| |
I'm going to pull my hair out over this one... Displaying 1-8 of 8 total.
1
Technetium
|
After getting nearly 10,000 lines of code done for my FF6-style battlesystem, I have come to recognize flaws in my design. There are a number of "rule-breaking" things I want to do, meaning that I didn't think of them originally when I made the system, and to implement them now requires huge workarounds that aren't pretty, either in code or on screen. These flaws can only be corrected by rewriting about 90% of the battlesystem code. :-(
Those of you that do programming outside of Verge, do you run into this situation often? How do you stay focused? I feel like ditching the whole project right now because of this. x_x
Posted on 2004-08-26 16:45:45
|
Omni
|
...er...if you don't like it, why would you want us to convince you it's good?
You said in code and screen. Have you abstracted the graphics code from the data management? If so, you could just rewrite the data management part, perhaps, though I've got no clue what you're doing.
Of course, if you desperately need to be satisfied with something you're not satisfied with (...), just keep telling yourself how awesome it is, and what it can do that other battle engines can't, and how cool the original idea was. Also, how much cooler you can make it.
If you completely think up new ideas that you can't implement in your current system, then obviously the solution is to make a new system that can. But the question is, what will you do with the old system? You seem to feel kinda bad that you don't see any value in it.
Posted on 2004-08-26 17:00:00
|
rpgking
|
I had a complete Dragon Quest VI-style system done in Verge2 that is atleast 20000 lines of code(probably more) that I ditched to completely rewrite the thing in Verge3... Usually, when you do something a second time, you know exactly what to expect and will generally come out with something much better than before. That's what motivates me to completely restart this...
But...if you feel that your system is good enough already, there's probably no need to rewrite it just for perfection. As long as it works smoothly and has an easy interface, that's all that really matters.
Posted on 2004-08-26 17:34:06
|
Mythril
|
Hmm, I think I did this for the first Intense Flight I made (which started out as a tech demo, and then I patched some excuse of a game on top of it).
Although I didn't have to rewrite so much, the code just didn't look very pretty.
Bah.
Posted on 2004-08-26 20:15:24
|
zonker6666
|
I've run into those kinds of situations time and time again - of course every time i take the hit and make the necessary changes (be it to 10, 40, 80 even 100% of the code) i end up learning something new.
The key is perseverance. You must not allow yourself to trash the project - maybe take a week break from it tho... you may benefit from taking a fresh look at the code.
Posted on 2004-08-26 20:27:28
|
mcgrue
|
Tech. There is nothing wrong with redesigning partway through. In fact, refactoring is essential to creating a solid library that won't destroy your soul if you come back to it after leaving it out of your active brain for a few weeks.
It's a foregone conclusion that unless you are the Seasoned God of Allmighty Design, you will need to fix flaws in your design. Most probably several times. The sooner you accept this as part of the programming process and do not let it deter you, the better off you'll be.
Even the most experienced coders I know have to refactor their large projects to fight against code entropy. It's nothing to be ashamed of. Hell, half the reason I decided to do a very simple v1-like Sully with Zip was the realization that even though I was almost done with one of the major systems, the Sully Complete codebase needed a massive refactor to be publically consumable.
Posted on 2004-08-26 21:32:22
|
Sungam
|
This is exactly why I love OOP so much.
You start a huge project, not really entirely sure of where it's going to end. Then halfway through, you realize that what you're working toward isn't actually what you wanted. Then, as you're opening up that bottle of rat poison, you realize that due to OOP structure, you can still reuse 75% of the code without any modifications. Then you start feeling depressed about biting your arm off earlier.
Posted on 2004-08-26 21:36:04
|
mcgrue
|
While that is one of OOP's great strengths, I cannot enforce how just a bit of functional decomposition can go a long, long way in a procedural language.
Posted on 2004-08-26 21:40:57
|
Displaying 1-8 of 8 total.
1
|
|
