Ad
Ad
Ad
Pages: [1] 2 »   Bottom of Page
Print
Author Topic: Printing from Lightroom 2.1 v. PS CS3 - Some Observations  (Read 11758 times)
Josh-H
Sr. Member
****
Offline Offline

Posts: 1910



WWW
« on: October 26, 2008, 07:20:08 PM »
ReplyReply

Firstly apologies for what is probably going to be a reasonably lengthy post - but there really isn't a short way to put this on 'paper'. And there probably are not even any definitive answers - but the observations are certainly worth discussion I believe.

I have been printing from CS2 and CS3 for a few years now - my workflow has been to do as much as possible in Lightroom [previously DPP] - convert to 16 bit Tiff in Pro Photo and then softproof in CS3, make any corrections required and output sharpen with Photokit. Pretty much the workflow in 'From Camera to Print' - and it produces wonderful results.

Post the release of Michael and Jeff's Lightroom 2.0 Tutorial I decided to try printing from Lightroom as the workflow simplification implications are considerable. I finally made my first prints last night and here is what I have found.

I used a Canon 1DSMKIII file that I shot with the Canon 50mm F1.2L at F5.6 of my 6 month old daughter - this particular capture is extremely sharp, but has wonderful gradations of skin tone. I processed the file with the normal sort of corrections I would make and capture sharpened it in LR2.1 to what I believe was optimal. I have had some experience at using LR's capture sharpening with 1DSMKIII files and feel I have it pretty much nailed. I am always a little light handed with my capture sharpening if anything as I cant stand the artifacts created by over-sharpening.

I then converted to Tiff 16 Bit Pro Photo to try and keep the results the same between LR2.1 and CS3.

I first printed the file from CS3 using the workflow I had been used to. I printed on a Canon IPF5100 on Ilford Gold Fibre Silk with a custom made profile with a X-Rite Spectrophotometer that I am very familiar with and that produces great accurate results. I did NOT  softproof the image other than to choose the better rendering intent - in this case - Relative Colorimetric. I made no adjustments as a result of the softproof. The result was a lovely A4 print made at 500ppi [output sharpened at 480 - but then resized without resampling to 500 to make it fit on the paper]. I used the Canon printer plugin for CS3 in 16 bit.

For the Lightroom print - I took the same Tiff file and printed it using LR's output sharpening of 'Standard' and 'Glossy'. I chose Vendor matching in the Canon print dialogue and turned off Color Correction to avoid the LR bug of double profiling with IPFx100 series printers from macs. I obviously selected the same custom profile in LR. I did not resample the file in LR print module - it was printed with Lightroom showing the resolution at 480+. I selected 16 bit printing.

The result - the Lightroom print is significantly sharper - The color between the two prints is for all intent and purposes 'identical' and accurate with the on screen file.  So I know I have all my color settings correct. Its arguable that the print from CS3 shows slightly smoother gradations on the skin tones - but its a quibble and may just be the result of the overall  print being less sharp, really to hard to tell. The CS3 print on its own looks great, but next to the LR print does look a little soft.  The LR print is really tack sharp - not overly sharp, just dam sharp. Its also arguable that the LR print is ever so slightly darker than the CS3 print [maybee...] - but again its a quibble and may just be viewer dependant.

What isnt a quibble is the noticeable difference in print sharpness. And I am wondering if this is a result of improved sharpening algorithims in LR as opposed to those used in Photokit. The curve ball of course being there are 3 levels of user selectable sharpening in LR print module and I have not yet experimented with either low or high. But would be very interested to see what other people are seeing.

Clearly all of this is going to be image dependant, and results will vary depending on the quality of the capture and the output sharpening chosen. I would very much like to hear the experiences of others though - have they found similar results?

