Ok, so, I want to make games. I have the capability of learning everything (or I'd like to think so), I just have absolutely no direction.

I want to make games, be that writing storylines, scripting, actually programming, engine development; I just want to make games. I read somewhere, that its worth dabbling with modding for a while until you understand what makes a game and how to structure a game/project. Would any of you agree with this? Any further advice on this would be great too (where to begin, tut sites, games to mod, etc)

The ideal is that I will learn c++, then dive into working with opengl/directx until I am comfortable to begin "from scratch" projects. As i said earlier though, I have no idea where to begin. Iv been rereading over c++ primer and getting into the nitty gritty of it and feeling comfortable. Is there anywhere I can go or anything to read that will guide me on some sort of path to learning what I need to?

I understand that I'm not going to make a game over night, I just need some direction on the learning. Thank you in advance.

I just want to make games. I read somewhere, that its worth dabbling with modding for a while until you understand what makes a game and how to structure a game/project.[/quote]
I was the same way and that's why I'm creating some tutorials (see my signature) that just focus on games and not the ins/outs of every little caveat to c++ or DX. C++ primer I think is a giant book about all the caveats of c++. I have a free tutorial in my signature you can check out.

Your opinion of modding is actually wrong. Modding will only teach you 1 thing: how to use a specific game engine tool (UDK/Source/CryEngine) and how to design some levels, and iterate on them to make them better. Modding will not teach you how to do coding, 3D art, how to actually structure a game project. You can't know how to structure a game project until you actually go and write a simple 2D game. Once you do that you can go: "ok I did this once, if I had a team of 5 people and we are doing a 2D game, I think I can structure and designate work".

And when I used to try and make teams with people on Moddb.com, those modders never ever got better at anything or really made even the simplest game. So I would stay away from modding. It is much better on a resume later to have written your own tools in say c++ and know how the game engines work. If you already took ambition to learn c++ then you are on your way to be a c++ engineer. And one more thing, I don't know any engineers that actually modded a game.

This is a career question, so this is moving to Breaking In.
Morzor, please have a look at this forum's FAQs, including the "Is It Worth It" FAQ (#66) and the "Passion" FAQ (#40).
The FAQs link is above.

This is a career question, so this is moving to Breaking In.
Morzor, please have a look at this forum's FAQs, including the "Is It Worth It" FAQ (#66) and the "Passion" FAQ (#40).
The FAQs link is above.

Surely this is a beginners question, am I'm asking about a route of study (although I agree, the title is misleading given the actual queries)

Your opinion of modding is actually wrong. Modding will only teach you 1 thing: how to use a specific game engine tool (UDK/Source/CryEngine) and how to design some levels, and iterate on them to make them better. Modding will not teach you how to do coding, 3D art, how to actually structure a game project. You can't know how to structure a game project until you actually go and write a simple 2D game.

No offense, but that's complete BS.
If modding requires you to write thousands of lines of C++, how could you nor learn coding? If it requires you to make 3D art, how could you not learn how to make 3D art? If you're exposed to the internal structures of a real commercial game project, how could you not learn about how to structure game projects? Actually working on a real, professionally-written code-base is going to teach you much more about game-structures than your own unguided attempts at making your own code-base from scratch.

And when I used to try and make teams with people on Moddb.com, those modders never ever got better at anything or really made even the simplest game. It is much better on a resume later to have written your own tools in say c++ and know how the game engines work. If you already took ambition to learn c++ then you are on your way to be a c++ engineer. And one more thing, I don't know any engineers that actually modded a game.[/quote]You can't say that you know "how the game engines work" unless you're experienced with them, which means working with a game that uses that engine. Modding an existing game is a lot easier than making a game from scratch, so it's a good way to gain experience with an engine.

