of course, but I could see a situation where the tools
might become more like the palettes in the palette well -- hovering over the
left of the options bar could drop the list...and clear some screen space. I'm
sure some would prefer that -- especially those who use the keyboard shortcuts
anyway.

There are probably ways to improve every feature in Photoshop -- and every other
program on the planet. Griping about it isn't going to help; accurate and
helpful criticism might. Adobe will have to choose their battles as there is
only so much time between cycles, and so many hands for programming and testing.
That is a reality. The other possibility would be to hire droves of programmers
to make all the possible changes, and increase the cost of the program,
encourage layoffs, and, frankly, become more business sloppy. I find it costly
enough, thank you, and don't see that a situation which inspires temporary help
as one that will entice the best programmers.

If you really have a solution to a problem with Photoshop, you might try making
suggestions for change through the Feature Requests page at Adobe:

http://www.adobe.com/support/feature.html

These will probably do more to affect change (so long as the requests make
sense) than some random complaints on any forum...

Hope that helps.

Richard Lynch
http://hiddenelements.com




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

> To extract wispy hair right you'll need a plug-in
> program such as Knock Out 2.

I disagree. To do it right you'll need patience. Knockout and other such
masking programs don't do any magic, and they cost quite a lot (see
http://www.amazon.com/exec/obidos/ASIN/B00005R7YY/newwriting/). All they really
do is allow you to select a color range to knock out (or something similar to
that). I've never used a program that doesn't make a mask I have to adjust. If
you see areas of the image that are over the type of background that are easy
to extract, I'd learn to use either Extract or Blend If (I prefer the latter)
and save the cost. What I'd be more likely to do is some careful manual
masking, and perhaps trim wispy hair.

Most images not taken specifically with the purpose of cutting the hair from
the background will prove difficult or very difficult to extract -- no matter
what you use. Learn to create manual masks either using alpha channels or via
Quick Mask. Don't be afraid to adjust those masks.

There is a tutorial on ps6.com for using the Extract tool...you might check it
out to learn more about it. Personally I more often will use a manual masking.

Richard Lynch
http://hiddenelements.com
http://ps6.com

> So if you use a color management workflow, what
> you see on your screen will be what you see when you print.

I think the statement above really makes profiling sound like a magic wand.
WYSIWYG is the goal whether you use a color management workflow or not. You can
work without profiles embedded and do a fine job -- but it is still a benefit to
calibrate, test, check, and know how your service handles color. Note that you
should calibrate, test, check and know how the service handles color whether
embedding or not. I've been working without embedding profiles for years, and
only rarely consider using a profile. The benefit of profiling is really
something you will achieve AFTER you can get decent results without them. And
then it might be that you may not need them.

> "In the real world of press output, sRGB is
> not a huge problem, although many make a big
> fuss about it's reduced gamut."

Huzzah. People claim to see and need a lot of things, claim results by fire or
ice, by witches brew and dominatrics...some do get all the right results for all
the wrong reasons. I have seen more disasters with profiling and people who
half-know theory than I bet there would be without. A conversion is a
conversion. CMYK is still CMYK -- generated from sRGB or Adobe RGB. If print
color matters and is critical, learn to create your own separations and use the
CMYK preview with an accurate monitor profile to preview from RGB(whatever).
Roughly: If I control the CMYK, this shouldn't change; If I supply the
RGB(whatever), I'm at the mercy of the service and their practices for the
conversion...someone called it a 'crapshoot'? If you want an RGB print to match
what you see on the monitor, have a calibrated monitor and use an RGB process
(LED/CRT). It isn't all that expensive, and can be great for single prints.

Workflow isn't necessarily one thing. You can have several workflowS. The right
approach is to start knowing where you hope to end up, and prep the image
accordingly...whatever the voodoo you prefer to use in between. If the results
don't match from screen to print and are unpredictable, you need to rethink your
workflow. Profiling is a potential complication -- and should be the first thing
to go when trying to solve these problems. Some make it the starting point --
perhaps for the wrong reasons. Calibrate, sure (so your screen is right). Create
an ICC profile, sure (so your previews are right). But just cause there is a
profile doesn't mean you are obligated to embed (just as your service is not
obligated to honor). Color conversion is more horseshoes or handgrenades, for
the most part, than sniper-shooting. More than one method can produce an
acceptible result. Knowing the benefit of color spaces and the reality of their
advantages can help you make good decisions. some decisions will make more or
less of an impact on the result...and some will be impossible to distinguish.

When in doubt, simplify.

Hope that Helps!

Richard Lynch
http://ps6.com
http://hiddenelements.com

Action Fx Photoshop Resources Summer Discount!

Your downloads are ready...NOW

Hello everyone...Al Ward here. I'm busy writing again, and there are a lot of
exciting things in the works regarding all things Photoshop. One announcement
I'd like to bring to your attention: for a few more days I'm running my summer
special for access to the Photoshop and Elements downloads at actionfx.com. I
literally have over 10,000 custom add-ons for Photoshop available now, and will
be greatly expand the tutorials available to members as my writing projects
allow.

Check it out on the main page:

http://actionfx.com

If you have any questions, feel free to drop me an email from the site.

Also, I hang around the Photoshopcafe forum...I'll see you there!

Take care,
Al

> I forgot to ask if there is a shortcut to switch to the eraser?

Just press E.

If you have Special Edition Using, the shortcuts are all on the card on the
inside front cover of the book. But you don't need the book for them. If you
have PS7, just roll your cursor over the tool and leave it there. A tool tip
will pop up (unless you shut them off) and these tell all the shortcut letters.
In PS6, you may have to click and hold on the tool to open the fly-out
menu...the menu has the shortcuts listed. Scroll like shortcuts by holding the
Shift key. For example, Shift+T will change to the Type tool and then the
Vertical Type and Type Mask, etc.

Hope that helps.

Richard Lynch
http://ps6.com
http://hiddenelements.com

> Let's say that if I had previously used the eraser in brush format at 70pts
> then move to pencil then hit F10, which I have assigned and recorded for a
> hard edged pencil-type eraser, I get the eraser set to 5 pt in size and a
> brush but not pencil.
>
> I still actually haven't a clue how to cycle between brush and pencil in
> eraser mode!

In PS7, record a preset for the eraser using the pencil with the brush size and
shape you want. Then record another preset for the eraser using the brush and
brush size you want. Switch back and forth using the presets palette by clicking
the choices.

