So You Want to be a Game Developer? Tips for Coders

So You Want to be a Game Developer? Tips for Coders

After last week's downer on why NOT to make the video game industry a profession, we get a little more optimistic view this week for those interested in coding.

Read Full Article

Yup going to have to agree here. Trying to be a games developer is like trying to make it in Hollywood. A very few do but most end waiting tables. A more general degree is a better choice because it allows you options outside games.

A few points about going indie. You have to learn marketing, budgeting, project management, a few bits of employment law, contractual law and tax laws. Don't go into indie development and just expect to be coding, you have to know general business skills and you will have to spend significant effort on them. Seriously, whatever you do, don't issue a death threat at Gabe Newell, it will not end well.

Being the next Notch has worse odds than the lottery? How many coders are there since the odds of winning the lottery (such as the powerball) is like 1 in 175 million. I'd have to assume that the odds are significantly better and are furthered by the possibility of making a return on investment even if it isn't in the multi-billion dollar department.

Game developers seem to make an average of $83k as of 2013. That's a pretty respectable salary with even QA Engineers making around $55k.

I'd say it's pretty darn hard work with long hours. So I totally understood what you were saying last week. But pay? The pay is pretty good. I work in software (not gaming software) and in my area the pay is great. Enough to work into old age with despite being lower than the average numbers from that study.

Perhaps there is a significant number of unemployed developers though.

Lightknight:
Game developers seem to make an average of $83k as of 2013. That's a pretty respectable salary with even QA Engineers making around $55k.

Game developers often feel they get paid badly because they compare themselves to their silicon valley counterparts. I used to be a post-doc in science and made basically half what average developers make and then I moved to work at a software company and saw my salary shoot up. It's all a matter of perspective and if you've been privileged your whole life, $83K may not seem like much. On the other hand, Indie devs often make pay equivalent to what a grad student might make.

Lightknight:
Being the next Notch has worse odds than the lottery? How many coders are there since the odds of winning the lottery (such as the powerball) is like 1 in 175 million. I'd have to assume that the odds are significantly better and are furthered by the possibility of making a return on investment even if it isn't in the multi-billion dollar department.

The odds of being the next Notch are better than the lottery, but then again, so are the odds of being the next Taylor Swift. Both fields are crowded with a lot of people who spend everything and come out with little to nothing, for a massive opportunity cost.

However, making an indie game or a serious mod makes total sense as a resume builder or learning tool. I would much rather hire someone who's put their game on Greenlight, even if it was a terrible zombie crafting simulator, than someone with a computer science degree and 2 years working as a file clerk.

Lightknight:
Being the next Notch has worse odds than the lottery? How many coders are there since the odds of winning the lottery (such as the powerball) is like 1 in 175 million. I'd have to assume that the odds are significantly better and are furthered by the possibility of making a return on investment even if it isn't in the multi-billion dollar department.

How many coders there are has little to nothing to do with being the next Notch. It'd be more accurate to look at how many games get released every year by single person developers that go on to make the creator hundreds of millions of dollars, never mind billions, compared to how many games that get released. The answer is basically that in the entire history of making games, Notch did it with pretty much no money and hit it big and basically no one else ever has.

If you like those odds than feel free, but buying lottery tickets with the money you'd develop with might actually have a better chance of getting you that jackpot. Terrible way to earn a living though.

grigjd3:

Lightknight:
Game developers seem to make an average of $83k as of 2013. That's a pretty respectable salary with even QA Engineers making around $55k.

Game developers often feel they get paid badly because they compare themselves to their silicon valley counterparts. I used to be a post-doc in science and made basically half what average developers make and then I moved to work at a software company and saw my salary shoot up. It's all a matter of perspective and if you've been privileged your whole life, $83K may not seem like much. On the other hand, Indie devs often make pay equivalent to what a grad student might make.

That makes good sense. Thanks for the information.

