GTA V the Most Expensive Video Game in History :-o
Ridonkulous… and I’m sure it will better be 100x better than Knights of the old Republic $200m fail…
…and they are the first. More and more AAA titles will join them and go down this path as they throw more and more money and development time into their project for the next gen (and GTA V is not even for next gen!). Expect more studios go bust and massive layoffs and this vicious cycle continues.
From the article:
Right before read this post, I was looking at some fb analytics, Candy Crush currently has 50 million DAU (that is unique players, not session).
They just struck me as interesting player numbers. Clearly the two products have virtually nothing in common other than they are games. But, today alone, twice as many people will play Candy Crush than will purchase a copy of GTA over the course of a year.
No point or anything, just interesting numbers.
That makes sense though, right? I mean, Rockstar make money when you purchase a copy of their game, where the people behind Candy Crush make money when people play it. So, one’s “premium” (aimed at people who aren’t particularly price sensitive, goes for a comparatively lower number of high-priced sales) and the other is “freemium” (finds where people are already hanging out and actively shoves it in their faces with the lowest possible barriers to entry).
When you’re playing GTA Rockstar have already made their money from you, so as long as you like it enough to play their next game they’re happy with giving you 10 to 30ish hours of entertainment. When you’re playing Candy Crush, on the other hand, the people behind it make money by keeping you coming back for an extended period.
Also no particular point, just observations about the very different approaches that each is taking.
What is interesting is how the size of games doesn’t scale profit, COD games make more money and are cheaper to make, probably the same with lots of IOS games and smaller games.
For the love of all that’s holy - while I don’t agree with the statement, I really hope you mean TOR, not KOTOR! 
Indeed. I find it kind of fascinating as Candy Crush and GTA really only have two things in common, they are games and they are both very successful. But nearly every other aspect of these two are polar opposites. Demographic, platform, content, goals, etc.
GTA is a massive production, focused, crazy budget, building on, and dependent on a brand they have carefully crafted over the years.
CC was basically a shotgun approach. It was one of a 150, single level, mini-games posted to their site over the course of a decade. Very small games, shared backend and reusable code. minimal time cost/investment, when it gained interest, they added more levels. It kept growing so they built it up and shifted to facebook. The approaches are so radically different, and yet both still widely successful.
That is a great observation!
If you know you can do something which will genuinely amaze people you will get the snowball running for sure and after that nothing can stop the money flying in . But I bet the guy in charge of that circus will sweat bullets no matter what they’ve got when the release date is approaching.
GTA is somewhat of an outlier in this case though and they’ve already made their budget back on pre-orders.
Wrong thread. Ooops.
Almost! I expect them to finish with 4 million preorders. That’s around 240 million. That’s an amazing feat! GTA is super hyped up, lets hope it blows everyone’s expectation out of the water! 
I thought I’d input here that the reason it cost 266 million is because it’s very difficult to upgrade RAGE, so they essentially had to work with a toolset that’s nowhere near as good as say, unity, and do it the hard way. Up to 800 employees were hired during development, to brute force completion.
This isn’t a remark on the quality of the game, just a window into the development processes. I’m pretty sure it will be an excellent game. 266 million seems expensive but also budgets in publishing, marketing and distribution fees (rockstar self publish).
So once those factors are taken into consideration, you’ll see it’s possible to create a comparable game as developer only, with say, a quarter of the budget if your tools are all cutting edge and you have the right people. This again, is just a personal observation and not anything else.
I’ve no doubt they’ll make the money back with ease.
Well… cutting edge tools have to come from somewhere. GTA has cutting edge tools because they built them, and as you remarked, a big chunk of the budget was in engine investment.
I love Unity and certainly in its market and price range it is cutting edge, but it the reality is that out of the box (Unity Pro) isn’t capable of doing a game with the scale and needs of GTA. But that isn’t what it designed for. Sure, with Unity source, it could be tailored to handle those needs, and from there it would be possible. But then you are looking a significant budget for development costs to build it up to support something like GTA V.
