Printer / Monitor profiles

"Fruit2O" wrote in message

I have a LaCie monitor which I keep calibrated at least once a week using a Lacie calibrator which adjusts the color guns. I am using Ian Lyon’s printer profiles for the Epson 1290 (I use the Epson 1280 but Ian says they’re both the same). If I understand correctly, when using soft proofing, if the monitor profile and printer profile are accurate, both my original and soft proof should look identical. Is this true or should I expect to see something else? I am willing to pay for more accurate printer profiles for my Epson 1280 if they are much better the Ian Lyon’s profiles. Thanks.

They generally won’t look identical, any more than a light bulb and a printed picture of a light bulb will appear identical. I question the usefulness of the soft proof for matching what you see on the screen to a paper print. Having said that, there are some things you can do to make the two look *more* identical.

Gernot Hoffman has been experimenting with this, and perhaps he will respond. Start by calibrating your display to be approximately the same brightness as a nearby sheet of paper. When viewing a soft proof, it is important to remove all other items – toolbars, window margins, etc. – from the screen. The resulting image will be very drab, so allow your eye to accommodate to the displayed image for 30 seconds or more. You will get a much closer match to the appearance of the printed image.

Many people make a practical compromise by un-checking the option to simulate the paper white. This gives you a sense of which colors will be most affected in the final print (generally blues and magentas will be less saturated) while still displaying a more or less normal looking image. —
Mike Russell
www.curvemeister.com/forum/

They generally won’t look identical, any more than a light bulb and a printed picture of a light bulb will appear identical. I question the usefulness of the soft proof for matching what you see on the screen to a paper print. Having said that, there are some things you can do to make the two look *more* identical.

Gernot Hoffman has been experimenting with this, and perhaps he will respond. Start by calibrating your display to be approximately the same brightness as a nearby sheet of paper. When viewing a soft proof, it is important to remove all other items – toolbars, window margins, etc. – from the screen. The resulting image will be very drab, so allow your eye to accommodate to the displayed image for 30 seconds or more. You will get a much closer match to the appearance of the printed image.
Many people make a practical compromise by un-checking the option to simulate the paper white. This gives you a sense of which colors will be most affected in the final print (generally blues and magentas will be less saturated) while still displaying a more or less normal looking image.

Thanks, Mike. One thing I don’t understand is – if the soft proof looks so drab and you want the resulting print to look like the original (with all the modifications you’ve done to it), and the soft proof looks somehwat like what the print will look like, you must have to also adjust the soft proof – and print IT instead of the original. Do I understand this correctly?

"Fruit2O" wrote in message

[re soft proofing]
….

Thanks, Mike. One thing I don’t understand is – if the soft proof looks so drab and you want the resulting print to look like the original (with all the modifications you’ve done to it), and the soft proof looks somehwat like what the print will look like, you must have to also adjust the soft proof – and print IT instead of the original. Do I understand this correctly?

As you may have gathered, I’m not much of a fan of soft proofing, though it can be useful provided you have a good printer proof, and you realize the limitations inherent in the process. In particular, if paper white is enabled, the image will look so drab that you probably will not want to show it to the customer. That said, there are any number of reasonable people who find that soft proofing gives them greater control over the final result. If you do find soft proofing useful, of course continue to use it.

IMHO the main use of a soft proof is not as an editing space, but to vet the appearance of the final output, and anticipate any problems in the final print. For example, you may want to give the customer an idea of what coated versus uncoated stock will look like.

You may also use the soft proofing feature to compensate somewhat for the appearance of the final image, and yes, you would then print this modified version of your image instead of the version that looks best on your monitor. You are not printing the soft proof, but a modified version of the original image.

It’s easy, IMHO, to bear hug the concept of soft proofing, and rely on the soft proof to do all your work. This is IMHO less useful than adjusting your printer to match the overall brightness and neutrality of the display, and trusting to the printer to render the image fairly accurately. If soft proofing is used, it is in place of, or in addition to, a small test print as preparation for printing a final large image.
—
Mike Russell
www.curvemeister.com/forum/

On Sun, 26 Nov 2006 10:06:13 GMT, "Mike Russell" wrote:

"Fruit2O" wrote in message

[re soft proofing]
…

Thanks, Mike. One thing I don’t understand is – if the soft proof looks so drab and you want the resulting print to look like the original (with all the modifications you’ve done to it), and the soft proof looks somehwat like what the print will look like, you must have to also adjust the soft proof – and print IT instead of the original. Do I understand this correctly?

