Designing, CMYK or RGB?

[jmcnicoll]

I like cmyk because of the control in color correcting raster images. I like everything 4 color for our solvent, and don't care if its RGB or cmyk for our aqueous printer.
Jim

 

[luggnut]

when i send files of to 4over or others i use CMYK... but for all in house stuff i use RGB.

just set you RIPs rendering intents to relative colormetric for rasters if you are rasterizing your files to print. if its just a photo though use perceptual.

and if i pick a color in photoshop i keep an eye on the CMYK values to get an idea of if my yellows will have cyan in them or reds might look too orange. photoshops color picker displays LAB, RGD,CMYK and HSB values even when you are in RGB mode.

 

[eye4clr]

and if i pick a color in photoshop i keep an eye on the CMYK values to get an idea of if my yellows will have cyan in them or reds might look too orange. photoshops color picker displays LAB, RGD,CMYK and HSB values even when you are in RGB mode.

So then i'd guess you are not applying an output profile in your RIP? If you are, the "purity" of your solid colors will be determined by the ICC in the media profile and the Photoshop numbers are not really relevant.

 

[Rooster]

Design in RGB and set-up your applications to soft proof in the output profile.

 

[luggnut]

So then i'd guess you are not applying an output profile in your RIP? If you are, the "purity" of your solid colors will be determined by the ICC in the media profile and the Photoshop numbers are not really relevant.

i am using the values of the other color modes present in photoshop to judge some of the colors... if i chose the brightest yellow i can in RGB it prints with a little cyan in it. but if i chose a RGB color while watching the other color modes i can see the CMYK,LAB, & HSB values that mimic my RGB choice... so if i choose a RGB yellow that has in the CMYK area say 100y 5m then i know i won't have a greenish yellow.

so yes i am using an output profile... but from my experience it is better to make sure my RGB color choices get me the best results.

 

[Letterbox Mike]

OK, so here's a question (mind you this is coming from someone who only understands color management on a basic level)...

This thread prompted me to experiment a little with RGB vs. CMYK files. So tonight I printed my Pantone swatch file 4 times. Twice as a vector file, one RGB and 1 CMYK, and twice as a raster file, again one RGB and one CMYK. Both have the output profile of relative colormetric. The file was originally an Illustrator .eps file. To rasterize it I simply opened the .eps in Photoshop. No other adjustments were made.

I printed all four side by side on 3M IJ180C using the apprpriate 720x1440 8-pass bidir profile. The input profile was relative colormetric.

All four were dead on identical. There was absolutely no color shift between any of them. In fact, I couldn't tell which one was which, I forgot to note on each file which was vector and raster and CMYK and RGB.

This sounds like a stupid question, but what does this mean? I assume it means we're doing something right to manage our color. Am I also safe to assume that I'm just as well off designing everything in CMYK as RGB, since I do have a much firmer grasp on CMYK color manipulation vs. RGB?

 

[luggnut]

This thread prompted me to experiment a little with RGB vs. CMYK files. So tonight I printed my Pantone swatch file 4 times. Twice as a vector file, one RGB and 1 CMYK, and twice as a raster file, again one RGB and one CMYK. Both have the output profile of relative colormetric. The file was originally an Illustrator .eps file. To rasterize it I simply opened the .eps in Photoshop. No other adjustments were made

your pantone swatch... was it created in CMYK? if so that would explain it..

 

[Letterbox Mike]

Couldn't tell you. They're not process swatches, they're all spots. I tried the same experiment with a couple other files this morning too with the same results. One file was mostly vector art, no PMS spot colors, with a couple placed images. The other was just a raster photograph. Printing as both RGB and CMYK vector or raster files resulted in no discernible difference between any of them... I'm not complaining, I'm just curious why, it seems everyone says RGB nets better more vivid prints, wondering if I'm accidentally doing something right, wrong or if my color management skills are better than I thought...

 

[eye4clr]

All four were dead on identical. There was absolutely no color shift between any of them. In fact, I couldn't tell which one was which, I forgot to note on each file which was vector and raster and CMYK and RGB.

This sounds like a stupid question, but what does this mean? I assume it means we're doing something right to manage our color. Am I also safe to assume that I'm just as well off designing everything in CMYK as RGB, since I do have a much firmer grasp on CMYK color manipulation vs. RGB?

Yes, that means your color management is working. Simply put, you tested the same file ran through different workflows. Some don't make sense from a production perspective, but make sense when you are testing/learning.

Also, if your color is really working, you should get very close results between sRGB and US Web Coated SWOP since they are not that far apart in gamut to begin with.

Comparing the different output from Adobe RGB or ProPhoto RGB with US Web Coated SWOP would show a greater difference. Thing is, each file needs to be CREATED in the respective working space to actually represent it. Starting from a single file and converting it into different "containers" of different profiles should not give noticeable different results.

Thanks for running an actual test instead of spouting half baked information that's been regurgitated over and over. Lots of this color discussion is nonsense that just perpetuates more confusion.

