Ad
Ad
Ad
Pages: [1]   Bottom of Page
Print
Author Topic: Camera2Print Question: PK Sharpener and B&W?  (Read 3314 times)
TaoSpace
Newbie
*
Offline Offline

Posts: 7


« on: August 13, 2007, 09:04:37 AM »
ReplyReply

Hi,

Like many here I bought Camera to Print and thought it was excellent. Learned a fair amount of new tricks and finally splashed for PK Sharpener too.

One question pops up though: If you are using Camera RAW to convert to B&W, using the nice new Lightroom-like colour-specific HSB sliders, you end up with a 8 or 16-bit greyscale image in Photoshop. This you cannot sharpen with PK Sharpener (automate menu options are greyed out).

What is the best alternative in that case?
1. Do the conversion to greyscale as normal in Camera RAW, then set the colour mode back to RGB, then run the various PK sharpener steps, or...

2. Don't convert to grey in Camera RAW, just get the tonality etc. right, and have it open up as a colour image in Photoshop, then run the sharpening steps, and finally convert it back to B&W in Photoshop (would be a shame since the Camera RAW greyscale conversion seems to have more control then it photoshop equivalent).

Bonus question: In the video it is said that it is best to let the printer handle greyscale and that one therefore has to turn off colour management all together in Photoshop. This seems to work OK, but I noticed a colour shift (to blueish) on some papers (Innova F-Type I think) that way. Anybody have any thoughts?

Thanks so much,

Tao.
Logged
madmanchan
Sr. Member
****
Offline Offline

Posts: 2110


« Reply #1 on: August 13, 2007, 12:36:34 PM »
ReplyReply

How come you can't convert to gray in LR/CR and also have it open in PS as a RGB color image? That's what I do ... I use the LR grayscale controls but then when I open the image in PS it's a color RGB image (in the ProPhoto RGB color space) which happens to be neutral (or near neutral if I added some color toning).
Logged

picnic
Sr. Member
****
Offline Offline

Posts: 574


« Reply #2 on: August 13, 2007, 02:16:19 PM »
ReplyReply

Quote
How come you can't convert to gray in LR/CR and also have it open in PS as a RGB color image? That's what I do ... I use the LR grayscale controls but then when I open the image in PS it's a color RGB image (in the ProPhoto RGB color space) which happens to be neutral (or near neutral if I added some color toning).
[a href=\"index.php?act=findpost&pid=133021\"][{POST_SNAPBACK}][/a]

I need to check Eric's site again (a great job, BTW) to see his recommendations on b/w, but before I bought the 3800, I asked around from some folks that had them (Andrew Rodney was one that posted here about his), and there were quite a number of people whose opinion I respect that simply work with the K3 inks and WYSIWYG.  Jack Flesher was another.  I was printing with a 2200 at that time (and the very old 1280 dye printer) and monotones had been the bane of my life---I used QTR (Quadtone RIP), but still felt I did not have the control I wanted---all of the toning--or not--was done in the RIP and even with many many samples on many many papers, it was still sort of a guess as to what you would get.  I really like QTR--but there is less control than using an RGB image 'as is'.  

This is the same feeling I have with the b/w feature on my 3800.  So--I just print RGB files and they match my calibrated monitor with paper/printer profiles.  I rarely do a 'neutral' mono--almost always from a very slightly toned to a split tone (usually done in ACR 4.1 or LR--but sometimes even in the b/w plugin from Harold Heim).

And--I do just import from LR as a Prophoto RGB 16 bit file if I process as a mono in LR.  

Diane
Logged
TaoSpace
Newbie
*
Offline Offline

Posts: 7


« Reply #3 on: August 13, 2007, 05:48:47 PM »
ReplyReply

Quote
How come you can't convert to gray in LR/CR and also have it open in PS as a RGB color image? That's what I do ... I use the LR grayscale controls but then when I open the image in PS it's a color RGB image (in the ProPhoto RGB color space) which happens to be neutral (or near neutral if I added some color toning).
[a href=\"index.php?act=findpost&pid=133021\"][{POST_SNAPBACK}][/a]

Sure, I can do that, but that would be the same as changing the colour mode back to RGB after importing it as greyscale. What I was wondering about was whether PK Sharpener is better at sharpening real colour files (without any kind of B&W conversion) than it is at sharpening greyscale ones.

In other words, should sharpening happen before B&W conversion or is it OK to do it afterwards. Or is there really no difference?

Regards,
-Tao.
Logged
madmanchan
Sr. Member
****
Offline Offline

Posts: 2110


« Reply #4 on: August 13, 2007, 09:11:21 PM »
ReplyReply

Oh I see what you're getting at now. My impression is that you should convert your image to B&W first, then sharpen. I don't think it much matters which you do first in terms of practical results, but in principle if you sharpen in full color mode before component mixing to get your B&W result, then you might get some parts of the image with more sharpening effects than others (in an undesirable way).
Logged

picnic
Sr. Member
****
Offline Offline

Posts: 574


« Reply #5 on: August 13, 2007, 09:51:27 PM »
ReplyReply

Quote
Oh I see what you're getting at now. My impression is that you should convert your image to B&W first, then sharpen. I don't think it much matters which you do first in terms of practical results, but in principle if you sharpen in full color mode before component mixing to get your B&W result, then you might get some parts of the image with more sharpening effects than others (in an undesirable way).
[a href=\"index.php?act=findpost&pid=133089\"][{POST_SNAPBACK}][/a]
 

It also seems to me that since the accepted practice is to sharpen last after resize, all other processing for printing, web, that it only makes sense to sharpen after conversion to mono.  You may also wish to do some dodging and burning, perhaps some masking, etc.--after the mono conversion--and one would not want to do final sharpening prior to those processes.

