This is assuming anybody has ever had a question for a graphic designer, but you know, everybody’s doin’ it…
My qualifications: been in the biz for 10 years, worked on projects ranging from print, (a little) interactive, sign systems, brand identity, advertising blah-blah-blah. The best recognition I’ve gotten is that my work has been in Graphis, which is, for my money, among the best publications you can get in.
What’s the best file type for a cross-platform-readable image?
For a recent project, I’ve been asked to make TIFFs with layers intact on my PC, but Photoshop still seems to want to make it Windows-compressed, so it confuses me.
Also, for a single layer, what’s better - TIFF or BMP?
What’s the deal with all the images I generate being RGB but printers only printing CMYK? I just made a fancy ray-traced picture for a presentation, but it prints out like crap. Turns out many of the colors don’t exist in CMYK-land, so the shading gets all messed up.
IAAGD…
The reason is that your computer’s monitor shows you colors made of light, which generates a wider spectrum of colors than four colors of ink can produce. Particularly the bright shades of colors cannot be achieved. Hexachrome printing (the addition of orange and green inks) can liven things up a bit if you are willing to pay for it.
To make sure your layouts match your expectations, don’t pick colors that are out of your printer’s range. When you pick an out-of-range color on your color palette you will see an exclamation point within a yellow triangle… click on the symbol to find the closest color match). Happy designing!
IANAGD, but PNG is a really nice compressed format - it has the advantages of transparency and losslessness, and it compresses real nice. PNG will flatten layers, though.
BMP is annoying, if only because it is a huge file size for minimal payback.
What DaveRaver said. When a monitor mixes Red, Green and Blue it is adding light to the equation. You can get some very bright cyans and greens. When you print on paper, the lightest value you will get is the white of your paper and any ink you add to it will only make it darker. The kicker is that even RGB monitors don’t do an adequate job of reproducing the number of colors that the human eye can percieve. Try scanning in a flourescent or neon color and see how poorly the reproduction is.
The fact is, despite great advances in technology, maintaing consistent color from real-life to digital input to output is greatly inconsistent. For some problem colors, it frequently takes professionals and specialized software to do the job properly
There is no one perfect file format, it all depends on the usage. Does it need to be manipulated by someone else? If so, always check and see their needs. Will it go to print or web? If you need layers intact, why not keep it as a native Photoshop file? To echo magog’s statement, I’ve never used .bmp for pro stuff. Print stuff is saved as either native Photoshop (unless you’re using Quark), in which case I use .eps or .tif. If it is destination: web, it needs to be .jpg or .gif.
(What does Windows-compressed mean exactly? I’m not familiar with that term.)
Sorry. I meant it asks for this
Actually, the files in question are intended for DVD, and they asked specifically for TIFF so that’s what I’m giving them. But some people ask for PSD (which I don’t think are cross-platform (i.e. will work in Windows and Macs)) and sometimes all is needed is a pure single image, so a BMP should be fine… but maybe a TIFF is better for cross platform compatability. Hence the question.
Forget PNG, I have never had a need for them, and they haven’t been very good in the compression department at all in my experimentation. I may be missing a step in the process, though…
If a manufacturer creates software for both Windows and Macintosh, the file can be used cross-platform so long as Mac users name it properly (use the filename extension and avoid any non-alphanumeric characters.) Summary: I create a Photoshop file on my Mac and name it “Image.psd” and I send it to a Windows user, he can open it with his Windows version of Photoshop.
OK, that answers my question very nicely. Now I have a follow-up: are there any colors or shades that are typically more problematic to automatically map correctly from RGB to CMYK? I know it’s a complicated issue, but I’m just looking for any rules of thumb or bits of common knowledge that people have.
Because my program does all the shading and lighting, I can only specify the colors of the objects before generating the image (e.g. a red sphere lit with a spotlight) – the shading and shadows produce a range of colors. So, if greens are more likely to be troublesome, I can just start with yellows or blues – the specific colors I use aren’t that important, as long as they are bright and easy to see.
Giraffe
Well, I don’t know if I know how to describe which values of which hues should be avoided in words exactly, though I fear that it’s those bright colors that you seem to want that will be the most problematic.
The best way for you to know is to see it. My suggestion is to get yourself in front of Photoshop and do the following:
Not a pretty sight is it? Welcome to the wonderful world of color-management.
Hey you!, thanks for the answer. I’ll use Photoshop to do some experimenting…
Do you still use mediums such as Prismacolors or watercolor to any large extent, or are you all computer driven? How big is your monitor?
Quick background for the non-artist types:
Before you make your final production-ready artwork, a graphic designer needs to produce a “comp” to show the client what solution the designer prescribes. In the days before computers, the only way to do this was to draw the layout with color media such as markers, colored pencils, watercolor, etc. Prismacolor is a brand name for one of the major producers of such color media.
