Project basics

[Beric Veincrusher]

Would those Cisco Networking classes give me the skills to do the networking?

[Thirlan Tyrandor]

Would those Cisco Networking classes give me the skills to do the networking?

I don't know ^_^
Let me go check out what they teach to see.

And the answer is... I don't think so. Seems like they would train you with Cisco tools and not teach you how to develop such things with C++.

By the way Fishbarrel, you should check Angeleon's account to see if he has rights to academics because I've seen him around but he hasn't showed up if you get what I'm hinting at. Once Angeleon shows up we can start figuring things out 8P

[Angeleon Askaroth]

Yay! Access at last!

Okay first things first we need to establish some basics to make sure everyone is on the same page.

1. We've agreed this will be a strategy game and an RTS, so I'm assuming no turn based?

2. The language will be c++ and the program will be written from scratch.

3. Will we use opengl or directx?

4. Windows or Unix?

5. We are going to need a version control system (aka CVS) for something this big or else we're bound to run into issues and rolling back becomes a pain in the ass.

6. We're going to have to agree to a universal programming standard, which would include variable/funtion/class naming conventions.

7. We will have to maintain a loose documentation of the code and classes so that we all know and agree what a certain class does.

8. To manage the project's progress we will also have to establish milestones (not date milestones) so that we know what we're striving for.

1. Correct. Not turn based. Real-Time.

2. Again Correct. However, if you've been assigned a part, it is up to you how much you want to use others parts to finish it.

3. Dosn't matter, and if we feel like it, we won't really have to choose. Basically we make a wrapper class that'll handle all the functions that we want and then we can have both. However, for starters, even if I think the wrapperclass is a smart move to start with, making it for only one is better, and then it don't matter for me.

4. Windows. Since well, it'd be fun if people actually could play it. =) If we want it portable, we should do it in Java.

5. Yeah. I think it might be good idea, but honestly I'm not sure if the overhead is worth it for only two developers. I've used a CVS-thingy before called subversion, but I can't say I've gotten the hang of it. We should probably do it though, in case it will grow later.

6. Definitly. I could type one up today. It won't be anything special and so, but at least we'll have something to start with,

7. Sounds like a good idea.

8. Definitly...

Now to the question about Art. I'll talk to some people I know that might be interested, but any and all contributions in this area is welcome and will be credited ingame.

Any dedicated server will be for unix, since well, windows isn't really reliable enough. =) However, at this point it would probably just consist of a battlenet like thing. Just a place to meet and stuff. But really not neccessay since there exists things like Hamachi.

[Angeleon Askaroth]

A game engine for a RTS isn't that much work for a basic thing. It is all about decomposition and ambition.

Personally, I'd be happy if we could strive for something a little bit more ambitious when it comes to the engine, so that we can have some more advanced commands rather than just click to move and attack.

Then if we can just divide the engine into good classes we'd be able to start working on it.

[Angeleon Askaroth]

Ok, decomposition and plans has been discussed and some actual code has been started. It is merely a shell at this point and a crucial decision has come up: I loathe windows. From the bottom of my heart I detest windows and that's why I am a bit divided. I have dabbled with windows before and its syntax really bothers me. How a OS with that kind of syntax can have been allowed to grow that big I'll never know.

On one hand, I want us to do this completely from scratch, utilizing windows and OpenGL together. But then, on the other hand, which might take the upper hand, I feel that it would be wrong to not utilize the existing wrapperclasses for openGL so that it not only is portable(if not platform-indepentent), but also without most of the windows syntax.

I feel that using a wrapper class such as SDL or GLUT/GLtk could be acceptable. The bonus it adds could easily be greater than the feeling of not having done it ourselves.

Any opinions on this?

[Surly von Fishbarrel]

I'm not the one developing your wrapper classes, so I don't care what you do with it honestly. If it's easier to use pre-made shit, then do it. Unless you feel you're losing a tremendous amount of learning from taking a short cut, then do it.

[Angeleon Askaroth]

Actually, the thing that gets avoided is mostly programming with windows syntax and api and a part of me feels that it is kinda silly to use windows when you can use something that is almost platform-independent.

What I'd like is for someone to tell me is whether windows is the bullshit I think it is nor not. If knowing its syntax like the back of your hand is really important for programming as a career. I can't believe it is but I havn't had a real job yet, so for me it remains to be seen, I guess.

[Surly von Fishbarrel]

I have no idea if that's the case, ask Thirlan.

[Thirlan Tyrandor]

Well finally I'm back on to a normal sleep cycle again...

From what I've seen and dabbled in you do not have to "use" windows. You only need windows to start a windows program and manipulate the windows screen you open for the game and that's about it... but I think it is possible not to even use windows. Opengl is setup very much the same way windows is setup with it's window generating class hence we could very well make it only using OpenGL. You only want windows if you want to do things with the operating system which we would if all the mouse clicks and keyboard pushes weren't handled by OpenGL. On the other hand we do need UDP protocols and sadly... we need to work with windows for that one OR develop a special dual unix/windows class that will compile differently depending on the operating system. Doing this will be both tricky and buggy because you need to know C compile codes and other ancient tech.

As for window's syntax system I agree it looks a bit bizarre but you're only at odds with it because it's not the same as the C syntax. The way you were first exposed to something will always leave a bit of bias but in this case it isn't so underserved

[Angeleon Askaroth]

actually, windows sockets aren't that hard to handle, not even the UDP. It is quite possibly the only thing they've done right, as it is practically the same syntax as in *nix.

I've got a class that handles networking rather well and compiles in both windows and unix, but I can't say that I've used UDP for much. You think it is preferable to use UDP over the reliable one?

Well, then... What do you say about using GLT or GLUT instead of windows for most things? I think we would still have to use windows for threads, but by just adding a #ifdef we could still make it portable. Threads aren't that much work. But apart from that, GLT or GLUT should handle everything else (well, not networking

Right off the bat, I'd say GLT out of the two, since it is only relying on openGL which makes it more portable...