Diane
Logged
Ernst Dinkla
Sr. Member
****
Offline Offline

Posts: 2868


« Reply #6 on: August 14, 2007, 02:37:34 AM »
ReplyReply

Quote
One question pops up though: If you are using Camera RAW to convert to B&W, using the nice new Lightroom-like colour-specific HSB sliders, you end up with a 8 or 16-bit greyscale image in Photoshop. This you cannot sharpen with PK Sharpener (automate menu options are greyed out).

What is the best alternative in that case?
1. Do the conversion to greyscale as normal in Camera RAW, then set the colour mode back to RGB, then run the various PK sharpener steps, or...

2. Don't convert to grey in Camera RAW, just get the tonality etc. right, and have it open up as a colour image in Photoshop, then run the sharpening steps, and finally convert it back to B&W in Photoshop (would be a shame since the Camera RAW greyscale conversion seems to have more control then it photoshop equivalent).

Bonus question: In the video it is said that it is best to let the printer handle greyscale and that one therefore has to turn off colour management all together in Photoshop. This seems to work OK, but I noticed a colour shift (to blueish) on some papers (Innova F-Type I think) that way. Anybody have any thoughts?

Thanks so much,

Tao.
[{POST_SNAPBACK}][/a]

There are no real differences between color sharpening and B&W sharpening according to Mike Chaney. He gave the following answers to my questions on the Qimage's sharpening functions.

">Remains that question whether the interpolation and sharpening
>> qualities of Qimage are as effective in B&W (quad) printing as
>> they are in color. Or are there better algorithms for B&W only ?

The techniques should be identical.  I don't know of any reason to treat
B/W sharpening and interpolation any differently than color.

Mike

>With a B&W image, you are altering only luminance values, so you should
>> be able to use higher levels of sharpening..  That leads to a question
>> of whether QImage currently defaults to higher sharpening values when
>> sharpening B&W images?


Shouldn't matter.  Qimage sharpens based on the luminance channel to
begin with, even in color images.

Mike"               End of quotes.

With Qimage a greyscale file treated to one of its features will be saved or goes en route to the printer as an RGB file, extrapolation and sharpening is done after the conversion to RGB and it will not convert the image back to greyscale. It is possible that your software tries to avoid that conversion and greys out the function for that reason.

There are 2 or 3 stages recommended (Bruce Fraser) to sharpen in image processing so your question is a bit more complicated.  The first sharpening part could be done in the Raw conversion itself but I understand that isn't the way you want it. I would say go to greyscale 16 bit right away with Gamma 2.2 embedded, convert to RGB with AdobeRGB embedded do the sharpening, convert back to greyscale for the rest and print with Qimage for the extrapolation + last print sharpening. If you are on a Mac you have to find another last part which will be more complicated.

I would say keep CM on in Photoshop and use an embedded greyscale profile that links to the gamma of the color space you normally use: Gamma 2.2 <> AdobeRGB.
The problem with Photoshop's CM is that it never is really off and it it is hard to say what still works and what not. If you switch CM off in Picture Window Pro it is OFF.
CM on in PS + a good monitor profile + printer profiles (including QTR and PrintFix Pro profiles) keeps the softproof features + normal tonal range control on the monitor and it is a reliable CM for the conversion of the image to the printer profile.

Ernst Dinkla

try: [a href=\"http://groups.yahoo.com/group/Wide_Inkjet_Printers/]http://groups.yahoo.com/group/Wide_Inkjet_Printers/[/url]
Logged
JeffKohn
Sr. Member
****
Offline Offline

Posts: 1671



WWW
« Reply #7 on: August 14, 2007, 07:48:04 AM »
ReplyReply

The fact that Photokit Sharpener can only sharpen the background layer of an RGB image is a limitation of the fact that it's just a script, and not a real plugin. (Just one of many reasons why I think Photokit Sharpener is overrated).

You can still do your B/W conversion in ACR, just do Image -> Mode -> RGB before sharpening (might also want to switch back to grayscale afterwards to make your file sizes less wasteful).
Logged

madmanchan
Sr. Member
****
Offline Offline

Posts: 2110


« Reply #8 on: August 14, 2007, 10:09:01 AM »
ReplyReply

Quote
The fact that Photokit Sharpener can only sharpen the background layer of an RGB image is a limitation of the fact that it's just a script, and not a real plugin.
[a href=\"index.php?act=findpost&pid=133170\"][{POST_SNAPBACK}][/a]

I don't think that's right. It is true that PKS requires the image to have a background layer, but PKS always operates by creating a new topmost layer that effectively contains the rest of the layers, flattened, before applying the sharpening. In other words, PKS always creates a new layer and then sharpens the new layer -- the original background layer isn't touched.