On my first resume for a games job, my main experience was movein.moddb.com, a HL1 total conversion that a team of newbies built over ~4 years, involving countless lines of C++, and a huge list of new game assets of every type. A bunch of my university classmates applied for the same job with their portfolios containing their home-made "game engines" -- guess who got the job? They guy who could explain the content-creation pipeline of every HL1 asset type and the inner-workings of every system in the HL1 code-base (including AI, rendering, effects, networking, audio, physics, gameplay, UI, persistence, asset compilation, etc), or the guys who can show off their D3D 101 classwork?

[quote name='dpadam450' timestamp='1321834124' post='4886026']Your opinion of modding is actually wrong. Modding will only teach you 1 thing: how to use a specific game engine tool (UDK/Source/CryEngine) and how to design some levels, and iterate on them to make them better. Modding will not teach you how to do coding, 3D art, how to actually structure a game project. You can't know how to structure a game project until you actually go and write a simple 2D game.

No offense, but that's complete BS.
If modding requires you to write thousands of lines of C++, how could you nor learn coding? If it requires you to make 3D art, how could you not learn how to make 3D art? If you're exposed to the internal structures of a real commercial game project, how could you not learn about how to structure game projects? Actually working on a real, professionally-written code-base is going to teach you much more about game-structures than your own unguided attempts at making your own code-base from scratch.

And when I used to try and make teams with people on Moddb.com, those modders never ever got better at anything or really made even the simplest game. It is much better on a resume later to have written your own tools in say c++ and know how the game engines work. If you already took ambition to learn c++ then you are on your way to be a c++ engineer. And one more thing, I don't know any engineers that actually modded a game.[/quote]You can't say that you know "how the game engines work" unless you're experienced with them, which means working with a game that uses that engine. Modding an existing game is a lot easier than making a game from scratch, so it's a good way to gain experience with an engine.

On my first resume for a games job, my main experience was movein.moddb.com, a HL1 total conversion that a team of newbies built over ~4 years, involving countless lines of C++, and a huge list of new game assets of every type. A bunch of my university classmates applied for the same job with their portfolios containing their home-made "game engines" -- guess who got the job? They guy who could explain the content-creation pipeline of every HL1 asset type and the inner-workings of every system in the HL1 code-base (including AI, rendering, effects, networking, audio, physics, gameplay, UI, persistence, asset compilation, etc), or the guys who can show off their D3D 101 classwork?
[/quote]

So what advice could you offer, if any? Thanks in advance.

involving countless lines of C++[/quote]
Well again from what I have seen, if you take out the opinions I posted. I have never met an engineer that said "I worked on a mod." I also see the trend from moddb, that nobody knows how to code, they just build walls and throw texures on them. If you look on the art posted there, its generally pretty low quality. Also most artists will be like "Why would you want to code anything just use UDK". So again when you are talking about 3D art, generally those people don't learn anything about coding or game engines, they just say "UDK does everything"----> from what I have seen. If you show someone a mod, then how do I know what you already did? All the rendering engine is done for you, most if not all of the physics. So when someone says here is my engine and game, then I know they did everything and I can say ok they know networking and rendering etc.


Maybe you actually coded and made a complete mod, and sure you can code in your mod, but again back to opinions: I don't feel that learning how to mod is anything other than a side step. He might learn some things, but is it faster, or is it going to accelerate him when he makes his own stuff? His question is what should I do and is modding worth it. My answer: I have never met someone who modded, other than people that are level designers. Sounds like he wants to engineer though. His other question was would it help in his understanding game project management: to which I said not really, the entire engine/mod tool manages everything.

Really it comes down to whatever you want. How old are you? Are you jumping into CS is college in a year or two? Maybe you want to get a head start in c++ etc. Or learn some coding and work with some mod tools/level editors at the same time.

