|
What's more interesting is that they're writing their own "open" API.
I put open into quotations because it doesn't really do anybody any good for it to be open, except AMD. While nVidia's free to adopt it, it's not really to their advantage to do so, since any new feature sets are at AMD's behest, so they'd be stuck in the same position they were in in the early part of the 2000's, where AMD was the de facto standard architecture for DirectX 9 by virtue of being first to market. Anybody who remembers the GeForce FX series can tell you how that shakes out.
The part that amuses me here, though, is that AMD has essentially decided it wants to be late 90's 3dFX and Turtle Beach all in one.
The latter decision is somewhat questionable to me. If gamers cared about sound, Turtle Beach wouldn't be defunct and Sound Blaster would still be relevant. Instead, Intel came out with a cheap, good-enough solution in AC'97 (and put the nail in the coffin with AudioHD) and Sound Blaster's been selling almost exclusively to low-end sound production hobbyists ever since.
On the other hand, Mantle has something Glide never did: a huge market segment to hold hostage for the next 5 years to make developers care enough about it to code a separate code path. If you want your engine to run on consoles, you're coding a Mantle code path, now.
The good news for nVidia is that they've got Valve on their side. Gabe Newell's crusade against Microsoft might deflect this move from AMD, turning a potentially bad situation for nVidia into the death knell for DirectX, instead. nVidia has two options: they can hastily write their own low-level API, or they can keep pushing and advocating OpenGL as they help Valve preach Linux for gaming.
Either option presents developers with similar scenarios:
1) They can code for Mantle and DirectX, resulting in grossly lopsided performance profiles based on brand (low level API on Windows vs. high level API on Windows; the high level's overhead causes performance to suffer). This is a great outcome for AMD but gamers and the developer lose out because it sacrifices the majority of the current PC market's performance to make AMD look good. This might happen if AMD throws money at a developer via their Gaming Evolved program.
2) The developer codes for Mantle and OpenGL, resulting in good performance on Mantle, good performance on nVidia Linux OpenGL platforms, and marginally better performance on nVidia Windows machines than DirectX. Here, AMD and Linux nVidia solutions are.. we'll go with comparable, since we don't have any benchmarking to fall back on. Essentially, it comes out to which set of overhead combinations is lower: low-level API + Windows or mid-level API + Linux. Either way, the big loser here is Microsoft.
3) If nVidia decides to produce their own low-level API, the developer can code for that and Mantle. See above, only the edge is more likely to go to nVidia on Linux in this situation, and if AMD manages to step their Linux driver support up several notches, it becomes a great arena for the two companies to really go head to head in an escalating innovation war.
Scenario 1 seems least likely. Newell's got some influence, even with competing publishers (who must be having the same fears and gripes with Microsoft, though they're in less of a position to voice them since they have to be politically neutral lest they rile the feathers of the owners of the X-Box One), and the crowdfunding movement has been a great soapbox for gamers to advocate platform neutrality in their PC games. Kickstarter is clearly the best thing to happen to Mac and Linux gaming since.. well, ever. That Valve has stepped up to carry the torch for Linux just makes this an even more inevitable loss for Microsoft when it comes to DirectX giving them influence and de facto platform preference.
Scenario 3 is possible, but I don't see it happening right away, if at all. AMD's been working on this for a while, clearly, and nVidia's going to need to match that time -- and woo partners to its side -- to make their own proprietary thing stick.
Fortunately, I see good competition coming out of Scenario 2, as well.
That said, this does really cement my dislike of AMD's business strategies lately. They look likely to keep them relevant and successful in the near- and mid-term, but I would really have rather seen them step up their OpenGL game and join in an open standard that is actually a collaborative effort rather than figure out clever ways to force their own proprietary **** down other's throats and look like the good guys in the process.
Oh, and on the audio front.. I've always been pretty unimpressed with "virtualized" positional audio. Those surround sound headsets? Bullshit, AFAIC -- the entire benefit of surround sound is that it positions stuff statically so I can pick up on cues when I shift my head. That doesn't work when that "static" position is strapped to my head. On the other hand, tracking positional information and then downmixing to stereo makes a lot of sense in an era where we're anticipating low-latency head tracking via things like Oculus Rift. Throw that stereo signal through the headset I'm wearing with my Rift, and suddenly I get per-ear dedicated channels with positional information virtualized based on where I'm looking. So that part's cool, has promise, and I hope to see other people pick up on that in a solution that doesn't lock me into one brand's offerings anti-competitively. But then claiming you can get elevation cues into my 7.1 setup? This is my skeptical face.
_________________
"Aaaah! Emotions are weird!" - Amdee
"... Mirrorshades prevent the forces of normalcy from realizing that one is crazed and possibly dangerous. They are the symbol of the sun-staring visionary, the biker, the rocker, the policeman, and similar outlaws." - Bruce Sterling, preface to Mirrorshades
|
Login
Register
FAQ
Search
