I usually find myself creating scenes along with Lua on a separate text editor window
side-by-side with Garry’s Mod. If I want to, say, spread rocks in a specific manner in
the scene, I can save hours of work with several lines of code.
DAL: Right, I see. So kind of like Vue’s Ecosystem materials. Sounds also like a sort
of advanced free-flow version of the new FlowScape. And then you use Photoshop for
postwork. Could you tell our readers what the complete workflow would be, in
outline, to make a picture in Garry’s Mod and Photoshop?
Vioxtar: Sure — when I’m done building the scene, it’s a bunch of 3D models in this
black, dark void with no light. So, the first thing I do is light it: I place a bunch of
lamps in the scene, and for each lamp, render its own light separately in its own
picture (with the rest turned off). So, say, seven lamps give me seven different
renders. I do this because later, in Photoshop, I am then able to take all these
individual renders, stack them additively on top of each other... and then I tweak their
brightness values more carefully. That way I get more control over the lighting
setup, and can maximize the impact of the scene. Sometimes finding the right
balance between lights takes me hours.
DAL: Yes, I found that with iClone. With something like iRay in DAZ Studio or
SuperFly in Poser, you tend to just slap on a preset light, do a little test render and
then and if that doesn’t work you try another preset. But becase iClone is “what you
see is what you get” real-time, the temptation is to keep pushing and tweaking your
custom lighting for hours until it’s perfect. But, back to Garry’s Mod...
Vioxtar: ... so after the lights, I add atmospheric elements – sky, fog, clouds or
smoke. At this point the picture becomes almost whole, and I can start tweaking the
picture more ‘conventionally’, that is: colour grading, removing render artefacts,
adding effects, polishing, etc.
In between all that, I may sometimes extract isolation passes (where every model is
coloured differently), or depth passes, or material overlay passes from Garry’s Mod to
aid the editing process. You can see a video of all this here:
DAL: Thanks. You mentioned the next version, but is development still continuing on
Garry’s Mod, is it now in a fairly mature and stable state? Setting aside its qualities
as a game —which seems to be how it’s marketed, at a similar $30 price to a game
— would you still recommend it to artists looking for a lower cost alternative to
iClone, and/or who are worried that the ‘free’ Unreal/Unity engines will start to slip
over into some paid / subscription version. Or should they wait for S&box?
Vioxtar: The current version of Garry’s Mod, which is 13, is quite mature and well
established. However, that doesn’t stop a dedicated team of Facepunch developers
from continuing to improve it with casual updates... and then there’s the whole
community development side to it. For example, just recently a community member
created an impressive VR addon for the game.
DAL: Right, so there’s still a lot of work being done on it. That’s good to hear.
Vioxtar: As for a recommendation, even though Garry’s Mod answered my artistic
needs, I would be hesitant to recommend it to others. Garry’s Mod is a game more
than it is an artistic medium, and the 15-year-old engine is bound to ‘fight back’
when you make technical demands on it. To ‘beat it’, I find that I have to get into
quite a fight with it, to make every new picture I work on. The workflow revolves
around workarounds. It’s a challenge.