GI: What are your thoughts on episodic gaming?

Carmack: You know, it's one of those things that sounds like a good idea initially, but I'm not sure that-it'll probably have some degree of success-but I don't think it's necessarily a method that's going to cover the majority of the gaming market right now. I think it does tend to feel a little bit-it's all in market psychology and all of that, and I won't claim to be any kind of an expert on that there. It would be nice if games could be developed more piecemeal like that, but it still is so heavily front-loaded to build your basic core on there. It would be great if you could go ahead and release episodes regularly after that, but I think the customers feel to some degree that if they buy a game and they can get episodic additions to it, that there's this sense of, "Why didn't they just put that on there when I bought the main game itself?" I think there are some barriers to overcome. It would be nice when Internet distribution does become the main way that people get their games and you can just continue carrying out additional upgrades as long as you've got enough people willing to buy them. I think that would be good for game-development studios. But, so far, nothing's come along that really satisfies the same thing as the blockbuster commercial release. Of course, it's a tough market, with winners and losers, but even if you're in the winner's slot, that's still where the big successes are. But I do think it's possible for a lot of the less-conventional marketing strategies to get to a lot of game studios. There's probably a lot of good business there, but it's not the same thing as if you've got the possibility of doing a triple A title, there's nothing that's currently drawing you away from that.

Hollenshead: If you're doing a triple A title, I think there are substantive reasons as to why you would choose to go that direction. I mean, the first thing, as John said, is that the costs are front-loaded in terms of time and the expenses, and basically the game has to be done, especially from a programming standpoint, by the time you release the first episode. You can add some features as you go along, but the fundamental aspects of what you're going to be able to do from a significant level are going to have to be done. You incur all the time and expense to create that, and typically the media creation is not something that's gating you for release anyway. So, there's a bit of out-of-sync between how you incur time and cost and how you would actually recoup that through sales. Plus, there's the way that the market is set up and the way consumer expectations are set, certainly, publishers' market games are for games with a large launch with maybe some potential upside. You saw a game like Battlefield: 1942 that had a big launch, but then when it caught on it still had a crescendo of support. Typically, the game-sales curve is that you have your biggest month in your launch month, and it decays from there as retailers devote less and less shelf space to it, publishers devote less and less marketing dollars to promoting it and those sorts of things. So, talking about a long-term cycle of continuing to have to support that from a retail channel of support and a marketing dollar channel of support-it's very difficult for the market to understand how exactly that's going to work.

Carmack: Obviously, the retail chain's going to hate it.

Hollenshead: They're not going to like it, because you have effectively one product that gets split into, let's say, four boxes at lower price points, and that's the exact opposite direction that retail wants to be in. If you're talking about Internet distribution, there are issues there, but the main one is that even when you look at Half Life 2 and the episodic content, the biggest litmus test is taking a triple A title, where you have retail distribution and Internet distribution, and the choice that Valve made was to have both, which I think tells you that even with a company that's extremely motivated-because they have their own proprietary distribution system-to go exclusively with that distribution system understands that from a market reality standpoint that if you want the big dollars, you're going to have to be in retail stores, too, at this point in time.

Carmack: That is one of the interesting things about the cell-phone market, where instead of the fall off on sales curves like with PC titles, the cell-phone ones actually tend to grow with time as word-of-mouth gets around. And it's an interesting different sales dynamic, and I do always hope for something that-it's sort of a shame, where with a cell phone, most of the content comes over the air, which has this weird dynamic with the fact that memories are getting huge on the cell phones and you have a little straw to suck the content through, and that has its own set of issues. It has the potential there, where if people are always buying the games on there, where it should be this egalitarian system where you should be able to get any sort of content through there, but because of the way that it is tightly controlled by the providers on there-who are not exactly nimble, quick companies-and they have that aspect of that, dealing with the actual cell-phone companies does stink. It's interesting to play around with different distribution strategies and technologies on something like that where you're not talking about sinking thirty million dollars on development into something that you want to roll the dice on.