Monday, April 27th 2009
''Real Men Use Real Cores'': AMD
AMD finally stepped out of its shell after Intel's launch of its newest line of Xeon processors based on the Nehalem architecture. In an interview with TechPulse 360, AMD's Pat Patla and John Fruehe took on Intel's recent marketing drive for Nehalem Xeon products. The conversation revolved mainly around the issues of platform costs, and the features the new Xeon processors introduce (or reintroduce) to the server/enterprise computing industry, namely the company's proprietary FSB-replacement, QuickPath Interconnect, and HyperThreading.
The two first took on Intel's marketing, particularly on its material that said that the slowest Nehalem Xeon chip was faster than the fastest Opteron chip, saying that Intel's statements weren't backed by real figures. The two also alleged that Intel's server platform was too expensive and delivered lesser value in an ailing state of the economy. Perhaps the most audacious statement from AMD since the somewhat famous "only real men have fabs" statement by Jerry Sanders III, came from this interview, where AMD responded to a question on HyperThreading saying that "real men use real cores". "We've got real cores across our products. HyperThreading is basically designed to act like a core except that it only gives 10 to 15 percent performance bump for real applications workload." they said. Is AMD making a real point, or fighting fire...erm marketing with marketing? Find out in this interview.
The two first took on Intel's marketing, particularly on its material that said that the slowest Nehalem Xeon chip was faster than the fastest Opteron chip, saying that Intel's statements weren't backed by real figures. The two also alleged that Intel's server platform was too expensive and delivered lesser value in an ailing state of the economy. Perhaps the most audacious statement from AMD since the somewhat famous "only real men have fabs" statement by Jerry Sanders III, came from this interview, where AMD responded to a question on HyperThreading saying that "real men use real cores". "We've got real cores across our products. HyperThreading is basically designed to act like a core except that it only gives 10 to 15 percent performance bump for real applications workload." they said. Is AMD making a real point, or fighting fire...erm marketing with marketing? Find out in this interview.
71 Comments on ''Real Men Use Real Cores'': AMD
Morons.
What is good though, is that Intel may already have their 6 core "Dunnington" on the market, but it is using the older Core 2 architecture without a built in memory controller (let alone the triple channel one in the Nehalem architecture). This means that AMD may still have the upper hand in memory bandwidth on 6 core processors. Also, these are straight up drop in upgrades and are MUCH more appealing to those who run servers (especially who are already running Opteron) as opposed to the ship that has to be jumped through to go from the Core 2 Xeon to the Nehalem Xeon.
And funny AMD mocks Hyperthreading. Hyperthreading means virtually none of the processor sits idle. What's so bad about that? It's good because more work is done per clock without a significant addition of more power draw and heat production. That means it's great for servers. AMD needs a better line than that like, "We're second place and lov'in it!" Or, "AMD, the tortoise: slow and steady. But we'll win in the end!"
AMD/ATI has been sinking since the Core2s hit the market.
if you dont want to spend a lot then AMD is just as good sure it cant compete with the i7 but calm down man
An 8 physical core AMD would (should?) kick the living hell out of any (current) I7 but AMD isn't there yet, nor is Intel for that matter.
Many claim that Phenom IIs aren't meant to be competing with I7s and should be competing with Yorkfields instead but i disagree: It's not Intel's fault AMD hasn't been able to make a closer (or better) performer to it's I7 and Intel's top CPUs should be competing with AMD's top CPUs.
This is no cause for AMD to give up: quite the contrary. Intel may have thought they were miles ahead of AMD but AMD have responded well, though not as good as they might have liked, but they did closed the gap quite a bit, IMHO.
In these CPU wars, one manufacturer must NOT get too far away from the other or it will be us consumers that will pay the price, literally.
Anyway I find that AMD's current PII is hardly any better than my q6600. Not enough for me to give her up. I'd only jump to nehalem or if dual amd board which is unlikely at best.
What do you think the result was? They started getting far more costumers then anyone else and the competition started lowering their prices as well.
Who do you think won this? IMO, there were two winners:
- the dudes that started selling the gas @ 10 cents lower then anyone else which gave them hordes of costumers and, despite been selling @ a lower price, made them more money then those that sold @ regular price;
- those consumers that saved 10 cents per liter and ended up saving a ton of money.
The point i'm trying to put across is that, without competition, those that sell a product can set it's price freely but, when faced with competition, they are forced to make a competitive price. I have had the following CPUs, over the years:
Pentium 100
Celeron 300A
Duron 800
Athlon 1800
Athlon 64 3200
E6300
E6850
And my current one: E8400.
This is where Intel threw a curveball: instead of telling those working on NetBurst to just give up, they let them continue to peck away at the idea of NetBurst while they get another research lab in Israel to look at destops/workstations. Israel, deciding that NetBurst wouldn't work for mobile processors some time before because of heavy power demands and high wattage, went back to Pentium III to develop what would be sold as Pentium M and later Core. Intel saw how well the Pentium M's were doing (namely, weren't stuck in an infinite loop of trying to get higher clockspeeds) and had them adopt their yet-to-be-released Core architecture to desktop, and later, workstation use. It worked. Intel released a product that caught virtually everyone (especially AMD) off guard: the Core 2.
Meanwhile, the research firm back in the USA was still trying to turn Nehalem into a feasible product. It took them 4-6 years but they did it. AMD, already shamed, was shamed again.
Morale of the story, Intel hit their stride and they are simply outspending and out-innovating AMD at every turn. AMD is, again, relegated to K5/K6 territory where they make and sell processors that are cheap. Let's just hope they don't fall back into reverse-engineering Intel products like they did back in the 80's and 90's. Note the clockspeeds a lot of Nehalem processors can achieve on air. If Intel feels threatened, expect those stock clocks to increase by at least 1 GHz (1/4 greater stock performance than available now). AMD has to not only match Intel, but even do one better to threaten Intel's position. Yeah, it's great if AMD can manage it but, stating the obvious here, it is a very steep hill to climb.
Oh, and AMD fell way behind not because Phenom sucked--just AMD's/IBM's 65nm process was absolutely hideous. People kept buying 90nm Windsor processors because 65nm Brisbanes (remember the introduction of .5 multipliers?) were coughing and choking to keep up with the Windsors, let alone a Core 2. I'm glad they put that bad chapter behind them. That was depressing. Three words: Moore's Second Law