If you want, you can record Tool Preset choices as actions. Once recorded as an
action, it can be assigned to a keystroke. If you have a keystroke for each Tool
Preset you use, you can effective switch back and forth to exact parameters by
pressing assigned shortcut keys on your keyboard.

In my opinion, Presets was one of the most useful developments in PS7!

OK?

Richard

> Does anyone know of any articles that have put forward convincing evidence
> for scanning B&W in color?

I don't know that I would consider needing evidence but rather common sense
and/or knowledge of your equipment. My suggestion would be that you are not
just going to scan every B&W image in color if you don't have a good reason.
There should also be less good reason if you are working with a hi-bit
scanner (which should differentiate between fine levels of gray). However, I
might very well scan a black and white image in RGB that had been yellowed,
stained, water-damaged, has processing problems or where the quality of the
scanner and scan are themselves suspect. I would do this to get the most
information compare channel quality, and perhaps simplify what would end up
being color correction of my B&W (e.g., blue information seems correct while
green and red are damaged/lacking). This fits with all of Shangara's
exceptions, as mentioned.

The idea, I think, is the same as when scanning a color image to turn it to B&
W. I much prefer having control over that rather than letting a computation
make the artistic decisions for me. Some scanners, scanner software and
function may be better at this, and the process you choose will be best if
chosen depending on your equipment, how it works, and whatever gets the best
results in practice -- keeping in mind the intended use. Practice can require
testing.

Hope that helps.

Richard Lynch
ps6.com
Hiddenelements.com

> Is the fade filter a plug-in? Everyone talks as if it is a standard
> component of Photoshop.

Technically I don't believe it is a 'plug-in'. I believe Fade was in 6, but
under the Filter menu. It is definitely is in 7 as a standard. It is limited to
certain features and functions. If you have a version that does not have it, or
want to 'fade' a function that Fade doesn't handle, you can imitate it like
this:

1. Create a new layer at the top of the layers stack
2. Press CTRL+ALT+SHIFT+E/Cmd-Control-Shift-E (PC/Mac). this will copy the
current visible image into the new layer as a flattened result.
3. Run the filter you want to run on the layer.
4. to Fade, shift the opacity.
5. To affect the Fade mode, change the Layer mode.

It is really that simple.

So, say you have an image with a layer you want to apply a luminosity fade
adjustment to:

1. duplicate the layer.
2. Run the filter you want to run on the duplicate layer.
3. Change the Layer mode to Luminosity.
4. Shift the opacity to affect the fade.

Putting this in an action can simplfy the process and provide more control than
the function, as you can run additional filers and masking on the fade layer.

Hope that helps!

Richard Lynch
------------------------------------
The Hidden Power of Photoshop (due October '2003)
http://ps6.com
The Hidden Power of Photoshop Elements 2
http://hiddenelements.com

------------------------------------
PS -- for those that have noticed, I have been off for a while. Sorry for the
long hiatus. I hope to be back at this somewhat regularly over the coming
months. I look forward to your questions.

LET ME KNOW: If there is anything that you want in Photoshop (or Elements) and
don't have, drop me a line either on the list or through email and I'll see what
I can do to help make it happen.

------------------------------------

<< Reading Ben Wilmore's tutorial on Curves. It suggests using luminosity
blending mode in curves when brightening or darkening an image in order to
preserve the color balance.  Tried it, but dont know if I'm impressed. >>

The technique is actually a good one -- as long as there is a little background
on how it works. In both Photoshop and Photoshop Elements I often separate RGBL
information as a stack in layers (Red, Green, Blue and Luminosity). The idea
there is you get the color control of red, green, blue corrections for working
with color and dynamic range, and can affect the changes in color separate from
the image tone without making several conversions to other color modes -- and
with the advantages of both methods at the same time. It is also easier to work
with RGBL separated as layers, than to try and use the controls in Photoshop
such as the channel targeting -- or to work between palettes (channels and
layers). My book on Elements shows how to implement this for either Elements or
Photoshop.

I don't have Ben's book and can't be sure, but I imagine that he suggested using
the technique on an image that already had some basic color correction
completed? For example, you should have used Levels to achieve color balance,
and then -- after the basic correction -- you can use a curve to affect the
tone. In other words, I don't know that I'd be correcting tone first.

Also, he should have made abundantly clear that you will probably  want to use
separate layers. For example, from a *flattened* image, duplicate the background
to create a duplicate of the original, set the new layer to Luminosity and apply
a grouped curve (make the curve adjustment layer and press Cmd/CTRL+G; Elements
users can access curves using tools
provided in my book). If the curve is not grouped to the luminosity layer -- or
targeted to the luminosity -- the result will affect the color as well as the
luminosity, and you will probably get some pretty horrible results.

With luminosity isolated, you can do some pretty amazing things to an image --
which is what I am sure Mr. Willmore is suggesting. The separation works well
for implementing digital color noise reduction, for example. The Luminosity
layer will freeze the shape of the image -- so to speak -- and the color can be
blurred (Gaussian, Median) to reduce collected noise, like you might get in a
low-light digital exposure.

Hope that helps! If you have questions, let me know.

--
Richard Lynch
author, The Hidden Power of Photoshop Elements 2
http://www.amazon.com/exec/obidos/ASIN/0782141781/newwriting/

<< Suddenly, when printing from photoshop 7, my pictures look muddy.  I checked
to make sure it wasn't a bad pack of paper or something going on with my printer
by : (1) printing directly from the printer and (2) using Picture It to make the
same picture. Both times they came out clear and true to color. After checking
everything we knew how to do (my husband and I are novices and use your
Photoshop 6 book to get us through most problem - usually successfully), we
uninstalled Photoshop 7 and then re-installed it, thinking that would get rid of
anything weird that might have happened to the program.  But the situation
persists. Any ideas? >>

If you are getting results from another program, it suggests the problem is in
Photoshop. That is effectively testing that it isn't a problem with the ink or
mechanisms.

I would be suspicious of anything you have done recently to change the color
settings and/or views. You might also just try calibrating again (using Adobe
Gamma, for example). If you have changed any color settings it is possible this
can affect output. The change would obviously not effect output in another
program unless it was due to embedded profiles which the other program doesn't
honor.