You can't say that you know "how the game engines work" unless you're experienced with them,[/quote]
So again to Morzor, I never modded, I learned everything on my own 3D art and how game engines work. If someone can learn from modding, I just haven't met or seen many mod projects that were impressive or with people that actually coded inside of them. Just do what you want. You already sound like you want to learn c++ and then dive into learning math and what not to make your own game engine. So you can jump right into that. Thats what I did. One other thing about art as well is that making a mod or game right away your focus is on getting a game out with what you know, than really refining your art skills and increasing them.

They guy who could explain the content-creation pipeline of every HL1 asset type and the inner-workings of every system in the HL1 code-base (including AI, rendering, effects, networking, audio, physics, gameplay, UI, persistence, asset compilation, etc), or the guys who can show off their D3D 101 classwork?[/quote]
Well my expectations are that by the end of college graduation are that if you can't build your own engine and do ALL of that on your own, then you are not addicted to making games. If you just do your classwork, yea you will have a crap 101 level engine. The point you made is that you were more dedicated than them, regardless of what was actually used.You also did prove another important fact: finish a game. The best thing to get you hired is to make a polished final product/demo of a game that is actually fun and not buggy as hell.

So what advice could you offer, if any? Thanks in advance.

As dpadam450 is alluding to, there's many different things that you can call "modding" -- from editing a few textures, to using a level editor to make some missions, to re-writing an entire rendering engine. The thing that makes these activities "mods", is just that they're based on an existing game instead of being done from scratch.

Some mods just add a new coloured monster to a game, other mods take a game like Half-Life and reinvent it as Counter Strike.

In the first extreme (just tweaking content), you're not going to learn much other than how content files are created for that particular game. In the second extreme (making a whole new game), there's really no difference between it and professional development.
So what you get out of it depends on what kind of modding you're doing, and where it lies between these two extremes.


For me, when I was first learning C++, I wasn't very familiar with how to structure projects, or how to write anything bigger than simple text-based games. When I downloaded the C++ source code to Half-Life, it made for a great learning environment for me -- for example, I could take existing weapon class and play around with it's code to see how it works (e.g. making machine guns fire shotgun shells, etc)... From there I'd eventually learn how their whole weapon/inventory system worked, and how it was structured.
Keep repeating this play/exploration for long enough, and I'd read/learnt the whole C++ code-base for a real commercial game. It was as if I had been a part of their programming team, and had worked on the code-base alongside the original programmers - I knew how it all fit together, and which parts I needed to change to make a particular modification to the game. N.B. this was a long process, I played with that code-base for about 5 years in my spare time, while attending high-school/university.

After a few years of "modding" (by which I mean, "writing C++ and making content for the goldsrc engine"), my confidence as a game programmer was through the roof, to the point where I could happily write my own engines from scratch without assistance, and intuitively know how to structure things.

So if you learn well from playing with other people's code, then go ahead and download some real game code and mod the flip out of it. Alternatively, if you learn better from books/tutorials, then stick with those instead.

I want to make games, be that writing storylines, scripting, actually programming, engine development; I just want to make games.

It's always helpful to be as multi-skilled as possible, but if you're interested in a career, then you need to pick one of these and specialize in it more than the others.
IMO, modding lets you be quickly exposed to many different responsibilities (designing levels, writing game code, writing new engine systems), so it might be a good way to find what you like the most.

Regarding the required programming skills:
Designer, Level-designer - often no "scripting" knowledge required (just design knowledge), sometimes something like Lua programming would be required.
Gameplay programmer - Usually a higher-level language, like Lua or UnrealScript is required. Sometimes C++ is also used. Whether or not C++ is used, it's probably assumed that you're proficient with it.
Engine programmer - A high level of proficiency in lower-level languages like C/C++ is required, plus some technical speciality is appreciated, like knowing DirectX/OpenGL, or knowledge of physics engines, streaming systems, audio libraries, etc...

[quote name='Morzor' timestamp='1321838043' post='4886041']So what advice could you offer, if any? Thanks in advance.