In general, $83K is fantastic for one person to make in the US. The median household (we have to use median because we have a significant number of outliers that skew the data significantly thanks to them making like $72 million per year on average) makes around $50k and 66% of American workers earn less than $41,212. If a household is making $75k or more then they're in the top 33% (as of 2012) of Americans.

But we're talking about just one person making $83k, not even the full household income.

I get that there are people that are privileged but is the average developer really so privileged as to not think that amount is a good wage? I've been to Seattle quite a bit and $83k means a lot there too.

Vivi22:

Lightknight:
Being the next Notch has worse odds than the lottery? How many coders are there since the odds of winning the lottery (such as the powerball) is like 1 in 175 million. I'd have to assume that the odds are significantly better and are furthered by the possibility of making a return on investment even if it isn't in the multi-billion dollar department.

How many coders there are has little to nothing to do with being the next Notch. It'd be more accurate to look at how many games get released every year by single person developers that go on to make the creator hundreds of millions of dollars, never mind billions, compared to how many games that get released. The answer is basically that in the entire history of making games, Notch did it with pretty much no money and hit it big and basically no one else ever has.

If you like those odds than feel free, but buying lottery tickets with the money you'd develop with might actually have a better chance of getting you that jackpot. Terrible way to earn a living though.

Ok, and the odds are still better than the lottery by a significant portion. Do you realize just how unlikely winning the lottery is?

We're talking about two very unlikely things but one (winning the lottery) is astronomically more unlikely than the other (making a wildly successful game).

Robyrt:

Lightknight:
Being the next Notch has worse odds than the lottery? How many coders are there since the odds of winning the lottery (such as the powerball) is like 1 in 175 million. I'd have to assume that the odds are significantly better and are furthered by the possibility of making a return on investment even if it isn't in the multi-billion dollar department.

The odds of being the next Notch are better than the lottery, but then again, so are the odds of being the next Taylor Swift. Both fields are crowded with a lot of people who spend everything and come out with little to nothing, for a massive opportunity cost.

Sure, I don't disagree. But we're talking about the difference of dying from a snake bite in the US (1 in 50 million) compared to dying in a train crash in the US (1 in 156,169). Both are unlikely but one is FAR more unlikely.

Shamus Young:
So You Want to be a Game Developer? Tips for Coders

After last week's downer on why NOT to make the video game industry a profession, we get a little more optimistic view this week for those interested in coding.

Read Full Article

Once your game is on greenlight, can we expect a feature?

Also, I have an idea for a game. It is seriously amazing. I also already have a well-paying job as a biostatistician, but little to no formal coding training. Where do I start? What environment do learn? What tools do I need? What platform do I target? Where does one start on all the wikis, tutorials and blogs? These are questions I have, but haven't dared ask yet, probably for fear of finding out how deep the rabbit hole goes.

Lightknight:

I get that there are people that are privileged but is the average developer really so privileged as to not think that amount is a good wage? I've been to Seattle quite a bit and $83k means a lot there too.

So, clearly, the set of assumptions I am about to list don't always apply, but consider the following:

1. Game developers likely have had access to a computer and video games for most if not all of their lives.

2. Most of them have a college education.

3. Most are people who are willing to get paid less to work on something they believe they love.

These are all indications of someone with a privileged background. If your family had little money, you likely could not have afforded a Nintendo or had a desktop to code on as a kid. College costs a lot of money and being able to go to college is largely connected to how well off your parents are. Lastly, people who are used to scraping by rarely take a job "because its their passion" but far more often because "it pays the bills". Granted, these are gross generalizations, but for the most part, these are all indicators of being from at least an educated upper middle class family. I'm not griping. I came from an educated upper middle class family myself and I appreciate the privileges that has bought me. So my entirely non-scientific set of assumptions convinces me that game developers are likely to be fairly well off to begin with.

Honestly, I don't know how good picking up coding without instruction really is. Right now I'm getting my CS degree, and from what I can tell most people have picked up a ton of terrible habits from their teen years, while my total lack of experience really hasn't been too much of an issue comparatively. Maybe developing some familiarity with algorithms and the like with something like Python might not come back to bite you, but getting set in your ways without an instructor is a huge problem with some people, and leads to some really bad code sometimes.

