Forum Moderators: not2easy
My graphic tool of choice is also Paintshop Pro (an older version). My designs are generally pretty simple (I am no graphic artist!)
I needed to crunch image files (gif and jpg) for minimum size without destroying clarity as well as a preview of their presentation on the html page.
I ended up with "WebGraphics Optimizer Graphics Compression Wizard" from Plenio.
It does the job I need and the crunched file it shows me always seems to look the same on the html page.
Tedster
"I've used BoxTop -- it's good. Also, Spinwave has both a plug-in (HVS JPG) and a stand alone desktop utility (JPEG Cruncher). I use the plug-in, and love it."
Do you consider HVS JPG the best at crunching jpg's? That's what I'm using now, but I'm always on the lookout for something better.
I'll tell you how much I like SFTW, I've begun redoing images on a page I maintain for a professional phographer. I remember well how I struggled to maintain the appearnce of his work and still get respectable file sizes. With SFTW in Photoshop, I am routinely chopping another 30% off his files with no obvious degrading of the images.
The last step I take before saving a JPG is taking the image into L*a*b color space. This space separates all the black-to-white Lightness detail into one channel (L). The other two channels have color information, (a) is green-to-red and (b) is blue-to-yellow.
Then I run a Gaussian blur on the two color channels, but not on the (L) channel. I can often run a very large blur (2 to 3 pixels) with little or no visible effect, because the fine detail is in the black-to-white channel, not the color ones. If there is a sharp red/green or blue/yellow edge, I just exclude 3 pixels or so around that edge before the blur.
When I return to RGB space, the JPG compression can reduce the file size even further because of this blurring.
Bingo! the exact problem I'm trying to solve. Thanks much.
jk321
Very nice! How long did you use PS before you figured that one out? (Or did you cheat, and RTFM or something?) I've been using it since v2.0.1, and I still hadn't figured out any good reason for using LAB color space....
A couple of other good uses for L*a*b:
1) Creating a black & white image. The isolated L channel is often a better choice, or at least a better starting place, than a conversion to grayscale or a 100% desaturation.
2) Sharpening sometimes works best in L*a*b, using just the L channel. In fact, sometimes all an image needs is not sharpening but more contrast in the L channel, or a Curves adjustment there. That way you get a sharper appearance without adding sharpening halos.
These L*a*b techniques can bring out subtle gradations of color that you just can't enhance in RGB, because you are working directly with the Luminance.
L*a*b is certainly under utilized, but it is the native Photoshop color space, and the space utilized by KODAK photodiscs, so it won't be going away anytime soon. If you have an image with a color cast, this can also be an easy place to fix that.