New to Epson. Could use some assistance.

Discussion in 'Digital Darkroom' started by alex, Nov 29, 2009.

  1. Yesterday I got and Artisan 810. Up to now I was sending out the prints but this is going to be a good oportunity to do my own up to 8x10, but mainly 4x6's.
    I printed about 30 pictures and I'm far from happy. I now is not the printer since I've seen somem wonderful images coming from those. Shaded areas particularly in the hair or things with that texture look terrible like made out of grans of sand, their moody and for the most part bery few pictures came out ok.
    My monitor is calibrated with Eye 1 display 2 so I doubt that is the ofender. All images were taken in raw then proccesed in ARC and final adjustments in CS3.
    For color space I use
    RGB: sRGB IEC61966-2.1
    CMYK: Web coated (SWOP) V2
    Gray: Dot gain 20%
    Spot: Dot gain 20%
    Soft Proofing is set to custom
    Device to simulate: Artisan 810 Photo paper glossy.
    Rendering intent: Relative colorimetric and black point compensation is checked.
    I did noticed that the images I had already proccesed when soft proofing them the darker areas/shadows show some bluish overcast that I could minimise but can't get rid of with an exposure layer. Even so I would be ok with a print like it but most of them are not even close, the images are not consistent the one with the truck (see link) doen not even show the front bumper. The proofing is not helping me a lot.
    Going to the print panel setting
    I selected color management
    Uncheck the Match print colors box make the image look a bit darker on the preview
    Then proof (profile) Artisan 810 Glossy photo paper
    Color handling: Photoshop manages colors
    Printer profile: Artisan 810 Premium glossy
    Then I set the printer
    Color management to: ICM and check (off) no color adjustment. set the paper the size and quality. Click preview and print. The preview looks awefull a whole lot wors than the actual print so I don't see much value in it.
    Last step I clik on print and wait about 15 to 20 sec for the disaoppointment.
    Could someone point me in the right direction? Here is one of the images I tried to print.
    I tried 4 times with different setting and one was acceptable but not even close to the image you see on screen.
    This one out of 3 same thing 2 are garbage and one I couls keep but miles away from the monitor.
    The air on this las one is awefull, the shaded part of the face is terrible I don't know hoe to discribe it.
  2. The printer is new? All prints have been off? What model Epson?
    First I'd run the nozzle check utility to make sure all colors are printing. You can access it in the Epson software.
  3. Try to print with printer color management. Send sRGB.
    If the print is good. You have a starting point.
  4. Alex
    Then proof (profile) Artisan 810 Glossy photo paper
    Printer profile: Artisan 810 Premium glossy​
    Epson has both Glossy photo paper and Premium Glossy photo paper. There seems to be a mismatch.
  5. Can you post screenshots of the printer driver settings? Your Photoshop print settings seem fine so I'm guessing that's where the problem is. I'd ignore the suggestion to print with printer color management. Also, search back posts here for Patrick Lavoie's How to print on an Epson guide.
  6. Thanks for the advise so far. Here is a picture that shows the screen captures on every setting I have gone through. Second and third sreen shots shows the difference with preview on and off. I hope this could give you an idea of what is going on. I like to get a book or some material about color management (suggestions welcome) until I get one, read trhough it and implement a system I like to be able to print something that does not look miles away from the monitor.
  7. Alex, from your color sequence:
    In the soft proof dialog, select the option "simulate paper color"
    Then, in the printer driver window, I see that at the bottom right the color management option is "ICM". This shoul be "No color management" (or equivalent)
  8. Try the advanced settings within the Epson print driver. The settings you show in your next to last screenshot don't tell enough info (i.e. is ICM= ICM OFF?)
    Also, you don't want to use "photo" quality but rather "best photo." This mistake will screw up your prints look.
  9. The ICM is off. (no management) I'll try the photo quility and see what difference it makes. Also I'll try the same picture with the sumulate paper color. The little preview seems better without but I'll try it both ways.
    Thanks so much so far for the assistance. Keet it coming Please.
