AMD's Quad FX: Technically Quad Core
by Anand Lal Shimpi on November 30, 2006 1:16 PM EST- Posted in
- CPUs
More Sockets, but Lower Performance?
When AMD briefed us on Quad FX, the performance focus was on heavy multitasking (AMD calls this "Megatasking") or very multi-threaded tests. We figured it was an innocent attempt to make sure we didn't run a bunch of single threaded benchmarks on Quad FX and proclaim it a failure. Given that the vast majority of our CPU test suite is multi-threaded to begin with, we didn't think there would be any problems showcasing where four cores is better than two, much like we did in our Kentsfield review.
However when running our SYSMark 2004SE tests we encountered a situation that didn't make total sense to us at first, and somewhat explained AMD's desire for us to strongly focus on megatasking/multithreaded tests. If we pulled one of the CPUs out of the Quad FX system, we actually got higher performance in SYSMark than with both CPUs in place. In other words, four cores was slower than two.
| CPU | SYSMark 2004SE | Internet Content Creation | Office Productivity |
| 2 Sockets (4 cores) | 261 | 373 | 182 |
| 1 Socket (2 cores) | 288 | 393 | 211 |
You'll see that in some of the individual tests there is an advantage to having both CPUs installed, but in the vast majority of them performance goes down with four cores. It turns out that there are two explanations for the anomaly.
| CPU | Internet Content Creation | 3D Creation | 2D Creation | Web Publication |
| 2 Sockets (4 cores) | 373 | 245 | 514 | 411 |
| 1 Socket (2 cores) | 393 | 364 | 453 | 369 |
First, in Internet Content Creation SYSMark 2004SE, there appears to be an issue with having two physical CPUs in the system that results in the 3dsmax rendering test only spawning a single thread, lowering performance below that of a normal dual-core processor. This problem may be caused by a licensing violation within the benchmark where it is expecting to see one physical CPU with multiple cores and isn't prepared to deal with multiple CPUs. Regardless of the exact cause of the problem, it doesn't appear to be anything more than a benchmark issue. It's the performance in the Office Productivity suite that is far more worrisome because there is no issue with the benchmark that's causing the problem.
| CPU | Office Productivity | Communication | Document Creation | Data Analysis |
| 2 Sockets (4 cores) | 182 | 171 | 259 | 137 |
| 1 Socket (2 cores) | 211 | 187 | 285 | 176 |
The Office Productivity suite of SYSMark 2004SE wasn't the only situation where we saw lower performance on Quad FX than with a single dual core setup. 3D games seemed to suffer the most; take a look at what happens in our Oblivion and Half Life 2: Episode One tests:
| CPU | Oblivion - Bruma | Oblivion - Dungeon | Half Life 2: Episode One |
| 2 Sockets (4 cores) | 67.3 | 78.3 | 155.8 |
| 1 Socket (2 cores) | 75.2 | 90.9 | 165.7 |
Once again, populate both sockets in the Quad FX system and performance goes down. The explanation for these anomalies lies in the result of one more benchmark, CPU-Z's memory latency test:
| CPU | CPU-Z Latency (8192KB, 128-byte) |
| 2 Sockets (4 cores) | 55.3 ns |
| 1 Socket (2 cores) | 43.3 ns |
With both sockets populated, memory latency goes up by around 27% and thus in applications that are more latency sensitive and don't necessarily need all four cores, you get worse performance than with a single dual-core CPU. The added latency comes from the additional probing over the HT bus that's done for coherency whenever a memory request is made to see where the latest copy of the data resides.
It's a problem that will go away if you have a single quad-core CPU with one memory controller, but one that makes Quad FX a tougher pill to swallow compared to Intel's quad-core offerings.
Facebook
Twitter
RSS
88 Comments
View All Comments
JarredWalton - Friday, December 1, 2006 - link
Yay for speech recognition!indigenous = enthusiast
we are see = we are seeing
Scratch a couple "basically" from the last paragraph.
yyrkoon - Friday, December 1, 2006 - link
I though Gary was the "one" with the Texas "Twang" ;)laok - Thursday, November 30, 2006 - link
The current 4x4 does not look to be a mature system to me. Wait until 65nm 4x4 comes out and hopefully a better chipset will be available at that time. 65W x 2 is reasonable, 130W x2 is kinda too much.And I also want to know how 4x4 compares to dual dual-core opteron with the same frequence: performance, power consumption etc.
DigitalFreak - Thursday, November 30, 2006 - link
It's the same chip as the Opteron, with the memory controller changed to work with unbuffered memory.JarredWalton - Friday, December 1, 2006 - link
The first word of his subject is the best advice: WAIT! Even if the future might get better, does anyone want to spend $1000+ on what may or may not turn out to be the better platform? When the new CPUs are available, then we can evaluate and decide. Of course, once AMD launches their quad core processors, I'm almost certain that our advice will be that most people only need a single socket motherboard and CPU anyway -- if that. Many people still get by a single core CPU, and the number of people that actually need more than dual CPUs is very small, at least in the desktop workspace.DigitalFreak - Thursday, November 30, 2006 - link
This is the best AMD can do against Kentsfield? They get stomped on in every benchmark, cost more, and draw nearly twice as much power. No one in their right mind would buy this over a Core2Quad. Whomever came up with this product should be fired.The last time I laughed this hard at a CPU/platform launch was when Intel rolled out the P4 dual core CPUs, and at least they came out on time in some of the multimedia benchmarks.
For the record, my current system is running an Opteron 165, so I ain't no Intel fanboy.
photoguy99 - Friday, December 1, 2006 - link
You're right it's actually embarrasing isn't it?If AMD's next-gen architecture improves performance by 30% at same clock, which is huge, they still won't take the lead.
And it seems Intel is done sitting on their hands, they are working like hell to dominate again by the time K8L ramps up big.
People hate hearing this but think it's over for AMD.
And like you, for the record, my current system is an FX-60 so I'm also no Intel fanboy.
DigitalFreak - Thursday, November 30, 2006 - link
...came out on top...photoguy99 - Thursday, November 30, 2006 - link
Seriously, I'd like to know who is selling them...Furen - Thursday, November 30, 2006 - link
That such a badly engineered product was rushed out to reviewers just to have a paper launch. Did AMD believe that no one would make a big deal about the power draw? Or maybe it expected no one to even look at power draw. I was actually impressed by what AMD had accomplished with 4x4, after all, the 3.0GHz Quad FX parts were close to the QX6700, until I saw the insane power draw. Two loaded FX-62 systems (whole systems, mind you) draw about the same power as 4x4 does IDLE!