You might try trashing the Photoshop preferences. These sometimes get corrupted
for whatever reason, and for the most part they don't get over-written in a
reinstall. Hold Command-Option-Shift (Mac) or Control-Alt-Shift (Windows)
immediately after you start launching Photoshop to rebuild prefs. Keep holding
the keys till the program is open. You may have to arrange things the way you
like again, but that's better than bad prints -- if it works. You could always
backup the prefs before deleting and restore them if the rebuild doesn't fix the
problem...

Hope that helps. Get back to me if not.

Richard Lynch
--------------------------------

<<  I am a photographer and use photoshop frequently.  I am having a problem
with the color of my work area in photoshop 6.0.  It is currently white and I
need it to be grey.  I have tried re-setting all the defaults, but nothing is
working?  Any suggestions?  I appreciate your time and any help you can offer? 
>>

The matte color around an image is probably more often set inadvertently by a
user and then they wonder what they did to get it that way. Set the foreground
color to the color you want the matte to be (you may want to consider something
around a 20% gray, or 191,191,191 RGB). With an image open, view the matte area
by enlarging the image window or zomming out from the image. Choose the Bucket
(paint bucket) tool. Hold down the Shift key and click in the matte area. This
will fill the matte area with the foreground color.

This and a zillion other things can be found in Special Edition Using PS6, which
should be going out of print soon. The PS7 version is quite another animal, by
different authors.

http://www.amazon.com/exec/obidos/ASIN/0789724251/newwriting/

Best of luck!

Richard Lynch
© 2003

<< I'm pretty much a beginner. How can I make an object in an image stand out
from the background without a lot of effort? Which tool do I use?>>

Regretfully, this is no push-button solution. You need to bring some relative
skill to the problem. There are several solutions, including
replacing the background, blurring the background (faking depth-of-field) and
altering the background in another way to create separation (e.g.,
shadowing/darkening). I'll give an overview of the process...

All of the techniques for creating separation really require the same first
step, and that would be to somehow separate or select the people in the
image so they can be worked separately from the background. To do this, you can
use any of the selection tools (the selection brush might not be a bad choice in
this case), or create a mask. If you are an absolute beginner, you might try the
freehand lasso and a little feathering (see the options bar when you have the
tool selected).

You see this is already starting to grow a little complicated. I would suggest
playing with the different selection tools to see how they work -- the two
mentioned here probably being your best bets. I don't mean to make light of
this...it may take several hours to make a good selection depending on the
complexity -- and it will be especially difficult without having some more
advanced technique under your belt.

Once you have made the selection, copy and paste the selected area to a new
layer (using Copy/paste, the new layer will be created automatically with what
you have selected and copied). Then invert the selection and copy/paste again;
this will fill another layer with everything you did not select. Most
instructions will leave out that second step, but it is very important if you
are to isolate your image components properly.

Save the selection.

At this point, you might just try blurring the background. Activate the layer
that has the area outside the people (not the background layer) and
choose Gaussian Blur from the Filter menu (Filter>Blur>Gaussian Blur). Set the
Radius to a few pixels, and look at the preview. When you get an effect you like
you can flatten the image and save.

The other effects I mentioned are discussed in my book, as well as some more
advanced selection and isolation techniques.

If you have additional questions, please ask!


Richard Lynch
rl@...
http://hiddenelements.com
------------------------------------

I have made an interface available to Elements users that can be
customized to run pre-recorded Photoshop Actions in Elements. This is
available for free. It also has a demo of the Hidden Power tools
included with my new book (The Hidden Power of Photoshop Elements 2).

Download the action tool here:

http://hiddenelements.com

Follow the Tools link and then the Free Tools.

I will be building a list of actions that users recommend on the
website to go with the free tools. Please feel free to recommend
actions -- and let me know if you find some that may need to be
adjusted for Elements users!

Richard Lynch
http://hiddenelements.com
-----------------------------

<< I have a Hp 7350 printer which will not print edge to edge.
How do I make a white border on an 8X10 Picture. >>

Printing to the edge and having a border are two mutually
exclusive things. Let me handle them separately.

First, you really don't ever want to print to the edge of the paper.
You might be able to do this on some printers by tricking it (and I hear there
are printers that will do it), but I'd not recommend it. See, if you were to
print over the edge of the paper, what would the ink/pigment stick to? Chances
are it would get on rollers, inside other stuff, would taint later images
passing through and just generally gum up the works.

Most manufacturers correctly give you a little edge around the image so that you
DON'T make a mistake (well, that's part of the reason). To make an image look
like it is printed to the edge, you print it large and then crop it down by
cutting the paper. This is done with books and magazine printing, and is called
bleeding an image...you print it an eighth of an inch too big over the edge you
want the image to print over, then crop back the paper so it LOOKS like it was
printed to the edge.

The other option you can get is pre-perferated paper, which allows you to print
with an overlap of the perforation, and then you can tear off the edge to crop
the image back. Again the appearance will be that you printed to the edge when
you really didn't.

Putting a white border around your image is as easy as using white paper. This
is just the opposite of printing to the edge in that you just leave the image
uncropped so the white of the paper acts as a border. You may need to crop an
edge if you print very large on a sheet as printers tend to have one long 'grip
edge' (often used as another name for the non-printing border). Your inks in
most printers will just be CMYK...that is Cyan, Magenta, Yellow and Black. You
add inks to darken images, not to turn them white. While there are special cases
where you can use a spot white ink (for example, printing anything white onto a
clear plastic bag), these are more expensive printers and printing jobs. Spot
color is usually a color added to a print job used to get custom results that
you wouldn't otherwise get with CMYK proess.

I hope that helps a bit. If I misunderstood the question, please come back at me
again.

Richard Lynch
author, the Hidden Power of Photoshop Elements 2
http://hiddenelements.com
-------------------------------------------------------

<< Is there a way to make a clipping path soft-edged? >>

Paths describe a hard-edged mathematical line (or shape) and are never really
soft on their own. In Photoshop 6 the ability was added to combine masking and
clipping to get combination results (e.g., soft edges and transparency with
absolute clipping edges). This will not work with Clipping Paths because you
can't apply a mask to the background layer, but does with vector layers (so
called in PS7).