Of course I cant finish this post without screaming for soft proofing in LR - so PLLEEEEAAASSSE  
« Last Edit: October 26, 2008, 07:24:18 PM by Josh-H » Logged

Schewe
Sr. Member
****
Offline Offline

Posts: 5499


WWW
« Reply #1 on: October 26, 2008, 08:43:34 PM »
ReplyReply

Quote from: Josh-H
I would very much like to hear the experiences of others though - have they found similar results?


Well, that pretty much mirrors my experience as well (course, I guess I might be a bit biased since Bruce and I had a little bit to do with the outcome).

My workflow for final high quality output is indeed to go from Lightroom to Photoshop for final imaging (including soft proofing) and do final printing from Lightroom for the workflow benefits (and to use the sharpening).

The upside of this is when the final tiff image has been made in Photoshop and reimported into Lightroom, I can repurpose that single image to just about any print size and paper (by popping the layer image back into Photoshop to re-soft proof for different paper) without having to do multiple iterations from that file.

So, I wonder why you are surprised that this whole thing works?

:~)
Logged
Josh-H
Sr. Member
****
Offline Offline

Posts: 1910



WWW
« Reply #2 on: October 26, 2008, 10:31:23 PM »
ReplyReply

Quote from: Schewe
Well, that pretty much mirrors my experience as well (course, I guess I might be a bit biased since Bruce and I had a little bit to do with the outcome).

My workflow for final high quality output is indeed to go from Lightroom to Photoshop for final imaging (including soft proofing) and do final printing from Lightroom for the workflow benefits (and to use the sharpening).

The upside of this is when the final tiff image has been made in Photoshop and reimported into Lightroom, I can repurpose that single image to just about any print size and paper (by popping the layer image back into Photoshop to re-soft proof for different paper) without having to do multiple iterations from that file.

So, I wonder why you are surprised that this whole thing works?

:~)

Thanks Jeff - Its very comforting to hear I am seeing results that mirror your own.

As for being surprised - pleasant surprises like this are more than welcome :-)

Logged

AndreG
Jr. Member
**
Offline Offline

Posts: 84



WWW
« Reply #3 on: October 29, 2008, 06:38:42 PM »
ReplyReply

Quote from: Josh-H
Thanks Jeff - Its very comforting to hear I am seeing results that mirror your own.

As for being surprised - pleasant surprises like this are more than welcome :-)

I dare to ask a nagging question that I have: would one have more control and precision using PKSharpener (the underlying technology of the Sharpening in LR 2.1) in Photoshop? I do have a sense that it does.

Thank you Jeff for the 64 bit version. It works great under Windows XP 64 bit.
Logged

Schewe
Sr. Member
****
Offline Offline

Posts: 5499


WWW
« Reply #4 on: October 29, 2008, 08:20:05 PM »
ReplyReply

Quote from: pratic
I dare to ask a nagging question that I have: would one have more control and precision using PKSharpener (the underlying technology of the Sharpening in LR 2.1) in Photoshop? I do have a sense that it does.

Control and precision? No...flexibility, yes.

Since you end up with layers in Photoshop, that give you flexibility. As far as precision? I don;t think so...while I have often tweaked the capture sharpening in PKS, I rarely ever bothered to tweak the output sharpening. And, you have the ability to tweak the capture sharpening in CR/LR to your hearts desire albeit with only local adjustments. You can tweak the local sharpening which is a single parameter sharpening only but offers local control.
Logged
AndreG
Jr. Member
**
Offline Offline

Posts: 84



WWW
« Reply #5 on: October 30, 2008, 09:15:00 AM »
ReplyReply

Quote from: Schewe
Control and precision? No...flexibility, yes.

Since you end up with layers in Photoshop, that give you flexibility. As far as precision? I don;t think so...while I have often tweaked the capture sharpening in PKS, I rarely ever bothered to tweak the output sharpening. And, you have the ability to tweak the capture sharpening in CR/LR to your hearts desire albeit with only local adjustments. You can tweak the local sharpening which is a single parameter sharpening only but offers local control.