 

[sjm]

If your goal is to use all the colors you can capture when you output your files, the role of the working space is important. However, it should be clear that a very large gamut working space isn’t automatically the best option for all situations or something you can “set and forget.”

http://www.adobe.com/digitalimag/pdfs/phscs2ip_colspace.pdf

 

[Rooster]

This sounds like a stupid question, but what does this mean? I assume it means we're doing something right to manage our color. Am I also safe to assume that I'm just as well off designing everything in CMYK as RGB, since I do have a much firmer grasp on CMYK color manipulation vs. RGB?

It really depends on what was contained with on the file you sent as a test. If the pantone swatches were left as spot colors then all the files will reference the spot color LAB data in the RIP and convert them identically as you were using the same input data (LAB) and the same output profile.

Now if the original file was created as a CMYK file then converting it to RGB should create a file with essentially the same color as your mapping a small gamut into a large gamut.

The test should be done with an RGB original that you know has out of gamut colors. Try lot's of oranges and greens as they're the weakest part of the CMYK gamut. Pick pantone spots you know you can't hit and then change them to Process RGB. This will turn the spot color LAB data into standard RGB data linked to the embedded profile and not LAB equivalents for spot colors that get sent and processed independently of the RGB profile.

And then of course the reason why you're seeing no difference could be that your color management is working correctly.

 

[Letterbox Mike]

Good to know.

The file had both spot PMS colors and process colors. One of the colors, PMS 151, I converted to a process in one part and left as a spot color in another to see what happened. None were a dead on match to 151 orange, but they were very close given what I'm working with. There was a minor difference between the process and spot version of 151, but it wasn't significant or really noticeable unless you held the two right next to each other, I'd say either was a very acceptable CMYK match to a difficult spot color, and if I had to quantify the difference between the spot and process version I'd say they were within 2-3% of each other if that makes sense. What I guess surprised me was that the process version of the color was identical from the CMYK and RGB file, both vector and raster.

Woo hoo! It's starting to sound like my color management is indeed better than I thought it was. Thanks for the help, I really appreciate it. And Anotherdog, sorry to hijack your thread!

 

[Neil]

So tonight I printed my Pantone swatch file 4 times. Twice as a vector file, one RGB and 1 CMYK, and twice as a raster file, again one RGB and one CMYK.

I am wondering how you printed the Pantone vector swatch as RGB?
Did you convert each patch individually to RGB?

Just asking coz that's how I would have to do it in my sign proggy and if I did it would alter each color significantly.

And if you didn't, then maybe the files were actually the same when printed??

 

[Rooster]

I am wondering how you printed the Pantone vector swatch as RGB?
Did you convert each patch individually to RGB?

Just asking coz that's how I would have to do it in my sign proggy and if I did it would alter each color significantly.

And if you didn't, then maybe the files were actually the same when printed??

The easiest way to accomplish this would be to open the vector file in photoshop as an RGB file. The spot LAB data will convert to RGB during the rasterization process.

 

[eye4clr]

The easiest way to accomplish this would be to open the vector file in photoshop as an RGB file. The spot LAB data will convert to RGB during the rasterization process.

And, it would allow you to test different RGB spaces as well.

Perfect suggestion Rooster.

 

[Neil]

The easiest way to accomplish this would be to open the vector file in photoshop as an RGB file.

Oh yeah, just rasterize the whole page and save as a tiff.

I suppose I could do that in my sign program too - convert to bitmap. Photoshop would handle it better though and like Eye4clr says you could assign different profiles if you wanted to test them.

I thought Insignia may have had a way of converting the CMYK vectors into RGB vectors en masse.
I'm about to print out my CMYK Pantone chart again and would like to be able to print it as RGB vectors too.

Anyone know of an RGB Pantone chart?

 

[sjm]

Pick pantone spots you know you can't hit and then change them to Process RGB.

Interested in learning more about Process RGB.

 

[sjm]

Oh yeah, just rasterize the whole page and save as a tiff.

I suppose I could do that in my sign program too - convert to bitmap. Photoshop would handle it better though and like Eye4clr says you could assign different profiles if you wanted to test them.

I thought Insignia may have had a way of converting the CMYK vectors into RGB vectors en masse.
I'm about to print out my CMYK Pantone chart again and would like to be able to print it as RGB vectors too.

Anyone know of an RGB Pantone chart?

http://www.pantone.com/pages/MYP_mypantone/mypInfo.aspx?pg=20525&ca=75

No guarantees though.

 

[Neil]

Thanks, actually I realised I have Pantone palettes supplied within Signlab.
Solid to process is CMYK as is the Process coated pal.

Solid Coated pal is in LAB mode.

I can create a palette swatch for each one and print them. Still no RGB though, but some of the supplied vinyl palettes are RGB and I can create a complete RGB color swatch within Signlab.

I know this is about CMYK vs RGB, but is there any valid reason to use LAB??

 

[Rooster]

Interested in learning more about Process RGB.

You mean the combination of RGB primaries using additive colour?

Like say in a projector (monitors and TV don't use real additive color) or a Lightjet printer that uses RGB lasers to image photographic medias?

Or are you being a smarty pants?