You are right, undoubtedly a similar type game could be made significantly cheaper. But GTA is become big IP, and with that comes with expectations and a ton of associated costs… By that nature alone, costs quickly escalate, and as great and successful as GTA IV was, GTA V has to be bigger and more advanced and more cutting edge just as a starting point. GTA I was much cheaper to build because there wasn’t a “GTA Brand” to live up to. Big IP has it’s upsides and downsides for sure. One of the big downsides is the amount of additional people/approvals/meetings involved, that seriously affects costs/budgets. The beauty of a small team/single dev is a streamlined process, and little time wasted on small issues… sigh. That is one aspect I do miss.
It’s pretty sad, a lot of good people will lose their jobs . Ohh well , time to work in normal software and do games on the side
Is it really “nowhere near as good as Unity”, though?
I have no idea, and from what I’ve seen in the past of in-house tools I wouldn’t be surprised if it was someone’s badly designed baby which the company only stuck with due to a combination o f “the sacred cow” and “sunk cost fallacy”… but really, I’m not sure I’d want to make something like GTA in Unity or any other general purpose engine without investing heavily into tools anyway (possibly including investment into source level licensing).
If the $266 million includes publishing, marketing and distribution then I think that someone else being able to do just the development for a quarter of that might be true regardless of the toolset. I can’t remember any figures well enough to quote them, but it’s not at all out of the ordinary for a successful “AAA” game to have more spent on marketing than on development, or for publishing (both physical copies and various platform licensing) to cost a mint in and of itself.
Plus, a lot of that “brute force” stuff would have been content development. And depending on the level of uniqueness of all of that content, that might just be a fact of life when developing that kind of game. (Though yes, certainly the available tools will have an effect on productivity.)
The estimated development cost is around $137 million. The rest is marketing and other.
Very true, if someone else set out to recreate something relatively similar, it could be done much cheaper. At least on the surface. But a few things to take into consideration.
- to reproduce something similar to GTA, you wouldn’t have all the concept, design, testing, iteration and prototyping costs associated with the original title. There would be some, but not anywhere on the same scale. Clones are always cheaper to build since the original figured out all the design.
- huge IP with lots of people involved, comes with a lot of overhead. I would love to know how much of that cost was meetings and approval process. It is mind-blowing how much time (and progress) can be lost when there are many stake-holders involved. A smaller team or new IP is much more efficient when it comes to making decisions. Lots of bureaucracy costs and business costs, not to mention talent costs are going to be higher just because of the name. (A lead on GTA V can demand a lot higher salary than a lead on pretty much anything else)
- long production cycle (5 years). The more time a game takes to build… uh well, the more additional time it takes. Tech, tools and platform changes, and even sometimes shifts in the market (or trends), add additional expense budget. Our current game lost time because the iPhone5/iPad3 were announced during production, and then Apple required Retina which we had planned for the future, but not for initial release. That requirement meant we had to invest a huge amount of time and effort that wasn’t planned for. Granted, console dev can be a bit more stable, but there are still system/network updates and significant tool advancements over 5 years.
Strip out all the costs associated original design/content and those that arise simple because it GTA™, and it could be done much, much cheaper. But by the same token, a game made at a quarter of the costs won’t be making even close to a quarter of the revenue of GTA.
Good points. I think the reality is that an internally developed engine and something like Unity cannot be accurately compared on an overall basis. While the both make games in the end, their value comes in how they service the needs of the development/project.
I don’t know any specifics about RAGE either, but I have experienced what the scenarios you mention. Currently, not so much, but definitely in past. A game I was on was the first product released on an internally developed engine that had so much effort and time was sunk into it that our team was required to use that (front end) engine. It did actually provide amazing performance, but the dev/production pipeline was a nightmare at the best of times, and broken more often than not. It fell under both “sacred cow” and “sunk cost”. It has since be mostly retired/parted out.
