Ask a game designer

[axell]

Hello people,


It's very nice to see so many people interested in making games here, reminds me of myself. But what it also reminds me is how hard it is to get info on how game development actually works and especially to get into game development.

So I thought I could help here. I 've been working full time as a Game/Level designer for 3 years now ( WHERE i work i won't say though).
I started out as a modder as well and over the 3 years i've been working in the industry i've improved a lot, and I still can improve even more.

So if any of you have any questions about making games and the game industry , feel free to ask and I'll happily answer when I get the chance.

 

[Dr Super Good]

Multi threaded engine design is still a bit beyond me at the moment. I know each thread need's its own set of resources to work on and that changes between sets need to be queued but the exact process of that is quite confusing. Depending on the type of threading you either require full scene graph duplication (console recommended thread scheduling system for small jobs) or you use a buffered system where you duplicate assets that one thread will use in read only mode (specific purpose threads, eg thread for game frames and thread for graphics). How are assets loaded into the system is hazy (probably using asynchronous I/O?).

 

[Xonok]

What does a modder need to do to eventually get in the industry and start making actual games?
I know how to perfect myself in the context of this game(WC3), but not how to move on from that.

 

[rover2341]

1) How Long Does the company work on a project start to finsh, From Concept to the point there receiving a decent amount revenue for the project.

6 Months, 1 year or 2 years? I mean I guess it depends on the project but ill leave that up to you.

2) Know your Customer, and getting feedback. How much time is spent into understanding who you customer is, what they do at home, what they enjoy doing ect. Before they take to much work on the project.

Also How do they find this people. Do you have any suggestions for this, for someone in a group of 1-3 people. willing to work on a project part time, on there spare time for 3-8 months on a smaller project.

Thanks in advance, any insight is great.

 

[axell]

Multi threaded engine design is still a bit beyond me at the moment. I know each thread need's its own set of resources to work on and that changes between sets need to be queued but the exact process of that is quite confusing. Depending on the type of threading you either require full scene graph duplication (console recommended thread scheduling system for small jobs) or you use a buffered system where you duplicate assets that one thread will use in read only mode (specific purpose threads, eg thread for game frames and thread for graphics). How are assets loaded into the system is hazy (probably using asynchronous I/O?).

My knowledge of multi threaded engines is kind of basic, but there are quite a few ways you can make an engine :

The first one you mention in which you build a graph with lots of small tasks to see what order they should be executed and run the tasks that are more or less independent in parallel ( you can run the sound mixer at the same time with the physics, or animation etc.


The second example you mention , you don't really have to duplicate the data. You no longer use small independent tasks to run in parallel, you can have interdependent tasks run in parallel the whole game loop and only communicate when needed. So for example the physics needs a rendering update...it just takes and copies the latest rendering update and uses that.
Timing here is very important since a lot of problems could appear if the physics simulation would get the wrong coordinates from the rendering system etc.


And the 3rd option is having separate threads for entire objects and running them in parallel ( example in WC 3 : you can have a thread for your units and a thread for the enemy units. Each has rendering,physics ,sound,animation etc tasks that are ran in parallel for each object). When the objects interact, they use the same latest updates as in the 2nd method.


Well that's basically the extent of my knowledge in the subject. Hope it helps and it wasn't redundant for you.

 

[axell]

What does a modder need to do to eventually get in the industry and start making actual games?
I know how to perfect myself in the context of this game(WC3), but not how to move on from that.

A big step from modding to working in the industry is learning to work with an actual game engine. Modding is a great starting point since you have a ton of assets, pre made scrips and systems, but a game engine doesn't.

And when you start working in gaming you'll soon find out that you have to design those systems and even program them. You won't have ready made assets, you'll have to use basic primitives and only then will the 3D artist start to create those models based on your specifications and the layout you did using the primitives ( unity 3D is a free option to get started, and it has a lot of assets and help available. Unreal engine is also free for non commercial use and very very popular).

Another very big plus is knowing basic programming ( if you want to become a game programmer you need to know a lot more than basic programming, but i'm referring mostly to becoming a designer). Mostly any language will do since a lot of games have their own scripting language and you just need to know the basics to get started.

And if you start working on a project, complete it! When you will be asked for a portfolio, it's much better to present small but finished projects than large,grand but abandoned projects.

Or you could make a very awesome MOD and be offered a job by a company. Gaming companies value modders and will not hesitate to offer the authors or very good MOD's jobs.

