What Percentage of a Typical Game is...

News & General Discussion
1

The modeling versus code and logic?

I was wondering as we design tools for 3D gaming level design use without much real world game design experience what ratio comprises a modern typical game? As in how much is the graphics design footprint versus the logic of the game in 1. end game size and 2. time for completion? Add any detail you want to your reply, as in what I am missing. This is actually for a good reason.

This research is being done to ID the model design bottle-necks from concept to refined 3D model to hopefully develop more tools that reduce the amount of indie and others’ needs to resort to “cookie cuttering” solutions. And we are focusing on the initial model concept (like from napkin to 2d to 3d first pass) and first layouts in specific as to how time consuming it is at present in general.

(In specific to gaming design, IF it is like say architectural 3D, then we know how slow it is, and we think it is just this in game design as well, btw, but I am looking for some specifics here and there)

We are trying to accelerate the first concept design stage by a new paradigm applied in the initial 2D-to-3D generation or projection for the concept to barebones model completion phase. In our case we are using AutoCAD for 2D and SketchUp for 3D due to ease of use, availability and base user size, and because the concept can end up being applied to any customizable 2D and or 3D program later; but we began with the AutoCAD to SketchUp synergy potential.

So a related question is what is the most time consuming portion of the model at concept level, not the tedium of detailing it when the base design is complete, but even coming up with a 3D “world” in its most basic form suitable for a game which employs such an environment at its most basic yet initially “complete” concept form? (before it gets detailed, the barebones 3D model)

Our goal is to speed up the initial design speed and ability to explore new designs all the time by making the 2D-to-3D phase automatic in much volumizing and profile projection. Your information here we hope will help us improve our future release of a tool more specific for game designers larger scale model and level design.

Thanks

2

I applaud you for trying something new and challenging old paradigms! However…

This is a red flag to me. You’d have to work very closely with people actually making games to reasonable address their needs in my opinion. Maybe you can find a few people here who would be interested to beta test and consult. I wouldn’t be the right one for the task though. I don’t really do leveldesign in any indoor/architectural sense.

As far as I know this asset is regarded as currently being the most streamlined solution for building levels in Unity. You should check it out:
https://www.assetstore.unity3d.com/en/#!/content/3558

I’m not so sure if that really is a solid strategy. If your background is ArchViz, it makes a lot of sense why you chose these tools, but I don’t think there is much overlap between the big AutoCAD userbase and gamedevs (please correct me if I’m wrong). I’m not quite sure what your target audience for your solution is. If you aim at the majority of Unity Indie-devs, then you are aiming at people with more hopes and dreams than budget. We regularly see people here talk about 100$ assets being too expensive for them, and - correct me if I’m wrong - AutoCAD costs 2000,- Euro per year and Sketchup Pro is 695$. Not even speaking about the issue of having to learn to use AutoCAD first, I think you are looking at an incredibly small potential userbase for any leveldesign solution that relies on using AutoCAD.

As I said I don’t really do leveldesign work, but I would speculate that the prime issue isn’t that people want to get from their first 2D scribble to the first 3D blockout faster, but that they want the iteration time between changes to be as low as possible. E.g. in a corridor based competitive first person shooter the line of sight from player perspective is super critical and moving an asset a few inches to the left can be quite a significant change under certain circumstances. I’m personally not aware (though that doesn’t mean much) of a solution for Unity that offers a good workflow to run through a level in play mode, allow live changes while playing the game, and gives the user the option to commit to certain changes to be applied to the level data when play mode exits. E.g. start play mode, try a change, discard, try a change, discard, try a change, commit to keeping it, try another change, discard that one, …, exit play mode, all commited changes are applied to the geometry/asset positioning, scene is saved.

Just my 2 cents. I’m a 2D/3D artist, so I hope what I suggest has at least some value to you, but if people, who actually do the architectural kind of level design that your tool might target, have contradicting opinions, you definitely should listen to them instead of me.

1 Like
3

There’s no simple answer. I’ve led 3D game teams where we have 2+ artists for every coder/designer, and 2D games with a 1:1, and currently, the 2nd most popular mobile games where it’s more like 1:2. I’ve led teams where the design was pretty obvious, and ones where finding the fun was a massive series of prototypes. There is no one-sized-fits-all structure. In general, 3D art implies higher overall visual quality, which means more work. Though, even that is not always true.

Good luck,
Gigi

2 Likes