Jump to content

Process Printing Problem - HELP PLEASE!


Recommended Posts

<p>Until recently I have always had my digital images process printed at Jessops. The resulting

prints have generally been 'acceptable', although still very different from the image shown on my monitor

at home. There have been quite a few prints that have been totally unusable and I've

had to 're-print' them on my Canon inkjet just to get a half-decent print to be able to use. So I thought I'd try

elsewhere and see if I could actually get a print that resembled the image that I could see on my monitor.</p>

<

p>After reading several reviews, I decided to send some images (the exact same files I had done at Jessops)&

nbsp;to DS Colour Labs. The prints cames back a few days later for me to compare with the ones from Jessops. T

o my disappointment, these prints were even more inaccurate. They seemed to be so dark and flat. So, I e-mailed D

S Colourlabs to see if they could help. They advised me to start off by downloading their own colour profile (specific t

o their printers). Then, to convert the image(s) to their profile before sending them off. (I'm using Photoshop CS2 by t

he way) So I then asked the question "at exactly what stage do I convert the image to their profile?". I was advised t

o convert & edit the image in their profile so I would be able to see what the print would look like. H

owever, when I opened the original image (in Adobe RGB 1998) and then converted it to the DS Colour Labs p

rofile - nothing happened. I was expecting the new image on screen to change dramatically and look something like o

ne of the prints. Now I feel like I've come to a dead-end and don't know what to do.</p>

<

p>I am using a Sony Viao laptop (which doesn't help I suppose because the screen is so bright and contrasty). I've t

ried to adjust the display to match one of the prints but can't get anywhere near. I don't have the option of buying a

ny colour calibration software.</p>

<

p>Can anyone give me any advice please?</p>

<

p>Has anyone any experience of using Photoshop and processing images on a Sony Vaio?</p>

<

p> </p>

<

p> </p>?

Link to comment
Share on other sites

<p>Their instructions are correct. You weren't sending labs images in AdobeRGB before, were you? You have to use a custom profile or sRGB.</p>

<p>You can use the soft proof feature to see what an image will look like (point it towards the DS labs profile) and see what's out of gamut. If nothing's out of gamut the image on the screen probably won't change much at all when you turn soft proof on.</p>

<p>What I do recommend is that you use an external monitor (I got a Lacie 19" CRT used for $75) with your laptop (the laptop screen is not acceptable) and purchase color calibration hardware to go with it (bought an Eye One Display for $60USD- keep an eye out for sales). Otherwise the problem with your prints isn't the printer, it's your monitor.</p>

Link to comment
Share on other sites

<p>I actually sent the prints to the lab in sRGB.<br>

So, let me get this right. If I convert the image on screen to a different profile, then nothing will happen to the image displayed. Is that correct?<br>

When I soft-proof the image to the lab's profile, I should see what the image will look like when printed at the lab (provided that my monitor is calibrated correctly). Is that correct?<br>

So, in order to work on/edit the image and see what it will look like when printed, why can't I use the lab's profile as my working space. Then I wouldn't have to mess about soft-proofing. Right?</p>

 

Link to comment
Share on other sites

<p>No, you shouldn't use a printer profile as a working space. Working spaces are designed to be well-behaved and predictable as you make edits.</p>

<p>Printing profiles reflect the limitations of printers which can display some colors much better than others and have fairly odd-looking gamuts.</p>

<p>I would just edit in your working space (AdobeRGB or ProPhoto), and if you have very saturated colors you might softproof to see how they will be rendered on your printer. Then just convert to the printer profile, send it to the lab and you will be fine</p>

Link to comment
Share on other sites

<p>One example of using an inappropriate working space is when you go into curves and try to lighten part of the image. As you lighten it you introduce a strong color cast. With a proper working space R=G=B and the image stays neutral (but part of it gets lighter) in this situation.</p>
Link to comment
Share on other sites

<p>You need to calibrate your monitor using a photometer and software designed for that purpose. That way what you see will be the same as on any other calibrated monitor. It's best that you use a desktop monitor with your laptop for editing. Laptop displays are hard to calibrate for several reasons - limited angle of view (for security and cost), limited color gamut and variations in brightness (power level). You can (should) work in one of the standard, device-independent colorspaces like sRGB, Adobe RGB or Profoto RGB (in increasing order of gamut).</p>

<p>Calibration of the printer is a separate issue. This is done by using a profile for that printer and particular type of paper. You can download profiles for popular printers from the paper manufacturer, purchase custom profiles from a third party service, or make your own (if you have a reflective photometer, $$). The profile is used by Photoshop (or some other color-management-system (CMS) program), and acts like a compound "correction factor" for three colors and a wide range of print densities. It is pointless to use "Soft Proofing" unless both the monitor and printer have accurate profiles.</p>

<p>Any image file you send out for printing should be set to a standard, device-independent color space, not some print profile. If in doubt about the lab's capabilities, make a copy of the image and convert it to sRGB. Use the lab-supplied print profile for soft-proofing (on a calibrated monitor), and make any fine adjustments with this tool.</p>

<p>It is a waste of time to try to calibrate a monitor from prints. First of all, this practice assumes the prints qualify as standards for comparison. Nothing in your post indicates you have calibrated the printer. Secondly, the manner in which the image is produced is completely different. Finally, the appearance of the print and monitor are strongly affected by the ambient light.</p>