The answer is quite clear. Thank you for your time and expertise.
Logged

Ernst Dinkla
Sr. Member
****
Offline Offline

Posts: 2867


« Reply #6 on: October 31, 2008, 03:17:37 AM »
ReplyReply

Quote from: Schewe
So, I wonder why you are surprised that this whole thing works?

:~)


Yes, it worked in Qimage for 5 years already, including a softproof, so why shouldn't it work in Lightroom ?



Ernst Dinkla

Try: http://groups.yahoo.com/group/Wide_Inkjet_Printers/



Logged
Schewe
Sr. Member
****
Offline Offline

Posts: 5499


WWW
« Reply #7 on: October 31, 2008, 11:44:28 AM »
ReplyReply

Quote from: Ernst Dinkla
Yes, it worked in Qimage for 5 years already


Qimage has automatic output sharpening for the resolution and media of the final paper of a print? You sure?
Logged
bjanes
Sr. Member
****
Offline Offline

Posts: 2823



« Reply #8 on: October 31, 2008, 01:56:45 PM »
ReplyReply

Quote from: Schewe
Qimage has automatic output sharpening for the resolution and media of the final paper of a print? You sure?

The link provided by Ernst Dinkla only points to a forum and not a specific post regarding Qiamge. I didn't do a search to see what claims were made for Qimage and I don't know if output sharpening was claimed.

If you look at the information in the sidebar on the Qimage web site, they make no claims of output sharpening, but do point out that their printing software sends the image to the printer at the native resolution of the printer, which they state is usually 600 ppi for Canon and HP and 720 ppi for Epson. This avoids resampling of the image by the printer driver.

It is often claimed that for optimum results, one should send data to the printer at the "native resolution" of the printer or even multipliers thereof. Bruce Fraser gives some useful comments in his sharpening book (pages 64-66). With continuous tone printers such as the Fuji Frontier (or the more upscale Océ LightJet), the addressable resolution (the resolution at which the printer lays down drops of ink) is the same as the pixel resolution. With these printers it has often been recommended to use even multiples of the printer resolution. Bruce reported that strategy was advantageous for some older printers but not for most current high end units. Presumably the resampling algorithm used in a printer costing hundreds of thousands of dollars is equal to or better than the one used in one's own image processing program.

With inkjet printers using error diffusion dither output, the relationship between addressable resolution and pixel resolution is rather indirect. Bruce stated the commonly accepted "native resolution" of Canon and HP printers was 300 ppi and that of Epson printers was 360 ppi. Bruce further stated that sending data to the printer at even multiples of the native resolution of these inkjet printers could give better results with line pair targets but not with most real world images. Within certain limits, he did not suggest resampling to the native resolution.

Going back to the Photoshop vs Lightroom printing differences, one should note that the resampling algorithm is different in the two applications (reference). For example, you are recommended to choose a Bicubic (smoother) when making an image bigger for smoother results, or Bicubic (sharper) when reducing the image size. There are no interpolation options available in Lightroom. Instead, a single ‘Lanczos kernel’ method of interpolation is used in both Lightroom and Camera Raw. If you resample prior to printing, this could make a difference. I think that Photoshop sends data to the printer at whatever resolution is in use; perhaps LR resamples to the native resolution ala Qimage.  There may also be significant differences between the PK output sharpening in Photoshop and the built in Lightroom output sharpening. Perhaps we need a new version of PK for use in Photoshop.

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

Posts: 2867


« Reply #9 on: October 31, 2008, 02:59:46 PM »
ReplyReply

Quote from: Schewe
Qimage has automatic output sharpening for the resolution and media of the final paper of a print? You sure?