Joe K
I was going through art school as designers were transitioning into using the Mac as the major design tool. As a result, I used color media a bit in school, but once I got into the working world, they were all but gone. Fortunately, I took to computers like a fish to water. There are still luddite designers who abhor the computer and rail against it at every turn, but they are in the minority. Some of the fortunate older designers who didn’t become computer savvy were far along enough in their career to become design directors which involves merely telling the younger whiz kids how they want the piece designed and leaving the technology to the underlings.
At this stage, the Prismacolors are probably more a part of the illustrator’s toolkit, though it’s worth mentioning that it is not uncommon for an illustrator to also be a designer and vice-versa. Charles S. Anderson is a great example of a famous and successful illustrator-designer.
I have a 17" monitor with an auxilliary 15" off to the side for floating palettes, email, iTunes, calendar, etc.
Why do JPEGs work great for photos, but seem to mess up and run colors together or blur areas of the image when it’s a non-photo (computer screenshots, .bmp drawings)? And what does GIF & TIFF do differently from JPEG that lets them compress these non-photos correctly?
Why aren’t there any low-end inkjet printers that do hexachrome instead of CMYK? I bought an Epson Stylus Photo awhile back and it had 6 colors (6 separate ink containers) but the extra two were variants on cyan and magenta, not orange and green. Thus not much increase in the gamut.
I’m a designer as well, so I thought I’d add an extra data point in this thread. . .
Neither I nor anyone in my company ever uses those physical media anymore. Except for a certain amount of pencil sketches, it’s all computer driven these days. Some of us paint/draw/sculpt/airbrush on our own time though. The world of fine art still has plenty of room for the tradition media, even if the world of commercial design doesn’t. It’s really much faster and more conventient to do a lot of design work on computer anyway. I don’t really miss traditional methods in that context.
I have two 21" calibrated monitors at my workstation. Now that I have gone dual-monitors, I would never go back! I even bought an extra 19" for my home computer (so now I have 2 19" at home). I would rather go down in size to something like two 17" monitors than go back to one 21". I run both monitors at 1600x1200 pixels, with 32 bpp color.
JPEG compression is actually much more complicated and sophisticated then the the methods used in GIF or TIFF files. JPEG is very well suited to photos in particular, as you know. It is basically designed to take advantage of the limitations of our visual perception to discard a cetain amount of data in the image. For instance, we are more sensitive to slight variances in luminance than in hue. One of the things that JPEG compression does is divide the image into groups of pixels, and this is where you often get artifacts with non-photo images.
GIF files use only 256 colors total (rather than the millions possible in a JPEG or TIFF), and then further compress the file by representing contiguous runs of a single color as one entry, rather than on a per pixel basis.
TIFF files can use a similar kind of compression, but also can handle many more colors. TIFF files also support other advanced features like 256 levels of transparecy (GIF supports one level). TIFF files are often very large, but they are the format of choice for many designers.
Those are somewhat simplified explanations, but I think they get the gist of it across (I hope!).
Very interesting Enigma42, I’ve always wondered that. Thanks!
I am using Illustrator 9 and I’d like to fit some type into a circle. I’m having a hard time trying to put this into words, but I’ll try to explain more. I am creating the cover to a booklet that is supposed to have the feel of an old circus flyer. I want to put “Feb. 19th” inside a circle and strech the type out so that it takes the shape of the circle.
I know there are two options that might give me something close to what I need:
A. I leaned a few years ago (on a much older version of Illustrator) how to merge two objects - lets say a red circle and a blue square. You could pick the number of steps, merge the items, and end up with a red circle on the left, a blue square on the right, and X number of objects in between, with the middle one being a purple shape that was halfway between circle/square. This might give me something usable if I converted my type to outlines, then merged with a circle, but I can’t figure out how to do this on v.9! (Admittedly, I have not yet consulted the manuals…)
B. I could create a circle, put the type converted to outlines on top of it, and manipulate the type by hand to fit the circle using the hollow selection tool on individual points. This would take a while and it would probably be hard to make it look just right.
::crosses fingers:: There’s an easy way to do this, isn’t there?
Enigma42
Thanks for the assist!!
AHunter3
You can’t find many low-end inkjet printers that print in Hexachrome because there aren’t many high-end printeres that print in Hexachrome. I’m not the final word on this, but I don’t know of any desktop printers that print in CMYKOG. Even in the professional offset printing world, Hexachrome (Pantone’s proprietary CMYKOG color-seperation technique that we’re talking about) isn’t taking off. We can only speculate as to the reason, but I would wager that the color gamut isn’t expanded enough to (1) give up two plates on press that we could be using for spot colors or spot varnishes and (2) make us give up the Cyan/Magenta/Yellow standard that we all know how to mix colors (which is an extension of the “Red paint and yellow paint make green paint” we learned in grade school.)