Our current engines are nothing like that (learned lessons and all that). Our main engine we use now is really just freaking incredible. Super high performing across platforms, equally great for 2d or 3d , wicked cool UI design app/system and designed primarily for a heavy data/content pipeline. It is primary built for games that have a constant influx of content and assets, and multiple builds . In many ways, Unity can’t even come close to what it does for its designed use. That said, it also doesn’t come close to doing what Unity is designed for either. There is no single app. It is a collection of services, tools, interfaces and libraries. Setting up a project is very complex and time consuming. It is very flexible overall, but that flexibility is all up front, not after production has started. Heavy pre-production and planning are required. But once in production (and live ops) it is amazing. It excels at processes that simply aren’t part of Unity. But because of the heavy setup, it really isn’t used for anything quick or prototyping. I would guess that RAGE kind of works along those lines, most of the successful internal engines do. The are focused on optimizing a specific production process.
Trying to compare Unity to RAGE to Octane to Frostbite or whatever just is really not a practical. Most often folks compare the end products, which really only reflects the ability of the developers. An engine’s primary purpose is facilitate and optimize the production process. You don’t need any engine to create gameplay or cutting edge graphics, they just make it more efficient. Unity is designed to facilitate a different production model than RAGE. I wouldn’t attempt to build GTA in Unity, and I wouldn’t consider RAGE for the many of types of games that I would use Unity for.
Its like trying to compare a Tesla Roadster with a Freightliner Semi truck. Any value is derived from need.
Sort of but not entirely. For many years now, much of professional game development is pretty nomadic. Working on a big title is really kind of like working on a extended contract. Many of the roles are hired for a particular project cycle. GTA had several hundred people people working on it. It won’t need that same amount of staff after launch. Staff requirements expand and contract. Everyone knows this going in to it. Pixar, for example, formalized it to a degree since they work in a similar fashion. When you are hired to work on a Pixar movie, you are only hired for that movie. Your job is over when the movie is done. You then re-apply to work another movie.
More than that though, it is just part of culture. Its not just about companies letting people go when projects end, developers also switch jobs regularly for a variety of reasons, more interesting project, better pay, closer to home, to work with friends/colleagues, startup opportunities, and even just boredom. In the last decade, there are people I have worked with repeatedly across three different companies, friends that have left to work with other companies which then came back, or their company was acquired by ours. A friend of mine a few years left to work as a concept artist as Lucasfilm. He is now back with us as an art director, but still Lucasfilm staff. One of our leads has worked at EA 3 different times in 3 different divisions over the years.
People move and change jobs a lot, for a variety of reasons. For good developers, even unexpected layoffs usually just mean a week or two off, they often walk right into another position elsewhere. And most companies that layoff for product/business reasons provide decent severance packages. While layoffs make the news, what usually isn’t mentioned is that most companies around here are in a perpetual state of hiring. It isn’t always happy results, but it isn’t always as dramatic as it appears on the surface.
This makes a lot of sense to me for a large (GTA scale) production. Once you know the kind of content that you need to make and you’re just churning out mass, I’d want a known process and I’d want it locked down. Flexibility is great in the early stages or for “small” projects (by which I mostly mean "not as comparably huge as GTA), but once a game’s design is locked down and you need to fill in the content you want development to be as straight a line as possible, with minimal opportunities for taking wrong turns.
One of my first experiences with game development was modding for an RPG. At the time I thought that the provided tools were crap, because they were very limited in scope. You could write your own scripts but the scripting was limited in odd ways, and the tools were very particular with what you could do and where. At the time it was annoying, because I didn’t want to just make more of the same, I wanted to do a bit of custom stuff (which I did manage with some hackery, but I was fighting against the tools). But looking back it makes perfect sense. The original developers didn’t want their content people running off the beaten track like I was, they wanted them to use a small number of known pipelines to create a huge amount of consistent stuff, and the tools supported that really well.