If you have images to combine and you are attempting it in layout (Illustrator,
Quark, InDesign), you may want to create the result using Photoshop rather than
trying to snap it together in a layout program. Photoshop will allow you to
create and save the Clipping Path (which will re-shape the image) and combine
Masking and Clipping in layers to control other results (the absolute shape of
additional objects using clipping layers (vector masks) combined with layer
masks that can control the feather). You can get results that would be difficult
or impossible otherwise.

Trying to save an image with a layer vector mask and no background will
insinuate a white background to fill the clipping path area if there is nothing
else there when you save the image from Photoshop. Be aware, also, that you have
to change the shape of the clipping Path to encompass everything you want to
keep in the final image (e.g., if you add a layer above a clipping layer that
sticks out from the layer content that you are using as clipping, you'll have to
reshape the path to encompass the new area as well -- IF you want it to show).
Layers that have content that falls outside the image clipping path will drop
that content.

A great way to test the result of using clipping paths is to work between
Photoshop and Illustrator. Generate EPS files in Photoshop, place in Illustrator
and generate PDFs to view in Acrobat Reader. Illustrator assumes you are looking
for postscript results, and you can preview what is going to happen.

Just to note, with my Hidden Power tools, you can edit vectors and apply
Clipping Paths in Photoshop Elements.

I hope that makes some sense and helps a bit.

Richard Lynch
The Hidden Power of Photoshop Elements 2
http://www.amazon.com/exec/obidos/ASIN/0782141781/newwriting/
-----------------------------------------------------

<<I am trying to create a layer mask using Elements. I would like to
blend two images ... then, I want to use a gradient (black to white or
transparent) on the layer mask to reveal the second image, in effect, gradually
revealing the image. With PS 7.0 there is a convenient layer mask icon at the
bottom of the layers palette. With Elements, I am able to create masks that are
adjustment layers, fills, and patterns, but I would like to have my image be the
mask.>>

You can do this using layer clipping. All you have to do is create your mask as
variable TRANSPARENCY and use that as the bottom layer of a layer grouping. The
bottom layer will be used as a cookie cutter based on transparency and will show
only those parts of the layers grouped with it that fall over the opaque areas.

There is a way to change image elements (e.g., grayscale, luminosity, channel
components. etc.) into transparent layers (rather than grayscale), but it
requires using Blend If...and Elements does not have the same controls for that
as Photoshop does. However, I added a tool I call Blend Mask in my new book
which uses Blend If properties to allow you to control transparency based on
grayscale -- without the complexity of having to understand how blend If works.
So while it isn't possible with the native Elements interface, you CAN do what
you want by employing some of the tools I include on the CD with the new book. I
can tell you exactly how it works, and it really doesn't matter -- you can't
create the solution with Elements alone because you need Photoshop to make
actions that run behind the scenes...I've already done all that for you.

In most cases the tool for Blend Mask will provide a one-click solution to the
problem you have. It won't be exactly like Photoshop in function, but the result
will be exactly the same.

Get the book here:

http://www.amazon.com/exec/obidos/ASIN/0782141781/newwriting/

Hope that helps!

Richard Lynch
rl@...
----------------------

<<Is it possible to do the nice pencil sketches with PS
Elements? I mean the really nice ones>>

I hope you don't mind my awful sense of humor, but usually one
uses pencils to make pencil sketches :-). There is a somewhat
more serious answer here...


<<I know that PSE has artistic filters for colored
> pencil, charcoal, etc. I've seen some posted on other sites
>that you can hardly tell whether its a photo or a real sketch.
>Is there any affordable (not Photoshop) program that has
>that capability?

There is nothing that Photoshop has in this way that Elements
does not. I am not sure from your post if you think the pencil
sketches that "you can hardly tell whether its a photo or a real
sketch" means that the sketches are too much like photos, but in
the end, whatever method was used -- if from Photoshop -- can
be done in Elements.

The tricky part to making a sketch would be applying filters so
that the result appears realistic. The unlikely part is doing that in
a fashion that requires only one filter. To get this effect and make
it what I would call realistic, you have to make multiple
applications of filters, masking and other tricks (e.g., layering,
blending, embossing, etc.) to really make the effect work. You
may also have to combine effects and masking with
randomization (which I show how to do in my new book). There
are probably some sophisticated masking techniques (e.g.,
edge masking) that may help you get the effect you want. I show
an effect to create a sketch in Adobe Photoshop 5 How-To...and
it is but one variation on thousands of possibilities.

The problem is that in a real pencil sketch, the artist who is
drawing, makes decisions on line, weight, color, etc., that a filter
or other simple tool application will not. Randomizing the
application of effects can really only get you part-way to reality.

What I guess I am driving at, is that while there are filters that
may say 'pencil' or 'chalk' or 'mona lisa' or whatever, filters are
not very smart critters. They don't look at an image and decide a
whole lot. It is up to the user to become enough of an artisan
with images that he/she knows what they want to accomplish,
how complicated they want the result to be, and either apply
developed expertise or learn the techniques to create the result.
The surprise may be that all $600 of Photoshop won't make it
any easier.

The solution I might suggest would be a terribly complicated
one. However, it would also depend on the type of pencil sketch
you want to accomplish. You might use techniques from those I
mention above to something like making output of the image you
want to turn to a sketch, sketching over it on tracing paper,
scanning the tracing, and using that to help mask and apply
color and tone. It all depends on how far you want to go.

If you have an example of a before and after image that was
accomplished with filter application only, I am pretty sure I can
tell you how the techniques were applied to achieve the result.
Can you send me an example you've seen?

Richard Lynch
author, The Hidden Power of Photoshop Elements 2
rl@...
--------------------------

<<My question requires that you examine two thumbnail images I've generated
-- one manually via Photoshop's "Image Size" wizard, the other
programmatically via use of an image manipulation component.  I've posted a
brief but concise description of my problem at the following URL:
http://www.nbsc.com/Photoshop/ >>

Essentially you used two different programs to resize the same image, and
each handled the content a little differently. In this case, you suggest you
like the results from Photoshop better. In cases where images have different
content and you are doing different things with sizing, there may be
advantages to using the other method of interpolation. I know that in
Photoshop, bicubic resampling/interpolation uses an averaging scheme by
looking at pixel areas and deciding on a new value (either exactly like
bilinear, or very similar) and then applies a bit of sharpening...I can't
tell you if it does this all at one time, or if it is a single calculation --
I don't think it much matters except from a programming standpoint. The
averaging scheme necessarily softens the image content, so the sharpening is
meant to counter that a little. The result is a slightly more edgy rendition
of bilinear sharpening.