In that 5 years I have used it every day

Quote from the manual:

Final Print Sharpening - Qimage offers the ability to apply different levels of unsharp mask to the final print. This unsharp mask is applied after all interpolation has been performed (after your print has been resampled to the DPI of your printer). Since this setting affects the sharpness of your prints, a setting should be used that makes your final prints match what is displayed on your monitor with respect to sharpness. The type of printer used and the print driver version can affect apparent sharpness of prints. It is best to leave this setting at the default slider position unless you prefer softer or sharper printed images. To make images look sharper, slide the slider to the right. To make images appear softer in print, slide the slider to the left. The following are notes that apply to the two different types of final sharpening available in Qimage:

Smart sharpening: Smart sharpening uses information such as resolution of the original image, the printer driver PPI, and the final print size to set the parameters of the sharpening algorithm to produce consistent, vibrant sharpening in every print.
Normal sharpening: Normal sharpening will sharpen only the finest details in the print since it sharpens the image based only on the print driver PPI. Since this method does not take all factors into account with respect to the appearance of sharpness, the effect of this type of sharpening is more subtle and less consistent, and will depend on the print driver PPI and size of the final print.
Note: The sharpening level (slider) applies to both types of sharpening above and simply controls the level for the selected type of sharpening.

End of quote.

[attachment=9352:QimageSharpener.jpg]


For more Lightroom inspiration:
http://www.ddisoftware.com/qimage/


Ernst Dinkla

Try: http://groups.yahoo.com/group/Wide_Inkjet_Printers/
Logged
Schewe
Sr. Member
****
Offline Offline

Posts: 5499


WWW
« Reply #10 on: October 31, 2008, 07:14:18 PM »
ReplyReply

Quote from: Ernst Dinkla
In that 5 years I have used it every day


That doesn't sound like automatic sharpening based on image resolution and media type...seems like ya still gotta finagle the results.
Logged
Ernst Dinkla
Sr. Member
****
Offline Offline

Posts: 2867


« Reply #11 on: November 01, 2008, 05:13:08 AM »
ReplyReply

Quote from: Schewe
That doesn't sound like automatic sharpening based on image resolution and media type...seems like ya still gotta finagle the results.


There's no way to know what Lightroom actually does under the hood with a setting like "Gloss" or "Matte" but my best guess is that it takes the higher dynamic range of most gloss papers into account for the sharpening degree. The true optical resolution possible on gloss and matte papers isn't that far apart and there's more deviation on that within the two categories instead so it is unlikely and possibly stupid to aim for a higher optical resolution of one category when applying the specific sharpening. In fact the possible optical resolution is already represented per media preset as a wider or smaller choice in driver quality settings and by that in the resulting native resolutions. Qimage Smart Sharpening algorithms take into account the original's PPI resolution/size, the printer's print quality settings DPI/unidirectional/strokes etc = native PPI resolution, the print size and the related up- or downsampling. There's a Print Filter in the print output line where you could add a custom filter for gloss or matte but most users will find a setting on the sharpening slider based on test prints easier. A far more subtle method in total than a choice of two categories, matte and gloss isn't representing today's media choices. The total of Qimage settings can be saved with an appropriate name and recalled for a similar job later on. For Web use the Print to File setting has a choice to go along the printer output channel or on a straight path without the on-the-fly print actions. Again it's easy to make a print settings shortcut for Web use and decide carefully the anti-aliasing on downsampling in that case as downsampling will most likely be needed.

That Qimage allows more control on the on-the-fly functions used later on isn't an indication that it isn't automatic at printing time. Far from it.

On this site it doesn't have the appeal of Apple, Adobe. Not a thing I'm concerned about.


Ernst Dinkla

Try: http://groups.yahoo.com/group/Wide_Inkjet_Printers/
« Last Edit: November 01, 2008, 05:15:39 AM by Ernst Dinkla » Logged
William Morse
Full Member
***
Offline Offline

Posts: 174


WWW
« Reply #12 on: November 01, 2008, 09:08:31 AM »
ReplyReply

No, Jeff, it's much better! It starts out with "automatic sharpening based on image resolution and media type", and then allows you to actually make aethetic choices based on your image and style requirement.

While I'm sure that your product works very well, It seems that you're trying just a little bit too hard here. Frankly, you only lose credibility in the process, from my point of view.

