nVidia just launched the GTX580 upping the ante in the high-stakes game for your DX11 love. But the question you have to wonder is; does the average [or even high level] gamer really care? Let?s take a look at the top selling games recently.
According to most internet sources [and Steam] the following PC games are listed as top sellers as of this writing.
From Steam you have;
1 ? Call of Duty Black Ops [DX9]
2 ? Risen [DX9.1]
3 ? F1 2010 [DX9] [a patch was released on the 4th of November to add some DX11 functionality]
4 ? Football Manager 2001 [DX9]
5 ? Fallout New Vegas [DX9]
6 ? Sid Mier?s Civilization V [DX9]
7 ? Left 4 Dead 2 [DX10]
8 ? Divinity II: The Dragon Knight Saga [DX10]
9 ? Counter-Strike: Source [DX9]
10 ? Battlefield: Bad Company 2 [DX11]
Now, let?s see what we have here. The top 6 games are DX9 we do not even see a DX10 game until the 7th place slot and no DX11 until 10th place. This is a huge indicator that while AMD and nVidia are fighting for the fastest DX11 GPU crown the market just does not care enough about this to buy into it.
We are seeing something of a trend and one that we have talked about on more than one occasion. The hardware is outpacing the money and time available to the developers coding these games. Because of this we are seeing halfhearted implementations of Direct X10 and 11 features. Games like Bioshock and Bioshock 2 only have DX10 surfaces. Left 4 Dead also implements only part of the API. We are seeing the same thing with DX11. The costs involved in building a fully DX11 game are very high not to mention the time to implement them; meanwhile the market wants these games now. Because of this game developers are coding for DX9 as it is easier and it also spans the console and PC worlds. This mind set allows them to build two games at once. They save money on not needing two game engines. True you will have some developers that will code for both DX9 and DX11 we see this with Hawx 2 but to do this they have two different executables; which we have also seen with other games.
But the market has not warmed to DX10 much and is still very cool on DX11. There is a growing number of consumers that feel there is very little visual difference between DX9 and DX10/11. This hesitance to accept the new APIs have reduced the perceived demand for DX10 and 11. No demand [real or not] makes the bean counters at the development shops not want to invest time and money in the new APIs [they do not want to waste money on a complete or partial rewrite]. So we have a vicious cycle, the game developers do not want to spend the time and money to build a game that fully implements DX11 due to lack of consumer demand. Consumers do not see the need for DX11 because there are no games that make full use of it that are at all compelling.
In the end we find that no matter how good the DX11 API is, it is still subject to the whims of the market. So if you ask your average PC gamer, not the nut with tri-SLI or crossfire but a regular gamer, what version of DX his favorite game is running you will probably get an answer something like ?don?t know, don?t care, but it is fun to play?. It is now up to the game developers combined with the hardware manufacturers to come out with multiple games that truly take advantage of the DX11 API. Not just Tessellation or a couple of features, but the whole suite. This game also needs to be a compelling and immersive game; one that grabs the players and pulls them in. Until that time the consumer will blissfully ignorant of what they are missing and no amount of benchmarking will make them feel otherwise.