As you may have gathered, I’m not much of a fan of soft proofing, though it can be useful provided you have a good printer proof, and you realize the limitations inherent in the process. In particular, if paper white is enabled, the image will look so drab that you probably will not want to show it to the customer. That said, there are any number of reasonable people who find that soft proofing gives them greater control over the final result. If you do find soft proofing useful, of course continue to use it.
IMHO the main use of a soft proof is not as an editing space, but to vet the appearance of the final output, and anticipate any problems in the final print. For example, you may want to give the customer an idea of what coated versus uncoated stock will look like.

You may also use the soft proofing feature to compensate somewhat for the appearance of the final image, and yes, you would then print this modified version of your image instead of the version that looks best on your monitor. You are not printing the soft proof, but a modified version of the original image.

It’s easy, IMHO, to bear hug the concept of soft proofing, and rely on the soft proof to do all your work. This is IMHO less useful than adjusting your printer to match the overall brightness and neutrality of the display, and trusting to the printer to render the image fairly accurately. If soft proofing is used, it is in place of, or in addition to, a small test print as preparation for printing a final large image.

Hi Mike! I umderstand what you’re saying – but I’m still missing something. I’ll try to restate it: Let’s say I make all my adjustments to a scanned photo and I like the way it looks. That’s what I want my print to look like. Then, I turn on my soft proof – and it looks drab. Of course, that’s not what I want to give to a customer. Her’s where I’m confused: Then I go ahead and make adjustments to the original so that it looks like the proof. I don’t understand how ths results in a finished print that looks like the original I saw on the screen before further adjustments. Now, I have an original that I don’t like and an (old) proof on the screen that I don’t like. Those adjustments I made to the original to make it look like the proof must have done something to the original to make a ‘new’ proof look like I want the finished product to look like. The problem is: I don’t see the logic in how the layer group does this. Thanks.

On Sun, 26 Nov 2006 10:06:13 GMT, "Mike Russell" wrote:

"Fruit2O" wrote in message

[re soft proofing]
…

Thanks, Mike. One thing I don’t understand is – if the soft proof looks so drab and you want the resulting print to look like the original (with all the modifications you’ve done to it), and the soft proof looks somehwat like what the print will look like, you must have to also adjust the soft proof – and print IT instead of the original. Do I understand this correctly?

As you may have gathered, I’m not much of a fan of soft proofing, though it can be useful provided you have a good printer proof, and you realize the limitations inherent in the process. In particular, if paper white is enabled, the image will look so drab that you probably will not want to show it to the customer. That said, there are any number of reasonable people who find that soft proofing gives them greater control over the final result. If you do find soft proofing useful, of course continue to use it.
IMHO the main use of a soft proof is not as an editing space, but to vet the appearance of the final output, and anticipate any problems in the final print. For example, you may want to give the customer an idea of what coated versus uncoated stock will look like.

You may also use the soft proofing feature to compensate somewhat for the appearance of the final image, and yes, you would then print this modified version of your image instead of the version that looks best on your monitor. You are not printing the soft proof, but a modified version of the original image.

It’s easy, IMHO, to bear hug the concept of soft proofing, and rely on the soft proof to do all your work. This is IMHO less useful than adjusting your printer to match the overall brightness and neutrality of the display, and trusting to the printer to render the image fairly accurately. If soft proofing is used, it is in place of, or in addition to, a small test print as preparation for printing a final large image.

Mike, If I’m still not clear about my confusion, I would be willing to call you – of course, at our convenience, if you agree and if you are in the U.S. This issue is very bewildering to me.

"Fruit2O" wrote in message
….
[re soft proofing]

Hi Mike! I umderstand what you’re saying – but I’m still missing something. I’ll try to restate it: Let’s say I make all my adjustments to a scanned photo and I like the way it looks. That’s what I want my print to look like. Then, I turn on my soft proof – and it looks drab. Of course, that’s not what I want to give to a customer.

