Recent content by seebs

What got me was that, when I cited the relatively low rates at which Cell could access the GDDR3 memory on the RSX (assuming the old slide with the "not a typo" comment is legit, and it seems reasonable), he quoted the Samsung GDDR3 spec sheets at me. A few paragraphs later, he was explaining...

I was recently reading some arguments from someone who concluded, among other things, that RSX is probably a multicore architecture like Cell, only Sony's "locked" the other cores until they need to show off a big performance boost, Cell has access to GDDR at the full rated bandwidth of the...

In most cases, sure. What about the case where the layout of the cores makes the difference between a given set of >8 transfers happening simultaneously, and them having to share bandwidth? If you have EIB saturated, then you could see issues. I agree that this may not be able to occur...

That's the point at which, being sort of a pragmatist, I scrap the box and get one with ECC memory. :p

Hmm. Roughly in order: * Reseat both ends of SATA2 data cable, and any available ends on SATA2 power cable. * Connect SATA2 to SATA1 cable, see whether anything shows up in BIOS. * If I have an SATA enclosure handy, try it in that, too. * Feel SATA2 disk to make sure it feels like it's spinning up.

I'd like to second that evaluation; if the PSU fried, REPLACE IT IMMEDIATELY. Not that I've ever zorched components using bad power supplies. No, of course not. I would never be that careless.

Yeah, I've thought about it. It's not hard to design a program which should in theory be able to saturate EIB; it won't do much, but who cares? It's a proof of concept. Hmm. Well, it's on my list. If I ever find myself inexplicably not having enough to do, I might have a go at it. It...

I love it, and I don't know whether I'm considered "hardcore" or "casual". "Wrote and maintained local patches to WINE to add support for additional function keys to map to buttons in WoW" strikes me as fairly hardcore, but "doesn't like shooters much" probably doesn't. I note that, at $DAYJOB...

Here's what I end up wondering about: From some of the IBM research papers, it looks like affinity CAN make a noticable difference. A couple-few percent, anyway. Now, imagine that you code without any awareness of affinity issues. And imagine that, as it happens, your code ends up...

Zack & Wiki is the most frustrating, aggravating, game ever. Not the gameplay; no, the gameplay is awesome. The puzzles are brilliant, with enough meat in them to satisfy even people who sit around chewing through Infocom games without hints. No, the frustrating part is that I keep...

That's one of several programming models for Cell. It's not the only one. Six, unless Sony stopped taking one. (I thought it was for the hypervisor, but DeanA says it's for "security stuff", and he's probably right.) The thing is, workload ought to be reasonably predictable. So I'd...

I would expect the Wii to price drop once they've got products reliably in stores and sales are slowing down; that would let them keep sales up so they don't have to ramp down production just to ramp it up again fifteen minutes later. I'd been thinking next spring; possibly as late as mid-May...

I think it's early to make solid predictions about third-party sales; until the Wii catches up with console demand, there are millions of people who want the console, and might buy some games for it, who aren't in a position to buy games yet. I'd guess that the broader market base will mean...

Well, yeah. But I think the rush-to-market is a big part of why we see so many games that look like utter crap.

My assumption is that the main issue may be time -- a whole lot of Wii games thus far were rushed hugely because everyone was assuming the Wii would be irrelevant in the market, and then suddenly BANG, "it's the fastest-selling console, RELEASE THE GAMES!" For an example, compare Zack & Wiki to...