Ad
Ad
Ad
Pages: « 1 2 3 [4] 5 »   Bottom of Page
Print
Author Topic: Printing advanced black and white from LR4?  (Read 16456 times)
S Kale
Jr. Member
**
Offline Offline

Posts: 72


« Reply #60 on: April 26, 2012, 09:18:47 AM »
ReplyReply

Whatever. It is used, perhaps incorrectly, by the B&W community to describe the issues I referenced. Call it whatever you want. How about "problem x"?



Case in point. It appears that anyone who disagrees with your rather weak arguments must somehow be affiliated with Epson. You affiliated with HP or Canon?

Not at all. I merely asked a question because the tone of response was so blinkered away from alternatives to Epson products.  No I am not affiliated with HP or Canon.  As noted above, I use an Epson.  If I were affiliated with HP or Canon I might well use their products but I am not and I don't. I would have thought that relatively clear.

Guys, you seem to be getting your knickers in a twist and so vehemently opposed to anyone who suggests that Epson's product offering is lacking in any way. I have merely made a few basic statements which I will repeat once more for clarity:

- Epson ABW is no more a "black box" than any other driver
- QTR Create ICC isn't a "hack"; it improves the output of Epson ABW (and any other B&W printing workflow whether involving an Epson printer, HP, Canon or otherwise) and is available to anyone with a densitometer or spectrophotometer. We should be thankful it's available. It's progress. It allows us to use those aspects of colour management we care most about in B&W printing. To call it a "hack", particularly when you are involved in colour management as a profession is not only disappointing but, frankly, puzzling.
- another example of improving Epson ABW beyond the Epson package is to use third party papers. Very simply, there are papers out there that are better than those provided by Epson. That's good for us users.
- the improvement of B&W printing quality should be encouraged and not discouraged. The OEM solutions are ok for some but lacking for many also. We should encourage manufacturers to keep improving their products. We should also encourage other third party vendors of software, ink and paper to continue to provide solutions that can enhance the OEM solutions.


I think Epson ABW is a fantastically convenient product. I use it (as I have said many times). I like that it deploys LK and LLK alongside K and that it dramatically reduces the use of Y, M and C. I like that I get a decent B&W solution and a colour solution in one box. I dropped using dedicated B&W ink sets when it came out - I made a decision to sacrifice a degree of quality for convenience. (If I had the time and space I would prefer to run a dedicated B&W ink set because I have seen the quality that it can produce.) With very little effort, I do, though, address some of Epson ABW's shortcomings by using QTR Create ICC and third party papers.  I don't discourage people from adopting such enhancements. I encourage people to explore options and demand more from the out-of-the-box offering. Were it not for people doing so ABW wouldn't exist.  Hopefully Epson and the colour management community will continue to improve on the offering.

Ciao.



Logged
RFPhotography
Guest
« Reply #61 on: April 26, 2012, 09:26:20 AM »
ReplyReply

It can be controlled and soft-proofed with ABW and QTR Create ICC.  I can also convert images to the profile I use to proof that toning and save this toned image for electronic display.


Converting the image to the profile is not following a properly colour managed workflow.

Please explain how you are softproofing toned images that are to be printed via ABW?  That is, how you are softproofing without converting the image to the paper profile.  How are you previewing the tone that will be added to the image?  How are the profiles created?  Are you printing the QTR step wedge by adding a tone in ABW then creating the profile from that toned step wedge?  How are you determining what the readings from the step wedge should be compared to in order to create the profile since, as far as I know, Epson doesn't provide the numbers at various toning (hue) settings?  
Logged
digitaldog
Sr. Member
****
Offline Offline

Posts: 9226



WWW
« Reply #62 on: April 26, 2012, 09:27:13 AM »
ReplyReply

To call it a "hack", particularly when you are involved in colour management as a profession is not only disappointing but, frankly, puzzling.

I did not call QTR a hack. Read the sentence again: The vast majority of users don’t seem to have any problem with either the design or the results without the use of a 3rd party ‘hack’ or drivers.

That last part of that sentence can include any solutions or options that are not the result of printing using ABW as designed and used by the vast majority of ABW users.

