Cost of Game Development is 'Crazy' Says EA 321
GamesIndustry.biz has the word from Alan Tascan, general manager of EA's Montreal studio, who has gone on record saying that development costs are 'crazy' in this next-gen world. From the article: "When asked whether he'd agree that it's larger companies like EA which are driving bigger game budgets, Tascan replied, 'I think a lot of [other companies] are spending even more money. It's people who want that, it's not EA per se ... I said to some of the guys here, "The gamer is not buying lines of code; you have to promise him enough entertainment for him to put his hand in his pocket and buy the game." It's a lot of money, so you need to give him a show, and we're just here to deliver the show.'"
Being a Spelling Nazi (Score:4, Informative)
Umm, it's "per se".
I realize this is how different flavours of languages propagate over the ages, but I'm all in favor of keeping English as unified as possible.
Solomon
Re:No Problem (Score:4, Informative)
Graphics are easy to see without playing. However, I don't see how you can deduce the gameplay characteristics of a game series you've never played. As a semi-regular Madden buyer, I'll address the issue anyways. Those improvements are incremental but if you look at how long it ususally takes to develop a sequel to a game (2-3 years) and what Madden has done in that amount of time, the changes are typically quite drastic. That would explain where the money went. I still maintain that I'd rather play Madden 07 than Madden 06 with 07's roster and that's been true every year except for IMHO some exceptionally poor showings from 2001 to 2003 (in Madden years).
Cost factor is the same old same old... (Score:5, Informative)
- Unrealistic schedules: A marketing dweeb decides when a game should be released without taking into consideration the developer's experience level, console manufacturing requirements, and whether enough QA will be available to adequately test the title. As a lead QA tester, I routinely add two months to the schedule and my time estimates are usually 90% correct.
- Bonus Structure: The producer's bonus is tied to the unrealistic schedule and a lot of decisions are made to compromises the game so the producer can get his bonus. As a lead QA tester, I was routinely accused of denying a producer his "hard earned" bonus.
- Unrestrained QA Overtime: If a game is not properly scheduled and managed, a tremendous amount of QA overtime will go into trying to save the game and, almost always, is shipped regardless of the final quality. As a lead QA tester, I worked 28 days straight on my last project because the schedule was cut by one-third and I was not notified until half-way through the project.
I'm not holding my breath that the video game industry will one day figure out that there's a saner method for developing a video game that doesn't blow the schedule and the budget like a bad lunch at Taco Bell.Re:Meh. (Score:4, Informative)
Bah... it doesn't have to be that expensive. I've plugged them before here and I'll plug them again because I think that the company is amazing: Stardock [stardock.com]. They're a tiny, independent developer/publisher about 30 minutes from Ann Arbor, Michigan. Their most notable game is Galactic Civilizations 2 [galciv2.com], which includes 3D modeling, professional music score and sound effects, an insane amount of detail, excellent replayability, challenging AI, and very balanced gameplay. When I think "independent game developer", lame little Flash-based games are what come to mind. However GalCiv2 *fully* competes with anything EA has ever put out. Stardock also has a very "pro-customer" stance [galciv2.com] on copy protection too.
Re:Cry me a river...all the way to the bank! (Score:5, Informative)
Re:Nintendo (Score:1, Informative)
DS/GBA development can be done for under $100/$50 respectively. The software is free, you just need to buy the hardware (flash cart and a NoPass if doing DS dev) and get to work.
Re:Cry me a river... (Score:3, Informative)
Yes yes, and if I built a PC I'd put a 256MB video card in it and I'd have 768MB. But you can't lump the memory together, and doing so displays your ignorance of the topic we're discussing. Now granted, there are UMA solutions that use part of your system memory as video memory, but due to the performance impact, most of us try to avoid doing such things anywhere other than servers and such where it doesn't matter.
The fact remains that the PS3 has only 256MB of memory for programs and data. Period. Not 512, 256. You can run shader programs from the video memory, maybe, but again, that won't help you with general purpose computing. Neither will the Cells, which is as you say why the system benchmarks like what is today an incredibly low-end processor. The Cell in the PS2 has one PPE which is based on a 64 bit PowerPC core, without a vector processor, and it has six or seven SPEs which are the cell cores. The PPE is there to do anything that can't be passed off to a vector processor, and to feed the SPEs. As such it's potentially incredibly powerful but is instead incredibly weak for general purpose computing until strategies are found to optimize general purpose code to the processor.
I'm very happy that the Xbox runs XBMC beautifully in 64MB. I have one under my TV and I use it all the time. However, I said general purpose computing. XBMC is an appliance-type program, so it is not any kind of useful counterexample.
Anyway, "weak comparisons"? The assertion was that it was a dandy PC. It is not. It is a shit PC. It may be a great game console; I won't know because I'll never buy one, because Sony can go fuck itself. However, that does not at all change the validity of my argument.
If you don't believe me, try this experiment: downgrade your computer to 256MB RAM and tell me how you enjoy the experience. If you only have 256MB now, I hope you have an Amiga or something. XP became about twice as fast when I bumped up to 1GB memory from 512MB. 256MB to 512MB is something like a fourfold boost in performance because you get out of swap-land. Oh, it's possible to get things done with that little memory, but given (for example) a choice between a 2GHz machine with 256MB, and the same machine with half the clock rate and twice the memory, I'll choose the memory every time. Unless you're performing a single optimized task that somehow does not go past that line, the machine with more RAM will always be faster.
Re:This has been bothering me for a while. (Score:3, Informative)
I can't believe cynical overgeneralizations still get modded up around here. There's a lot of noise out there, but also some killer fun games. The emerging cinematic elements aren't a substitution for fun, but a huge addition to the ones which do it right.
Re:So? (Score:3, Informative)