The soft proof will always look drab because a monitor is able to display a better image – particularly in the brightest areas – than a printer. Once printed, the drab image will appear normal.

Her’s where I’m confused: Then I go ahead and make
adjustments to the original so that it looks like the proof. I don’t understand how ths results in a finished print that looks like the original I saw on the screen before further adjustments. Now, I have an original that I don’t like and an (old) proof on the screen that I don’t like. Those adjustments I made to the original to make it look like the proof must have done something to the original to make a ‘new’ proof look like I want the finished product to look like. The problem is: I don’t see the logic in how the layer group does this.

Take a specific example, where a blue sky becomes too faded, or worse yet takes a turn toward purple. You could make two compromises: 1) add a curves adjustment layer to bump the brightness of the sky – if it can’t be pure blue at least make it brighter, and 2) add more cyan to the sky to get rid of the purple. BTW, when making adjustments, its best to base your color adjustment decisions on the info palette in addition to the visual appearance of the image.

Thanks.

You’re very welcome. Once again I would add that I’m not really an advocate of relying on a soft proof for evaluating your work. There are just too many things that can go wrong. I think the best uses of a soft proof are comparison of various output media, and as a replacement for a test print when the situation does not permit making one.

You may want to wait for others to chime in before deciding on your own workflow.
—

Mike Russell
www.curvemeister.com/forum/

"Bob Onysko" wrote in message
….
[re soft proofing concepts]

Mike, If I’m still not clear about my confusion, I would be willing to call you – of course, at our convenience, if you agree and if you are in the U.S. This issue is very bewildering to me.

I’m not sure a phone call would be that useful. What would probably be more useful would be a discussion with specific images with colors that show a significant difference on screen versus a soft proof.

0) The working space may be thought of as an idealized way of looking at the image – perfect in every way with no inaccuracy. For example, in sRGB RGB(12,32,234) is a blue-cyan color of a particular brightness, hue, and saturation.

1) During editing, the image is converted from the working space (typically sRGB or Adobe RGB), using the color profile of your particular display.

2) During printing, the image is converted from the working space to your printer’s color space, using the color profile or driver settings of your printer.

3) For a soft proof, the image is first converted using your printer’s profile, and then converted to your display using a special mode called the Absolute Colorimetric (AbsCol) intent. This intent has the effect of using your display to simulate your printer’s colors as accurately as possible. This step works, at least in principle, because your display is capable of displaying more colors than your printer. The image appears drab on the screen for exactly this reason.

In the description above, I’ve left out the concept of the profile connection space (PCS). A profile is generally capable of converting colors in two directions, into and out of the profile’s color space. The PCS is a sort of "super" color space (either Lab or CIEXYZ) designed to allow profiles to be plugged into one another, to accommodate both of these directions. So the profile for sRGB would be used to convert your image’s colors to the PCS, after which your display profile would be used to convert from the PCS to your display device’s RGB values. Cool, huh?

It’s worth mentioning here that printer profiles generally put most of their mojo into the conversion from the PCS to the printer color space, and not the other way around. The reason for this is obvious once you think about it. After all, printers are never used to produce color, are they? Well, almost. There is one case where printers may be thought of as producing color, and that’s during a soft proof operation, when printer colors are converted to screen colors in AbsCol mode for the simulation Obviously, this undercuts the accuracy, and the usefulness of soft proofing.

More discussion of the PCS and other profile related concepts here: < http://download.microsoft.com/download/1/6/1/161ba512-40e2-4 cc9-843a-923143f3456c/ColorMgmtConcepts.doc>
—
Mike Russell
www.curvemeister.com/forum/

Hi Mike! I umderstand what you’re saying – but I’m still missing something. I’ll try to restate it: Let’s say I make all my adjustments to a scanned photo and I like the way it looks. That’s what I want my print to look like. Then, I turn on my soft proof – and it looks drab. Of course, that’s not what I want to give to a customer. Her’s where I’m confused: Then I go ahead and make adjustments to the original so that it looks like the proof. I don’t understand how ths results in a finished print that looks like the original I saw on the screen before further adjustments. Now, I have an original that I don’t like and an (old) proof on the screen that I don’t like. Those adjustments I made to the original to make it look like the proof must have done something to the original to make a ‘new’ proof look like I want the finished product to look like. The problem is: I don’t see the logic in how the layer group does this. Thanks.