You clearly lost the point of the sentence to aim onto your preferred output methods while implying that those who don’t follow your workflow are incapable of recognizing or producing quality output. Some call that mindset elitist...
« Last Edit: April 26, 2012, 09:29:23 AM by digitaldog » Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
S Kale
Jr. Member
**
Offline Offline

Posts: 72


« Reply #63 on: April 26, 2012, 09:31:09 AM »
ReplyReply

I did not call QTR a hack. Read the sentence again: The vast majority of users don’t seem to have any problem with either the design or the results without the use of a 3rd party ‘hack’ or drivers.

That last part of that sentence can include any solutions or options that are not the result of printing using ABW as designed and used by the vast majority of ABW users.

You clearly lost the point of the sentence to aim onto your preferred output methods while implying that those who don’t follow your workflow are incapable of recognizing or producing quality output. Some call that mindset elitist...

Oh boy, I not even going to bother with the above.
Logged
S Kale
Jr. Member
**
Offline Offline

Posts: 72


« Reply #64 on: April 26, 2012, 10:15:58 AM »
ReplyReply

Converting the image to the profile is not following a properly colour managed workflow.

Please explain how you are softproofing toned images that are to be printed via ABW?  That is, how you are softproofing without converting the image to the paper profile.  How are you previewing the tone that will be added to the image?  How are the profiles created?  Are you printing the QTR step wedge by adding a tone in ABW then creating the profile from that toned step wedge?  How are you determining what the readings from the step wedge should be compared to in order to create the profile since, as far as I know, Epson doesn't provide the numbers at various toning (hue) settings?  

Let's use an example. I have a nice greyscale image and this image is tagged with my greyscale workspace of Gray Gamma 2.2. I have profiled a particular set of ABW settings (let's say 10-15 with the colour picker, a somewhat subtle shift, and everything else at default i.e. darker) with QTR Create ICC. Yes, I did this by printing a step wedge with those settings and reading the results with my i1 spectro and dropping those results into QTR Create ICC. (I don't need to know what Epson hoped the results would be, only what they actually are for my printer, driver settings, ink and paper combination.) I now have, in an ICC profile shell, something I can use (1) to manage luminance transition from image file to print and (2) to soft-proof the hue of image as printed. 10-15 is a warm brown hue. I will see a warm brown image when I set up my soft proof. (I do this in the same manner as I would any paper profile soft proof. Personally, I don't find it useful to simulate paper colour or ink black. As you likely know, PS can show a soft proof without actually converting the profile of the image file.) If I have profiled a number of different colour picker settings, I can quickly run through soft proofs of what the image will look like when printed.

Now, if I wanted to place a copy of this image on a website or any other digital display platform, placing a copy of the original image would obviously display a neutral grey image. However, I want people to see the image toned as it is when it comes out of the printer i.e. a warm brown. I can take a copy of the file (I don't want to convert my master file to an output space, as you note) and convert it to the QTR ICC profile I made for 10-15 settings. My image now takes on the hue of the Epson ABW driver set at 10-15. It can do this because I have measured the output from the printer for stimulus values from 0 to 255 (typically 25 steps is enough with Roy's curve fitting) when I made the profile. (I might then convert the file to sRGB because there's a lot of people out there with browsers that don't support colour management...)

Here's a simple example. (I hesitate to post an image as I don't want this to be about whether I can take a photograph or not. There are many, many thousands of better photographers than me.) If I recall correctly, this is "10-15" on Hahnemuehle Photorag 308gsm. (I say "recall correctly" because I just grabbed an existing jpeg rather than going through the conversion process on a master file.)


One can do exactly the same with any printer/B&W workflow.

Personally, I like the convenience of all of this. I have a few favourite hue settings in Epson ABW which I have profiled for my favourite papers. I keep my image files in greyscale and have a degree of confidence (admittedly untested away from a neutral print) that Epson ABW is minimising the use of C, M and Y and maximising the use of LK and LLK as much as possible (versus the colour driver) even for toned prints (of course some hues will require more cyan and magenta but they tend not to be ones I like). I can soft-proof the output and display/send an electronic version of the image with the hue of my selected printer settings. I also have a means of more accurately getting the image into the narrower luminance gamut (again, forgive me if this is improper use of a technical term but I think you get my drift) of the printer/paper/ink/driver settings combination.

