AMD Dual-Core Performance Revealed

Timmus writes "In two separate articles, FiringSquad takes a look at the performance of AMD's dual-core Opteron CPU. The first article examines the performance of dual-core in scientific computing applications (MATLAB and LS-DYNA) as well as digital photography, while the second story focuses on the performance of dual-core Opteron paired against Intel's dual-core Pentium Extreme Edition in video encoding, Cinebench, and a few other applications. The performance improvements are pretty impressive in multi-threaded applications that take advantage of the technology."

Comments

full article mirrors

[winkydink]

here and here.

Chance for someone to karma whore...

[Amiga Lover]

OK. Anyone have a quick simple explanation of why Dual Core over Dual CPU motherboard? are there inherent advantages to dual CPUs so close together?

1 Dual Core vs Dual CPUs?

[uofitorn]

Wouldn't a better benchmark be to compare a dual core setup to a similarly configured dual processor workstation?

Re:1 Dual Core vs Dual CPUs?

[timeOday]

I think the benchmark for comparison should be whatever else you can get at a similar price.

Dual processor motherboards and CPUs were never priced to make them attractive for widespread use, whereas dual core chips supposedly will be. We shall see.

The simple future

[caryw]

Why stop at dual core?
Once a way to link multiple cores of a CPU is firmly implemented scaling the chip to 4, 8, or even 32768 cores should be relatively easy.
With chip dies getting smaller and smaller the only real reasons not to continue this multi-core scaling would be physical space and power usage.
Perhaps they could scale multiple cores vertically instead of just making the chip wider and longer.
And perhaps the cores could only be "turned on" when called for instead of using up juice all the time.

Interesting look at the future of chips.
Sony's Playstation 3 is using a "cell processor" or similar multi-core design that has already been covered here in the past.

Arstechnica article on the cell processor here.
--
NoVA Underground: Fairfax County, Loudoun County, Arlington, Price William chat and local forums

Re:The simple future

[jd]

IIRC, the latest generation of Sun UltraSPARCs has 6 CPU cores. An alternative approach is to have "virtual" cores - have a stack of registers and pools of computational elements. This does require some extra element of sophistication, to share out resources, but if you have two programs with very different CPU needs, both programs should run faster. Also, if you have fewer programs than there are virtual cores, but instruction parallelization can be performed, you still get a speedup.

The idea of turning off parts of the CPU would work, if you have a large enough cache. What you would need to do is prefetch all possible paths far enough ahead that you could turn on any deactivated part of the CPU before the instruction needed to be executed. You then have an independent "monitor" processor (an MPU?) which purely scans the cache and turns off all elements on the CPU that aren't needed within the lifetime of any of the contents of the cache.

Another poster noted the bandwidth issue between processor and main memory. That is certainly a problem, but one that may be fixable. One way is to sped up memory (and the bus). The other is to look at ways of reducing the amount that needs to be transferred, by putting some of the CPU in memory. (The technique is called "Processor-In-Memory", and has been around for about 10-15 years.)

Re:As for gamers (from TFA)

[millennial]

I suppose that makes sense. The question this raises, though, is whether there are any games designed to work better on hyperthreaded/multiprocessor systems.

Wow!

[truesaer]

This is pretty disasterous for Intel. The game benchmarks show significant performance penalties for dual core chips, as expected. Intel launched its dual core specifically as an Extreme Edition for games.

On other benchmarks the AMD dual core gets 10-20% better performance! SiSoft Sandra is an exception, where there is a mixed bag between the two processors.

This pretty much verifies for me that Intel did a seriously rushed cludge to get this thing out the door. The only reason I can think of to target this to gamers is that no OEMs would want to buy them for server or desktop use, so you have to target people who like the latest technology even if it isn't that great.

AMD on the other hand seems to have a pretty good product here. I can't wait until the desktop versions come out.

Re:Wow!

[Anonymous Coward]

My understanding of Intel's hyperthreading is that it is essentially an intelligent fast context switcher running in hardware. It still has to swap out registers and empty or suspend instruction queues. This means that threads are not actually executed simultaneously.

Nope. Hyperthreading can handle instructions from two different threads at the same time. If one thread stalls on a memory access, that leaves more execution units available for the other thread. It's especially important for Intel because they put their memory controller way off in a separate chip, so their memory latencies are horrible.

Anyone realize that suddenly all P4's disappear?

[denominateur]

Now, it's struck me as very peculiar that the benchmarks where the dual dual core setup from AMD really shines leave out any comparison whatsoever to the Intel dual-core offering. This begs the question whether the person doing this review is a journalist or a marketing represenatative of AMD.

