PC versus console

[Jonathan]

That is precisely what the PC game industry faces right now. Either you're selling subscriptions, you're Valve and you have Steam, or you're moving to consoles.

I can't think of any big name upcoming PC releases apart from MMOs. There's Alan Wake, but that's a 360 title too. There's Spore, but I can't find the launch platform list. Maybe I just don't follow the news as closely as I did years ago, but I think there is a narrowing of genres on the PC. There are plenty of games for the PC, but very few from major developers.

[quantus]

This is obviously also driving the shrinkage of the Halo Effect. Without an uber-cool game that is really suffering from lack of frame-rate or something, people just won't shell out extra cash for better hardware to have a better in-game experience. There has to be some sort of in-game advantage that makes it worth it to shell out that cash.

Theoretically, Vista could've driven a hardware upgrade cycle, but it would've had to be released like 3 years ago. Now the hardware is more than powerful enough at the mainstream levels. Besides, the XP upgrade cycle was kinda annoying (especially with ME in there too) so people are not going to want to jump to Vista too soon.

I'll reiterate my previous view that hardware engineers were signaling a change to multi-core for a while. There were no easy solutions for how to code on these types of machines, so people stuck their head in the sand hoping someone else would solve the problem, and guess what, no major progress. Interestingly, the dot-com-bust might make the situation worse according to EETimes. We may be heading into a period of high demand for software engineers.

[Jonathan]

The multicore argument makes no sense in the context of PC versus console. The 360 and PS3 have more truly simultaneous threads of execution than all but the highest end PCs.

Anyway, for floating point (graphics, physics), performance scales with the number of execution units. Integer really doesn't unless you do something clever with the algorithms, so there's a real limit law-of-diminishing-returns style on the number of cores a CPU manufacturer can throw at its customers and still expect them to swallow. Consoles have it relatively easy as long as no one is attempting to do very tricky AI logic. They can go on throwing cores at graphics and physics for a very long time, and the developers will only grumble a little bit.

[Jonathan]

This is obviously also driving the shrinkage of the Halo Effect. Without an uber-cool game that is really suffering from lack of frame-rate or something, people just won't shell out extra cash for better hardware to have a better in-game experience.

This is a very interesting idea. I've seen a lot of folks say that the game industry is as hit-driven as music or the recording industry. I think this is mostly hogwash, but when it comes to driving hardware adoption you may have a point. That is, you can have a Sims-style hit and you won't sell any more computers, but it is difficult to create a performance-motivated hardware upgrade cycle without a demanding hit like the first generation of hardware accelerated FPSs. Hmm.

[George]

This is obviously also driving the shrinkage of the Halo Effect. Without an uber-cool game that is really suffering from lack of frame-rate or something, people just won't shell out extra cash for better hardware to have a better in-game experience. There has to be some sort of in-game advantage that makes it worth it to shell out that cash.

Oblivion. I wouldn't be surprised if 50% of the high-end video card purchases in Q2&3 of last year were to play that game. Mine was (at least partially). I seem to recall BF2 being the big driver before that, but I wasn't interested, so I didn't follow it that closely.

An interesting point in that Oblivion was released without content protection. Today, a year after release, it still commands full price (well, $40) in stores. People are still buying it, even though it's been pirateable since day one.

I have a theory that piracy-to-sales ratios are proportional to a game's hype-to-quality ratio. When someone chooses to acquire a game, legally or otherwise, they make a decision based on their anticipated enjoyment vs the cost. Anticipated enjoyment is based on direct experience with the game, experience with similar game (or by the same developers), comments from trustworthy sources (friends' experiences), and comments from less trustworthy sources (publishers, review sites that get ad revenue from the publishers), weighted in approximately that order.

On release day, only similar experience and untrustworthy comments are available. The latter is overwhelmingly positive, but it's limited by the reach of the publisher's marketing department. So, purchases and piracy on day one are largely proportional to the quality of marketing and are completely independent of quality of gameplay. Big name games from big publishers sell better one release day. No revelation there.

As time goes on, the forms of information that are actually related to the gameplay become available. Perceived enjoyment will gradually approach realized enjoyment, and the sales will change accordingly. Hence, sleeper hits and bombs where the initial perceived enjoyment is radically different from later perceived enjoyment.

Now, that's what happens on the legal side, where there is a significant cost to acquisition. On the pirate side, there is almost no cost to acquire (aside from the time to find a copy). Thus, perceived quality isn't likely to fall below the cost, even when it is well below the initial perception. So, piracy rates don't drop much when a game turns out to suck.

Let's pick on Doom 3. Highly anticipated. Lots of piracy (because it's anticipated). Game sucks, so it doesn't sell. Publisher observes high piracy and low sales and hypothesizes a correlation that doesn't exist.

Conversely, Oblivion. Highly anticipated. Lots of piracy. Game is adequate (bordering on good). Sales rate is good enough that publisher doesn't bother to look for a scapegoat.

[George]

Hmm, condensed theory: Piracy is proportional to hype (awareness of game), sales are proportional to quality.

[Jonathan]

I have a theory that piracy-to-sales ratios are proportional to a game's hype-to-quality ratio.

Yeah, this is what I was saying, but snappier!

[Dave]

Well considering how the local best buy had 8 PS3s and 0 Wii and assloads of 360s, I'd say pie > cake.

[Jonathan]

Valve thinks the PC market is too fractured with the DX9/DX10 split.

http://theinquirer.net/?article=41937

Seems like there's not much left for OpenGL to take over, even if they could. I suppose leaving PC gaming in Microsoft's hands is just asking for trouble.

The sad thing is we'll probably have games coded to DX9 standards for years now.

[Jonathan]

http://www.extremetech.com/article2/0,1 ... 339,00.asp

This guy has 5 myths of PC gaming that he attempts to debunk. I think he may have a good point with the third myth. As a business, PC gaming is not as big as console gaming. However, as one of many platforms, PCs are doing fine. Perhaps not in the lead, revenue-wise (although WoW gives PS2 a run for its money), but definitely a strong showing.

[Jonathan]

http://www.theinquirer.net/gb/inquirer/ ... agued-bugs

And, lo, it turns out supporting several different SKUs of multiple incompatible platforms is pretty difficult to do regardless of whether those platforms are "consoles" or not.

[Jonathan]

http://www.anandtech.com/cpuchipsets/sh ... i=3312&p=2

AMD Game: gives a simplified brand name to computers that can be used for gaming (made of AMD parts, of course).

I think AMD has got the right approach here. Contrast with the Windows Experience score (does anyone use that?). If you buy the Ultra, you know you can play new games at 30fps 1600x1200. That's a useful label.

[George]

No one's got Vista, so no one uses the Experience scores. My secondary PC rates a 5.0, limited by its CPU, oddly enough.

The games they listed aren't current. They've got stuff like Sims 2 and World of Warcraft, which are certainly popular, but weren't even pushing the envelope when they came out a couple years ago. They've left off the aggressive stuff like Crysis.

And my main computer doesn't even meet the minimum specs for the base Game label. Yet it plays Witcher at 1920x1200 just fine. Except for the dialog sections, which for no apparent reason drop to about 1 Hz refresh. So, in conclusion, marketing BS.

[Jonathan]

Well, the idea is sound in principle unlike the Experience junk. Perhaps a competitor will execute better...

[Jonathan]

http://dealbook.blogs.nytimes.com/2010/ ... on-paypal/
http://investor.ea.com/releasedetail.cf ... eID=436876

EA's annual revenue is reportedly only 3-4x the annual revenue of the Farmville/Mafia Wars guys. I suspect the Farmville guys make just as much profit as EA, however. I have seen the future of gaming, and it is suburban housewives.

[Jonathan]

I think SteamOS is the closest analog to what I had originally envisioned here. It's a fairly good analog, too, at least in the context of Steam Machines. I think there's a market for a Steam Machine in a virtual box for, say, Alienware or somebody to step into, because as it is I have a hard time visualizing a future where SteamOS is useful for stuff that isn't gaming.

It is odd to realize that the iTunes app store + iOS actually is a competing vision here, one that has been quite successful at taking market share from Xbox and Playstation and Nintendo. And, of course, they've cornered the market at doing stuff other than games.

[Jonathan]

And of course, Early Access and Greenlight (plus, to some extent, Kickstarter) have really lowered the price of PC games. All my friends have a couple console games they played intensely for a couple months and then got tired of, and huge catalogs of Steam games they bought on sale or Early Access that they'll never get around to exhaustively completing.