  10. You're doing an awful lot considering it's a new printer - how can you tell which step is getting it wrong? Try without the soft proofing and all the advanced options, set Photoshop to Printer Manages Colors and set up the options in the printer driver. Confirm that that can get a usable print. Then add complications one at a time until something break.
  11. The ICM is off. (no management) I'll try the photo quility and see what difference it makes. Also I'll try the same picture with the sumulate paper color. The little preview seems better without but I'll try it both ways.
    Thanks so much so far for the assistance. Keet it coming Please.
  12. Andrew I thought I was actually doing that. Aside from soft proofing, that as far as I understand does not afect the way the printer does it's work, but let's me know how far off am I from the output. Besides entering the print quality and type of paper the ICM is off on the Espon no color enhancement or management seems to be going through the printer, unless there is something I'm missing.
    I tried this picture checking and unchecking Match print color and it did make a slight difference. Check looked better. Also chequing best photo helped.
    I detect a color cast when soft proofing that I can't tell what color it is nor I can get rid of it. It looks like a bluish purple and it's mainly on dark areas or shadows. I kind of mutes the dark tones.
    So far the images are lot better than yesterday.
  13. What paper type and print quality was recommended for the Artisan 810 Photo paper glossy profile? (e.g. Epson Premium Glossy as the paper type and Best Photo or PhotoRPM as the right amount of ink). It's important to use the exact settings the profile was intended to be used with. If they don't list the settings, or if the profile's not particularly good, you might just try Epson's premium glossy photo paper profile and settings and if the paper's reasonably similar it may look better.
    I do this with Epson premium semi-gloss, luster and some Ilford papers- they are so similar that the same profile prints identically on all of them.
    Different rending intents may make the shadows look a bit different. Relative colormetric can yield weaker blacks with some papers and profiles. I often use Perceptual instead- soft proofing should show subtle differences between it and relative colormetric.
  14. Roger thanks for the imput.
    I used Epson Premium photo paper Glossy. I need to find out about Epson's requirements for for paper and print quility.
    The profiles I suppose came with the drivers Disc, unless they were part of CS3 and I saw them now that I got the printer and looked for EPSON for the soft proofing. For now I guess I'll stick to Epson papers until I feel confortable with the management proccess.
  15. Sorry, I thought the Artisan was a paper type- it appears to be the printer model.
    Try here for media types for Epson papers:
    You can download manuals and more information about your printer here:
  16. Hey, Alex, can you throw up a scan of a strip of one of the prints that looks bad? It looks like you're posting up the pics before they hit the printer. What do they look like after?
    And, is the print better or worse compared to a D50 look on the monitor? The whole point of D50 is printing.
    If you post a sample of the mistake (maybe a strip of it, considering your subjects) the troubleshooting might go easier.
  17. Daylight 5000K (D50) and Daylight 5600K; you've seen those forever in reflective prints. The 6000+ temps are for staring straight at the lightbulb of the monitor.
    I'm willing to bet that you're editing this while looking at a monitor, set for being a good looking monitor, and then wondering why your print (viewed under different conditions) doesn't match.
    Hey, there is no monitor lamp shining through that print.
    I know you know this; but, think on that for a second. Now think about what you've been doing while editing the image.
    [Try this: Make a big white rectangle on your monitor. Hold the print up to it, so that some of the light might help the paper to glow some. Try to get the monitor to illuminate the print from behind. Look at the print in a dark room in this way. If that print were a transparency, the temp of the monitor would change the appearance of the print.]
    If the print is going to be the final product for that version of the image, edit it under parameters that are similar to the conditions under which it will be viewed. If the final product or primary product is a print, use D50. If you need a print and a monitor image, run one version under D50 and another under 6000+. D50 is about printing.
    D50 is unbleached paper under noon daylight. It's been the standard for a bunch of printing stuff for a very long time. The monitor calibration is about changing incident light to coordinate with reflected light. Try one simplified: just do the whole thing in D50 and see if you can get it on paper.
    D50 is about printing. Those other temps are about monitor manufacturer's limitations. Use D50.
  18. The monitors are incident light sources. The prints are reflective. Even if that thing is "calibrated to match" (ha!), there is no way an engineer in some lab far away from you could have predicted the difference between your monitor and your light sources in the room where you are viewing the print now back when he coded in those software math standards for those applications.