What the other result you got seems to suggest is either a different
calculation scheme is used, or that resizing uses some elements more akin to
Photoshop's Nearest Neighbor resampling. I'd really have to do some
experimentation to find out, but my guess is that you'll see a marked
resemblance if you use NN in Photoshop and compare. You are left with a
somewhat more jagged (less blurred) result, which could actually be better in
some cases (e.g., where the content has naturally sharper edges). My guess is
that if you slightly blur the result from the other program, you will have
similar and perhaps better results in some cases than you got in Photoshop.

All interpolation is not alike -- even when using the same or similar
terminology. All the word means is that image information will be added or
removed. Bicubic suggests that more than one pixel will be used to determine
a pixel value, but seems to have no specific meaning in and of itself (it is
perhaps a technical or mathematical term?). It is possibly that the other
program borrowed the term and not the idea of the functionality. Sometimes
even when using different versions of the same program interpolation by the
same type will have changed (hopefully to improve as versions progress).
Interpolation is not a standardized or exact science when resizing images, so
it won't be accomplished in exactly the same way between programs -- unless
they use exactly the same procedures and equations. In a similar vein, it is
possible that using two digital cameras with the same CCD taking the same
picture to get different results because of the way they interpret the raw
data that is captured. As far as answering the question what is the one
program doing incorrectly that the other is doing properly...it is impossible
to answer, because one isn't necessarily right and the other wrong. Result is
a matter of preference.

Anti-aliasing is something you generally want to consider as part of masking
and selection edges, and SHOULD have little to do with the result here. If,
however, you mean that the program has no ability to calculate based on
adjacent pixels, then you may not have the ability to interpolate in it as
Photoshop does. I guess my next question would be: If you have Photoshop, why
worry about a plugin that does something Photoshop does yet doesn't do as
well? There may be other, better automated means within Photoshop itself to
accomplish what you need.

Hope that helps!

For those who are interested -- I got a copy of my new book, so they should be
available soon. You'll get them sooner (and support the author while getting the
best price) by ordering yours from Amazon.com here:

http://www.amazon.com/exec/obidos/ASIN/0782141781/newwriting/

The tools and  techniques work in Photoshop as well as Elements...The book has a
pretty good discussion of interpolation.

Richard Lynch
ps6.com
-----------------

<<I'm sure you already know this but just in case I thought I'd pass it along.
On a Mac, press Command-Shift-4 and the pointer turns into a crosshair and you
can select exactly what you want to capture. If you hold down the control key
while you're selecting, the image is captured to the scrapbook and you can then
paste it directly into a new Photoshop file or other image editing software.>>

Yes -- though I don't know if you can select 'exactly'. I used Command-Shift-4 
when shooting screens on OS X and all the other utilities I was trying failed. I
very much like Snapz Pro for Mac and use something called Microsoft® HTML Help
Image Editor for captures on the PC (a free-bee that came with a Microsoft
product). These products go a little beyond just snipping out a block of the
screen that you draw. They automate selection of palettes, menus, and allow
options for file type and appearance of cursors. If you shoot a lot of screens
(e.g., for tutorials and books?), something more powerful than the native
functions is almost a necessity.

Maybe Adobe will include a screen utility with Photoshop and Elements
sometime...Would be a nice step-saver.

Richard

<<I have about 15 images that I scanned and edited in photoshop. I'm making
a report for my employer for an upcoming digitization project and they want
histograms included in the report. Is there a way to print out the levels or
histogram of an image in photoshop? I've tried saving it and printing it, i
can't get it to work. >>

As you can't treat a program screen like an image, the idea would be to take a
screen shot and use that. When you SAVE the histogram using a save function, my
guess is that you are on the Levels screen using the Save? If you are, you are
saving the levels setting, not the screen itself. The purpose behind doing that
is to load the Levels settings at another time without having to manually set
the sliders.

To take a screenshot on a PC press Print Screen on the keyboard and then go
into Photoshop and create a new image then paste...you'll need to trim the
image.

To take a screen on a Mac, press Command-Shift-3. You have to open/paste/trim
as with a PC.

Both platforms have screenshot utilities that can be used instead of the
default...often these utilities have added useful features like automatically
capturing a single screen, rather than the whole monitor. You might take a look
at www.download.com and search for "screen shot" under your platform to see a
sampling of what is available.

The only real trick to using a screen shot in print is attaining the right
resolution. Most people look at 300 ppi as some sort of holy print resolution.
If you are familiar with my books, that just isn't true. However, lets say you
were printing on a printing press and were running 150 lpi, you'd need between
233 and 300 ppi in your image. Trouble with that is, screen shots are 72 ppi. If
you just change the resolution, it will come out about 1/4th the size you get on
screen. That's potentially pretty tiny. Your only choice is to upsample. But if
you just upsample a screenshot, you'll note that it gets a little soft around
the edges, unless you are unlike most people and use Nearest Neighbor for your
interpolation type. That is, Bilinear and Bicubic will both blur your image when
you resize; Nearest Neighbor is more likely to hold an edge...in fact, it has to
because it chooses from colors already in the image, rather than making new
blended ones.

The real trick to resizing a screenshot is to make sure you choose the right end
resolution. You want to square the number of pixels in the image, rather than
just picking a number at random. The reason for this is: you want each pixel in
the image to turn into 4 (2x2=4, or 3x3=9, or 4x4=16). If they don't, you'll get
a result that doesn't do its best in replicating the way the original screen
looks.

An increase of size by EXACTLY 4 times (4.167 just won't work) using Nearest
Neighbor interpolation will be increasing the resolution to 288 pixels
(4x72=288). Luckily that falls right in the range you are looking for. You'll be
able to output high-resolution slick sheets on an offset press without any
distortion. Build a quick action that resizes the screens to 288 ppi using the
NN interpolation -- and batch process them to save time.

Hope that helps!