I hope this helps.
« Last Edit: April 26, 2012, 10:33:16 AM by S Kale » Logged
Nora_nor
Jr. Member
**
Offline Offline

Posts: 51


« Reply #65 on: April 26, 2012, 10:38:51 AM »
ReplyReply

now what does apple do when sending a file from Lightroom to the advanced black and white mode of the HPZ2100 printer? Note this printer is capaple of making its own paper profiles. I have access to a HPZ2100 sometimes.
Logged
S Kale
Jr. Member
**
Offline Offline

Posts: 72


« Reply #66 on: April 26, 2012, 10:43:07 AM »
ReplyReply

I don't have an HP but any greyscale file is converted to "Generic Gray Gamma 2.2" by OS-X before it gets to the printer. So if you have profiled the HP's ABW mode you will have issues. It will complicate - modestly - how you print targets for ABW profiling and how you print the image itself (as described in earlier posts here). It's not difficult to work around the issue but it is rather bizarre that OS-X makes us jump through these hoops.
Logged
RFPhotography
Guest
« Reply #67 on: April 26, 2012, 11:45:04 AM »
ReplyReply

Thanks for the follow up.  What you're doing is what I expected.  You're not creating an ICC profile, you're creating a curve that's embedded in, as you say, an ICC shell.  It's still a bit of hit an miss since you may need to make multiple prints with various toning to come up with the one you're happy with.  And in that sense we're back to ABW being a black box because you don't know till you actually make a print what the toning will look like. 
Logged
S Kale
Jr. Member
**
Offline Offline

Posts: 72


« Reply #68 on: April 26, 2012, 12:20:45 PM »
ReplyReply

Well a greyscale profile isn't much more than a curve. Plus black point compensation and soft proofing etc. I do know what the toning will be like but, yes, only because I have printed a range of step wedges (and profiled them) and figured out a set of favourites. (One has to do the same (and more) with a B&W RIP.)  I don't need to do this for each image. Hey, I, of all people, never said ABW was perfect. But hopefully this demonstrates, at least in part, that there's a lot more that can be gleaned out of Epson ABW with a little effort.

As I have said, the disadvantage with a colour workflow is that you're back into the 'improper' use of inks*. If you want full control over ink channels and hue then you're into using a RIP like QTR or the remarkable stuff that guys like Jon Cone, Tyler Boley and Paul Roark produce with dedicated inks. Of course, Epson could (and should) develop a more sophisticated driver which analyses all the pixel values for a particular image and places constraints on particular inks as a result. That would be very cool. We wouldn't need ABW for a start.

* Just look again at the analysis I posted. When toning B&W images we rarely step too far from centre. Neutral is typically cooled pigment, naturally warm etc. I'd love to see the same calculations that Roy did for ABW in Neutral mode for the other preselects. (I have asked him if he ever did this.) One would print a step wedge with, for example, the warm setting and analyse the ink usage (via the commands sent to the printer). Then you'd convert the step wedge to the QTR ICC profile for that warm setting, print it with the colour driver settings and run the same analysis.
Logged
digitaldog
Sr. Member
****
Offline Offline

Posts: 9226



WWW
« Reply #69 on: April 26, 2012, 12:26:09 PM »
ReplyReply

Well a greyscale profile isn't much more than a curve. Plus black point compensation and soft proofing etc. I do know what the toning will be like but, yes, only because I have printed a range of step wedges (and profiled them) and figured out a set of favourites.

BPC isn’t part of a profile, it is part of the CMM.

You are getting a soft proof or not? Knowing via a step wedge isn’t soft proofing.

Quote
If you want full control over ink channels and hue then you're into using a RIP like QTR

It is a RIP? Really? Postscript comparable?
« Last Edit: April 26, 2012, 12:28:39 PM by digitaldog » Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
S Kale
Jr. Member
**
Offline Offline

Posts: 72


« Reply #70 on: April 26, 2012, 01:01:24 PM »
ReplyReply

BPC isn’t part of a profile, it is part of the CMM.

You are getting a soft proof or not? Knowing via a step wedge isn’t soft proofing.

It is a RIP? Really? Postscript comparable?

Yes and having an ICC profile gives access to the CMM. BPC deployment is still one of those things that isn't as clear to me as I would like it to be. I recall reading a number of years ago a number of papers, including some by Adobe, regarding the maths involved with BPC. I no longer recall the details. I know that there are aspects of BPC and OS-X's CMM that Roy would like to understand better for another project he is working on.

