Pretty straight forward, I just was wondering what everyone feels they could use some practice in.
And yeah, you need to always be learning and what-not.
I’m just currious what people feel they are struggling with. Maybe we can feel a little better about it if we see we’re not alone on something.
Coding structure. I try hard to keep code organized properly, to follow specific patterns, etc. But I still tend to end up with a mess. I do a lot more refactoring than I probably should have to…
Marketing and sales. I’m by no means perfect at the stuff on the list, so I’m always looking for ways to improve, but I’m confident enough that I can handle any not-insane task in any of those areas. Actually selling the thing I’ve made, though, is an area I’ve very little direct practice in.
Shaders.
I’m still wrapping my head around them. I’m almost at the point where I can make them do what I want , but even when I can, they don’t often play well with other shaders. (Or work the same on all platforms). They mock me at every turn.
That’s why I bought Shader Forge. I figured at the very least I can simply implement the free shaders people made for UE4.
Ugh. Same here. But like music/audio, I simply gave up on it. 
I could use learning more of everything. There are some things I can’t stand learning, but until I start collaborating with people who can fill in the gaps, I have to learn that stuff too.
Ugh! How did I leave that off the poll?
Yes, right there with you. I know very little about shaders. Like I don’t even know where to start with them.
Yeah, I’ve tried to pick up shaders half a dozen times. They are in this weird area that is not true coding and not true art. But you have to understand both to make them.
Multithreading might have been a good one too. Not to mention spending less time on the forums. 
It may well be a case of “you don’t know what you don’t know”. The more you learn about any topic, the more complex questions there are to ask, and the more gaps in your knowledge you become aware of.
They’re definitely ‘true’ coding. (This does not mean they aren’t also art.) They’re a bit odd to learn because you have to understand the full graphics pipeline, otherwise there are just these huge gaps of “magic” between the bits of code you write, made even bigger by the parts your engine helps you with. The code typically ends up being relatively simple albeit highly mathematical. The challenge comes from understanding the vast number of different areas that come together to make it all work.
The couple of times I tried to just generally “learn shaders” I never got far. The bunch of times I’ve attacked them with a specific purpose in mind I’ve nailed it, and learned a bunch extra along the way. So that’s my advice - learn it through specific exercises. But before that, make sure you really understand the basics of the rendering pipeline (eg: so you have full step-by-step understanding in your head of how you get from “verts + a texture” to “a rasterised image on the screen”).
Count me in on the shaders one as well.
Yea, there is this strange elegance to them. And, at least for me, it seems the learning curve is inverted. Normally, when I start learning something, my comprehension accelerates at each step. With shaders, every step seems to open a door to more complexity. While fascinating, it’s a little frustrating, I don’t like not “being good” at something. It feels like each time I think I’m close to grokking it, the goal shifts.
Our main shader guy has been mentoring me, and that has helped a lot, but I’m just not getting to the point where I can “see” through the pipeline yet. My goal is to “get it” by the end of summer, but its going to be a challenge to get there. I am currently building a shoreline shader which does breaking waves based on colliding geometry. It mostly works, but frankly some parts have been blind dumb luck. Which makes refinements a nightmare, as I am not entirely sure how I got the other parts working. I do need to go back and, as you said, better understand the steps at each stage.
I’m glad to see that I’m not alone on code structure.
Planted about 2 dozen fruit & nut trees so far since my computer shuts down because it can’t keep cool when the ambient temperature hits the mid 70s. So it’s shiver in winter to make a game or watch my computer shut off in in the middle of editing so Unity scene in summer.
Hands down completion. I have a plethora of gameplay prototypes holding the highest hopes of grandeur, then I get struck hard by the idea fairy (See comic below from TheMeatly.com) and the prototype starts collecting layers of dust. Provided I ever finish something, I know I will be in the same boat as @angrypenguin and need to focus on improving my marketing and sales tactics.
@lorenalexm : Yep, that about nails it. Let’s see, currently I have 12 future projects I can think of at the moment that I’d love to do.
I used to have that issue. I’d spend several hours a week working on personal game projects and never have anything to show for it. One day I decided to be disciplined about it, pick one, and finish it. And I did.
Between projects I let myself prototype different things or make half games or experiments I never intend to show off. That’s important, so we should all do that kind of thing. If nothing else it lets us make better informed decisions about what deserves more work and what doesn’t. The key is that after doing that for a little while I pick a project to finish, and when I’ve done that I’m disciplined about putting all of my game dev time into that.
That’s what makes me shake my head at the “idea guys”, or when people ask for others to help them make their game for them. Most of us who are into making games usually have more ideas than they have time for.
