OS WARS MEGA THREAD (Now debating proprietary vs. open-source!)

Ummm...

I hate to throw cold water on things, but game development is actually pretty tricky, especially if you're new to developing games, and super especially if you dont know how to program. Its definitely a great experience, but I wouldnt describe it as easy...

On the other hand, creating mods for existing games is often a lot easier, definitely would be my first suggestion for an up & coming game developer.

But hey, that's just me. Its not that people shouldn't try, just don't expect to be rivaling commercial games on your first try...

Ummmmmmmmmmmm...... no?

Nowadays you don't need to know a line of code to create a game, it just isn't an absolute requirement anymore. Obviously this depends on your toolkit and you have more options available to you if you can code. Game Maker and Construct 2 are examples of engines you can use without needing to program. As for the original context of using game development as a way to learn a bit of programming, GM and Construct etc have scripting too, which is an easy way to learn a bit of code without having to have most of your game reliant on it. Game dev gives you something more interesting to work on that C++ commandline calculator boring tutorial project kind of stuff. My first exposure to programming was robotics which was fun, then I wanted to continue and did some boring C++ projects and totally lost interest. Took me years to touch code again and I did so primarily through game dev.

I never implied you can create a AAA game on your first go. You probably wouldn't regardless of programming experience. The hardest part of game development is game design. Eventually you'll smack your keyboard enough to make the code come out right, but if your game's design is bad, it will still be a bad game. Game jams are neat in that they serve as an excuse to finish something, not about creating a monolithic AAA title. Quick iteration, yo. Like I said, game development has a low barrier to entry, which is absolutely true, I said nothing about the skill required to create a fantastic game. Though, I have played some really fun games from Ludum Dare and other jams that were made by non-coders.

I'm sure some of it is nostalgia, but still, there are a lot of older games released well before my time that I find terrifically entertaining. Say, NHL94 and Rollercoaster Tycoon?
Again, not saying at all that games weren't good before, it is just wrong to say that because of graphical fidelity games are now all worse. RCT2 is tremendous.
 
Last edited:
Capability isn't everything. You also have to consider ease of use, driver support, etc.

And infrastructure. Especially the XNA stuff really had a huge effect on the game industry (and a negative one on the Quality of the games, the only good one is Space Engineers)

Also, DirectX is more suited to gaming, while OpenGL is more suited for CAD. Typical CAD Tasks are much simpler in OpenGL than in DirectX, while it takes a lot more code making game engines in OpenGL, that run well on different Computers.
 
A pretty good rundown of the history on this is on stack exchange: http://programmers.stackexchange.com/a/88055
Nice read, thanks.

My favorite part:
Microsoft had decided that they weren't going to be late again. So instead of looking at what NVIDIA was doing and then copying it after the fact, they took the astonishing position of going to them and talking to them. And then they fell in love and had a little console together.
A messy divorce ensued later. But that's for another time.
At least the subsequent children turned out okay...
 
Ummmmmmmmmmmm...... no?

Nowadays you don't need to know a line of code to create a game, it just isn't an absolute requirement anymore. Obviously this depends on your toolkit and you have more options available to you if you can code. Game Maker and Construct 2 are examples of engines you can use without needing to program. As for the original context of using game development as a way to learn a bit of programming, GM and Construct etc have scripting too, which is an easy way to learn a bit of code without having to have most of your game reliant on it. Game dev gives you something more interesting to work on that C++ commandline calculator boring tutorial project kind of stuff. My first exposure to programming was robotics which was fun, then I wanted to continue and did some boring C++ projects and totally lost interest. Took me years to touch code again and I did so primarily through game dev.

That was my exact point, creating a game in an external engine is basically a mod, as opposed to creating an engine from scratch. Most peoples conception of creating a game will often clash with working inside a preexisting engine, due to wanting feature a, b, or c. Usually the only way around that is to write an engine specifically for that type of game, and writing a proper framework is very difficult. :rolleyes:
 
That was my exact point, creating a game in an external engine is basically a mod, as opposed to creating an engine from scratch. Most peoples conception of creating a game will often clash with working inside a preexisting engine, due to wanting feature a, b, or c. Usually the only way around that is to write an engine specifically for that type of game, and writing a proper framework is very difficult. :rolleyes:
So game dev is hard, but game dev with an engine is a mod, and a mod is easy so game dev is easy, but game dev is hard. Got it. Simple.

Also that is like completely wrong. I can't think of a popular engine that doesn't have a scripting or plugin system. It would be hard to run into something your engine couldn't do that would warrant a new engine. There is almost no reason someone getting into development would write an engine from scratch unless you are trying to learn. Nearly any game you encounter today would be a "mod" under your definition. So yeah, I guess game dev is hard if you are going out of your way to make it so, but [spoiler alert] it is not.
 
So game dev is hard, but game dev with an engine is a mod, and a mod is easy so game dev is easy, but game dev is hard. Got it. Simple.

