Archive for August, 2005
IBM et al have released a lot of information about the Cell processor recently. Unless you’ve been living under a PS1 for the last five years, you know that the Cell processor will be powering the upcoming PS3 gaming console from Sony.
One of the most interesting bits is IBM’s claim of superior memory access. Currently the processors spend far too much time waiting on memory, so the Cell tries to ameliorate memory access problems with a new memory architecture.
This three-level organization of storage (register file, local store, main storage) — with asynchronous DMA transfers between local store and main storage — is a radical break with conventional architecture and programming models because it explicitly parallelizes computation and the transfers of data and instructions.
The reason for this radical change is that memory latency, measured in processor cycles, has gone up several hundredfold in the last 20 years. The result is that application performance is often limited by memory latency rather than peak compute capability or peak bandwidth. When a sequential program on a conventional architecture performs a load instruction that misses in the caches, program execution now comes to a halt for several hundred cycles. Compared with this penalty, the few cycles it takes to set up a DMA transfer for an SPE is quite small. Even with deep and costly speculation, conventional processors manage to get at best a handful of independent memory accesses in flight. The result can be compared to a bucket brigade in which a hundred people are required to cover the distance to the water needed to put the fire out, but only a few buckets are available.
In contrast, the explicit DMA model allows each SPE to have many concurrent memory accesses in flight without the need for speculation.
The most productive SPE memory-access model appears to be the one in which a list (such as a scatter-gather list) of DMA transfers is constructed in an SPE’s local store so that the SPE’s DMA controller can process the list asynchronously while the SPE operates on previously transferred data. In several cases, this new approach to accessing memory has led to application performance exceeding that of conventional processors by almost two orders of magnitude, significantly more than anyone would expect from the peak performance ratio (about 10x) between the Cell Broadband Engine and conventional PC processors.
First mover advantage. Yup, gotta get that Xbox 360 out there before Sony gets theirs out there. It’ll make all the difference in the world! So says J Allard, at least. We’ve heard it so many times from all sorts of different sources that we know Microsoft believes this to be one of their two aces for the Xbox 360 launch (the other being Xbox Live). And frankly, I think they have a point, everything else being equal (which it rarely is).
Someone over at Microsoft forgot to tell Chris Lewis, regional vice president, Xbox, Europe, Middle East and Africa (EMEA). He says that the reasons for the release date are:
This is something that comes up a lot and I want to get this straight. The timing’s not about the competition. The timing’s about listening to what gamers want, and listening to what the market wants right now. It’s about giving people the next-generation of gaming, now. People want high definition, they want deeper gameplay, they want more riveting stories, and people want to do a whole lot more things. They want to connect their music, they want to rip tracks, they want to plug their iPod or PSP in, and I think it’s about giving people what they want, right now, in a scaleable way. If that works well in terms of timing, then great.
Well, I certainly want HD, but with only 10% or so of American homes having an HDTV, I don’t think it’s something gamers want right now.
But anyway, maybe Mr. Lewis just hasn’t read the memo yet.
My wife is having a baby.
[Update: My son Grayson Heinrich Hoffmann was born on Thursday August 25th at 3:27pm weighing 7 lbs 13 oz. He and his mom are both doing well. He’s so cute!]
With all the talk about how next-gen development costs are going to skyrocket, it’s nice to see a developer actually give some numbers.
While discussing possible PS3 support by Tecmo, Tecmo president Junji Nakamura stated that Xbox 360 development will cost 20% more than for previous platforms. While 20% isn’t anything to sneeze at, especially since costs to produce a game are often multi-million dollar affairs, it’s still a number that’s a lot less than analysts have predicted.
Does this mean that prices will go up by 20%? Probably, yes. But if developers can increase the quality in the process, they’ll sell more copies, and therefore will not need to increase prices?
Wanna make any bets about the general quality of next-gen games versus current-gen?