AMD Launches Counterstrike Against Core 2 Duo

DigitalDame2 writes to mention a PC Magazine article about the AMD 4x4 enthusiast platform, which is meant to counter Core 2 Duo. The article observes that AMD is now facing many of the same business practices it used in its war against Intel. From the article: "While imitation may be the sincerest form of flattery, improvement can often be a slap in the face. Intel's C2D was designed with both low power and performance per watt in mind, two key design metrics that helped AMD cut into Intel's market share with the Athlon 64 and Athlon 64 X2. And, as preliminary numbers have indicated and final performance reviews now show, the C2D has learned its lesson well: its performance now tops AMD's Athlon 64 architecture by a substantial margin."

Comments

4X4 is more a marketing ploy than anything else

[Harry Balls]

4X4 sounds more like a marketing ploy to me than like a feasible solution for Joe Average or even Joe Gamer.

Why?

Consider the cost of Athlon X2 processors:
http://www.pricewatch.com/cpu/442067-1.htm
The least expensive Athlon X2 costs a cool 300 bucks, while the mid-range Core 2 Duo (Conroe) E6600 costs $315 (projected wholesale price).

Now factor in a more expensive (because of 2 processor sockets) 4X4 motherboard, two Athlon X2 chips at $300, and you wind up with a $350 to $400 surcharge for being an AMD fanboy.

The situation gets worse if you want a high-end system:
Two FX-62 will set you back $1045 + $1045 = $2090
http://www.pricewatch.com/cpu/992212-1.htm
and while this combination is expected to outperform a single Core 2 Duo at $1057
http://froogle.google.com/froogle?q=E6800&btnG=Sea rch+Froogle&lmode=online&scoring=p
factoring in the more expensive two-socket motherboard expect to pay a cool $1100 more than for the E6800 system.

Personally, I'll probably buy an E6600 ($315) or an E6400 ($240) as soon as they become available.

Re:4X4 is more a marketing ploy than anything else

[TTK Ciar]

I am a software engineer working at The Internet Archive, and I write parallel software every day (sometimes with PVM for "real" applications, but more often as throwaway perl slammed out on the command line, using open3() to open several simultaneous subprocesses, sometimes fed data by the parent but more often each reading from a different data file). Much of what I do is "trivially parallelizable", meaning it's pretty easy to make scale across multiple processors or machines. It is my impression that most real-life problems seen by most businesses are trivially parallelizable, with the rare exceptions hogging all the attention by dint of being more interesting.

My workstation is a single-processor machine, but I have at my exclusive disposal a dual-xeon machine and two AMD dual-core machines. I'm always scp'ing my work up to them from my workstation so I can take advantage of their multi-process goodness. (Developing while ssh'd into those machines is usually not a good idea, since the network likes to go down or slow down a lot between Archive HQ and our datacenters, and our HQ firewall blocks PVM so I can't just make my workstation the PVM master node with the other three machines slaves.)

When I read this article, my initial reaction was "Enthusiasts, hell! I want as many of these as I can get for servers!" (assuming this 4x4 product is significantly cheaper than current dual-opteron products -- we're a non-profit, without a lot to spend on hardware, and we're always running on the edge of starvation. But maybe that's a bad assumption and these will be prohibitively pricey).

If someone offered me a 4x4 or 8x8 for my desktop, though, I'd accept it gladly, and make good use of it, parsing/analyzing Archive metadata, processing multiple simultaneous http streams (we use a lot of http-rpc here, and xml data representation which means each http-rpc stream can suck down a lot of processing power), md5'ing multiple files in parallel, and the like. I'd probably also make more extensive use of bzip2 than I do currently :-)

My datasets commonly consist of hundreds or thousands of files, each of which can be processed in parallel, so I can keep throwing cores at the problem with near-linear scalability until I grind against disk or bus bandwidth limits (at which point the data needs to start out distributed in order to keep scaling).

Just my $0.02

-- TTK

Performance improvement?

[Atroxodisse]

Not if you ask hard OCP http://enthusiast.hardocp.com/article.html?art=MTE wOCwxLCxobmV3cw==/

Re:Part of the vicious cycle in Tech

[Uryene]

...A cycle I learned my lesson about many, many moons ago.

At home, I keep a $640 check I wrote back in 1990 for a 486 CPU.
It's framed and visible on top of a bookcase to serve as a reminder.

At the time, I thought it was a great deal; screaming processors were
never going to get much cheaper than that!

These days, last years tech (or even two years ago tech) is usually
MORE than sufficient. Except for games, which always seem to
need NEXT years processor in order to be playable... ;-)

Re:"well.. my dad can beat up your dad!"

[uop]

Well, having the strongest offering usually does count for something (especially if you ask the marketeers).
The 4x4 initiative basically looks like DP for the desktop, which Intel offers as well (although Xeon only).

imho, the really interesting thing about 4x4 is the possibility of plugging in a coprocessor in the future.
For example, you may settle for a single Athlon64 X2 in a 4x4 board for now, and add a physics/video/dsp/whatever coprocessor in the future.
That's wild speculation, of course, but it does make the 4x4 setup intriguing as a future-proof product.

--

Everyone not getting it

[charnov]

Okay, I'll give it the Slashdot norm, but nobody gets what this is. It a hypertransport socket for not just another CPU, but ANYTHING you would want to connect directly to memory and CPU. No PCI or other slow bus.

There are already Xilinx cards available because this has been used in Cray supercomputers for a while (the Opteron ones anyways). This means AMD can counter ANYTHING Intel puts out because you can just slap a $20 speciality DSP on the mobo which could easily be 100x faster than that Intel chip at whatever small set of functions it needs. Video cards are already in the works for this along with all kinds of audio and video stuff. I seem to remember one manufacturer has a RAID processor. The possibilities are endless.