And last but not least, you must love games. It's a very important step since working in the games industry involves lots of overtime, lots of last minute changes and you really have to be passionate about games to motivate you.

 

[axell]

1) How Long Does the company work on a project start to finsh, From Concept to the point there receiving a decent amount revenue for the project.

6 Months, 1 year or 2 years? I mean I guess it depends on the project but ill leave that up to you.

2) Know your Customer, and getting feedback. How much time is spent into understanding who you customer is, what they do at home, what they enjoy doing ect. Before they take to much work on the project.

Also How do they find this people. Do you have any suggestions for this, for someone in a group of 1-3 people. willing to work on a project part time, on there spare time for 3-8 months on a smaller project.

Thanks in advance, any insight is great.

Well it really depends on the game, but an AAA game can take 2-3 years to be completed or more.
For iOS/android games (the more modern ones) it takes roughly around a year( for a team of about 20 -30 people).

The target customer is a VERY big factor when designing the game. This is adressed in the preproduction phase : What type of people do i to play the game? Hardcore, casual, time rich, money rich etc? . Trying to make games that appeal to all the markets( casual,hardcore, willing to purchase a full game, willing to pay in a F2P game etc) rarely works and you mostly focus on making the game for 1-2 types of costumers.

Market research is very important. There are lots of ways to see what games are successful, why they work and who they target,what market is already saturated and what market has untapped potential.

Companies use focus groups to see the reaction to the game. From time to time they make a polished version of a part of the game and a group of 20 outside people play it and complete surveys after to see what they liked and what they didn't.

For a small team it's not really an option, so the best course of action is to release an early alpha/demo and see it's reception. This allows you to change your game before it's too late.

As a final tip for you and your team. 3 months part time for everything but the smallest project is not a lot of time. You will probably spend way more time working on the game than you anticipated, but it happens everywhere so don't be discouraged. You will reach points when you'll be discouraged and will lose the will to continue ,but if you really want to make the game do it! Don't be afraid to ask for outside help or take break for a while


And don't expect to make considerable money with your first game. Most don't! If you manage to make an instant hit ,be very happy since it doesn't happen often. If you have an inexperienced team, your first game will likely be not very good, but that's ok. See what you did wrong with it and what people liked or didn't , learn from your mistakes and your next game will most likely be much better and successful!

 

[Drunken_Jackal]

First, you can use the multiquote feature and don't triple post

Questions:

1. What position exactly? Are you a lead or something?
2. How do you exactly make the levels? In-house editor?
3. How did you get your job?
4. Tips?

 

[axell]

First, you can use the multiquote feature and don't triple post

Questions:

1. What position exactly? Are you a lead or something?
2. How do you exactly make the levels? In-house editor?
3. How did you get your job?
4. Tips?

I wanted to answer in just one post but it got too long so i thought i'd respond to everyone.

Back to the questions :

1. Lead Designer but only for about 6 months.
2. In - house editor.
3. I always dreamed of becoming a game designer ever since i made my first starcraft 1 map , but most of the spots required you to have experience or spend a long time in QA first.
In my final years in college , the company wanted to open a new studio and came to my college to search for programmers. I went to the presentation and asked them if i could apply for a game designer.
They told me to send a CV ( mostly gaming related, i listed the mods i've done and my experience in games ) and after that I had a couple of tests ( one required me to make a playable stand alone game in about 2-3 weeks). And 2 weeks later i was hired.

4. The usual don't chew off more than you can bite. Try to finish your projects, if they are too big ,cut features but don't leave them unfinished. Don't be afraid to copy a cool feature from another game if it fits in yours. People prefer a more simple yet polished game, rather than a very original but buggy and unfinished one .
Game designers are the jack of all trades of the industry. You need to know some programming, some 3d/3d art , how to write and present a story. Always keep an eye for aesthetics : make sure every level has a theme. Don't just fill it with assets because they look good, make sure everything fits and has a nice flow.
Try to do as much as you can with as little as possible (reuse assets, zones scrips as much as possible).

There is a lot more, but I think it's enough for now. If you want more specifics on a certain area, feel free to ask.

 

[Tanasren]

Is it a good idea to publish your first indie game in your native language first then translate it if it sells ?

Especially a game which has humor(humour) in it and a refined language ? humour is fuckin hard to translate and harder when you have no money I bet

 

[Magtheridon96]

In the Actor model in which each Actor has one thread with a task list, how exactly would you get threads to communicate with one another? (I'm mainly a C++ programmer, but I know how to read most languages you can throw at me.)