Jump to content
  • Advertisement
Sign in to follow this  
Protocol_0

Array Oriented Programming Tutorial

This topic is 4766 days old which is more than the 365 day threshold we allow for new replies. Please post a new topic.

If you intended to correct an error in the post then please contact us.

Recommended Posts

Advertisement
An interesting approach to the problem indeed. A valuable learning experience for the one who writes it, but I would advise others against using it. I intend no offense, but rather some clarifications, by dealing with the following details:

A simple example (Grouping)
1. When you move your character around, the common solution is not to change the coordinates of all objects in the world (which would require lots of work), but only the coordinates of objects drawn to the screen. To do this, you store the offset, and apply it each frame.
2. Usually, code does not make any assumptions about how many objects there are in the game. Object-oriented programming would make all objects derive from a "Movable" class which has a Move(x,y) function. An iterator would allow you to iterate over all objects in the game (no matter how many there are) and call their Move member function with the arguments (-1,0), by using an adapted functor (named "move" here), resulting (after a short bit of preliminary work) in the code:

std::for_each(objects.begin( ),objects.end( ),move(-1,0));

3. Also, the container for objects is usually not an array, but often a list (so you can remove objects easily when they die). Adding objects to the list is as simple as a call to:

objects.push_back(object( ));

4. Your array-based code deals badly with changes in implementation (such as changing the objects so they don't have an "x" coordinate anymore, but rather a positional matrix), and also does not allow you to remove dead objects from the array easily.

A simple example (Linking)
Good application of arrays, although some of the repetitive work you do would be easily replaced with for loops.

A simple example (Optional Inheritance)
There is no need to code the fact that something doesn't exist (such as a field, for instance): you just don't create that something. Usually, when a field is not provided, it isn't created, so it isn't displayed either. No checking is therefore necessary to determine if a field should be displayed (if it exists, it should be). Even then, the "does not exist" information can be stored as a simple pointer (NULL if it does not exist, or the value otherwise).


Share this post


Link to post
Share on other sites
Thanks for the input.

Do you think that if I used different examples, than this would be a great learning experience for someone? Because with the grouping concept, wasn't it just that I used a wrong example for that method? Like using it for moving the x coord? What if I did it on something different, surely it would be valuable.

And with the optional inheritence, I understand, but again, I think I just rushed for an example. What if I were inheriting another array with usefull data, instead of zeros? What If the array was optionally inheriting two different arrays, but this time instead of one being full of zeros, it also contained inportant information, but only needed at a certain time (optionally)?

Same with the linking, wrong example? Too redundant?

Share this post


Link to post
Share on other sites
Quote:
Original post by Protocol_0
Do you think that if I used different examples, than this would be a great learning experience for someone?

Quote:
wasn't it just that I used a wrong example for that method?

Quote:
What if I did it on something different, surely it would be valuable.

Quote:
but again, I think I just rushed for an example.

Quote:
Same with the linking, wrong example? Too redundant?



Interestingly I did read this thread earlier this morning, looked at the webpage but did not have any time to answer (expecting someone else to do it instead), what I DID think of was that when someone had some negative comments to these methods, the answer would be "Maybe I used a bad example". I did not expect to get that explanations THAT many times. And especially without an instantenious -better- example.

Why is this important do you think? There are very few, if not not zero methods & products out there that can be regarded as good or useful without a good example. That is, there is no point whatsoever with a method/product unless there's a good example of where it could be used.

How many T-Fords do you think Henry Ford would have sold with the arguments
"Buy this machine, you can have it standing there beside you and it makes plenty of noise. Pretty cool huh?".
Instead of: "Use this vehicle to get from point A to point B without having to walk. And when you have arrived, no need to look after it (like horses). And best of all, you can do all this whenever you wish (in contrary to trains) and wherever you wish (contrary to trains).


So, a method is useful only when there's an example that shows it, otherwise it's useless. Don't construct something and then find a place where to use it, but the other way around.

It's good to explore and try different things, often when you do that (explore new things) things will go wrong, which is not a bad thing as it teaches you what works and what does not. :)

Share this post


Link to post
Share on other sites
rofl everyone's so damn pedantic. I'm SORRY I wrote a tutorial, and Caminman, I promise I will NEVER make a car! Dear oh dear...

Now, if you don't mind, I'm going to go play Mario Gold: Toadstool Tour. I hope NONE of you tease the way I walk xD

Share this post


Link to post
Share on other sites
Sign in to follow this  

  • Advertisement
×

Important Information

By using GameDev.net, you agree to our community Guidelines, Terms of Use, and Privacy Policy.

We are the game development community.

Whether you are an indie, hobbyist, AAA developer, or just trying to learn, GameDev.net is the place for you to learn, share, and connect with the games industry. Learn more About Us or sign up!

Sign me up!