Richard Lynch
http://ps6.com
author, The Hidden Power of Photoshop Elements 2 (out before christmas!)
------------------------------------------------

<< The problem I am having is I work with a lot of glass items. The colors of
the resulting .jpg (save for web) are drastically shifted in some cases. Is
there something I may have messed up in my settings?  >>

Whether you work with glass, razor blades or fine art, the color of your
final print or output shouldn't have 'drastic' shift (depending on how you
define 'drastic'). But certainly you could have messed up a few things...it is
actually pretty easy to do, especially if you don't have any guidelines (or a
LOT of time to test and reinvent a system). You'll have to check a few things.
I'm not going to write out here because each of these deserves detailed coverage
(I'm getting the big obvious ones first):

1. Calibrate your monitor (Use Adobe Gamma)

2. Set up color management correctly.

3. When saving, be sure you are saving to JPEG format so you don't
incidentally reduce color -- and don't compress it all to hades!

Information on Adobe Gamma can be had on your Photoshop program disk. It is a
utility for calibration and creating an ICC profile. Setting color management is
a more-or-less personal thing based on how you like to work with images and
whether or not you trust profiling. Personally I use minimal color management
and work with Adobe Gamma to be sure I've calibrated (in the instances where I
don't have a dedicated calibration device). While there is a big hubbub about
either using or not using color management, the real choice is: do what works
for you. Don't just dive in headlong to embedding color profiles in everything
if you don't understand what they are and how they work, even if you've read an
article about it. Regardless of what pro-profiling people say, you can get
results without profiling...and in either case I'd test to be sure you are
getting what you want.

You will want to select the ICC profile you create for RGB to give you the best
previews on screen. If you don't know about color management, you'll need to
read up on it...start with Adobe's suggestions, and browse the PLETHORA of
information on the net. Watch out, though, there is a lot of misinformation,
close information, good information and esoteric. Some better information
appears in books I can trust (The Hidden Power of Photoshop Elements). You might
look at the ps6 archive for some other ideas.
(http://groups.yahoo.com/group/ps6/)

It almost sounds as if you are saving as a GIF if the color is changing that
far...but you weren't specific in saying where you note the change...is it
between computers? on computer and in print? on the same computer but in a web
browser? (If the latter...yikes. Try reinstalling the browser first.) Without
more information about the color shift, it is really tough to go any further
without making this reply enormously long.

If none of this helps...or if it is just to much to digest/fathom, email me
direct and we'll see if we can get you moving. I need to see specifically what
the problem is, so I'll need some sample art and a few more details, as noted.

Richard Lynch
rl@ps6
http://ps6.com
The Hidden Power of Photoshop Elements 2 (Dec. 2002)
Special Edition Using Photoshop 6
------------------------------------------------

<<I am trying to isolate an object by blurring the background...I make the
selection and then blur, but the blur appears fake because the foreground object
bleeds into the bacground making a sort of halo or glow. Any way to fix that?>>

If I am getting the problem, you are trying to make everything happen on the
same layer. What Photoshop (or Elements) does is look at all the image
information on the layer you are blurring...even what isn't selected. That's
why you are getting a halo. Your selection is not totally isolating the image
elements. To do that you have to get everything on its own layer.

I believe if you do the following it will fix the problem:

1. Make the selection of the subject (should be tight and accurate).

2. Copy the subject to a new layer (Copy>Paste works fine). Name the layer
Subject.

3. Copy the background to a new layer (Activate the Background, invert the
selection, then Copy>Paste). Name the layer New Background. Be sure the
background falls below the Subject layer in the stack.

4. Blur the New Background layer.

This isolates the subject and the background from one another so you can make
changes to each without worrying about the other. The problem with other
solutions here (e.g., resizing) is that the resizing will interpolate and
compromise the subject somewhat...If it isn't absolutely necessary, I would
avoid that because it leads to blurring of parts of the image you'd prefer to
keep sharp.

This also offers the opportunity to make additional adjustments, and play
some blending tricks with the New Background layer's opacity. Separating
elements in an image is a focus (...ahem...) in my new book The Hidden Power
of Photoshop Elements 2.

Hope that helps!

Richard Lynch
http://ps6.com/
rl@...
------------------------

<< I would like to place an extracted subject onto a background image.  I could
just place a sharp subject on a sharp background, and I've done it a bunch of
times, but it appears false.  I would like to adjust the background so that it
appears to gradually blur from the foreground (edge of the image) to the
distance (center of the image, assuming straight-on perspective) to give the
illusion of depth-0f-field and a realistic image.  I have no idea how to combine
the Gaussian Blur Id need with a gradient to provide the fading effect. >>

Believe it or not, I cover this in Special Edition Using Photoshop 6 in the
section called Selective Blur (p. 612-615). What you are really asking to do is
make a mask for an application of the blur filter. This is far easier to do than
you might think. But here's the short of it:

1. Open the image background you want to apply the effect to.

2. Flatten (if necessary), then duplicate the background (choose Duplicate Layer
from the Layers menu on the menu bar or off the Layers palette).

3. Apply the Blur filter (Gaussian) so that the blur is applied to the whole
layer with the maximum strength you want to use. We'll edit it out to make it
less intense in a minute.

4. Add a Layer Mask to the layer (choose Add Layer Mask> Reveal All from the
Layer menu).

5. Fill the layer mask (it will be active) with a black-to-white gradient from
the center to the edge using the Gradient tool and the Radial Gradient option
(see the options bar).

I am guessing that this will not be entirely satisfactory, but it does exactly
what you asked. To move the 'far' point, just move where you apply the Gradient.
What I think will be more useful is selectively blurring. You can do this by
applying the gradient (either the same way or another way) and then painting on
the mask with black to sharpen or white to blur. In other words, you'll be able
to use the mask to paint in the filter effect. This is a really useful technique
and it is something you can use for any filter or adjustment to selectively make
corrections.

Once you've got the background the way you want, copy and paste that foreground
object in as the top layer. If you leave the layers, you can continue to make
adjustments to the mask to affect the look you want from the blur and to add
separation where you need it.

Hope that helps!

Richard Lynch
author, Special Edition Using Photoshop 6
The Hidden Power of Photoshop Elements 2 (Dec. 2002)
-------------------------------------------------------------------

<< why is that when I make a selection and duplicate it I am left with a slight
line from the previous layer? It is on the underneath layer and I
have to use the eraser tool to get rid of it. This a copy of three layers with
the line I'm talking about. I used the oval marque tool and feathering is set to
zero. [file attached]>>

I don't think this JPEG shows me everything I need to know. However, I can see
the problem. There are actually two things that Photoshop does to blend
selections: Feathering and Antialiasing. Though you have Feathering off,
Anti-aliasing is on. Photoshop does this to make the edge seem smooth...and does
so by blending the very edge of the selection, just like a slight feather (about
1 pixel).