As dpadam450 is alluding to, there's many different things that you can call "modding" -- from editing a few textures, to using a level editor to make some missions, to re-writing an entire rendering engine. The thing that makes these activities "mods", is just that they're based on an existing game instead of being done from scratch.

Some mods just add a new coloured monster to a game, other mods take a game like Half-Life and reinvent it as Counter Strike.

In the first extreme (just tweaking content), you're not going to learn much other than how content files are created for that particular game. In the second extreme (making a whole new game), there's really no difference between it and professional development.
So what you get out of it depends on what kind of modding you're doing, and where it lies between these two extremes.


For me, when I was first learning C++, I wasn't very familiar with how to structure projects, or how to write anything bigger than simple text-based games. When I downloaded the C++ source code to Half-Life, it made for a great learning environment for me -- for example, I could take existing weapon class and play around with it's code to see how it works (e.g. making machine guns fire shotgun shells, etc)... From there I'd eventually learn how their whole weapon/inventory system worked, and how it was structured.
Keep repeating this play/exploration for long enough, and I'd read/learnt the whole C++ code-base for a real commercial game. It was as if I had been a part of their programming team, and had worked on the code-base alongside the original programmers - I knew how it all fit together, and which parts I needed to change to make a particular modification to the game. N.B. this was a long process, I played with that code-base for about 5 years in my spare time, while attending high-school/university.

After a few years of "modding" (by which I mean, "writing C++ and making content for the goldsrc engine"), my confidence as a game programmer was through the roof, to the point where I could happily write my own engines from scratch without assistance, and intuitively know how to structure things.

So if you learn well from playing with other people's code, then go ahead and download some real game code and mod the flip out of it. Alternatively, if you learn better from books/tutorials, then stick with those instead.

I want to make games, be that writing storylines, scripting, actually programming, engine development; I just want to make games.

It's always helpful to be as multi-skilled as possible, but if you're interested in a career, then you need to pick one of these and specialize in it more than the others.
IMO, modding lets you be quickly exposed to many different responsibilities (designing levels, writing game code, writing new engine systems), so it might be a good way to find what you like the most.

Regarding the required programming skills:
Designer, Level-designer - often no "scripting" knowledge required (just design knowledge), sometimes something like Lua programming would be required.
Gameplay programmer - Usually a higher-level language, like Lua or UnrealScript is required. Sometimes C++ is also used. Whether or not C++ is used, it's probably assumed that you're proficient with it.
Engine programmer - A high level of proficiency in lower-level languages like C/C++ is required, plus some technical speciality is appreciated, like knowing DirectX/OpenGL, or knowledge of physics engines, streaming systems, audio libraries, etc...
[/quote]

Thanks very much! So what games would you advise looking at, apart from your example of HL1?

I have never met an engineer that said "I worked on a mod."

How many engineers have you met that are really passionate about making games? It may surprise you to know that not many of them really are.
None of my coworkers play games at all, except one creepy guy who lists hentai games as his hobby. I never talk to him.

And did you specifically ask them? No one has ever asked me that question, despite knowing my passion for game development.
I wouldn’t have asked Hodgman had I met him. How do you know?


My life basically began to show its real direction when I started making mods for Starsiege: Tribes. It was an extremely valuable use of my time.
It allowed me to not only practice coding, but also explore the structure of the engine, create and import assets (I learned 3D modeling, building modeling, and map design specifically for my mod projects, skills I still have today), and provide me with an outlet for my designs (I am a designer at heart, even though I actually enjoy programming more).


At GDC a few years ago a prominent lead designer, I believe from Valve, gave a presentation on deciding which game designers to hire.
He showed a bunch of slides describing the traits of a good designer. Some were:
Ignores girlfriend/boyfriend to design a game.
Stays up late to work on his or her designs.
Makes mods.

Guess that means I could get a job at Valve pretty easily.


I wouldn’t list my mod on my resume, but it is an extremely helpful part of one’s own development.
Unless of course you really don’t have passion for game development.


L. Spiro