The requirement of a background layer, as explained by Bruce Fraser in his Real World Image Sharpening book, is because of a Photoshop technicality -- it guarantees the ability to create a new topmost layer that has the contents of all of the other layers.
Logged

JeffKohn
Sr. Member
****
Offline Offline

Posts: 1671



WWW
« Reply #9 on: August 14, 2007, 10:20:26 AM »
ReplyReply

Quote
I don't think that's right. It is true that PKS requires the image to have a background layer, but PKS always operates by creating a new topmost layer that effectively contains the rest of the layers, flattened, before applying the sharpening. In other words, PKS always creates a new layer and then sharpens the new layer -- the original background layer isn't touched.

The requirement of a background layer, as explained by Bruce Fraser in his Real World Image Sharpening book, is because of a Photoshop technicality -- it guarantees the ability to create a new topmost layer that has the contents of all of the other layers.
[a href=\"index.php?act=findpost&pid=133201\"][{POST_SNAPBACK}][/a]
Didn't mean to imply that it sharpens directly on the background layer, only that you have to have one (which sometimes I don't). I still don't like the way PK Sharpener works. Most real plugins will operate on whatever is the current layer (or even portion of a layer if you have something selected) which IMHO is more flexible.
« Last Edit: August 14, 2007, 10:22:22 AM by JeffKohn » Logged

madmanchan
Sr. Member
****
Offline Offline

Posts: 2110


« Reply #10 on: August 14, 2007, 03:32:26 PM »
ReplyReply

Quote
Most real plugins will operate on whatever is the current layer (or even portion of a layer if you have something selected) which IMHO is more flexible.
[a href=\"index.php?act=findpost&pid=133203\"][{POST_SNAPBACK}][/a]

Sure, but on the other hand, what happens if the current layer is an adjustment layer (e.g., a curves, levels, or HSL adjustment layer)? Or, what happens if the current layer is a standard pixel layer but contains a mask? Wouldn't a new pixel layer be needed anyways for the purposes of sharpening?
Logged

Charles Gast
Sr. Member
****
Offline Offline

Posts: 250


« Reply #11 on: August 14, 2007, 04:22:42 PM »
ReplyReply

Have you tried FocalBlade?  I am using Nik right now and wonder if FocalBlade might do better.
Logged
JeffKohn
Sr. Member
****
Offline Offline

Posts: 1671



WWW
« Reply #12 on: August 15, 2007, 10:45:54 AM »
ReplyReply

Quote
Sure, but on the other hand, what happens if the current layer is an adjustment layer (e.g., a curves, levels, or HSL adjustment layer)? Or, what happens if the current layer is a standard pixel layer but contains a mask? Wouldn't a new pixel layer be needed anyways for the purposes of sharpening?
[a href=\"index.php?act=findpost&pid=133271\"][{POST_SNAPBACK}][/a]
Just because added flexibility means it's possible to do things that don't really make sense, doesn't mean that added flexibility is a bad thing. I'd rather the filter let me decide what type of data it can be used on. There are some things you just can't do with PKS because of the way it works: you can't sharpen individual channels, you can't use it in LAB mode, etc.
Logged

JeffKohn
Sr. Member
****
Offline Offline

Posts: 1671



WWW
« Reply #13 on: August 15, 2007, 10:49:35 AM »
ReplyReply

Quote
Have you tried FocalBlade?  I am using Nik right now and wonder if FocalBlade might do better.
[a href=\"index.php?act=findpost&pid=133286\"][{POST_SNAPBACK}][/a]
For landscapes where I'm doing "detail" capture sharpening (small radius, large amount, no edge mask), I prefer good old USM. But for images where I want to use edge sharpening I prefer FocalBlade to PKS because it's a real plugin, and IMHO it gives you more control: you have independent control of the surface and edge sharpening parameters, you have full control over the edge mask, and it also has other parameters such as halo fixing, etc. I'd suggest giving the demo a try.
Logged

madmanchan
Sr. Member
****
Offline Offline

Posts: 2110


« Reply #14 on: August 15, 2007, 10:54:51 AM »
ReplyReply

Jeff, we're probably getting a little too far from the OP's preferred topic, so I'll make this my last post on the subject of PKS here ... I agree that PKS doesn't directly let you do those things you mentioned, but I don't see why you'd want to apply PKS in Lab mode (for one, PKS already limits itself to the luminance information, which is close though not identical to the L* channel of Lab, and second, converting from RGB to Lab and back is a lossy process). I'm also unsure of the benefit of the sharpening on individual channels. Presumably you're referring to wanting to not sharpen the blue channel so much, since the blue one is more prone to noise. But if you sharpen the luminance channel (which PKS does) then it's already weighting red and green much more than blue (it's roughly a 30/60/10 weighting for RGB to get luminance) so again I don't think it's necessary ...

Anyways ...  
Logged

Pages: [1]   Top of Page
Print
Jump to:  

Ad
Ad
Ad