    The print has to be built to look good where it is viewed now. That could be under a bunch of conditions. It has absolutely nothing to do with software parameters. When the image leaves the computer, the computer has no more influence over the print. None. Print color will be influenced by local light.
    You need to run your edits so that it looks good in the final viewing's illumination circumstances. The computer has no idea what those circumstances are.
    What you are seeing is a reflective photo that was edited while it was an incident light source image. Too far off. Set for D50 when editing for printing.
    And I don't know about that suction cup calibration thing. Newspaper graphic designers and press operators never used those. Unless this thing fetches coffee, I'd recommend unplugging it permanently. You can use your eyes and your brains and your hands to make your image. I have confidence in you.
  19. When you are using that suction cup calibration thing you are trying to get the computer to predict for you what the difference is between light from your monitor and light in the room where you are viewing the print. Is that logical?
    No, you are in charge of the print. You match it with your decisions. Fire the suction cup!
  20. For glossy photos I too, always set the printer to the highest quality, the "best" setting. Also, Epson printers prefer a print resolution of 360 dpi, not the 300 shown in your dialog box. But, the problems you describe, for me, have almost always been a mismatch of paper to ICC profiles. And like Ron said, make sure the nozzles are clean.
  21. John, you're overthinking this a bit. First he needs to get the basic driver settings right for the printer. You're right that if the prints then look too dark it may be necessary to recalibrate the monitor at a lower brightness setting.
    That's one reason calibrators are needed- to drop the luminance without skewing all the other colors. Many LCDs work better (less banding) with native white point than specifying D50 or D65. The second reason is softproofing.
    I don't think the advice to abandon color management is remotely helpful. It's also unnecessary to calibrate once for prints and a second time for web or "a monitor image." We have more control now than newspaper press people every did (not that newspapers were ever state of the art for image quality...) Why give it up?
  22. Okay, good points. But, there is an old fashioned logic to the appearance of the final print guiding a manual color adjustment workflow. Less passionately, I would suggest that workflow would go like this:
    Set monitor for D50, only for making the print. Pull up the original, edit in D50 to make it look the way you want. Print it out. Look at the print. Does it look as you want? Probably not on the first pass. Compare the print to the D50 image on the monitor. Adjust and refine the computer file, to generate the kind of change that you want to see in the print, even if that change looks horrible on the monitor. Print it out again. Recycle until you get the print to look right under the display viewing conditions.
    That basic workflow guided printing for years. The overarching concept is that it is the appearance of each of the two kinds of final images on which performance will be judged. Since there are two kinds of images, each with their own physical advantages and disadvantages, it pays to edit each version for its own final image.
    You are not giving up control with this method. No, no, no! You are having to take responsibility. You are having to do the job of the pressman who was not hired when you printed the image yourself. You are having to take the control that was not given to him. Use his methods.
    In that comparison part, the answer is going to be something like "Add red, subtract yellow." Just like a pressman at the controls. How many inks in that printer? Four? Then you have four options to adjust. That's it.
    Two image forms: incident light and reflected light images. Two appearance standards. Two editing methods. If you go one size fits all, something is going to fall short.
    The final image's exposure to light is predicated by all preceding steps used to make the image.
  23. Like, in the example above, Alex said he had a tinge of magenta or blue in the shadows he didn't like. So, maybe a touch of increase in the magenta or blue to darken it, maybe a corresponding increase in the black to help cover it up. Or, maybe hold those magentas and blues, and touch in a complement with or without black. It's going to take some experience to figure out how to get it right; this is why those pressman get paid the big bucks.
    The monitor shows four colors. He's probably got about four inks. If we edit the image to look good under a million incident colors, is it really plausible to expect that same image file to look exactly perfect in four reflected inks on paper?
    Two sets of edits. The print is a tangible object. It can only be pushed so far before it breaks. Entropy is going to getcha. There's imperfection in making an object. If the editing is focused on making that final object be a final object, within the limitations of its structure, it will work out better in the end.