Bill

Quote from: Schewe
That doesn't sound like automatic sharpening based on image resolution and media type...seems like ya still gotta finagle the results.
Logged

Wm. Morse Editions
A Fine-Art Digital Printmaking Studio
Photography, Limited Editions, Film Drum Scanning
Restored Early 20th Century Photos of China
www.MorseEditions.com
Schewe
Sr. Member
****
Offline Offline

Posts: 5499


WWW
« Reply #13 on: November 01, 2008, 12:34:11 PM »
ReplyReply

Quote from: Ernst Dinkla
There's no way to know what Lightroom actually does under the hood with a setting like "Gloss" or "Matte" but my best guess is that it takes the higher dynamic range of most gloss papers into account for the sharpening degree. The true optical resolution possible on gloss and matte papers isn't that far apart and there's more deviation on that within the two categories instead so it is unlikely and possibly stupid to aim for a higher optical resolution of one category when applying the specific sharpening.


Well, actually, there is a way to know because I worked with the engineer who wrote the code, so, I know EXACTLY what differences are taken into account when applying the matte and glossy settings in Lightroom. And no, I can't tell you other than to say that using the matte on glossy paper produces poorer results while using the glossy on matte paper, the high detail sharpening is wasted. It doesn't look bad, but it's working on detail that the paper can't retain.

As far as the inherent differences that matte and glossy paper can retain, I don't agree with you at all. Due to both surface and ink set, "glossy" paper can resolved far more detail than "matte" paper can. Take Epson pro printers for example, many of the watercolor or fine art paper can't even be set to print at 2880 and are limited to 1440. A high resolution original properly sharpened and printed on glossy type paper can out resolve the high frequency textural on "matte" paper. Enhanced Matte or UltraSmooth Fine Art paper can hold quite a bit of detail to be sure, but not as much as EFP or Luster. I've done enough side by side prints to know that.

As far as the deviation within a category, to an extend that's true...Premium Glossy can hold more detail than Semi-Gloss but both hold more detail than Enhanced Matte. Enhanced Matte holds detail better than textured watercolor paper but neither are in the league of Exhibition Fiber Paper (EFB).

The whole reason that PhotoKit Sharpener developed a matte and glossy approach was that in testing Bruce Fraser determined that the media did indeed have an impact on the sharpening needed for optimal output sharpening. He also found no pressing need for a "range" of sharpening based upon the individual paper properties in a category–although at one point we were going to do an "uncoated" category for natural papers but there doesn't seem to be much call for that paper any more since there are now so many good, coated matte and watercolor papers available.
Logged
Ernst Dinkla
Sr. Member
****
Offline Offline

Posts: 2867


« Reply #14 on: November 01, 2008, 04:06:11 PM »
ReplyReply

Quote from: Schewe
Well, actually, there is a way to know because I worked with the engineer who wrote the code, so, I know EXACTLY what differences are taken into account when applying the matte and glossy settings in Lightroom. And no, I can't tell you other than to say that using the matte on glossy paper produces poorer results while using the glossy on matte paper, the high detail sharpening is wasted. It doesn't look bad, but it's working on detail that the paper can't retain.

As far as the inherent differences that matte and glossy paper can retain, I don't agree with you at all. Due to both surface and ink set, "glossy" paper can resolved far more detail than "matte" paper can. Take Epson pro printers for example, many of the watercolor or fine art paper can't even be set to print at 2880 and are limited to 1440. A high resolution original properly sharpened and printed on glossy type paper can out resolve the high frequency textural on "matte" paper. Enhanced Matte or UltraSmooth Fine Art paper can hold quite a bit of detail to be sure, but not as much as EFP or Luster. I've done enough side by side prints to know that.

As far as the deviation within a category, to an extend that's true...Premium Glossy can hold more detail than Semi-Gloss but both hold more detail than Enhanced Matte. Enhanced Matte holds detail better than textured watercolor paper but neither are in the league of Exhibition Fiber Paper (EFB).

