What Do You Spend Most Of Your Game Dev Time On?

What updated workflows? Things haven't really changed that much in the last decade, especially not in games.. Yeah most "art" infused engines are custom and some you could even license (for an astronomical fee)..

The state of Enlighten isn't an engine development issue, it's a UNITY issue.. There's nothing wrong with Lightmass in UE and it's far superior in terms of quality.. It's not like they just bought some middleware either, they made it...

You think making an art package is difficult? You should try making an engine, a lot of it overlaps anyway with basic engine functionality and AFAIK the teams that work on said tools are usually very small.

I also cringe at the thought of Blender slapped into the engine, irrelevant of what either one of us say that's the direction things are heading again with Source 2 / Epic with ART (Animation and rigging toolset) and CryEngines in engine design tools..

I'll just say it for once Epic are lightyears ahead in terms of engine development for their main platforms, it's not even a comparison at this point and it's not an opinion either.. An engine is nothing but a collection of tools, once they've finished optimising stuff and decide to finish off their experiment with 2D or Mobile then it could be a good move if they get a bit bored..

They already said they would update their BSP tools to something more intuitive for level design, so who knows.

 

You make a valid point about workflows - but it seems like every cycle there is a new process improvement and I think integrating things into the engine can upset the flow. The future is uncertain.

I think we are agreeing here. The limited time I spent with UDK I can honestly say was pleasurable - if not pretty overwhelming since I had just been laid off and was prepping for a job interview - witch ultimately resulted in little more than a modeling/UV onsite test - when the position was calling for a animator. Not to rant but who judges someone solely on a skill that is nearly completely unrelated to the job requirements? haha - that was a while ago (Rant about that experience over)

Epic has - since I can ever remember - had BSP tools right? I remember ever since the second Unreal Tournament - working with guys who were designing levels in the engine, but most of the art (90%) that went on top of these layouts and all dynamic stuff except for elevators was still imported. This was all mod work experience - not commercial game development, so my experience is limited.

I'm also - (maybe) becoming an old dog Set in my ways - because I'm comfortable. I still have no cloud based software - and prefer it that way.
But change is certainly not scary - I'm just a little cautious about it.
I'm sure the 2D guys might of felt uncomfortable when Unity brought out the new 2D tools - but I was like FINALLY ha.

 

When I start the project:

50% coding.
25% Fixing bugs and/or glitches.
10% polish, if it works I clean it up later and get it looking nice later.
15% Looking up the problem I'm having, most of the time I figure it out but some times you just gotta Google it.

When I'm at a stage where I'm very pleased with it:

60% cleaning up old code and reconstructing it entirely.
15% fixing bugs from the new code, the second time around it's less messy and I think about it more before I do it.
25% polish, making things look nice, adding new variables for easy customization and better variety, new art, etc.

That about sums it up for me.

 

Most of my time is spent in premature optimisation

 

On my last game I spent around 80% on coding. Then got some feedback which basically said the artwork was lacking. So then lately I've been spending 90% of my time making the artwork better. So overall it will work out maybe 50/50 code/art.

The moral of this story is yes, your code might be amazing and bug free but most people only care about what it looks like!

 

Tools, suckas! I spend a ton of my game development time making tools. In some ways, this is an end unto itself. I just really like making tools. And thanks to services like the Asset Store, it is entirely possible for me to package up my modular tools and call it a project. This approach is really nice for me.

Also, Unity's general-purpose design and excellent UI-scripting system makes it a really great engine for developing in-engine tools.

 

Modelling/Animating/Texturing takes most of my time, maybe because i'm not that good at it yet.

 

It seems like a spend an insane amount of time play testing my own games, iterating the gameplay.

 

Research. Design second. Nothing worse than sinking time into something and then hitting a breaking issue which forces kludges on an issue, that if it had been researched properly, could've avoided it.

 

I've been working for two years on my first game (yes, I'm one of those guys that starts to big.

first half year: understanding Unity, try out a few things, start basic coding. Decide what game to make, start with story / art direction. Model and rig Characters (3).
second half year: coding mechanics, writing shaders.
third half year: Try to get first chapter working (redoing a lot of code, adding features, removing features), also doing all the art for chapter one.
After that it takes around 2 - 3 month for a chapter, first modeling / concept (2 weeks), implementing (white boxing), coding (2-3 weeks), then texturing, shading, rendering, lighting (2-4 weeks), finalizing and testing (rest).
When I'm done with all 5 Chapters, I'm left to do all sound and some polish and more playtesting (adjusting dificulty).

I have some pictures on my twitter, if you want a visual for this.

 

Do you have a thread setup or something like that? I'd love to see your game.

 

doing class diagrams, i like to over engineer every system and make their design fit perfectly the game requirement which is absolute counter productive but it give a huge satisfaction when the code is really well organised.

 

Writing software takes up most of my time, fighting with 3rd party plugins takes a chunk too. Oh and actually playing my own games through development takes a fair chunk of time too.

 

It's still month away, so no, not yet. Will start one later this year. Check my twitter if you want to keep in touch.