What happens in the result is, you have cut a part of the background to reveal
the background color, and when you paste back in the lesser opacity of the new
layer at the edge of the selection lets the fully opaque background show
through. You can clean this up as you said by then fixing up the area that you
cut.

However, there are several better solutions. The first is, of course, to shut
off the antialiasing. However, in this case where you seem to be using only
simple shapes, why not:

1. Make a Fill shape layer.

Obviously it doesn't get any easier than that....I've probably never written a
one-step procedure before! This will allow you to have a scalable shape that you
can adjust more easily later. If you are making a logo or other art that would
be useful to scale (infinitely), creating it as vector art would allow you to
scale the result optimally, without losing sharpness and resolution.

You could also do this by using selection, but there are a few additional steps:

1. make a new layer
2. create the selection (feathering off)
3. fill with your desired color.

This will keep you from getting other stuff in where it doesn't belong, and
can let you have the advantage of anti-aliasing on so the shape will appear
smoother (an aliased selection will have a perfectly hard edge, and may not seem
as smoothly shaped).

I hope that helps! Feel free to send more questions. The new book is due out in
December, before the holidays, so if you know anyone getting Elements -- or
advanced Elements users (which is really who the book is for), order it here: 
http://www.amazon.com/exec/obidos/ASIN/0782141781/newwriting/

If you don't have my book on PS6 yet, get it before it goes out of print!
http://www.amazon.com/exec/obidos/ASIN/0789724251/newwriting/


Richard Lynch
http://ps6.com/
The Hidden Power of Photoshop Elements 2
-------------------------------------------------------------

<< In a painting I am reproducing for the artist, it was leaning backward
ever so slightly so the the top of the painting is just a bit narrower than
the bottom. It is possible to crop but part of the edges will be lost. Also,
a building near the edge may be leaning in a bit though it should be
vertical. Is it possible to grab the top corners and stretch them out just a
wee bit for a perfect alignment? I know Kai has a "goop" program which distorts
things but I don't know if  that would accomplish what I need. >>

Don't use Kai. This is not what it is for. This is really for making distortions
somewhat like what you already have in the Liquid filter right in Photoshop.
Other Kai tools are probably very powerful, but I never use any of them. At the
same time, I would think that most are more likely for artistic work and
creating image content than for more straight-forward repairs. Please don't
mistake that for an endorsement.

First, you could re-shoot the painting using some type of copy stand, or set-up
that helps you to keep the plain of the image the same as the plain of the
lense. Keeping the optics flat to the plain of the image should help remove
distortion of perspective. Use a longer lens, 52mm+ for 35mm (this can vary
depending on camera type, etc.), and perhaps keep the image from filling the
frame...of course stepping up to a larger format film could also help by
allowing more play at the edge of the image frame.

All you need to make correction on what you've already shot (if you are not
getting curvature from lens distortion via a  wide-angle lens) is the Transform
tool.

1. Duplicate the original to a new layer.
2. Select Transform>Persective.
3. When the bounding box handles appear, click and drag on the upper corner of
the box to straignten the perspective.

You can also drag the lower corners in. Please note, the image will have to go
through interpolation in any case, which is why re-shooting is recommended. If
you drag in the lower corners, you will have to probably distort the vertical as
well to make the image shorter.

If the problem has to do with lens distortion (objects curving through lense
distortion), you are going to have to try to reverse the distortion, perhaps
using the pinch filter. Of course, the old adage comes into play: the best final
images come from the best original shots...There may just be too much distortion
to fix easily, and at that point it might be easier just to re-shoot and create
a better shot to work from.

Thanks for the question! Hope that helps.

Richard Lynch
http://ps6.com/
The Hidden Power of Photoshop Elements 2
--------------------------------------------------------------

<<How can I extract red eye???>>

You don't really want to 'extract' red eye (extract is a process of
separating part of an image from another), what you want to do is tone it
down. The effect is simply this: eyes glow red in images taken with a flash
on camera in a darkened room...the subject has their pupils wide open because
of the darkness, and the flash goes right inside their eye-ball and bounces
off the blood-vesseled back wall of the eye. What you get on film is that red
reflection, and usually the subject comes off looking a little demonic.

In better lighting, what would happen is the pupil would be smaller and
darker. What you have to do to correct it is either sap the color from the
red, darken the red, or both. I recommend the latter. You could also do a lot
more strenuous things, like rebuild or replace the iris (so the pupil opening
is smaller), but for most cases, this isn't necessary.

Try this:

1. Open the image you want to correct.
2. Create a new layer at the top of the stack and change the mode to
Saturation.
3. Choose black as your foreground swatch color.
4. Select a painting tool, zoom in to the eyes, and paint in ONLY the red
part with the brush (you may have to select a brush that is the appropriate
size). Wherever you paint will desaturate the color.
5. Duplicate the layer, and change the mode to Overlay. This will darken the
dark area of the pupil while keeping the highlights -- which are important
for the result to look realistic. If you just nix the highlight by painting
in with black, the eyes will look dead and empty.
6. Adjust the opacity of the Overlay layer to effect the intensity of the
darkening.

You might also try Softlight, Darken or Multiply mode for Step 5. Each will
have somewhat different effect. If you want some of the red color to come
through, adjust the Opacity of the Saturation layer (Step 2-4). Some people
like to use selection rather than their skills with a paintbrush, and that is
fine. Just make the selection of the red area before step 2.

While there are other ways to do this, there are also less satisfactory
results...'magic' red-eye reduction tools always seem to go part way to
me...and part way to me is like partially setting a broken leg.

Hope that helps!

Richard Lynch
author, Special Edition Using Photoshop 6
The Hidden Power of Photoshop Elements 2 (Dec. 2002)
--------------------------------------------------------------------

<< My B&W images are not printing as well as they might...the problem is there
doesn't seem to be enough depth....give me some input on what I can do to ....
make the blacks blacker and sharper... >>

You'll need to do a levels correction (please see my tutorial here:
http://ps6.com/Tutorials/Color_Correction/intro.html).

