Ad
Ad
Ad
Pages: « 1 2 3 [4]   Bottom of Page
Print
Author Topic: sharpening using a layer  (Read 64926 times)
Stephen Best
Guest
« Reply #60 on: May 28, 2006, 02:01:55 AM »
ReplyReply

Quote
If there are features that others would like to see in the scripts, I am open to incorporating them in version 2. Contact me.
[a href=\"index.php?act=findpost&pid=66607\"][{POST_SNAPBACK}][/a]

I'm experimenting with moving the bulk of my work over to LAB. It would be great if your scripts/actions could accommodate this. A few of the procedures I use the most I've already rewritten (the changes aren't great) but the Blend If numbers will obviously need to change.

The more people working on this and the more sharing of information, the better off the Photoshop community will be.

Best wishes on a speedy recovery.
Logged
Mark D Segal
Contributor
Sr. Member
*
Offline Offline

Posts: 6767


WWW
« Reply #61 on: May 28, 2006, 08:20:17 AM »
ReplyReply

Regardless of the well-known pitfalls of monitors for displaying sharpness, it is nonetheless possible to narrow-down the sharpening problem by examining the monitor image generally at 50% magnification (from my experience). One can at least observe tell-tale signs of over-sharpening - when images look obviously too brittle or "crinkly" to be "natural". This is the more usual issue I happen-upon when sharpening images. When I see this, it is usually a reliable guide that the sharpening needs to be toned-down. One of the real advantages of PK Sharpener Pro is the tremendous flexibility it offers for achieving this at any stage of the sharpening process. Sometimes the adjustment is as simple as reducing the master opacity of the sharpening layer set; sometimes it involves adjusting the opacities of the light or dark contours; sometimes it involves scrapping the layer set and changing the choice of sharpener settings; sometimes I find it effective to simply erase the sharpening - say, over areas of skin that one may not wish to sharpen at all. Each image needs to be evaluated and treated on its own merits, but at least this tool provides the means for doing so most effectively.
Logged

Mark D Segal (formerly MarkDS)
Author: "Scanning Workflows with SilverFast 8....." http://www.luminous-landscape.com/reviews/film/scanning_workflows_with_silverfast_8.shtml
Ben Rubinstein
Sr. Member
****
Offline Offline

Posts: 1710


« Reply #62 on: May 28, 2006, 11:40:39 AM »
ReplyReply

Whether or not sharpening on screen is a viable method, when sharpening to taste, such as in the PK sharpener 'Creative' mode, you have to sharpen by what you see do you not. Painting in sharpness with maskes is impossible any other way. I did find my LCD far too sharp when I started using one (since reverted back to CRT's) but if you use what you see to sharpen based on relative sharpness, i.e. difference between OOF and focus, it does work, you learn to see the sharpness relative to the picture.

I'm still happier using CRT's, the sharpness is closer to how it looks in print.
Logged

bruce fraser
Jr. Member
**
Offline Offline

Posts: 51


« Reply #63 on: May 28, 2006, 01:10:00 PM »
ReplyReply

Quote
Regardless of the well-known pitfalls of monitors for displaying sharpness, it is nonetheless possible to narrow-down the sharpening problem by examining the monitor image generally at 50% magnification (from my experience).
[a href=\"index.php?act=findpost&pid=66779\"][{POST_SNAPBACK}][/a]

It's not that you can't make judgments from the monitor. It's that you have to learn the behavior of YOUR monitor, because it's probably a bit different from anyone else's. And the only reliable way to do that is to work backwards from the print.

Once you've learned the behavior of a particular display (or set of displays—I run a Sony Artisan and an NEC 2180 WG side by side, and they portray sharpness very differently), you can make judgments from various levels of zoom. Just avoid the "odd" zoom percentages—66.7%, 33.3%, etc.—because they're heavily antialiased and will prove misleading.

Quote
Each image needs to be evaluated and treated on its own merits,[a href=\"index.php?act=findpost&pid=66779\"][{POST_SNAPBACK}][/a]

If you're doing fine art or delivering final selects to a client, that's true. If you're shooting 500 widgets on white seamless, though, you may not be able to give every image individual attention. One of the reasons we sweated the presets so much is to let the folks who shoot 3000 images a day make all 3000 "good" automatically, with no user intervention. "Great" comes later in the workflow once the selects have been made.....
Logged
gmitchel
Full Member
***
Offline Offline

Posts: 151


WWW
« Reply #64 on: May 28, 2006, 04:17:20 PM »
ReplyReply

Quote
I'm experimenting with moving the bulk of my work over to LAB. It would be great if your scripts/actions could accommodate this. A few of the procedures I use the most I've already rewritten (the changes aren't great) but the Blend If numbers will obviously need to change.

The more people working on this and the more sharing of information, the better off the Photoshop community will be.

Best wishes on a speedy recovery.
[a href=\"index.php?act=findpost&pid=66768\"][{POST_SNAPBACK}][/a]

Thank you for the suggestion, Stephen. And the kind wishes re. my recovery.

I'm aware that L*a*b sharpening is popular right now. I see wisdom in sharpening in L*a*b when one is already in L*a*b for some other reason. But I sure would not put an image through the color engine twice just to sharpen in L*a*b.

My scripts use a Luminosity blend to avoid color shifts.

They use Blend If slider settings with the Luminosity blends for a couple of reasons. First, Lighten and Darken blends can quite easily get you color shifts along sharpened edges. Second, you have a lot of control over where the sharpening halos get applied and where they are not applied with Blend If settings. You get better protection for your shadows and highlights.

The Blend If setting adjustments work the same in both RGB and L*a*b. In fact, the scripts run just fine in L*a*b mode. They apply to all three L*a*b channels, rather than just the Lightness channel. But, again, they use a Luminosity blend. It would be possible to apply them to just the Lightness channel in L*a*b, and that is something I'll give more consideration.

Thanks again for the suggestion.

Cheers,

Mitch
« Last Edit: May 28, 2006, 04:18:08 PM by gmitchel » Logged
Pages: « 1 2 3 [4]   Top of Page
Print
Jump to:  

Ad
Ad
Ad