The whole reason that PhotoKit Sharpener developed a matte and glossy approach was that in testing Bruce Fraser determined that the media did indeed have an impact on the sharpening needed for optimal output sharpening. He also found no pressing need for a "range" of sharpening based upon the individual paper properties in a category–although at one point we were going to do an "uncoated" category for natural papers but there doesn't seem to be much call for that paper any more since there are now so many good, coated matte and watercolor papers available.


In a thread on the Digital B&W mailing list "How tiny dots determine real resolution in a B&W ink print" some qualified judges of (pigment) inkjet B&W prints had the opinion that it isn't so sure that gloss inkjet paper has the higher optical resolution. Gloss inkjet, not analogue gloss paper. That discussion was started after Tyler Boley published his page with Cone's K7 print enlargments. All pigment inks like we normally discuss here.
http://www.custom-digital.com/2008/09/bw-print-quality/
I have mentioned the greater dynamic range and inherent higher contrast possible in gloss prints. That's another parameter and could lead to different sharpening needed but shouldn't be confused with optical resolution. So I do not argue that different sharpening is needed but I also set it against the deviation in coating quality within the two categories and then I rather go by what the test prints with Qimage's method tell me.


Ernst Dinkla

Try: http://groups.yahoo.com/group/Wide_Inkjet_Printers/


Logged
alan a
Jr. Member
**
Online Online

Posts: 95


« Reply #15 on: November 01, 2008, 04:23:40 PM »
ReplyReply

I must be doing something wrong, and maybe Schewe can help me out.  (He might not be favorably disposed to me at the moment, but I'll ask anyway.)  

I have attempted, below, to ask specific questions.  I know my vague questions drive Schewe crazy, but this is as precise and specific as I can be:

For large 22x32 prints from a Canon 1dsIII, my experience is that the Lightroom sharpening is too soft as compared with USM in CS3.  I am printing very sharp landscapes.  So I never use LR for sharpening, and export to CS3 and use USM.

My normal USM settings are 300%, .8 radius and 2 threshold.  I don't use of the sharpening in LR and just export to CS3 and do it there.

Everyone else appears to be getting great results in LR so I must be doing something wrong.  So several questions:

(1)  How do the LR sharpening settings correspond to USM in CS3?  I know the settings are not the same, because Reichamann and Schewe explained that in their excellent LR tutorial.  But I don't understand how the numbers correspond.

(2)  If I want to print a 16x20 or 22x32 and have the same level of sharpening as I can get in one step using USM at those settings (300%, .8 radius and 2 threshold), what settings do I use for the capture sharpening in LR?  I assume I am using the wrong settings there, because that is the only variable that can be controlled.

(3)  If the radius in USM is reduced to .5 or .6, how does that compare to LR?  If the threshold in USM is increased, how does that compared in LR?

Finally, I still use various noise plug-ins in CS3 such as Noiseware or Noise Ninja.  They also appear to work easier and better than noise reduction in LR.
« Last Edit: November 01, 2008, 04:24:29 PM by alan a » Logged
Schewe
Sr. Member
****
Offline Offline

Posts: 5499


WWW
« Reply #16 on: November 01, 2008, 05:18:46 PM »
ReplyReply

Quote from: alan a
For large 22x32 prints from a Canon 1dsIII, my experience is that the Lightroom sharpening is too soft as compared with USM in CS3.  I am printing very sharp landscapes.  So I never use LR for sharpening, and export to CS3 and use USM.

then you are not optimizing the capture sharpening in the Detail panel...

Quote
(1)  How do the LR sharpening settings correspond to USM in CS3?  I know the settings are not the same, because Reichamann and Schewe explained that in their excellent LR tutorial.  But I don't understand how the numbers correspond.