So when your game is coming out?

PH3NOmenon:

Also, I have an idea for a game. It is seriously amazing. I also already have a well-paying job as a biostatistician, but little to no formal coding training. Where do I start? What environment do learn? What tools do I need? What platform do I target? Where does one start on all the wikis, tutorials and blogs? These are questions I have, but haven't dared ask yet, probably for fear of finding out how deep the rabbit hole goes.

The first step is simply learning how to program, in general. You could try to jump right into game dev, but I'm pretty sure that's trying to run before you know how to crawl. (At least, it is if you intend to do something physics based, or 3D; something text based might be more manageable)

What programming language you start with is a bit of a hotly debated topic. Python (and to a lesser extent Ruby) is a commonly recommended language, since it's powerful but fairly approachable. Other people take more of a "vegetables before desert" sort of approach and recommend something more in the vein of C++/C#/Java - in particular some consider C++ because of its more complex memory management. I'm a particular heretic in that I consider web development, primarily Javascript, to be a viable first choice. Advantages of JS/Web programming are that you already have all the tools you need (i.e. a web browser and a text editor... though you'll probably want to checkout a nicer text editor than the standard OS one), that HTML/CSS is the most widespread framework for graphical programming ever, and it's an built-in distribution platform as well. (Downsides are that JS has its quirks and that browser compatibility is famously a nightmare.. though someone learning doesn't need to be concerned with getting full browser compatibility)

Ultimately language choice is pretty arbitrary. The essential skills of a programmer translate across languages quite well, and those are the important parts. And if you get serious into programming, there's a good chance you'll end up taking closer to an "all of the above" approach, anyways. (I'm a relatively new programmer, a year or two out of college, and I've done fairly extensive work in every language I've mentioned so far except Ruby) I learned on an esoteric language for 2D games (making my "don't start with game development" advice somewhat hypocritical, but I don't think my route is necessarily the best) that I've virtually never heard anyone mention, much less recommend, and I don't think that's really hampered me as a programmer any, so the choice really is pretty much up to your preference/random selection.

For resources, http://www.codecademy.com/ is a good resource for learning the basis of Python, Ruby, or Javascript. Or if you prefer something more textual and you're doing the JS/web dev route, http://eloquentjavascript.net/ is a good ebook on Javascript.

Tips for coders: unlike other people in the septic games industry, you can choose to work somewhere nicer.

grigjd3:

Lightknight:

I get that there are people that are privileged but is the average developer really so privileged as to not think that amount is a good wage? I've been to Seattle quite a bit and $83k means a lot there too.

So, clearly, the set of assumptions I am about to list don't always apply, but consider the following:

1. Game developers likely have had access to a computer and video games for most if not all of their lives.

2. Most of them have a college education.

3. Most are people who are willing to get paid less to work on something they believe they love.

These are all indications of someone with a privileged background. If your family had little money, you likely could not have afforded a Nintendo or had a desktop to code on as a kid. College costs a lot of money and being able to go to college is largely connected to how well off your parents are. Lastly, people who are used to scraping by rarely take a job "because its their passion" but far more often because "it pays the bills". Granted, these are gross generalizations, but for the most part, these are all indicators of being from at least an educated upper middle class family. I'm not griping. I came from an educated upper middle class family myself and I appreciate the privileges that has bought me. So my entirely non-scientific set of assumptions convinces me that game developers are likely to be fairly well off to begin with.

Hmm, I don't know. My family was pretty low income but we had gaming consoles dating back to the original PONG. We were late to the computer arena but getting a poor grade on a paper about Russia since our encyclopedia was from 1960 prompted the necessity for the internet (The teacher relented once she learned we didn't have a computer and I'd clearly put a ton of effort into the paper including burning a skyline of St. Basil's cathedral into wood).

But still, that's gaming consoles my whole life and a computer with internet through high school. We had an old green screen computer before that too but it simply never occurred to me that there was a world of coding beneath it.

That was in a family with a single income in a non-managerial government job. My parents made some sacrifices for us to have them but gaming isn't that steep of a cost. A couple hundred bucks to keep your kids quiet for months? DONE

Me going to college had everything to do with me earning scholarships in school by working hard. My parents didn't and couldn't pay money for my college. For everything else there's student loan debt like most Americans that attend college.

As for getting paid less for something you want to do. This isn't necessarily to do with privilege. In fact, if you are used to a more lavish lifestyle then you're even less likely to work for less money. If you were poor then getting paid "less" to some is often a lot of money for you. Once all of your needs are basically taken care of then it's just a matter of evaluating the opportunity cost of different jobs. For example, my current job is 9 minutes from my home, allows casual wear 365 days a year, and is quite stable for me. What is the dollar value of this job compared to a higher paying one? There is something to be said for the non-salary parts of a job and you don't have to be rich to consider them. You just have to not be desperate for higher pay.

In my area a $40k/year job is getting paid VERY well if you manage your finances properly. I don't know if we even offer any jobs for less than that at starting pay. A $75k job or higher would be overkill here and I know a lot of my coworkers remote in to work so I'd expect other development studios also accept remote workers. I couldn't imagine remoting into a big Seattle company for work while living here. Half of my paycheck would basically just go into savings, investments or paying off my house. To put living costs in perspective, we got a 2,800 ft^2 home, 5 bedrooms/3 bath and 6 acres of land with a large private pond on it (dock included) for under $200k. A $80k+ job would be NUTS here.

I simply don't see how any but the most privileged would scoff at an average pay of more than $80k. You're talking about less than 30% of the population that makes that as a full household. A developer making that with a spouse that has another decent job would be in serious upper class territory. Heck, a part time job only raking in $20k would allow them to breach into the upper class. It's extremely cute when six figure income families call themselves "middle class" as if we aren't thinking of lower middle class anytime some just says "middle class". They're not. They're upper class even if they aren't the ultra elite "5 percent" class making millions. I guess people don't realize that upper middle class IS the upper class. The 5% group is even just $150k household income. So two developers would be in that group. 1% is just $250k.

People who think that only 1% of Americans are rich are truly deceived.

Lightknight:

I simply don't see how any but the most privileged would scoff at an average pay of more than $80k. You're talking about less than 30% of the population that makes that as a full household. A developer making that with a spouse that has another decent job would be in serious upper class territory. Heck, a part time job only raking in $20k would allow them to breach into the upper class. It's extremely cute when six figure income families call themselves "middle class" as if we aren't thinking of lower middle class anytime some just says "middle class". They're not. They're upper class even if they aren't the ultra elite "5 percent" class making millions. I guess people don't realize that upper middle class IS the upper class. The 5% group is even just $150k household income. So two developers would be in that group. 1% is just $250k.

People who think that only 1% of Americans are rich are truly deceived.

Again, I am clearly making a gross generalization. I am not accounting for everything - simply pointing out that there is likely a tendency among AAA game developers to have come from fairly well off backgrounds and that if all they do is compare themselves to people who work at google, then 80K may not seem like much.

Lightknight:
In general, $83K is fantastic for one person to make in the US. The median household (we have to use median because we have a significant number of outliers that skew the data significantly thanks to them making like $72 million per year on average) makes around $50k...

Do you realize you're comparing a bottom-cut-off (salaried-only) mean to a median?

Pyrian:

Lightknight:
In general, $83K is fantastic for one person to make in the US. The median household (we have to use median because we have a significant number of outliers that skew the data significantly thanks to them making like $72 million per year on average) makes around $50k...

Do you realize you're comparing a bottom-cut-off (salaried-only) mean to a median?

The data in the developers' salaries was more symmetrically distributed. You're supposed to use mean when it's symmetric and median when it's not (like overall salary in the US). As long as you are applying the correct measure to the subject at hand then comparison should be permissible even though the measurement differs since the goal is to get the real "average" as we mean when discussing it.

Is there something I'm missing? The median for application developers is actually around $92.6k in the same year (2013) that I'm discussing. They both suit the purposes of my point. Median would actually be a bit better. But due to the symmetry of the results I thought mean would be a more honest comparison in the developers field even if it was $8k less suited for my point.

Both values are well over the mean and median household incomes. This is a single income that any way you look at it, it far exceeds what we can typically expect a couple of employed people to be making together when looking at it both by mean and median.

Lightknight:
The data in the developers' salaries was more symmetrically distributed.

More symmetric than the overall U.S. curve? Not exactly an impressive feat. The data is not symmetric in any absolute sense (even a casual glance at the breakdowns confirms this impression), and they actually cap salaries at $200,000 to prevent the big earners from distorting the data too much - which really begs the question of why they're using a Mean at all. As far as I'm concerned, that 83K doesn't really tell us much of anything.

Lightknight:
You're supposed to use mean when it's symmetric and median when it's not (like overall salary in the US).

In truly symmetric data, the mean and the median will be the same. I see little reason to not use the median in wage data, and I'm truly curious why they did not, especially given the lengths they went to, to compensate. Why do they exclude hourly employees? Why is there absolutely no mention of hours worked for that money?

Lightknight:
Is there something I'm missing?

Sure. Look at the data, and in particular look at the non-lead artists, designers, and QA (the bulk of personnel for typical studios) in the middle experience categories. 38K, 55K, 65K. I'll bet 55K is a much closer approximation of the median than 83K. And I'll bet very, very few of these people meaningfully telecommute (that's more of a high-end consultant gig); instead these studios tend to be in major cities, where you've either got a long commute or a high cost of living. And none of that is even remotely trying to account for the fact that they're not including hourly employees at all (QA temps, anyone?).

Living on 83K on a country estate is just not what this life is like.

Pyrian:

Lightknight:
The data in the developers' salaries was more symmetrically distributed.

More symmetric than the overall U.S. curve? Not exactly an impressive feat. The data is not symmetric in any absolute sense (even a casual glance at the breakdowns confirms this impression), and they actually cap salaries at $200,000 to prevent the big earners from distorting the data too much - which really begs the question of why they're using a Mean at all. As far as I'm concerned, that 83K doesn't really tell us much of anything.

Wait, so you think that because they cap salaries at $200,000 and ergo eliminate the outliers that you don't think using the mean value is more appropriate than the median value here? Howso? Doesn't eliminating outliers make the data significantly more symmetric?

In truly symmetric data, the mean and the median will be the same. I see little reason to not use the median in wage data, and I'm truly curious why they did not, especially given the lengths they went to, to compensate. Why do they exclude hourly employees? Why is there absolutely no mention of hours worked for that money?

Huh? Their compensation is to make mean results more accurate. They specifically went through the extra effort so that the mean makes more sense. I think you're getting them backwards. It's the one where outliers are significant and still present where using the mean doesn't make sense.

Lightknight:
Is there something I'm missing?

Sure. Look at the data, and in particular look at the non-lead artists, designers, and QA (the bulk of personnel for typical studios) in the middle experience categories. 38K, 55K, 65K. I'll bet 55K is a much closer approximation of the median than 83K. And I'll bet very, very few of these people meaningfully telecommute (that's more of a high-end consultant gig); instead these studios tend to be in major cities, where you've either got a long commute or a high cost of living. And none of that is even remotely trying to account for the fact that they're not including hourly employees at all (QA temps, anyone?).

Living on 83K on a country estate is just not what this life is like.

You're just kinda cherry picking your numbers there. 38k? That's from a QA engineer with the lowest amount of experience and not a lead. Yeah, they don't make much money, that's not a secret as they are the easiest position to replace and a lot of QA's are mostly devs in training. I know that, I was one. QA's also aren't developers. As for the artists/animators the moment you get into tech animation it shoots right up there with the other developers and programmers.

55k and 65k? 50k is the lowest average for non-leads in any other area. Where you magicked the average being "55k" is beyond me when the programmers and engineers start at $71k in the lowest experience/non-lead average.

The majority of those surveyed were in the 3-6 years or 6+ years of experience categories.

Your averaging is way off and I would imagine is purely colored by your own assumptions. But you can see the approximate proportion of experience categories below each section as well as the average of each category. Assuming $55k doesn't match any of the data you linked.

There are certain roles where you can make something closer to the 50 range but those also shoot up in your 6th year to the $82k mark. So there's a lot better starting to ending salaries in those positions than average work.

And once again, these are all salaries that are higher than most household incomes. These are nearly all good paying jobs unless you're a non-lead QA engineer.

I'm also unsure why a lot of these devs wouldn't be able to telecommute. If my much smaller company allows it then why not others? I've known other studios to do it too like Double Fine.

But sure, if we only take the <3 years of experience positions and include non-development jobs in the conversation we're having about developers then I guess we can start getting closer to your $55k average. Not bad ground-level pay.

Here's a fun fact:

The Labor Department reports that software developers made a median salary of $92,660 in 2013. The highest-paid 10 percent in the profession earned $143,540 in 2013, while the lowest-paid earned $55,770.

http://money.usnews.com/careers/best-jobs/software-developer/salary

It's not specific to "games" but I'm afraid you've got very little evidence on your side here. Look, I specialized in statistics in college. I'm willing to believe you if you just present legitimate evidence that's not anecdotal estimation.

Well done, Shamus. Good advice for game coders, probably for many that want into games.

I only do software, interested in games but not enough in game coding itself. I don't want to remove my escape by turning games into work.

For coders some additional comments: Learn more than just software, in particular more than the software buzz word of the month. Learn how the OS works (Linux has source code), learn how a device driver works, learn assembler, see how the compiler translates some silly line into a mountain of shit - then figure a way around it. Each of those steps will make you a stronger coder, because you will know that the device driver cannot do that, the OS will prevent this crazy idea and you will know that the code is slow because of some heuristic the compiler made that you didn't know about.

For the things besides software, read Shamus' articles damn it! Learn how the art pipe line works, why it works the way it does. Read John Carmack talking about the mistakes in Rage (hint art pipeline) and porting the older games and Rage to iOS.

Find out what it takes to get a business loan, and what paying it back actually costs. Find out whether a publisher is a good one or even a good idea. Read the contract closely, take it to a lawyer.

Learn.

I've been a hobbyist game dev for nearly 3 years now and have yet to release a game. And shamus is totally right, your grand game idea probably sucks once you start implementing it.

Here is a screenshot from my most "professionally" made game: Valance (the lasers in that game are awesome if I do say so myself)

image

That said I don't regret the time I spent on learning dev. It makes me much more appreciative of games that work well as a player.

And I haven't given up my dream yet. Just toned it down to a more realistic level.

My games (such as they are) can be played from the links on this page http://www.twitch.tv/acp_twitch

I'm happy to answer questions over PM if anyone has any

I don't want to be a game developer, mainly because I'm a desolate tundra when it comes to training and talent in that area, but a little because In a way I am already.

The scale is much smaller of course, running table top games for my friends.
Almost a quarter of a century running various systems with a few small breaks.
The method and medium are completely different, but the end goal of providing an entertaining experience to a vocal and critical audience is the same.

All I can promise you brave, brave souls who take the path towards being a game developer is my undying gratitude.
I'll try to keep a lid on the "obsessively watchful harsh critical eye" that sometimes creeps along with the grateful part.

 

Reply to Thread

Log in or Register to Comment
Have an account? Login below:
With Facebook:Login With Facebook
or
Username:  
Password:  
  
Not registered? To sign up for an account with The Escapist:
Register With Facebook
Register With Facebook
or
Register for a free account here