  24. John, I print on an Epson 7900. My monitor is 30" at 2560 dpi. I use a COLORMUNKI spectrophotometer which also measures and compensates for ambient light at the monitor. After much trial and error prior to the Munki, I now print exactly what I see on my monitor. I don't make many mistakes and with 24" paper that saves a lot of money. Your exercise in color control provides excellent understanding of how color works, but I'll take the fast easy way.
  25. Good point, Dominick. Guys, I'm gonna close with my responses for a good while because I've written a lot. I guess I'm advocating a second workflow for prints; and, if I haven't convinced someone by now to consider that method, then I'm not going to be helpful by writing any further. Proceed with confidence.
  26. John: I trully appreciate your time on this post. The info is great. I think I'll read it once or twice more to grasp the whole concept.
    The part you're talking about the D50 or D65. I know it's going to sound to basic but. Where do you look that up? I will try it and see it for my self.
    As much as I favor new technology, hardware to calibrate software and those nice little gadgets, I also appreciate knowing what is going on and once I learn enough about it I go ahead I buy what I think I need. I can see the usefulness of such thing a the Colormunki and I might end up with one. For now I'll read and learn.
    Thanks everyone else for your great info so far. I is a good tread that is helping me a lot.
  27. John, I just want to say that I appreciated your comments as well since I'm just starting to establish a workflow that will end with a print off my local printer. You seem to think like Ansel Adams when it comes to making the print; I can relate to that.
    So while I'll be using a Canon Printer and have much to learn, I really like to hear different opinions about workflow and what to focus one, and conceptually separating the different processes, adjustments, and renderings.
    Good Stuff! I'm sure this thread will help a lot of people. Thanks to everyone!
  28. Alex, if you've got a Mac, it'll be right in "System Preferences."
    System Preferences > Displays > Color > Calibrate. Computer opens a dialog box entitled "Display Calibrator Assistant." You can choose "Basic" which is like a wizard, or "Expert" which will get you right into three dimensional graphs about color models.
    D50 is nothing but "daylight 50" or 5000K. D65 is bleached paper; 6500K.
    For the color modeling itself, the only references I could maybe point you to would be wikipedia (Color models). What little I know about publishing came from some years as a laborer on a factory line in a newspaper. Unless you know a guy nicknamed "Tugboat," I probably can't help you much there. I can't say I was ever really pointed to a printed reference outside of a college class. Errors were followed by cusswords and a pile of trash that needed to be picked up; I learned like Pavlov's dog.
  29. John,
    I want to echo Jake's comments. Sometimes on this site it seems that if you don't "fall into line" with the majority view, you're openly admonished for not doing so, yet I learn a lot from many of those posters.
  30. That basic workflow guided printing for years. The overarching concept is that it is the appearance of each of the two kinds of final images on which performance will be judged. Since there are two kinds of images, each with their own physical advantages and disadvantages, it pays to edit each version for its own final image.
    You are not giving up control with this method. No, no, no! You are having to take responsibility. You are having to do the job of the pressman who was not hired when you printed the image yourself. You are having to take the control that was not given to him. Use his methods.​
    This method does work. I used it for about two years when I first started scanning my film and doing all the post-processing myself. I got started reading Scott Eaton posts about reducing monitor brightness to match prints (see 2/3 the way down this thread) This process was fairly painful as I was sending out to Frontier labs at the time, so waiting for the prints to come back and then have to correct them and resubmit them took days.
    I find that even with a ICC calibrated workflow (monitor and printer) I am still tweaking and reprinting as John describes. With experience I know how different ICC profiles behave on different papers and also how they look using different rendering intents (relative vs perceptual vs saturation). The difference is the corrections I'm making now are modest and subtle as the initial prints are quite close to what I see on screen. I also like that I don't need to change settings on my monitor to post images for the web versus printing them- softproofing plus knowing your profiles through experience works pretty well to previsualize prints.
    For me the bottom line is that you can get decent results using any system if you are determined to have a decent print. Using modern calibration methods mainly saves you time and wasted materials. What's your time worth? Mine was worth more than the $70 I spent on an Eye One Display.
    Addendum : interesting link on D50- see the final post by Tim with a picture:

Share This Page