Also that is like completely wrong. I can't think of a popular engine that doesn't have a scripting or plugin system. It would be hard to run into something your engine couldn't do that would warrant a new engine. There is almost no reason someone getting into development would write an engine from scratch unless you are trying to learn. Nearly any game you encounter today would be a "mod" under your definition. So yeah, I guess game dev is hard if you are going out of your way to make it so, but [spoiler alert] it is not.

No that isnt what I meant. I was saying that its much easier for beginners to create projects working off of an existing engine or game, a mod, or essentially one.

Really...
 
To Windows XP....
picture.php
 
Linux is generally considered to be a much better development platform,

By whom? I've written code on both Windows and Linux (and I even wrote code on Windows with Linux toolchain). In terms of productivity, Visual Studio beats any IDE Linux has to offer. Automatic highlighting syntax errors in editor, before even compiling, is a huge time saver compared to digging through several screens of GCC errors just to learn that you have forgotten a semicolon after a class definition.

And don't even get me started on the autoconf/automake/libtool insanity.
 
I agree, Skipping Vista and wait for Windows 7 was a good choice by me so far, while I hated the Vista on my brothers PC, Windows 7 on my parents notebook works like charm and has some useful features.

I do not know why people hates both Vista. I have Vista since 2007 and still works perfectly (okay, maybe have some annoying bugs, but few yet). With Windows 7 I have had and still have very annoying errors, and some of which I do not find logical explanations. Still love the Aero interface and know well Luna (XP) and I also like. I particularly hate Modern UI, Metro or whatever. What I like about Windows 7 is the Windows Media Center. I know what brings you Vista, but I pre-installed Home Basic.

ZDnet reports:
Ubuntu is particularly popular in poor countries where most copies of Windows are pirated (not everyone can shell out several hundred dollars for an OS).

I agree with that, but in my country, a person can find Windows pirated everywhere, in government offices, schools, shops of good reputation, is very difficult to find the original Windows (in fact, I've only seen a Genuine Windows box once in a lifetime). In my country probably not pirate Linux just because it's free, because if they could they would.

I have read several pages of this ultra-thread and I think I approve with the:
cvjnuem7ll0g4b76g.jpg

:rofl:
 
By whom? I've written code on both Windows and Linux (and I even wrote code on Windows with Linux toolchain). In terms of productivity, Visual Studio beats any IDE Linux has to offer. Automatic highlighting syntax errors in editor, before even compiling, is a huge time saver compared to digging through several screens of GCC errors just to learn that you have forgotten a semicolon after a class definition.

And don't even get me started on the autoconf/automake/libtool insanity.

I really really like Visual Studio. But to be fair, for people who are really familiar with the gcc toolchain the power and flexibility offered more than makes up for the lack of a sophisticated IDE. It just depends on personal preference and how much you are accustomed to using the tools you use.
 
I really really like Visual Studio. But to be fair, for people who are really familiar with the gcc toolchain the power and flexibility offered more than makes up for the lack of a sophisticated IDE. It just depends on personal preference and how much you are accustomed to using the tools you use.

You can also do great things with the visual studio toolchain, if you spent equally long for learning its details, as you are expected to learn to just do the basic stuff with GCC.

Still, there is a huge problem with the IDEs under Linux. Even Eclipse has become pretty outdated, if you compared this to Visual Studio or Netbeans (which also can do C++ in Linux, but it works much better with Java)
 
You can also do great things with the visual studio toolchain, if you spent equally long for learning its details, as you are expected to learn to just do the basic stuff with GCC.

gcc itself is quite straightfotward. All switches you need to know are -c, -g, -o, -O, -W, -I,-L, -l, -shared, -m. (That's about the same number as basic VS (cl.exe) options, except that you can set these in GUI.)

On the other hand, the autoconf/automake/libtool toolchain is the true abomination. If you don't believe me, read this article describing how to build a shared library. Read it and weep.
 
Last edited:
gcc itself is quite straightfotward. All switches you need to know are -c, -g, -o, -O, -W, -I,-L, -l, -shared, -m. (That's about the same number as basic VS (cl.exe) options, except that you can set these in GUI.)

On the other hand, the autoconf/automake/libtool toolchain is the true abomination. If you don't believe me, read this article describing how to build a shared library. Read it and weep.

Yes, but that is what I mean with the toolchain - not just the compiler or compiler/linker tool. The whole infrastructure to get from code to product.

Which is VERY complicated - what I do in Java with my customized Ant-Scripts (and which is considered minor magic by coworkers) is just the very basic actions needed in any C makefile for a Linux machine before you can even consider handling different machine configurations (which is a major PITA), let alone including data from additional files to produce the full distribution. And often you also need pre-compiler statements to make everything fit.
 
Last edited:
But, strictly speaking, use of autotools is not mandatory. There are alternatives. And in my experience, GNU make by itself is sufficient as a build system. This can be understood in one afternoon. I usually target at most two platforms (desktop+embedded), so I can live with maintaining two makefiles.

Of course, someone will run into a problem if they attempt to build your code on OpenVMS without autoconf. But the truth is, most code is so full of Linuxisms that it will not build on anything else, autoconf or not...
 
Back
Top