PSP v. PS - Colours

[chilipie]

Weird thing just happened - I took a screenie, and copied it in to Paint Shop pro 7. It looked exactly as it did on the screen. Then I copied the same screen shot into Photoshop CS; it looked loads darker. Anyone else encountered this problem before?

[brothercake]

Have you done gamma setup on both programs, and have them both the same? That would be my guess..

[chilipie]

How would I do that?

[bradyj]

Photoshop has a lot more advanced color controls in it's preferences. On a mac, this is under file > color settings -- if you don't want any adjustments because you have no idea what it's doing, just turn all profiling off. If you want an explanation of what each thing is doing, just ask:)

either way, the photoshop is profiling your color when it brings it in.

[JamieR]

Ollie if you want my advice, ditch PSP and just use Photoshop - I know some things are easier to do in PSP, as I'm a ex-user of it..but at the end of the day, Photoshop is a lot better in everything it does:P

[chilipie]

On a mac, this is under file > color settings -- if you don't want any adjustments because you have no idea what it's doing, just turn all profiling off.

Edit > Color Settings on PC ;) . I turned all the color settings off, but it didn't look exactly the same as it does in PSP :( .

If you want an explanation of what each thing is doing, just ask:)

Well... :p I'm not really sure what any of it does, so everything you know :D .

Ollie if you want my advice, ditch PSP and just use Photoshop - I know some things are easier to do in PSP, as I'm a ex-user of it..but at the end of the day, Photoshop is a lot better in everything it does:P
Yeah, I agree with you, but I started out in PSP, and there's still some stuff I can't/don't know how to do in PS, so I just switch between the two sometimes :cool: .

[JamieR]

Well true, i'm more of a photoshopper now, as I call myself a ex-paint shop proer so to speak. However, when I'm saving a screenshot, I prefer to use PSP as you dont have to create a new image then paste it in which is what you have to do in photoshop.

Anyhow, overall, Photoshop is way better than PSP, although, I admit I do sometimes transition between the two.

[ronaldb66]

You seem to dismiss the fact that PSP costs about a sixth of what Photo Shop does...

[JamieR]

Yeah I know that....I still transition between the two.

[chilipie]

If you want an explanation of what each thing is doing, just ask:)

C'mon then Brady, bring it on :p .

[Bessieflint]

Post deleted - spam


// liorean

[JamieR]

I believe that that is a advertisment or spam. Unauthorised advertisments are not allowed on this forum, please read the rules.

Jamie.

[bradyj]

C'mon then Brady, bring it on :p .

Brief overview:)

It is estimated that of the millions upon millions of colors the human eye can see -- RGB can only reproduce 13% of that output, CMYK only 7% (at most) -- and this can ultimately varying upon the quality of your output. Ultimately, the goal for printing, and the color profiles you use, should be dependent on the output device you or your professional print uses (and that printer should have a profile for you, or will handle it yourself). How we control that output is typically the largest question, and the trickiest.

Most printers take a predefined color setting on a machine they print out (using something along the lines of this machine (http://www.colormall.com/acb/stores/1/product1.asp?SID=1&Category_ID=232&Product_ID=1237) ), and then use the said device to match each color perfectly (or as perfect as possible) and adjust their printer for any weaknesses (such as does the magenta drop too much ink - bring it's maximum ink level from 100% to say 95% - but obviously much more advanced). That profile would then be translated to work with your program so that the colors you build on screen, or mathematically like we recommend (never trust your eyes or your monitor, get used to figuring out what mix makes what by the numbers) -- this is where your photoshop color settings come into play, and what their usefullness, or not, is.

CMYK (Cyan, Magenta, Yellow and Black -- the K is not a B only because early printers had the bad habit of calling Cyan, Blue... even though it is not blue. To get away from this confusion, it was given the letter K instead) is the standard four process prints - you'll notice blues, oranges, some greens can be quite dull. Modern printers offer much more CMYKcmOG (Cyan, Magenta, Yellow, Black, Light Cyan, Light Magenta, Orange, Green) -- adding these extra colors ups the percentage to almost full RGB output -- and has earned the term 'Hi-Fi Printing in the industry'. Really cool printers (http://www.oceusa.com/main/product_detail.jsp?PRODUCT%3C%3Eprd_id=845524441805287&FOLDER%3C%3Efolder_id=2534374302161412) are high end lasers, and are almost perfect RGB output.

Color is an extremely long discussion (I could go on for hours) from the science to the psychology... but what you should know is that a profile controls how much or how little the colors output; and this is the same for your browser and your printer. RGB as a profile can be wacky, and is a big problem for a lot of retail shops -- what happens if a client orders a red sweater, but she thought it was more maroon, and it came to her as a redish-orange hue? Client is unhappy, and returns it (Amazon.com has invested money in methods to control color through screen, though much of us argue against it for good reasons still). All those profiles in there work at a different level of color spaces (http://www.drycreekphoto.com/Learn/color_spaces.htm), Adobe RGB being the largest output of possible color, and being my preference. Most digital cameras work in sRGB, as this is typically synonymous with most electronic devices. A users computer monitor will always adjust the colors to what the settings are specified -- but for those systems, Mac OS X cough, whose native browser, safari cough, can handle profiles -- when you output your jpeg with an embedded profile, it will look proper to what you have set (that is, if your customer has a proper monitor, which is highly unlikely).