Yes I can soft proof a given set of ABW settings having created a profile. (For example, I can produce a soft-proof of a greyscale image identical to the image I posted above to proof the ABW 10-15 toning.) The point Bob was mentioning is that I don't know what particular settings will produce until I print a step wedge, create a profile and soft proof the image.  The little image that accompanies the picker in ABW is useless. So there's a bit of work to do to identify a favourite.  In Bob's colour workflow, his colour, toned images are there on-screen in real-time as he produces them.  He just has to hope the driver can reproduce them without some of the carry-overs from a colour driver that have plagued B&W (neutral and toned).

Take a look at the QTR website.  It's a RIP for printing B&W images with any ink set and provides full control over individual ink channels and how they are limited and "mixed" to create toned output. It's been quite some years since I have used QTR itself. I decided to adopt the convenience of ABW with QTR Create ICC profiles (you can view this utility as completely distinct from QTR) and my favourite papers.

Logged
digitaldog
Sr. Member
****
Offline Offline

Posts: 9226



WWW
« Reply #71 on: April 26, 2012, 01:22:26 PM »
ReplyReply

Yes and having an ICC profile gives access to the CMM.

It is the other way around. If a CMM doesn’t use BPC, well you get no BPC. Apple’s CMM doesn’t. Adobe’s does. Roy could in theory create his own CMM (plugging that into the OS and thus Photoshop is another matter).

Quote
The point Bob was mentioning is that I don't know what particular settings will produce until I print a step wedge, create a profile and soft proof the image.

You use said useless little image in ABW picker to initially select a ‘look’ but the profile honors that when soft proofing. Now why the simulate check boxes don’t provide a better simulation is questionable and probably due to the lack of data about the paper white Lab values (an assumption, I haven’t looked into such profiles). Simulate check boxes should provide a better match.

Quote
It's a RIP for printing B&W images with any ink set and provides full control over individual ink channels and how they are limited and "mixed" to create toned output.

I don’t know what non raster data it is RIPping (or why the site uses lower cases to define the product as a RIP). I would suspect it is a substitute print driver. Or maybe you want to call the Epson print driver a RIP too (it really isn’t, we are feeding it raster data, it isn’t rasterizing anything).
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
RFPhotography
Guest
« Reply #72 on: April 26, 2012, 01:40:48 PM »
ReplyReply

I used QTR several years ago when I had a printer with dedicated b&w inks (MIS UT7, if I recall) so I'm aware of what QTR can do, yes. 

Andrew, I think your assumption is probably correct.  These aren't true ICC profiles that are being built.  It's really a measurement of ink density on the paper.  The paper white isn't taken into consideration.  And given that while the accuracy of the toning will be pretty good in many cases, on a paper with a heavy cast of its own the toning would be off. 
Logged
S Kale
Jr. Member
**
Offline Offline

Posts: 72


« Reply #73 on: April 26, 2012, 04:11:18 PM »
ReplyReply

I should mention that when I said I've never found it useful to simulate paper white or ink black that comment pertains to any proofing I do - colour profiles or ABW. Paper white is definitely recorded in a QTR Create ICC profile and checking simulate paper white or simulate ink black definitely does affect the proof. I personally just don't find that it adds much to my own workflow and I've never tested the proof on dramatically off-white papers. It may well provide a more accurate proof - I just don't take advantage of it. (Even with prints on matt papers where dMax is a lot less than on a photo paper, I find the eye is rather adept at adjusting for the darkest parts of the image and at adjusting paper off-white to white. I'm probably missing a beat here but, like I said, I don't find it affects my own work that much.)

On BPC, Andrew your comments are interesting. As I said, I don't recall the specifics of how QTR Create ICC dealt with or interacted with (or however else you might want to express it) BPC. I have the vague recollection that Roy eventually hard-coded the BPC translation into the kTRC curve in the profile but I could have that completely wrong. I do recall us discussing an Adobe paper on BPC and trying to understand its implementation by Adobe. (When converting an image to a QTR Create ICC profile in PS, checking and unchecking use BPC doesn't change the preview.)

This paper here may well have been one of the documents we were looking at years ago (certainly the maths on pg 8 is vaguely familiar to me):

http://wwwimages.adobe.com/www.adobe.com/content/dam/Adobe/en/devnet/photoshop/sdk/AdobeBPC.pdf

Your comments are particularly interesting, though, in relation to a new project of Roy's. Without giving away the details, here's a quote from an email he sent me:

"Note that CMM is not Adobe's -- I use Apple's ColorSync.  For the most part I believe its all
reasonably close.  One issue is BPC, Apple doesn't break this out.  I suspect it's done but
I don't know for sure."

