
While you still should keep those options at the ready, the big win here is that you can work much faster in unexpected ways.
Personally, I’ve been big on incorporating 3D into photographs since PS first introduced the feature in CS3. Prior to that, I just didn’t have the experience to pull off the stuff I wanted to in ‘real’ 3D applications. But the moment I saw it in CS3, I knew I had to own the capability to exploit it at every turn. Unfortunately, in CS3 and CS4, using 3D in photorealistic ways meant getting outside help from apps like StrataCX and 3DInvigorator. I love the programs, and have not replaced them, but now I can reserve them for the heavy lifting.
In CS5, I have been much more confident in the 3D capabilities all the way around. We now have Image-Based Lighting, too, which allows a whole new level of realism in lightfall and environmental cues. When I started with 3D in Photoshop, one of my goals was to use it for generating realistic shadows when compositing. In my book, I discouraged users from trying to use multiple objects in a scene because it was so difficult to control. And the shadows had no control for density, etc. There were other limitations, too, but that’s history 😉
Using the enhanced feature set in CS5, I can no easily mock up a ‘roughly similar’ shape to catch shadows. Using Repoussé and some primitives, I can actually get pretty darn close to most surfaces that I’d want to deal with. Painting in the remainder is quite easy, as it turns out, so as long as I can get in the ballpark with a simulated shadow, I can eyeball the rest. Previously, I had to eyeball everything, including density and soft edges.
Repoussé also allows the 3D neophyte to more easily manage elements like lighting and materials. There are presets built into the Repoussé dialog that let you do quite a bit with simple button selections. And when you return the object to the PS main window, you get a bona fide 3D object, just like any other imported object. Your edits are not lost, and the materials go along with the 3D layer. This makes it a lot easier to get quick 3D elements knocked out.
There are some shortcomings, though. One major annoyance for me is the sensitivity of the sliders for the Extrusion and other options. Given that you are taxing your GPU anyway, these overly sensitive controls can really tie things up and cause frustration. If you get a bit antsy while moving one, chances are you’ll overshoot by a LONG way, and have to wait for everything to settle down before cleaning up the adjustment. Another problem is the naming conventions. I’d love to see an input box where I can set the prefix or name of an object, and have all child elements named with that prefix and a short identifier. It’s easy to lose track of materials and objects because the names do not seem to be consistent, or even logical in many ways.
In the coming year, I will spend considerable time working 3D into my images, and I’ll share those images and techniques whenever possible. Look to CommunityMX.com and PhotoshopTechniques.com for tutorials and articles. Of course, I’ll post here and on Twitter, too!
