If you are converting your pdf to the same cmyk color space as you have been converting the individual images in the past, there is no advantage. You will be doing the same thing and saving time by doing this in the coversion to pdf.
If all you're doing is converting is just converting, then the result will be the same, but if you're interested in quality, you can almost always do post conversion tweaks that improve the image. And some images are better with Relative Colorimetric while others might need Perceptual rendering intent. Some might need a heavier or lighter black plate. And then there's sharpening. Sharpening should be done in CMYK at the final size for the best results. My normal workflow is to design in RGB, then convert all files to CMYK, then resize so they import at 100 percent and sharpen them as needed. Yeah, it's extra work, but it really pays off when you see the job coming off the press. What's right for you depends on how picky you or your clients are and if you need to squeeze that last bit of image quality out of your files. If "good enough" is okay for you, then the convert on export route might be fine.
Thanks for the advice. To be honest, I don't generally do a huge amount of image adjustment (in fact, you say "some images are better with Relative Colorimetric while others might need Perceptual rendering intent" – I don't know what this means, which might give you an indication of the level I'm working at). Where an image comes to me from a photographer, I often assume (perhaps wrongly) that any adjustments to the image have already been made by them, and that they wouldn't appreciate me doing anything further.
Now I think about it, I should probably do some research/learning with regard to optimizing images for print using Photoshop – the main focus of my work is on layout and typography, to the point where I am probably reglecting this area.
I was probably trying to assess where you were at in terms of what your jobs required and what you were used to and capable of doing. This is all good to know.
Whenever you convert digital color from one color space to another, and it can be from RGB to CMYK, RGB to RGB, RGB to Lab, CMYK to CMYK, CMYK to RGB, etcetera, there are different ways of applying the conversions. Photoshop allows you to use four different "Rendering Intents". They are Relative Colorimetric, Perceptual, Absolute Colorimetric, and Saturation. The two most common are the two I mentioned - Relative and Perceptual. The biggest difference between the two is what happens to really bright, saturated colors when you convert from one space to another. You can read up on that online more quickly than I can write about it here, but the difference can be substantial with some images and inconsequential with others - usually in whether there is detail in saturated colors or whether it all turns to one solid blob of color.
Being a commercial photographer in Los Angeles who also does a lot of high end prepress work for a variety of clients, and sees a lot of work from all sorts of photographers, I would say that most, including some very well known ones, have little idea of how to color correct, and that images provided by photographer can almost always be made substantially better. It's easy to think you know how to do it right until someone sits down with one of your images and shows you otherwise. I could tell you a hundred stories…
Everyone comes from there own perspective, and mine has always been to make the images I'm working on, whether they're mine or someone else's, print they best regardless of the type of printing - from inkjet to offset. I take an individual approach and fine tune each image for each catalogue, brochure, CD package or large print to be it's best. But I also realize that not everything needs this treatment and not all clients can tell the difference anyway. And then there's always the time vs. money factor.
Peter Figen Photography
5069 Exposition Blvd.
Los Angeles, Ca. 90016
What could help:
-Synchronize all application to the same color settings (be sure to select the proper Color Settings first -> 'Europe Prepress 3' if you're located in the EU for instance)
- Keep your images in RGB
- Place the RGB images in InDesign
- Export to PDF using proper PDF export settings (PDF/X-1a if it has to flattened (PDF1.3), PDF/X-4 if you want to keep live transparency)
- Never forget to 'live preflight' your document before converting it to PDF or sending it to your Print Service Provider btw :-): http://www.vigc.org/standard-preflight-profiles/
This approach assumes of course that you are content with the standard ICC based color conversion that InDesign will do for you during PDF conversion (similar to the color conversions you did in Photoshop) and that no manual tweaking of images is required to achieve better image quality afterwards.