Mike, I thank you for all your help – but, please re-read the above. I can’t think of a better way to explain my confusion. If I change my original with a Layer Group, it will look different than what I want to print. Since I am trying to match the original soft proof, it doesn’t change while I’m making the adjustments on the original. Now I’m left with two images on my screen – neither of which looks like what I want the finished product to look like. According to the protocol, I should print the adjusted original which now looks very similar to the original soft proof. So how is that supposed to result in a print that looks (somewhat) like what I want? In other words, what are those adjustments to the original really doing so that I end up with a reasonable result? Making changes to the original so that it looks like the soft proof will make the original ‘drab.’ So, if I proof it again, the proof will look even more drsb. I think I’m becoming a pain in the neck – but I’d really like to understand the process and the logic. It is not clear to me at all (yet). Thanks.

"Bob Onysko (Formerly Fruit2O)" wrote in message

Hi Mike! I umderstand what you’re saying – but I’m still missing something. I’ll try to restate it: Let’s say I make all my adjustments to a scanned photo and I like the way it looks. That’s what I want my print to look like. Then, I turn on my soft proof – and it looks drab. Of course, that’s not what I want to give to a customer. Her’s where I’m confused: Then I go ahead and make adjustments to the original so that it looks like the proof. I don’t understand how ths results in a finished print that looks like the original I saw on the screen before further adjustments. Now, I have an original that I don’t like and an (old) proof on the screen that I don’t like. Those adjustments I made to the original to make it look like the proof must have done something to the original to make a ‘new’ proof look like I want the finished product to look like. The problem is: I don’t see the logic in how the layer group does this. Thanks.

Mike, I thank you for all your help – but, please re-read the above. I can’t think of a better way to explain my confusion. If I change my original with a Layer Group, it will look different than what I want to print. Since I am trying to match the original soft proof, it doesn’t change while I’m making the adjustments on the original. Now I’m left with two images on my screen – neither of which looks like what I want the finished product to look like. According to the protocol, I should print the adjusted original which now looks very similar to the original soft proof. So how is that supposed to result in a print that looks (somewhat) like what I want? In other words, what are those adjustments to the original really doing so that I end up with a reasonable result? Making changes to the original so that it looks like the soft proof will make the original ‘drab.’ So, if I proof it again, the proof will look even more drsb. I think I’m becoming a pain in the neck – but I’d really like to understand the process and the logic. It is not clear to me at all (yet). Thanks.

Ah – I believe I understand now. Here’s how adjusting the soft proof to prepare for printing should work *in principle*. Rather than adjusting the original to match the proof, add an adjustment layer and make the proof look the way you want your final image to look, keeping in mind that it is impossible for it to have the zing of the original non-proof image. For example, you may want to brighten certain saturated colors that took a hit in the soft proof. Then you print your original image with this adjustment layer in place.

I say *in principle* because there are a number of reasons your printer profile may not be accurate. There is therefore some risk that the soft proof will be a poor guide to the final appearance of the print.

BTW – there is another use for soft proofing. You can predict what your image will look like on a Macintosh versus a PC screen – this can be valuable in itself, and for fine web work where you use a script to detect OSX versus Windows, and display a different version of the image for both platforms

Don’t worry about being a pain in the neck – remember this is Usenet and we’ve got petrof and everyone else trying to be as obnoxious as possible. It’s refreshing to have someone follow up on an actual question. Plus, I’ve been getting questions about soft proofing in the curvemeister forum, and the ideas from this feed into it.
—
Mike Russell
www.curvemeister.com/forum/

Mike, I thank you for all your help – but, please re-read the above. I can’t think of a better way to explain my confusion. If I change my original with a Layer Group, it will look different than what I want to print. Since I am trying to match the original soft proof, it doesn’t change while I’m making the adjustments on the original. Now I’m left with two images on my screen – neither of which looks like what I want the finished product to look like. According to the protocol, I should print the adjusted original which now looks very similar to the original soft proof. So how is that supposed to result in a print that looks (somewhat) like what I want? In other words, what are those adjustments to the original really doing so that I end up with a reasonable result? Making changes to the original so that it looks like the soft proof will make the original ‘drab.’ So, if I proof it again, the proof will look even more drsb. I think I’m becoming a pain in the neck – but I’d really like to understand the process and the logic. It is not clear to me at all (yet). Thanks.

