No, most of the costs do not come from these. Advertisement is expensive only on big ass IPs like call of duty, world of warcraft, etc.
The biggest cost is salaries, having a 200 people team that works for several years while being paid at least 2k to 3k a month is very expensive. Just do some maths:
200 * 2500 = 500.000 $ just for a month. And that's without counting the various taxes that apply to salaries.
So yeah, having a 50 man team or less is important, you reduce the costs and that means you have a less risky development.
The way i see it is that we're headed to more and more graphical *useless* features (like 4k definition) and that it'll make game development very hard because you'll have to pay an army of graphic artists to come up with insane definition textures and assets. John Carmack even mentioned "unique texturing"...i do hope this never happens.

If you think about it, game mechanics have barely evolved recently. Ok there are some new features, but overall a FPS is still a FPS. You shoot bullets with your gun and that kills people. It hasn't become insanely complicated to make an engine for this. What's become complicated as hell is trying to make it uber-realistic with dx11 features such as tesselation, big textures, high poly 3D models, lots of shader effects, etc.
Here's an example of what i mean:
http://treize.ryustemple.net/quake2-1.jpg
http://treize.ryustemple.net/callofdmw.png
do you really see a lot of differences between the 2 screenshots ?
Buildings will remain some kind of cubes with textures applied on them. The core gameplay hasn't really changed either. What has changed is that the player models use a lot more polys, there are a lot of shader effects, as well as dx11 effects, and the textures are really much bigger.
Ok...so this means not much more work for programmers, but a big increase in the amount of graphic artists required for the game. That's it.

edit: omg...this forum has issues xD going to host the screens somewhere else.

Yes i understand... However textures for example are mostly made high res (mostly) and then lowered. Yes i can understand things like shaders, light, particle, good use of tesselation might be time consuming. I dont have anything against better graphics... But i dont care about it that much either. I mostly play older games anyway.

Have you seen this:

http://www.youtube.com/watch?feature=player_detailpage&v=6KW73TZTAH8#t=380s

DAT lighting and human face O_O

Yeah graphics can stop now.

I was some exhibition thing and one thing shown was a PC with top of the range graphics, it looked good but only keeps you occupied/amazed for a few minutes at the most, puts graphics vs gameplay into perspective.

Yeah lets just ignore that the average programmer gets 5 time as much in the USA than he gets in Poland.

Well, you can go ahead and share whatever crazy-ass crack you're smoking, because I too desire a disconnect from reality once in a while.

Didn't say it wasn't well made, i said i didn't like it. Why do you feel the need to have everyone like the same game as you ? O.o
TO ME, the witcher series isn't fun and i'd much rather play skyrim or DA:O. Get over it.

OH absolutely, and most of the commenters on this thread don't realize that with more powerful hardware, all of this is achievable. Heck, developer tools also benefit from more capable hardware, seeing the UE4 demo highlighted at what speed game development is becoming simpler. The rumored 8 core cpu's of the new consoles aren't there to power all the gfx, they are there to propel the aspects you mentioned above into bigger and better heights.

I'm not sure the massive video game budgets are always due to graphics, though. Certainly it now takes a team to create the assets, so it's part of it, but AAA games now have real script writing, consultations with authors, they hire orchestras and real voice actors, etc. That stuff adds up quick in a budget.

Additionally the game play surfaces (i.e. the possibility of play) in many AAA games these days have increased a lot (Jeff Vogel has a pretty good blog post explaining what the game play surface is), as these expand, testing and development must expand, regardless of what graphics you're working with.

Certainly, and I'm all for examining all those expenditures. I guess my main point was that you don't need to have a super high budget to sell games, there are countless examples of this. So why spend a hundred and fifty million to make a game when it is irrelevant to selling a ton of copies?

Certainly you can make games for much less and folks often do, I'm not sure you can make the 150 million dollar games for a lot less (using today's tech anyway) and I'm grateful to play the expensive ones as well. I think it crosses the point of silliness when companies bankrupt themselves by "only" making the expensive games and refusing to produce other equally profitable (but not on the same magnitude, i.e. lower cost and lower total profit) games.

I always argue for as much as possible to contrast stuff. My brother (an art guy) was going on and on about some famous dude who signed a urinal as his art entry in a contest (art folks will know who this is, but it isn't germane to the story), he was pointing out how famous and important this was as well as how insulted the contest organizers were.

My only thought was that without the kind of art he was criticizing and contrasting with his signed urinal it couldn't have meant merely as much.

I guess that's my way of saying, I think we're all richer for having all sorts of shit, even the shit we don't personally care so much about. It's probably a bit of an ideal, but I kind of think it's true (and at least would want it to be true).

Indeed. I just think the days of Darksiders 2 selling 2 million copies and being a massive failure that puts a nail in THQ's coffin have to end.

The audience for "core" games is not large enough to plan on 10 million copies being sold of anything.

Darksiders 2 was a pretty good example of how to cut off your nose and pretend it's a "good thing". Yeah, that was pure jackassery. Though I'm not sure that all of THQ's woes were of their own doing, near the end many of them seemed to be.

Affordability comes into play here too, as well as the gameplay vs. graphics issue. Gameplay is OBVIOUSLY more important than graphics from a gamer's point of view, if not from a greedy publisher's point of view. When Crysis was released I wanted both to play the game and to see the marvellous graphics, and thankfully the game was very good and not just "all looks and no brains", but I was also worried about "future proofing". I was wrong : it wasn't worth the outlay when I look back on it.

The fact is that the release of Crysis made me believe that if future games were going to require such advanced hardware I would have to upgrade, and in my case that meant buying an entire new computer. So, like the mug that I am, I bought a new PC. Was buying a brand new PC justified in the end ? No, because the quality of games since then, regardless of graphics, has not been amazing. Most of the games I want to play are still oldies. That may be partly because I am OLD (for a gamer), but I know I'm not the only one who thinks like this, and plenty of younger gamers agree. Ultimately it wasn't worth the £1700 I shelled out for a new PC, even if it did allow me to play a few other excellent games, e.g. Bioshock, which my old machine wouldn't have been able to cope with.

The PC gaming industry started out as something creative, exciting and honest which (seemingly) didn't care so much about making money as being a medium for the creative imagination and an enjoyable pastime, open to all. Nowadays it seems to have become yet another cynical money-making "enterprise", driven by greed and aimed at the majority (which is clearly where the profits lie). Half Life 2's or Far Cry's graphics were good enough for me and I'm not really that bothered about any further advancements because when you play a game it's not merely the look of the game which makes you want to keep going back to it.