In an absolute coding slump!

Newbie & Debugging Questions
#1

Hey guys, I am in a coding slump! All my recent projects have been some form of a platformer or top down shooter but have all ended up just being “engines” of sort that end up getting thrown out and recoded a few days later with a new “smaller” idea in mind. It seems I am always confused about how to design my class architecture and what not.

Through my billions of projects, I have gotten a decent knowledge of designing classes and what not, but I feel like I need to learn more. Also, I decided to take a HUGE step back and code a simple game such as snake or something and I couldn’t even do that!

My game programming confidence is terrible right now and am wondering if you guys have any ideas of what I should do, any links or books to read, and if this has happened to any of you?

#2

[quote=“Swattkidd7,post:1,topic:35087”]
Phew! Glad it’s not just me! :slight_smile:
If you need focus, try doing a 4K gane - it makes you think about what really matters in a game!

#3

Build a game within your skills, eg Snake. Get it working, then refactor the code until it is extremely clean and nice, as much so as possible. The about how it is organized, what other approaches might make more sense, etc. Go down a few different paths, recognize when an approach isn’t really fitting, and try a dfferent one. Then post it here for us to pick apart. It will take many projects to be exposed to many different types of problems. Do this for all your projects, and the quality of the code you write the first time will improve. You can read books, study design patterns, etc, that is good, but you will learn the most through real experience.

#4

My approach is to draw some sketches of what I’m trying to achieve, along with some notes outlining gameplay. Usually I use powerpoint. Then I add some notes on key implementation algorithms. I try to fully define the user interface and then take a Use Case approach based on that interface to discover objects and subsequently classes. I rather like the UML approach as it actually pretty much fits in with what I do anyway.

I have lots of abandoned concepts, but mostly in the powerpoint stage, as I realised they weren’t right before cutting any code. My main problem is sizing my imagination to my available time.

#5

Thanks for the replies guys, glad to see I am not the only one and thanks a lot for the tips, I am definitely going to try and knock out some smaller games and fully complete them so I can just have completed projects, because although I have been programming for a while most of my projects are just “learning tools” which I made to learn a certain technique or something.

#6

Contest are good thing since something has to be done in due time. I have lots of projects and most are stopped. When I came back on them, I allways have new idea and I have to redo most part of them :-. But Java4k contest is really a great thing for this, I must focus on gameplay, simple graphics (I don’t spend week to try to do sprite or 3d models…) and to deliver it on time.

#7

Leaving stuff unfinished is the worst thing possible.
Pretty much like people said. Tighten scope, and Finish It!
Dont worry about speed and grace, or how hard it is sometimes to bash your way through some barriers. Ten years later, there will still be six hour fights with quirks in compilers and hard-to-trace bugs. If your brain just isnt working, walk, do normal stuff, then back to the grind.
Once you start getting comfortable in the programming environment some of the work may even become boring, like doing plumbing, but under pressure of timelimits even simple jobs can become horrible hacks.
Anyway. Take your time, but dedicate it to focusing clearly on tasks at hand. Never multitask, break jobs up into clear objectives, and do them one by one. Doing any more than one at a time will usually confuse the average primate male like me.
Try to find the flow. You’ve found it when youve listened to the same cd 6 times on repeat without noticing, and a sparkling conglomeration of harmoniously working pixels looks at your smile. Ommmm.

#8

Why not do the one hour Petite competition coming up?

http://www.java-gaming.org/index.php/board,66.0.html

There’s no way to even make an engine in an hour. You’ll be forced to focus on a game, and game only. And after only an hour, you’ve got something to look at! Not to forget you’ll be working in parallel with a lot of more experienced JGO dev.

#9

Great advice Karmington, you’re right, i need to just get down to it and finish these things off to at least have that completed project infront of me, no matter how badly it is “hacked” together. Thanks!

Hmm, I never really considered competitions before, they seem fun but never thought I was ready. But you sold me, hopefully if everything goes smoothly this Petite competition will be my first ever competition!

#10

Here are some exercises for practicing writing clean code:

Object Calisthenics - write 1000 lines of code following these rules. That will teach you something on object orientation. Of course in real life one would not use this extreme rules, but I’m guessing that compared to the code that most people write, good OO code is closer to Object Calisthenics than whatever most people are writing today.

Programming Without Ifs - that will teach you to write polymorphic OO code. Again, when you press yourself to the extreme (by not using any ifs), then after that you will better be able to decide that when to use or not use an if. Many times the code can be simplified considerably by avoiding ifs.

In addition to those code quality practices, I recommend learning the TDD programming technique. Here is a tutorial which I have written to help getting started with TDD. The tutorial has been tested by 40+ people on a university course and it appeared to work quite well. If you find something to improve, for example if it’s not clear that what should be done at some point, please leave me a message.

And some good books to read:
Clean Code: A Handbook of Agile Software Craftsmanship
Growing Object-Oriented Software, Guided by Tests

#11

how can a class be less than 50 lines???

that is riduclous, just making an applet takes up almost as much as that :stuck_out_tongue:

#12

So the extra 5 lines or so you have should be hooked into calls into another class. Therefore you know that all the Applet class is doing is making the Applet.

#13

fair enough…

#14

I agree with the poster who mentioned contests. I used to take part in a regular 24 hour game contest and learned tons.