Ah – I believe I understand now. Here’s how adjusting the soft proof to prepare for printing should work *in principle*. Rather than adjusting the original to match the proof, add an adjustment layer and make the proof look the way you want your final image to look, keeping in mind that it is impossible for it to have the zing of the original non-proof image. For example, you may want to brighten certain saturated colors that took a hit in the soft proof. Then you print your original image with this adjustment layer in place.

OK, now you understand what I’ve been trying to explain. So, I have on my left of the screen the original which I have the way I want. To its’ right is the soft proof which is ‘flat.’ To which do I apply the adjustment layer? I assume it’s the original, yes? But then, the proof on the right will stay the same unless I re-proof. Is there a way to apply the adjustments to the original and see the result on the soft proof while I’m making those changes?

That being said, now I would like to learn ‘your’ method since I now understand the limitations of soft proofing – especially since I’ve been told that my Epson 1280 is very tricky and unpredictable with soft proofing. Thanks…….

I say *in principle* because there are a number of reasons your printer profile may not be accurate. There is therefore some risk that the soft proof will be a poor guide to the final appearance of the print.
BTW – there is another use for soft proofing. You can predict what your image will look like on a Macintosh versus a PC screen – this can be valuable in itself, and for fine web work where you use a script to detect OSX versus Windows, and display a different version of the image for both platforms

Don’t worry about being a pain in the neck – remember this is Usenet and we’ve got petrof and everyone else trying to be as obnoxious as possible. It’s refreshing to have someone follow up on an actual question. Plus, I’ve been getting questions about soft proofing in the curvemeister forum, and the ideas from this feed into it.

"Fruit2O" wrote in message

[re soft proofing]

OK, now you understand what I’ve been trying to explain. So, I have on my left of the screen the original which I have the way I want. To its’ right is the soft proof which is ‘flat.’ To which do I apply the adjustment layer? I assume it’s the original, yes? But then, the proof on the right will stay the same unless I re-proof. Is there a way to apply the adjustments to the original and see the result on the soft proof while I’m making those changes?

Setting up a soft proof tells Photoshop to simulate the printed appearance on your screen. It does not change the underlying image, so you should* be able to edit the existing image until you are happier with the result.

That being said, now I would like to learn ‘your’ method since I now understand the limitations of soft proofing – especially since I’ve been told that my Epson 1280 is very tricky and unpredictable with soft proofing. Thanks…….

I used my Epson 1270 for years – until it fell victim to the ignominious magenta streaking problem -without an icc profile, setting it to PhotoEnhance4 mode. I produced hundreds of good prints that matched the screen quite well.

* the word "should", when used in connection with computers, generally means "doesn’t, and I don’t know why".
—
Mike Russell
www.curvemeister.com/forum/

On Tue, 28 Nov 2006 01:23:47 GMT, "Mike Russell" wrote:

"Fruit2O" wrote in message

[re soft proofing]

OK, now you understand what I’ve been trying to explain. So, I have on my left of the screen the original which I have the way I want. To its’ right is the soft proof which is ‘flat.’ To which do I apply the adjustment layer? I assume it’s the original, yes? But then, the proof on the right will stay the same unless I re-proof. Is there a way to apply the adjustments to the original and see the result on the soft proof while I’m making those changes?

Setting up a soft proof tells Photoshop to simulate the printed appearance on your screen. It does not change the underlying image, so you should* be able to edit the existing image until you are happier with the result.

I’ve tried editing the existing image – but the proof doesn’t change at all.

That being said, now I would like to learn ‘your’ method since I now understand the limitations of soft proofing – especially since I’ve been told that my Epson 1280 is very tricky and unpredictable with soft proofing. Thanks…….

Will you explain ‘your’ method?

I used my Epson 1270 for years – until it fell victim to the ignominious magenta streaking problem -without an icc profile, setting it to PhotoEnhance4 mode. I produced hundreds of good prints that matched the screen quite well.

* the word "should", when used in connection with computers, generally means "doesn’t, and I don’t know why".