Smasharoo wrote:
You're better off getting 2 dual cores then 1 quad core, and typically half the price.
Then do what, exactly? Duct tape them together? The percentage of the motherboard market that offers more than one CPU socket is about .001.
Hah. Um... Well. I don't buy computers for myself that often, so I may be a bit out of touch, but every computer I've purchased in like the last 3 years has dual cpu capability. These aren't huge server room systems, but standard case workstations.
I've actually spent quite a bit of time pricing these types of systems. Consistently, you're better off buying a dual cpu MB with two dual core's then a single quad core in terms of performance for the price. That's honestly an issue with the lack of support at the application level for quad cores right now and will certainly change, but it doesn't change that performance comparison
right now.
Which is what I was talking about. I was comparing the two lines of chips, not specifically limiting myself to the home market.
Quote:
The problem isn't the chip architecture here, and it has nothing to do with Intel specifically. The problem is developers aren't very good at coding for multi-core chips yet. AMD's quad core/three core line is a @#%^ing mess. Their space in the market is completely reliant on power/watt at this point which will keep them alive in the server space, but their share of the gaming/enthusiast market is disintegrating. They need something dramatic not to lose the consumer market completely in the next five years. An architecture breakthrough, or a new fab success. Considering the difficulty they had with the K8 die shrink and the Phenom launch, neither of those seems particularly likely.
Yeah. AMD is kinda in the doghouse right now. But as I pointed out, this has been the case at least twice in the past, and they manage to pull out of it each time. AMDs strength has been the cost/performance ratio, and making more efficient use of existing designs rather then pushing the envelope into new areas.
I disagree that they need to do something dramatic. That's actually what's got them in trouble. They'll start grabbing market share, think this means they can lead the pack, and end up goofing it up each time. AMD should stick to what it does best: Refinement of existing technology. When they try to push the envelope, they crash and burn every time. Each time in the past when they've been able to pass Intel, it's not been because they were first to market with a new generation of design, but because they refined the design/size and made a moire efficient product and allowed them to lengthen the effective lifespan of the line. Intel spends massive money on inventing the next thing, the next size jump, etc, then gets about 3-4 years of ahead time on it before AMD comes up with a better design in the same size/power range and takes it away for a bit.
This same cycle has been going on between them for nearly the last 20 years. I think that AMD's going to be behind on the next catch-up phase though, because after the last cycle, AMD shifted to attempting to challenge Intel in the "first to market" area, and basically got spanked. We'll see what happens though. I just don't think it needs to be dramatic. If next year, AMDs dual cores perform equivalently to Intel's, but are cheaper (in cost and/or power), you'll start seeing folks buy AMD. And if it takes Intel longer then that to get big market penetration with quad cores (or even if the cost prevents significant penetration), AMD will gobble up yet more of the dual core and/or dual cpu market. It's how they've competed in the past, and they'd be fools not to continue.