CMYK is more advanced -- if you look under those color settings in photoshop, and play with a 'custom cmyk' setting, you'll get all the glorious ideas as to what a printer has to deal with. Ink colors specifies the Inks used for the process. 'SWOP', though not right for digital printing, is the standard for much printing since it is the typical standard for screen prints (and they have set the bar for much of the other processes). All the others mean what they say. GCR to UCR are specific types of separations of color output -- but they are really there to control 'the gray'. The trick in printing is to make your grayscale perfect; processes like newsprint never get a good black with those types of inks and presses -- screen print is touchy. UCR = under colour removal
GCR = grey component replacement

UCR is pretty much a Light GCR in that it drops C, M and Y from dark, strong shadows; GCR is the process of gradually removing netrual tones and replacing them with black the closer it gets mathematically. In theory CMY should make black -- but this doesn't happen since inks are manmade, so they are imperfect by nature -- black has to be thrown in to stop it from being a puddle dark brown. Since your probably won't use UCR in your life, maybe one project, we'll discuss GCR.

The black ink limit, of course, can lower the amount of black in that process mix. If you change it to 90%, black will never be 'full black', CMY will take up the slack. The total ink limit is how much your stock can take -- processes like offset printing with thick material can handle a great amount of ink (resulting in more vibrant colors if done properly). The give or take in this is considering the printer. 'Neutral' tones of a photograph (midtones, quarter tones, three quarter tones -- stuff that isn't 100% shadow or 100% highlight in a photo) are controlled by CMY, nothing more. If you were to set your GCR controls to light, as opposed to medium or heavy, you give the printer more control to play with the ink on his end. You make it dark, you give him little control. So most people ask why would you use Medium or Heavy in the first place? Only if you think:
1) you're that good (you'll get there)
2) You think the company, and/or printer is untrustworthy and you don't want him ruining your work

Those two usually go hand in hand. But for safe sake, if you aren't knowledgable enough yet, back off on medium or heavy or maximum.

UCA controls the amount of under color in your prints (how much in the dark areas) -- and I've never known anyone who actually adds to this, but I wouldn't ever do more than 5-10%, and I honestly would never do it anyway.

The options for Gray and Spot settings for printing, can be summed up best here:
http://www.aols.com/colorite/dotgain1.html

These are the basics, ask questions as you see fit:)

The standards for color control electronically was in a company alliance that can't really seem to get a hold on anything without overcomplicating things:
http://www.color.org/

But it's getting there. At the moment, the mix of devices and processes makes it a fine art and a skilled process.

My recommendations if you're unsure:
- Use Adobe for RGB. If going to web, no need to embed the profile, if you take another persons file, and they know what they are doing, don't 'convert' their profile to your working space (same for CMYK)
- Use a simple CMYK mix:
Swop coated (most common)
Dot Gain 20%
GCR
Black limit: 100%
Total ink: anywhere from 300-320%
UCA 0%

The other twist to the story is the 'Lab' working space, which is meant to be a bridge for RGB and CMYK (as well as can handle both color settings while building). It's not necessary to use this in your work unless you're doing some high end prints; and even then, Lab is usually used as a 'bridge' between two profiles (secenario: I take my Photoshop file, embedded with an RGB profile, it goes to the que and embedds with a Lab profile to translate, then it goes to the printer and is embedded with a CMYK profile to print -- this allows all colors to be translated properly).

You also have different variations of color output (absolute colormetric, relative colormetric, saturation, perpetual, etc.) which can control the translation of colors -- but this is meant for printers to worry about.

My hand hurts.

[rmedek]

Geez Brady, what Liorean is to web, you are to print. :)

I was still struggling with the "yellow and blue make green" on my sandwich bags...

[Scootertaj]

Wow I think that is the longest post I have ever seen!

[liorean]

Wow I think that is the longest post I have ever seen!Ah, come on, I must have written a couple of posts that were longer than that - I've had to split a posting over several posts since I hit the character limit...

It is estimated that of the millions upon millions of colors the human eye can see -- RGB can only reproduce 13% of that output, CMYK only 7% (at most)Well, if you ask our professors here, they say that 8-bit-per-channel RGB is quite a lot more capable than the eye in number of simultaneous colours. Similarly, camera film can catch far more colours than the eye can. The eye can distinguish about 8-12k colours at a time, and has an RBG spectrum where you can distinguish more colours the smaller the entire contrast spectrum is (especially brightness differences makes it hard to see hue differences). Also, we can see far fewer simultaneous colours in the purple-blue-cyan and red-magenta-purple areas of the spectrum than in the green-yellow-orange area. (Also note that in very low lightning, we use a slightly different vision system due to a transition from mainly tri-colour cones to mainly single-colour rods.)

However, the eye is sometimes superior to both camera film and computer graphics, but that's mostly related to small time-position-lightning-balance changes - metallic surfaces is a typical example of this, they look metallic not because of their colour, but because of reflections and surrounding lightning, and also changes in this due to minute position changes of the viewpoints (note that the fact we have two eyes with slighly different images also are part of this).

Of course, there's another question here - the size of the smallest possible point that the eye can discern. This size is somethere in the area around a tenth of a millimeter at close ranges, and modern printers, especially lasers, have point sizes far below this. The computer screen still has a bit to go to reach it, though.