Pure Code

S

Stac

Guest
Hi all. I have 2 questions and they both about code.

1. Desire to rewrite code from zero is disease of all programmers? I want to do it everyday, because I'm think that my solvings and ideas is so stupidly and all project can crash after another line.

2. What does you do, if you can't find any algorithm to code your idea and community can't help you, because your code become so big and there is so many features, which even you by yourself can't fully understand?(lol) Does you continue to search solving in reddit/google/documentation or there is much easy to fully rewrite code and find another way to code your idea, on which you has been stuck?

Thanks in advance. I'm hope I'm not lonely with that problems
 

Xer0botXer0

Senpai
Buildings fall when the foundations are poorly built.

Don't write big when you don't understand the basics.
Take your time.
 
S

Stac

Guest
Buildings fall when the foundations are poorly built.

Don't write big when you don't understand the basics.
Take your time.
How bad is it to hear after 4 months of coding of your "magnum opus" x)
But you're right I guess
 
Last edited by a moderator:

FrostyCat

Redemption Seeker
My advice to you is to get a background in basic computer science. This will teach you how to organize data, measure your performance properly, what the reusable components and algorithms are and how they can be applied. As little as just first-year university CS is enough to be competent.

I spent 7 years bumbling until I found out what having a good footing in computer science could do. 4 months doesn't look that bad now, does it?
 

Xer0botXer0

Senpai
I've spent 3 years trying to make my current game, 3 retries and multiple other projects. I'm finally making progress.

I know my project will take another two years, but I'm okay with that. I don't rush anything.
 

sylvain_l

Member
Personnaly I have too many projects started, that of course never go near to any released state.
But GML is more a hobby, my metrics (if any) is mostly my pleasure. (or perhaps the number of projects started, instead of finished :D)

Just be coherent between your goal(s) and what you do to achieve them.
And even if you "fail", just move on. Change, refine, reschedule or rescale your goal(s) based on what you just learned.
Making it perfectly at first try is often impossible, pass goes trougth try, error, correction, success ;)


1. lots of programmer are perfectionnists, just learn to spot, when it's a good point and when it's getting you down. When to use that trait and when you must let it go. (easy say, hard to master; I'm not specially good at it :oops:)
2. for me, I have no problem to move to a new idea/project. I'm looking for the right spot of Flow (https://en.wikipedia.org/wiki/Flow_(psychology)) where the chanllenge is right for my skills (not to easy or too hard).
 
S

Stac

Guest
I've spent 3 years trying to make my current game, 3 retries and multiple other projects. I'm finally making progress.

I know my project will take another two years, but I'm okay with that. I don't rush anything.
5 Years is very long time. I hope it's worthy of it
 
S

Stac

Guest
My advice to you is to get a background in basic computer science. This will teach you how to organize data, measure your performance properly, what the reusable components and algorithms are and how they can be applied. As little as just first-year university CS is enough to be competent.

I spent 7 years bumbling until I found out what having a good footing in computer science could do. 4 months doesn't look that bad now, does it?
Is there exist any schedule for newbies in GML? Like "read and know by heart documentation(I still didn't, lol) and make 5 different tetris and mario clones"? : D
 

NicoDT

Member
Is there exist any schedule for newbies in GML? Like "read and know by heart documentation(I still didn't, lol) and make 5 different tetris and mario clones"? : D
There´s no magical formula. What works for some, may not work for others.
I started with my current game about 2 years ago (instead of starting with pong, breakout, or other recomended games ), and kept working on it while learning.
A few times I thought that there were some things that could have been done better, so if they were causing performance issues, or could interfere with other codes, I would change/improve them.
Just try to keep everything organized (learn to about finite state machine if you don´t know it), and it will be much easier.
 
S

Stac

Guest
There´s no magical formula. What works for some, may not work for others.
I started with my current game about 2 years ago (instead of starting with pong, breakout, or other recomended games ), and kept working on it while learning.
A few times I thought that there were some things that could have been done better, so if they were causing performance issues, or could interfere with other codes, I would change/improve them.
Just try to keep everything organized (learn to about finite state machine if you don´t know it), and it will be much easier.
Thanks. That's gives inspiration)
BTW, voted for your game in Greenlight. Hope you will pass through it while GL still open
 

JackTurbo

Member
Do you have code in your project that you don't fully understand?

I think a lot of new gm users tend to do a bunch on tutorials that perhaps show you how to code systems without actually teaching them in depth.

If you have anything like this in your project learn how it all works and why it works before adding a bunch of new stuff
 
S

Stac

Guest
Do you have code in your project that you don't fully understand?

I think a lot of new gm users tend to do a bunch on tutorials that perhaps show you how to code systems without actually teaching them in depth.

If you have anything like this in your project learn how it all works and why it works before adding a bunch of new stuff
Maybe I said it wrong. I meant that everyday I'm learning new functions and adding them to code, so my code, in my opinion, looks very dirty. I want to make it Object-Oriented and move it to Scripts to call it from objects.

And there is some problems with searching of algorithm. Because I think that my project on level, where all bugs I should solve by myself (because nobody wants to delve in stranger's code), but I can't do it, because I don't know how to do it.

Recursion?) I think best way is to read list of all functions (I still didn't do) and fully rewrite all code with new knowledges
 

Genetix

Member
Reading the manual definitely is valuable, but there is no timeframe. I've been using GM for over 15 years and am still learning new things. What really matters is how dedicated you are to solving whatever problems you may run into.

A sad truth is that game development is certainly not for everyone - you'll either love it enough to push forward and keep progress... or get frustrated/bored and walk away.

That part is unique to each of us, but I would recommend not getting to hung up on a single project, and just keep building and creating new things.

Good luck.
 
S

Stac

Guest
My advice to you is to get a background in basic computer science. This will teach you how to organize data, measure your performance properly, what the reusable components and algorithms are and how they can be applied. As little as just first-year university CS is enough to be competent.

I spent 7 years bumbling until I found out what having a good footing in computer science could do. 4 months doesn't look that bad now, does it?
Excuse me, universities in my country just gives you "status" and diploma instead knowledges.) Herewith "education" here costs like in top foreign universities.
Sooo, I need another way
 
In the past I have had to gut a lot of bad code from my inexperience coupled with listening to the wrong Youtube tutorials.
 

Coded Games

Member
I also advocate for formal CS education. I'm currently 3 years into a CS Bachelors and I learned more about programming practices in my first 2 quarters of Java at community college than my 5+ years of Youtube tutorials and making games in GML. I find learning in a classroom setting incredibly effective.
 
Look up CS50 from Harvard. Online, completely free and gives you a grounding in basic computer science that will serve you very well for any gaming projects you take on.
 
Top