I'm sure Roy would appreciate any help you could give in this regard.

Andrew, as to your last point, whether than means it isn't or is a "true" RIP I don't know and, to be honest, I don't care. I just know what can and can't be achieved with its help. I'm sure you are more than welcome to take a closer peak under the hood. It's shareware and nothing stops someone giving it a good trial before making the shareware payment.  Same applies to having a play with the independent (but bundled) QTR Create ICC module. (If you are on a Mac you just need to be aware of how to handle this rather annoying OS-X conversion to Generic Gray Gamma 2.2.)

Cheers

Steve
« Last Edit: April 26, 2012, 04:18:46 PM by S Kale » Logged
opgr
Sr. Member
****
Offline Offline

Posts: 1125


WWW
« Reply #74 on: April 26, 2012, 04:27:35 PM »
ReplyReply

I haven't followed this discussion but perhaps this may be helpful:

The Apple CMM does support BPC. I have implemented this several years ago, and it still seems to work in 10.7.

For programmers:
It's a setting in the CM2Header flags, look for:
cmBlackPointCompensationMask

Whether it actually conforms to Adobe's implementation guideline, I don't know.
Logged

Regards,
Oscar Rysdyk
theimagingfactory
Alan Goldhammer
Sr. Member
****
Offline Offline

Posts: 1711


WWW
« Reply #75 on: April 26, 2012, 04:57:47 PM »
ReplyReply

For those interested, Giorgio Trucco published a rather detailed explanation and tests of all this several years ago when the Epson 3800 first came out.  He shows how ABW profiles can be made and used (including soft proofing).  He has some good graphs showing the difference between the ABW driver and the normal Epson driver.  You will need to scroll down a ways to see things.  I believe that Eric Chan's ABW profiles work in the same manner and do know that he was disappointed that you cannot print in his recommended manner on the Apple OS any longer with the updated Epson driver.

Alan
Logged

digitaldog
Sr. Member
****
Offline Offline

Posts: 9226



WWW
« Reply #76 on: April 26, 2012, 06:07:28 PM »
ReplyReply

The Apple CMM does support BPC. I have implemented this several years ago, and it still seems to work in 10.7.

It may support it, I’d like to know where it is implemented. Not in Photoshop and not in ColorThink (which I believe calls the OS CMM). I ran a number of conversion patches though both and when plotting the delta’s, it certainly appears to be in an areas of color space where we’d expect to see BPC.

Maybe Preview or Aperture?
Logged

Andrew Rodney
Author “Color Management for Photographers”
http://digitaldog.net/
RFPhotography
Guest
« Reply #77 on: April 27, 2012, 07:36:28 AM »
ReplyReply

Steve, if all you're measuring are ink densities, or L values, how does the paper white get taken into account?  I mean, I can understand that it may have an impact on the density of the ink but just measuring L or brightness values shouldn't take it into account in the profile.  You have to measure and record a and b values for that, don't you?
Logged
S Kale
Jr. Member
**
Offline Offline

Posts: 72


« Reply #78 on: April 27, 2012, 07:41:53 AM »
ReplyReply

Yes for proofing purposes  - and I do with my i1.

If a* and b* aren't recorded (i.e. are set to zero) the soft proofing will suffer, as you note, but one still has the tone mapping from image file to print space. So, to get the most out of it one needs a spectrophotometer but one can still get a lot with just a densitometer.

Why don't you give it a go? Absent drying time, it only takes a minute or two to create a profile. Roy would recommend the 21-step x 4 target as Epson printers have a weird issue starting up and the first row of steps is generally lighter than the rest. I made a different target for use with Colorport and built in a "warm-up" for the printer.

You can then poke around the profile with ICC Profile Inspector (or similar)
« Last Edit: April 27, 2012, 08:06:41 AM by S Kale » Logged
RFPhotography
Guest
« Reply #79 on: April 27, 2012, 09:57:39 AM »
ReplyReply

Given that I like what ABW can do with a profiled print workflow I may.  But it'll have to wait till I can buy a measuring instrument.  A Colormunki is on the list of desired items but it's not high on the list.

That said, since I print on a 3800, I can use the profiles that Eric has so generously provided on his website.
« Last Edit: April 27, 2012, 10:36:54 AM by BobFisher » Logged
Pages: « 1 2 3 [4] 5 »   Top of Page
Print
Jump to:  

Ad
Ad
Ad