That will enhance the dynamic range of the image. Then you should sharpen. If
this were a color image I would suggest extracting luminosity and sharpening
that, but as you are in B&W, try this (there are more advanced ways):

1. Open the Unsharp Mask filter (Filter>Sharpen>Unsharp)
2. Set the sliders to 60%, 1.5 pixel radius and 0 threshold.
3. Click OK.

You may want to vary the setting for sharpness depending on the image, but
these are settings that will work much of the time.

The other thing you will want to do is learn more about scanning -- which is
really where you will improve your results because from what you are saying
it seems you can get more from the original scan. Specifically, look at your
manual for the scanner and see if there is a way to adjust white and black
points. In getting a better scan, you will start with more and better image
information and that's always a good way to go.

Hope that helps!

Richard Lynch
author, Special Edition Using Photoshop 6
The Hidden Power of Photoshop Elements 2 (Dec. 2002)
http://ps6.com/
-----------------------------------------

<< This is probably a stupid question. I would like to know if there is any
easy way to remove the background from images scanned from books. If I scan a
drawing or diagram from a book that is on a white background, when I paste
the image into a word document and print it, the seemingly white background
of the drawing prints with a light shading. Apart from taking the eraser and
removing all the white in the image (which takes hours) is there a simple way
to ensure that you get only the black and white outlines or any
greyscale/colour shading, without having the white paper of the book showing
up to? >>

Not a stupid question at all. Most people just ignore that stuff.

What should appear as white in your scans is, simply, what will print that
way. As there is no white ink in the color set of a standard printer, any
white that you get will be the color of the paper. Papers are not usually
really white -- there is some color to them. As the scanner is scanning a
color, it is really just doing its job in picking up the information. The
'white' areas of whatever you are scanning from print are actually the color
of the paper they were printed on (or heavily influenced by that). What you
have to do is adjust or eliminate that 'white' color to get better results.
The idea is to have the white in the image match the background, which in
this case will be pure white, or RGB 255,255,255.

Before I go on, a quick warning: be aware that things you are scanning that
have been printed already are, for the most part, copyrighted. You should not
be using them without permission if they will be distributed in any way.

You have several ways to go here, including using the scanner to make a
'better' scan. You can, with some scanning software, adjust your white point,
and in this case it might take care of the problem. If you purposely pick a
darker paper area as the white point when making the scan (you will make a
pre-scan and then find the target using the scanner software), you will
correct the paper to be white -- or registered that way. I'll assume,
however, that you do not have the option to adjust the scan.

What you end up wanting to do at this point is identical to what I've
outlined in Chapter 5 of my first book, and in chapter 25 of Special Edition
Using PS 6 (page 564).

First thing you should do is a standard levels correction for the scan. I
happen to have a tutorial about that up on the ps6.com site:
http://ps6.com/Tutorials/Color_Correction/intro.html

With that correction complete, you will have to do something to flatten out
the white areas that are still not white.  You can do this by creating
mock-transparency (matching the white color to 255,255,255 RGB) as outlined
in the tutorial on transparency:
http://ps6.com/Tutorials/Web_Transparency/steps.html

You might need to use selection and other methods to accomplish your goals.

One other solution that may also come into play is Blend If. Once you have
corrected the scan and taken some basic steps to match the white, there may
still be some very difficult areas to clean up (e.g., in between the halftone
dots in the print!). If you duplicate your flattened image to a new layer,
then create a white layer below, you'll be set to try Blend If.

1. Flatten your image.

2. duplicate the background by dragging it to the duplicate button at the
bottom of the Layers palette

3. Activate the background layer again.

4. Create a new, blank layer. This will fall between the duplicate and the
background.

5. Fill the new layer with white.

6. Double-click the duplicate layer...this will open the Layer Styles
dialogue.

7. Locate the This Layer white slider (right), and slide it in to 253. Let go
of the mouse button.

8. Hold down the Option/ALT key and click on the left side of the white
slider and slide it to the left...the white slider should split in half.
Don't go too far with the slider, and watch the image as a preview of the
result. All you want to do is take the hard-edge off the blend to white. If
the scan has enough resolution, this will work pretty nicely. A lower
resolution scan may perform with a little chunkiness.

As you move the slider to the left more of the lighter color in the image
will turn white because it will be blending with the layer below.

When you get the look that you want, flatten the image, save and apply. After
blending to white to match the paper color, your result should be a lot
better. Make that color correction first before using Blend If, or you will
end up using too much blending, and the uncorrected white color result will
cause problems.

That is, obviously, a lot of information to absorb. People sometimes think
that just because it looks like it should be simple, that it is. Others may
offer lesser solutions (e.g., making a quick selection with the magic wand or
whatever), but try it and see the result difference. One clever solution
might be to load image Luminosity as a selection to lighten white image
areas...What you end up doing really depends a little on the image.

I hope that helps. Let me know if you have more questions.

Richard Lynch
PS6.com
rl@...
----------------------------------

<<I find the Effects drop shadow too limited to accomplish what I want. Is there
a way to manually create a shadow?>>

If you have already selected the subject that you want to apply a drop
shadow to and it is in its own layer, try this (a variation on that other
technique):

1. Start with your object (the one that will be casting the shadow) on its own
layer. It should be clipped from any background.

2. Press Command/CTRL (Mac/PC) and click on the layer in the layers palette
with the object that will be casting the shadow. The object will load as a
selection.

3. Create a new, blank layer below the layer where your object is that is
casting the shadow (or create one above and drag it below in the layers
palette).

4. Feather the selection by several pixels by choosing Feather from the
Select menu (the larger the image and higher the dpi, the more you'll want
to feather). Or, you can blur after deselecting in step 6.

5. Set the foreground swatch to black (press D on the keyboard). Choose the
Paint Bucket tool and fill the selection on the blank layer (created in step
3).

6. Deselect (press Command/CTRL + D), choose the Move tool (press V) and
move the shadow into place.

Adjust layer opacity and mode (Multiply mode suggested) to make the shadow
look as intense as you want. You can apply a gradient mask to make it fade, and
distort or otherwise alter it to fit in with the image landscape or to make it
fall properly across objects.

I use this technique and several others to show how to build realistic
shadows in my new book The Hidden Power of Photoshop Elements 2, due out in
December 2002. Of course, all the technique there holds for Photoshop users...

Hope that helps!

Richard Lynch
rl@...
----------------------