Well, you could read my book or this aritlce About Camera Raw 4.1 to have a better understanding of the Lightroom capture sharpening controls...they are the same as Camera Raw's. As for corrosponding to Ps's USM, they really don't. But if you take the detail slider all the way to 100, it will "simulate" what USM does except that since CR/LR is working on luminance data in linear space, the effective radius is LR/CR is different. If you take the LR radius and multiply by either 1.3 or 1.4 or 1.6 (I really can't remember what the factor is for it's a multiplier greater than 1 and ess than 2) then you'll get the equivalent radius in USM. So, that means you need a slightly lower radius in LR/CR compared to USM. There is no function that works like the Detail slider in USM cause Detail doesn't do a threshold it's a halo pining multi-frequency adjustment between extremely high frequency (towards 100) and a lower frequency (towards zero). And the Mask function replicate a 21-22 step series of Photoshop steps in a single slider...something USM couldn't hope to do in a single step...

Quote
(2)  If I want to print a 16x20 or 22x32 and have the same level of sharpening as I can get in one step using USM at those settings (300%, .8 radius and 2 threshold), what settings do I use for the capture sharpening in LR?  I assume I am using the wrong settings there, because that is the only variable that can be controlled.

Don't know...depends on your image. The aim of the sharpening controls is to make the image "look good" at 100% zoom...what settings to use will depend on your image content and capture size and your taste. A useful starting point would be the Landscape preset included in Lightroom...

Really, USM is so last millennium...for digital capture, Lightroom and Camera Raw are pretty state of the art for image detailing. Once you learn how to use the controls, Photoshop USM seems so primitive.
Logged
bjanes
Sr. Member
****
Offline Offline

Posts: 2823



« Reply #17 on: November 01, 2008, 06:15:18 PM »
ReplyReply

Quote from: alan a
(2)  If I want to print a 16x20 or 22x32 and have the same level of sharpening as I can get in one step using USM at those settings (300%, .8 radius and 2 threshold), what settings do I use for the capture sharpening in LR?  I assume I am using the wrong settings there, because that is the only variable that can be controlled.

Finally, I still use various noise plug-ins in CS3 such as Noiseware or Noise Ninja.  They also appear to work easier and better than noise reduction in LR.

It sounds like you are using one step sharpening in CS3. You should describe your sharpening procedure in more detail. Output sharpening is normally performed as a final step prior to printing and assumes that capture sharpening has already been applied. One step sharpening is no longer recommended.

If you are using a Canon 1dsIII at a low ISO, I doubt Noiseware or Noise Ninja is necessary and these tools will degrade sharpness. If sky or other uniform areas need noise reduction, you might consider using a mask.

Bill
Logged
alan a
Jr. Member
**
Online Online

Posts: 95


« Reply #18 on: November 01, 2008, 06:48:06 PM »
ReplyReply

Quote from: bjanes
It sounds like you are using one step sharpening in CS3. You should describe your sharpening procedure in more detail. Output sharpening is normally performed as a final step prior to printing and assumes that capture sharpening has already been applied. One step sharpening is no longer recommended.

If you are using a Canon 1dsIII at a low ISO, I doubt Noiseware or Noise Ninja is necessary and these tools will degrade sharpness. If sky or other uniform areas need noise reduction, you might consider using a mask.

Bill
I'm not sure what I can add, because my sharpening technique it simple -- don't do any of it LR, export from LR, and do one step USM at 300%/.8/2.

(I could never understand the logic of doing capture sharpening in CS3, and then, a few minutes later, doing a second step of output sharpening.  I posted that question with Tim Grey and DDQ, and he agreed that one step sharpening is as good as the two step process.)

Anyway, my purpose here is not to debate the merits of one step versus two step sharpening.  I am happy to use the two step approach in LR.  And avoid exporting to CS3.

My question is why I can't get sharp results in LR using the LR two step approach, as compared with using USM in a single step, with settings of 300%/.8/2.  For a 22x32 enlargement.

