Linux: PS3s Cell is faster than i7 965 XE

[alephnull]

Deneidez said:

alephnull said:

Deneidez said:

Kynes said:

CGI-Quality said:
I'm baffled by the some of the negative reception to this. Why is this news so bothersome to some?

Not cheerleading the Cell is not saying that the processor is bad. What we need is some perspective, and realize the highs and lows of the chip. It's not the second coming, only that. Having a good FOH performance is not what we, as gamers, need. FOH doesn't translate directly to performance in games, that's all.

Well, I am not saying either that CELL is bad. Its excellent when it comes to heavily parallel FP operating. Thats why they are using CELL for scientific stuff etc. Unfortunately games aren't that simple and actually CELL is more encumbrance than useful in games.

(To be honest XENON isn't either that good for games. Its in order like CELL and that can actually nearly halve its efficiency in games. Both of them encourage to make simple games, because they can't really handle anything complex.)

To design a chip with OOE you must necessarily reduce the amount of cache chip has as it becomes dedicated to your instruction pool. Really you only want it for badly optimized applications.

Yes, you are right. OoOE make chips bigger and more expensive. Anyway, it haven't been invented for nothing. Also are you implying that most advanced AIs are badly optimized, because they might have too many branches (and way too big reserved memory block) Cell to handle? :)

(And before MikeB is going to dump some stupid crap about Cell simulating human brains. I know how neural networks work and what they are good for. You really can't use them for AI. Teaching would take forever and it would still be very simple vs traditional methods.)

MikeB:

I don't lack talent to use CELL. Unfortunately many others do and I can understand why.

1) OOE will make the highly branchy code you are talking about run slower. The purpose of OOE is to avoid wasting cycles on memory loads. The problem is if your code branches the wrong way an OOE architecture has to undo either undo all the damage it did, or just stall at every conditional. Either way you were better off without it.

2) Define AI. I don't think any games are using neural networks. Not that they require loads of branching anyway, in fact they are probably one of the inherently least branchy things I can think of which is why people do them on GPUs or even better FPGAs. But, Neual networks are non-rigourus silliness any way:P FSMs can easily be done with matrix multiplies. All graph operations have matrix analogs. Back in the day when I wrote an RTS the unit AIs were modeled as a particle swarm.

Now I assume you are talking about decision trees, but there are invariably better ways of doing clustering (which pretty much all AI boils down to) than that. And yes people who work with decision trees tend to write inefficient code. Hence LISP.

3) Just because there are conditional statements doesn't mean you actually need to branch. You can use masks and guards as I did in that sample code 1-2 nights ago in the Xenon vs Cell thread.