function said:
Okay, but I'm sure the AGEIA guy said they hadn't actually run the benchmarks on a 360. It was an early presumption or some such.
If you want the whole story on this, there are plenty of threads from the mists of time to consult.
function said:
I thought it was interesting in that it showed the gulf between theoretical perfromance and performance in a practical implementation by a (presumeably) skilled team from the company responsible for the chip.
Physics ain't their domain, though. I'd be much more interested to see what someone like Havok was doing with Cell than someone like IBM, tbh.
function said:
If reports are to be believed, PS3 developers have had as long with Cell as 360 developers have had with the final 360 CPU (or maybe longer).
Some developers have. It's still reportedly more difficult to tap, however, which would work against it.
function said:
I'm inclined to think that nothing, for either platform, has started to propperly plumb its respective depths yet.
I completely agree, and have said so on a number of occasions here now.
function said:
As I tried to point out in my last big post, the link between improved rendering power and improved visuals is a great deal simpler than between greater physics calculation and more fun, which is why direct comparisons to GPUs aren't helpful for that particular discussion.
This implies that the only way physics can be rendered is toward "fun" when it can be leveraged toward purely visual ends as much as GPUs are, hence the comparison to GPUs and graphics. But I see you go on to say..
function said:
And I'm not trying to talk down simulation for improving the way things look and move.
Which is great. But there have been others here who were, hence the argument.
function said:
I'm genuinely tired of the Carmack bashing from people that have never made a game, and never been involved in the process of making a game. He's talked about a number of gameplay related issues, none of which any of his detractors will engage with on a point by point basis, preferring to make generalised criticisms about his level of understanding, his approach to work, and the games he's worked on. No-one wants to pick him up on the points he made about physics and gameplay other than to say meaningless, generalised things like "more physics gives more options" or at very best "imagine a game where ... (random interaction X)" without trying to fit that into a fun single player game where it doesn't also cause gameplay issues.
I fully appreciate and understand his comments here. In fact, I'd agree that it can present challenges to use simulation in fundamentally gameplay-affecting ways. I may disagree with him on whether it's a worthwhile pursuit or not, but that's a different point. I agree that this is challenging. Carmack himself, however, even acknowleges the usefulness of simulation toward visual ends, and that's great, but he still makes a point of talking down all of these more CPU-orientated activities (physics/AI) because as he sees it, they don't (automatically) fundamentally improve the gameplay. YET, simultaneously, he thinks graphics are the best thing since sliced bread. In fact, he opened this very debate at Quakecon about physics and simulation and AI, and the focus being put on these things next-gen, by challenging the notion that realtime graphics has been "solved" and that we need to focus more on other things. He argues we should remain focussed on graphics, that there are big challenges to be resolved here and advances to be made. Yet the very same arguments he uses to try and take attention away from these other areas, could equally be applied to the area he wants us all to look more closely at again.
I think, however, he was fairly honest about this. He openly admitted that it was a pain in his ass when others came along and raised the bar in areas like physics and AI, and suddenly he needed to incorporate a better standard in these areas into his games. TBH, it sounds like he dreads the days when the bar is raised still further, such that advances in these areas become expected in all games, including his. But at least he was perfectly honest about it, and I can appreciate his position and why he takes that position. His company is small (in terms of headcount), he wants to keep it that way, and he'd rather focus on his biggest strength - graphics rendering.
Ike Turner said:
Totally different stuff.
The "car" and "gas station" stuff, if Harrison is to be believed, are taken directly from a game they have in production. He even went out of his way to explain the desert setting in the first instance by saying that they wanted to take as much as possible out except for the car so in order not to give away what game it was. You can believe him or not, but that's the claim.
Simulated wind would rock
