D

Al,

Thanks very much for the details and reasons. I know there must be some
logic to which kinds of things to do first, but I am too new at PS (and PS
is way too complicated) for me to understand what that logic might be. I've
printed out your suggestions and I hope to get into a routine. And then
make myself a nice action.

I take the photos at the highest resolution (just in case I get a really
great one which I might want to print and/or so I can size up an especially
nice area of a pic. But then they are very large to work on when I
eventually just want them for the web. I thought possibly that re-sizing
them first might alter or harm the organization of the pixels when I planned
to do other changes after re-sizing. So I am glad to hear that I can make
them a reasonable size first then play further.

All your tips are great. I read about LAB mode, but hadn't tried it yet.
Off to try all these suggestions now.

Thanks again.

Diane




"Al Denelsbeck" <news@wadingin.net> wrote in message
news:Xns93D698D1CEE99sandalsatwadinginnet@65.32.1. 8...
> "D" <d@sympatico.ca> wrote in
> news:zraZa.5841$ox5.646540@news20.bellglobal.com:
>
> > I'm playing with the images I get from my digital camera with
> > Photoshop 7 (just learning PS). I've done lots of tutorials (Janee
> > and RetouchPro, Russell Brown and more) and I have '50 Fast Photoshop
> > Techniques' book and now I am wondering which is the best ORDER to do
> > the various 'repairs' an image might need or I might want to apply to
> > an image for an effect. So far I am only making images for web sites
> > and posting images on the web, so it doesn't seem to make much
> > difference, but I feel I should get into the right practice,
> > especially since I am planning to make actions to batch changes and I
> > may in the future wish to make prints on other images.
> >
> > I guess I am generally confused about the theory of commands which
> > make changes to the image itself and those which add layers and in
> > what order these should be done, such as those below, which I seem to
> > be using over and over again.
> >
> >
> > A. crop
> > B. rotate
> > C. levels/curves/set white point
> > D. filters
> > E. skew or alter shape to fix distortion of buildings, etc.,
> > F. selection to alter only a portion of the image (darken, lighten,
> > blur etc.)
> > G. resize w x h
> > Y. sharpen (at the end, I know this much)
> > Z. save for web (if appropriate).
> >
> > Is there a list or cheat sheet somewhere? Send me off for more
> > reading, please.
>
> Well, as, uh, "n8" said, there isn't a definitive way, at least not
> that I've found. However, there are some things that may make it easier.
>
> If your end result is *only* intended for web use, resize first. This
> saves a tremendous amount of time with all other changes because they have
> fewer pixels to screw with. BUT, this is shooting yourself in the foot if
> you want multiple uses, such as web and printing, out of it, since you'll
> end up doing it all twice.
>
> If I know it will always need distortion or perspective adjustments
> for whatever usage, I do this first, then crop off the skewed edges. This
> way you know what you're keeping in the image, and you don't waste time
> cloning out some bit of trash or scratches that later get cropped (I've
> done this a few times).
>
> Brightness and contrast before color, since they can seriously affect
> saturation, shadow detail, and so on. Both of these before dodging, etc.
>
> Resize and sharpen *before* esoteric effects or laying in text.
> Sharpening can do terrible things to them.
>
> Oh, and as a side tip, I always sharpen in LAB mode, Lightness
> channel only, viewed at 200% and watching the highest contrast portion of
> the image. If you're getting halos, back off. Works wonders. Occasionally
> sharpening in one of the color channels (like Blue in RGB mode) adds a bit
> of oomph to tricky areas, and many scanners don't respond well in certain
> channels.
>
> Hope this helps,
>
>
> - Al.
>
> --
> To reply, insert dash in address to separate G and I in the domain