I assume it is because I am not using high enough capture settings in LR.  So I wondering how the numbers translate -- for a sharp landscape, to be printed in 16x20 or 22x32, what capture sharpening in LR should be used?  When those settings are combined with the correct setting for output sharpening in LR, will that result in the same level of sharpening as I obtain in CS3 with USM in one step?

Thanks for the tip on noise reduction.  I will give it a try.
Logged
Josh-H
Sr. Member
****
Offline Offline

Posts: 1910



WWW
« Reply #19 on: November 01, 2008, 08:01:40 PM »
ReplyReply

Quote from: alan a
I'm not sure what I can add, because my sharpening technique it simple -- don't do any of it LR, export from LR, and do one step USM at 300%/.8/2.

(I could never understand the logic of doing capture sharpening in CS3, and then, a few minutes later, doing a second step of output sharpening.  I posted that question with Tim Grey and DDQ, and he agreed that one step sharpening is as good as the two step process.)

Anyway, my purpose here is not to debate the merits of one step versus two step sharpening.  I am happy to use the two step approach in LR.  And avoid exporting to CS3.

My question is why I can't get sharp results in LR using the LR two step approach, as compared with using USM in a single step, with settings of 300%/.8/2.  For a 22x32 enlargement.

I assume it is because I am not using high enough capture settings in LR.  So I wondering how the numbers translate -- for a sharp landscape, to be printed in 16x20 or 22x32, what capture sharpening in LR should be used?  When those settings are combined with the correct setting for output sharpening in LR, will that result in the same level of sharpening as I obtain in CS3 with USM in one step?

Thanks for the tip on noise reduction.  I will give it a try.


I think you missunderstand the purpose of multi-step sharpening (It is not just to try and achieve 'better on screen' sharpness, or rather that is the first step - not the second.)You also definately seem to have missed the concept of how to use sharpening in Lightroom.

Multi-step sharpening is a process whereby sharpening is applied for capture and then for a specific output. Example: I import my .CR2 Canon 1DSMK3 RAW file into lightroom and apply sharpening in the develop module so that it looks good at 100% on screen. That is capture sharpening - beginning to end. End of step 1.

The capture sharpening done in Lightroom is really just to remove the inherent softness introduced by the cameras anti alias filter. It is not sharpening for a specific output - wether it be print, jpeg whatever [thats step 2]. It isnt even creative sharpening - which you can do as a 1.5 step using the new local adjustment brushes in Lightroom - very cool!

Now - If I want to make a print, I need to apply the second round of sharpening specifically for the print and specifically for the paper matt or glossy - optimal print sharpening is not going to look good on screen, but will in print.

If you are doing one round of sharpening in CS3 with USM and as you say and using that for print and getting good prints then I will wager my 1DSMK3 that your on screen image is viciously over sharpened. It has to be if you are getting good properly sharpened prints. This is ok if all you want to do with that file is make a print. But it makes a lot more sense to use the multi-step method and create a properly capture sharpened file that can then be output to both print and web in the future. All it needs is the right output sharpening post the capture sharpening.

The two step sharpening process is an absolute MUST - it gives a file that is properly capture sharpened, and can then be sharpened for whatever output is required.

As to what setting to use in LR for 1DSMK3 files - I shoot every day with my 1DSMK3 and feel I have the capture sharpening totally nailed in Lightroom. Now.. it IS image dependant - so these settings should be considered as a basis or starting point only. Your mileage may vary. Also, I am usually a little light handed with my capture sharpening as I hate the artifacts created by over use.

I find for Landscapes in LR I use amount 60, Radius 1, and Detail 40 for 1DSMK3 files. That is pretty much my starting point for capture sharpening and tweak from there. Add masking as/if required.

Then Ill do a round of output sharpening depending on wether I am going to print / web etc.

Edit - I find low ISO 1DSMK3 files need NO noise reduction other than lightroom's default of 25.
« Last Edit: November 01, 2008, 08:12:22 PM by Josh-H » Logged

Pages: [1] 2 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad