Why do games cost so much to develop?

What I want to know is why despite such huge advances in tech A.I is still pretty bad; yet games made many years ago have better A.I.

AI is one of the tougher things to program. Making code act like a human is difficult, and can be extremely costly cycle-wise.

Another issues is that companies don't tend to share with one another. You have one company that comes up with some sweet AI, but they aren't going to share it with other companies. Each company has to make these strides on their own. Yes, devs sometimes move from one company to another, but they can't just take the implementation they came up with at the first company to their new employer. Unless there is a really good open source / previously documented AI code, the devs on a project essentially have to rebuild it for each game, or try and tweak an existing code the company has the rights to in order to have it work for the new game. Reusable code is also something that is difficult to come by. The more reusable you make something, generally the more difficult/expensive it is to implement it the first time.

It is similar to game physics. These concepts involve difficult computations. You need people who really know what they are doing, and that is expensive. It is why game companies resort to things like Physx or Havok instead of spending months/years and a lot of money on an in-house implementation. Most of the in-house physics stuff you see is generally pretty basic, as the algorithms and computations representing natural physics can be difficult to represent in a game; you really need someone who understands both physics and software engineering.
 
One thing you didn't include was the liscening fees for the PS3/xbox 360 development kits and engine lisencing costs if they used a third party engine. Back in the day, you used to have to pay Sony $10,000,000.00* for the 'honor' of 'attempting' to make a game for thier system + a percentage of sales. Even if your game tanked or sold 1 copy, you still had to pay them money up from so they made a lot off of everyyyy ps2 title. Even shovelware from startup companies.

Cool eh? ... ....

Some developers at that time perferred developing for the original xbox as they only changed around $2,000,000.00* in lisencing fees which is obviously less. However, when your paying $50,000,000.00 in employee wages, benefits and other costs. +/- 8 million in liscencing fees isn't that big of a deal if the PS2 has 6x the install base.

Did I mention the white powder frequented by gaming companies's CEOs? Its the magic-sauce that makes management decisions work or so it seems when you look at some of their decisions. I can't help but think $200,000,000.00 on advertising for CoD:MW2 might have been overkill consider that was 1/5th of sales. Only because I think call of duty: mw2 kind of sold itself after Cod: MW1, worth of mouth, a few dashboard advertisments, few tv spots and you'd be good to go for $100,000,000.00. Advertising your game at a cost of $25 mil in a 3 minute e3 spot during the xbox 360 press conference. Maybe not such a good use for 1/8th of the advertising budget. Why? b/c $25 million dollars of tv advertisement would be a lot more exposure to an audience that *shocking* *wouldn't already be buying the game* unlike the majority of the E3 watchers whom probably already had plans to purchase.

If they had spent that $25 million dollar e3 advertising splurge on game development, they would have increased their development budget by 50% and who knows how much better MW2 could have been and how many sales that would have translated into? Even if it just meant more content like a second disc with 10 more hours of campaign/story/whatnot. People would feel like they were getting more value.

*I'm not sure the exact figures but it was extremely high and xbox original was substantially less to encourage developers to choose their platform. In the range of 80% less.
 
Last edited:
Honest question, why? I dont get it... movies ok I can see why they cost a lot, you have sets and huge amounts of support staff to pay, hiring and buying of equipment and cameras.

How does a game like Gran Turismo 5 cost $80 million? Where does the money go? PD apparently have ~110 employees, and Kaz said 150 people worked on the project. Over the course of 5 years of dev time, assuming they're getting paid $50k each (which would be a lot for sitting infront of a PC designing a game, I'd think), that's $37.5 million in wages. A million bucks would buy you enough high end computers for them to work on (workstation graphics cards, high end CPUs, etc). A couple of million would buy you volume copies of all the graphics modelling programs.

Where did the other $40 million go? Or am I underestimating some of the costs?

I just used GT5 as an example, but there a few games in the $40-100 million range.

50k is a VERY low salary for software engineers. VERY low.
 
No its not. thats fucking high for a software engineer. thats senior level salary over here in the uk. 30k is the midrange around the uk here.

Keywords "in the UK". And to ask a possibly stupid question is that Euro or Dollar amounts?
 
Last edited:
Keywords "in the UK". And is that Euro or Dollar amounts?

I'd assume that would be British Pounds, which would equate to rougly $80000 per year. £30000 though is only $48000, which is very low for software developers. Granted, the video game industry pays peanuts compared to other software development positions, but you're still looking at averaging $75000 per year. Partly though, I think that's because the main hubs for software developers are on the west and east coasts in expensive areas, so the salaries need to be higher. I've seen tons of jobs offered in San Fransisco, New York, Dallas, and not so many jobs offered in Jacksonville, Louisville, Cheyenne, etc.
 
If you really wanted to compare salary in a dick size-esque manner, my instincts tell me that you should try looking at Norwegian software development salaries. I got an acquaintance out in Norway and he's getting a super-low-not-even-people-working-at-Mcdonalds-gets-as-little-as-you-do wage, and he's at roughly 45k USD a year. He works IT at some school and that wage is so low for him that he's thinking about quitting because it's just not worth it for him.
 
I don't believe they do.
EA says that up to 80% of all cost go to advertising.
Personally they spend millions trying to get an explosion that gets ooohhhh and ahhhhhh
10cents on game play and fun
I put the blame right where the blame belongs
consoles
 
Most money is spent buying 9/10 reviews at IGN and the likes

that is the best thing you could invest in when making a game
 
alot of things cost more than you expect.

in drug research a typical drug that makes it to market cost about $800 million in a 7-10yr span. Not all those drugs make it to market because it doesn't work or has safety issues. But those drugs that don't make it still cost money.

The pharmaceutical companies just like the game companies have to make a profit to stay in business.

Also think about games that are canceled and a few million went into them before they were canceled. There's alot of costs that most people don't realize are factored into market prices.
 
Most money is spent buying 9/10 reviews at IGN and the likes

that is the best thing you could invest in when making a game

i don't think it works like that... if one site had an abnormally high review people would catch on, and from what i've seen scores are pretty universal across the board
 
i don't think it works like that... if one site had an abnormally high review people would catch on, and from what i've seen scores are pretty universal across the board

Except that we live in a time when game scores go from 6-10, so it's not like there will be much variation.

Also, they can pay multiple companies to review a specific game better, and many corporations own multiple different sites.
 
Yep, most of their budget goes on marketing the game so don't believe their BS when they say X game cost millions to make because 80% of that is marketing the game which is not part of production costs.
Marketing is their choice and not a necessity.
 
Most money is spent buying 9/10 reviews at IGN and the likes

that is the best thing you could invest in when making a game

Not really, because review scores don't really affect purchasing rates much. With games like Black Ops there doesn't need to be a single review out before it sells 2 million units. People buy that shit up regardless.
 
i don't think it works like that... if one site had an abnormally high review people would catch on, and from what i've seen scores are pretty universal across the board

With an advertising budget of $200,000,000, you can buy pretty universal across the board scores with the left-over pocket change.
 
because there's a lot more involved now to create a game. the non labor costs such as royalties and ad space eat up much more than you think. It's like the typical McDOnalds strategy, create a cheap crappy product and blow the entire budget on marketing it.
 
Lol @ OP thinking $50k would be a lot for sitting in front of a computer designing a game. $50k is/was right around average for a first year programmer entering the field. And I'd bet there aren't many 0 year experience guys working at PD.

I love how people think programming is simple. I mean it's just sitting at a computer typing right. Jane SuburbanMom does that all day too at her secretary/typist job (not putting down secretary/typists). Little Joe does almost the same thing searching wikipedia and typing up his report for school, no?

There's way more to it than just what most people see/think. It often is one of the most mentally fatiguing tasks you can perform. Sure it seems simple when you see someone who's already done what he/she is trying to accomplish before. But set a programmer on a new task, which happens more frequently than not, and things can become extremely draining. Try coding things that don't have a reference. Often you have to derive your own algorithms, data structures, etc. for things as they either aren't widely documented or haven't been publicly exposed before.

It's fucking hard, demanding, and stressful work in almost every facet. Yeah, most of us love and strive on that, but that by no means makes $50k a lot for what we do. We are developing all the things people take for granted in this digital world, right along with all the mathematician's, hardware engineer's, etc.

Sorry. I just get frustrated when things come off as though people think our job is just as simple as the kid that watches youtube all day, or the 24/7 forum troll, because we get to stare at a computer all day.

I, for one, thank you for everything you do. Keep up the good work my friend!
 
Really makes me sad to see how some think how easy it is to make a game and our pay level.

I worked in the game industry for a few years and it is far from easy work. Things are very complicated and it takes teams to accomplish things. I lol'd at your 50k estimate, I don't think I know a single guy in the art dept of major studio that would work in the industry for that. You see, the game industry requires a certain set of skills but the pay for the amount of work put in isn't normally all that good. The good thing for the devs are is that those same skills are needed in many of fields where the pay is better as well as the hours. It is becoming unusual for older people outside of a director position to even work in games anymore due to its demands, and with so many kids thinking it is great to go make a game and go to school for it they burn out incredibly fast.

Just because you watched a behind the scenes video from your collection edition for a game doesn't mean you even understand any of what goes on into making a game. 3D scanners to pop out a perfect game asset, companies don't give you their precious files of their items, you just don't take a picture and boom it is a texture. There is tons of man hours put into that.
 
What I want to know is why despite such huge advances in tech A.I is still pretty bad; yet games made many years ago have better A.I.

There is AI that learns, or tries to, and there is AI that is just pre-programmed behavior. I don't know of any games that implement actual AI that learns.

One is significantly harder to implement than the other.
 
you guys make it sound like 50K a year is mcdonalds pay. Most jobs start at 40-45K range anyway, Mcdonalds pay is 25K or less a year.
 
because game programmers are not like "most jobs". Software engineering in general is often listed as one of the top 10, some times even top 5 most skill demanding professions.
 
AI is one of the tougher things to program. Making code act like a human is difficult, and can be extremely costly cycle-wise.

Costly? I don't think they put that much money into developing the AI. We already know that the most money goes into marketing. Second is probably pretty graphics. Do you really think that a big budget game even spends 1% of the money on the AI?

I haven't seen much of an improvement in AI since the old DOS days. I think the AI in M.A.X. is still better than most games today. A little more recent is Galactic Civilizations 2. There the AI was developed part time by the companies owner. So in his spare time after running the company, he managed to make an AI that is better than in most games. You'd think big budget game could do better, but they haven't. So I'm guessing it's not a priority for most games, and a very small percentage of the money gets spent on it.

Making an human like AI might be expensive, but no games ever made one.
 
Costly? I don't think they put that much money into developing the AI. We already know that the most money goes into marketing. Second is probably pretty graphics. Do you really think that a big budget game even spends 1% of the money on the AI?.

If you read my original post I wrote "cycle-wise". That has nothing to do with monetary cost. The poster I was responding to was asking about the lack of advances in AI. "Cycle-wise" I was referring to the computational costs involved with writing algorithms that perform AI.

This is a huge topic and I know people who have spent a huge chunk of their lives working on AI. The issue is that in order to make AI in which the agent acts like a human, one that is intelligent, you get into algorithms that can push past the boundaries of what even the most powerful computers are able to compute in a timely manner, nonetheless you're average quad-core rig. Some of these algorithms push into np-complete / AI-complete the more like a human you want the agent to react.

EDIT: Obviously they don't pour a ton of time and money into AI, as they aren't going to make significant advances. You tend to go with what gives a decent appearance of AI, but isn't going to nail a cpu.
 
I haven't seen much of an improvement in AI since the old DOS days. I think the AI in M.A.X. is still better than most games today. A little more recent is Galactic Civilizations 2. There the AI was developed part time by the companies owner. So in his spare time after running the company, he managed to make an AI that is better than in most games. You'd think big budget game could do better, but they haven't. So I'm guessing it's not a priority for most games, and a very small percentage of the money gets spent on it.

Making an human like AI might be expensive, but no games ever made one.

Even with Galactic Civilizations 2, the AI got somewhat broken by the Twilight Of The Arnor expansion. There is hope, though: Frogboy recently posted they're going to do some work on GC2.

Game companies have decided it's easier to just give you real human opponents with multiplayer than try to make an effective A.I. That's a reasonable decision because making an effective A.I. is all but impossible for most games more complex than chess.
 
AI is one of the tougher things to program. Making code act like a human is difficult, and can be extremely costly cycle-wise.

No programmer has accomplished realistic AI yet IMO. That's my biggest complaint about modern games, nice shiny graphics with AI that is no better than games from 1993.
 
because there's a lot more involved now to create a game. the non labor costs such as royalties and ad space eat up much more than you think. It's like the typical McDOnalds strategy, create a cheap crappy product and blow the entire budget on marketing it.

Tell me about it, there is no other explanation for the-worst-burger-known-to-man to be so successful. I can make a better burger blind folded.
 
Back
Top