Link to comment
Share on other sites

<p >Right - I'm just about getting this now.</p>

<p >Unfortunately due to my current financial situation, I do not have the option of buying any new equipment, be it a monitor or calibration software. I have to make the best out of what I've got for the moment.</p>

<p >So, I send the image to the lab (with their profile assigned) and a few days later I get the print back. Now, if I open the exact file in Photoshop CS2 then soft-proof it to the lab’s profile, I should now (in theory) see an image onscreen that is similar to the print I have in my hand. The bigger the difference – the more ‘un-calibrated’ my monitor/screen is. Anyone disagree with this part?</p>

<p >So then I have a whole bunch of images of my daughter’s birthday party printed. My family come round one afternoon and I decide to show them all the prints, which look fine because there is nothing to compare them with. Then I decide to show them the exact same images on a slideshow on the laptop. Surely it will seem like we are looking at two totally different sets of pictures. The prints will look nothing like the ones on screen. Anyone disagree with this part?</p>

<p >So, what I can’t seem to get my head around is the fact that you can’t edit the image(s) using the print lab’s profile to be able to see exactly what you are going to get as you are making the adjustments. And, you would have to save two versions of each image – one for viewing on the monitor, and one for sending to the print lab. The ones sent to the print lab will obviously be unsuitable for showing (say as a slideshow) on the monitor because the colours and tones will be way too strong. Anyone disagree with this part?</p>

Link to comment
Share on other sites

<p>You don't get it. Soft-proofing is worthless on an uncalibrated monitor. You can't calibrate a monitor against a print. If your monitor is too "strong" (bright, contrasty), then don't expect to get a good display of your images.</p>

<p>If you can't afford a calibrator, then don't edit your pictures. You will get better results if you send them out just as they come from the camera.</p>

Link to comment
Share on other sites

<p>Edward,<br>

Thankyou for your comments. I am fully aware that with the current 'uncalibrated' monitor I have, I will struggle to match any print to any image onscreen.<br>

What I was trying to establish is whether or not I now have an understanding of the 'theory' of how your workflow/equipment should be correctly set-up.</p>

 

Link to comment
Share on other sites

<p><em>So, what I can’t seem to get my head around is the fact that you can’t edit the image(s) using the print lab’s profile to be able to see exactly what you are going to get as you are making the adjustments. And, you would have to save two versions of each image – one for viewing on the monitor, and one for sending to the print lab</em></p>

<p>My suggestion is that it is better to do nothing than to do harm. I try to avoid making extensive corrections in the field instead of my work station and controlled environment, even though my laptops are calibrated. This is an argument for saving JPEG files in parallel with RAW files in the camera - the RAW files are your digital "film", while the JPEGs are "ready to eat", since they are processed and enhanced in the camera according to preset conditions.</p>

<p>The starting point is a normalized image file, consisting of a set of a set of pixels represented by numbers which, together with the embedded color space, specify how much red, green and blue should be present in the display. If you "Convert" to another color space, the numbers are changed to conform to the new color space. If you "Assign" another color space, the numbers remain unchanged, but the way colors are displayed changes, according to the new color space. The job of a CMS (Color Management System) compliant application like Photoshop is to read which color space to use to determine how the color(s) should be displayed. (That's why files look the same in Photoshop even when you change to a different, even incorrect, color space.)</p>

<p>A monitor profile is a complex "correction factor" - a set of curves which alter the signal sent from Photoshop to the monitor for each color (RGB) as a function of the intensity of that color. That way, the color interpreted by Photoshop (or other CMS program) is displayed correctly. The image file is not changed, only the way it is displayed.</p>

<p>A print profile acts in the same manner, except that it corrects the colors sent from Photoshop to the printer. Again, the image file is not changed.</p>

<p>Soft-proofing is a way to see how the limitations of paper and ink affect the way the image will look in print. (<a href="http://www.luminous-landscape.com/tutorials/soft-proofing.shtml">http://www.luminous-landscape.com/tutorials/soft-proofing.shtml</a>). If you have an accurate profile, the differences will be very subtle from that with soft-proofing turned off.</p>

<p>Now to answer your final question - you may have two or more copies of the same file for different purposes. It is common to create a separate file strictly for printing, based on your master file. The "print" file will be of the exact size as the final print, with the appropriate resolution (e.g., through resampling), with sharpening and (perhaps) edge darkening to match the final aspect ratio. If your calibration and profiling is accurate, the "print" file won't look much different than the master, but it is a separate file in any case.</p>

<p>Similarly, you may edit and archive your master images in a wide-gamut color space, like Adobe RGB or Prophoto RGB, but find that these images look flat and unsaturated when viewed in a non-CMS application like Internet Explorer. The solution is to convert a copy of that file to sRGB for viewing purposes (e.g., a file uploaded to PNET, or a slide show to be viewed on a TV screen).</p>

<p>You asked for the time and I built you a watch. Oh well!</p>

Link to comment
Share on other sites

<p>I agree with Edward. <br>

Your choices are to not use what you see on the screen as a guide when you edit your jpegs (forget raw) as you need to go by the numbers or you can make changes, hit print, see what you don't like about the print and edit again (an iterative approach).<br>

I suggest the cheap but good approach I took with my laptop: used CRT + reasonably priced calibrator.</p>

Link to comment
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now
×
×
  • Create New...