"We did not have time to evaluate the Intel platform with the Intel MKL, the P4 3.0GHz is an older reference measurement." is a very cheap excuse and indicates either lazyness or bribes on the side of AMD... I hate hardware review sites!

Re:As for gamers (from TFA)

[99BottlesOfBeerInMyF]

I suppose that makes sense. The question this raises, though, is whether there are any games designed to work better on hyperthreaded/multiprocessor systems.

I very much doubt it. I've always thought of Blizzard as being one of the better companies when it came to "doing it right" with regard to coding their games. I know playing Warcraft III it always consumed 100% of one processor and did not put a dent in the other. I have not noticed any games that do a better job.

Re:As for gamers (from TFA)

[DigitumDei]

Depends I guess. I know I don't have the luxury of keeping my gaming machine seperate from all other applications I use, so my gaming machine is also my work machine and it tends to have a lot of stuff running at any given time. Now when playing shooter games I often notice a sudden drop in fps when some service or other decides it needs to do something. A dual core machine would be a lot less prone to this I guess.

Also, from the article. "And although the company says dual-core isn't for gamers quite yet, perhaps it is, only in a different usage model. Alan Dang and I were discussing processor benchmarking moving forward and he came up with the idea that we don't run compute-intensive tasks in the background today because we think they can't be done. However, if a dual-core processor enables a DVD encode while you're playing Half-Life 2: Deathmatch, there's a good chance that the way we think about demanding tasks may change. Even though games aren't currently threaded, the background processes a dual-core processor enables may very well catapult the technology into favor with game enthusiasts."

Re:All windows, all the way.

[bersl2]

Well, at the very least, Anand put some Cygwin/gcc tests in there.

Most Windows users don't know what it means to multitask. It's much harder to do when you don't have multiple desktops, virtual or otherwise.

Re:OK then.

[Bwian_of_Nazareth]

Oh my, here it comes again. Comparing pears to apples. If I install MS Windows, what do I get? Operating system with a few (let me say - lousy) applications. If I install Linux distro of my choice, what do I get? Depending on my choice, it can be a full blown suite of application ranging from development to office apps to video processing.
And further more, e.g. KDE has been quite successfull at speeding up between 3.2 and 3.4. I am not so sure about the memory print, but that is no concern for me today (RAM is abundant).

Re:As for gamers (from TFA)

[i.r.id10t]

IIRC Quake3 used separate threads for sound and video.

In my own experience with both a dual P2-450 system and a dual AMD 1.2ghz system is that the game will run on one cpu and the OS will use the other... simply balancing the load between both processors. Quake2/3/UT/ET ran on the dual 450s with comparable oomph to a single 1ghz system.

Re:As for gamers (from TFA)

[Mancat]

That just means that WC3 is not multithreaded.

Re:As for gamers (from TFA)

[tokabola]

The reason games are written without multithreading is because most games are written with consoles in mind. While the PS 2 has a type of parallel processing, neither XBox nor Nintendo support any kind of SMP.

The next generation of consoles from Sony and MS are supposedly going to be fully SMP capable, so game developers will start taking advantage. That will make multithreaded ports to PC a no-brainer.

There's nothing about gaming that makes multithreading less usefull - in fact the need to run a real time physics engine as well as an AI make most games excellent candidates for multithreading. The only reason it hasn't been done is because most hardware (console and PC) doesn't support it and it doesn't make sense (financially) to write a second version of the game that can, just for the 30 or so people who could actually use it. As dual core cpu's, along with the PS 3 and XBox 2, become more common, the software will begin taking advantage.

Re:OK then.

[r_naked]

As for the "more than one CD" complaint, Linux distributions these days come bundled with about all the software there is. Windows XP doesn't even include a C compiler as far as I know.

I am sure Microsoft would love to include a stripped down version of VS.NET with XP. With all the grief they are given for including a web browser, what kinda headaches do you think they would get if they started shipping a compiler?
I hear compliants on here all the time about how you get so many more applications with a Linux distribution, and in the same breath how Microsoft should remove things from their OS.
I would LOVE to see Microsoft be allowed to ship more software with the OS. A compiler, an office suite (not just wordpad and Outlook Express), a decent DVD authoring package, something better than MSPaint etc...etc... but people would scream UNFAIR!
To me unfair is Microsoft trying to compete with free when the free side of things doesn't have the same restrictions. Hell even Apple ships most of the popular GNU software with OSX.
Please, someone tell me how that is fair. Did anyone stop and think that maybe if Microsoft didn't have to keep hacking their OS to make various groups / government bodies happy that the OS might be more secure? I mean they already use BSDs ftp (and probably several other BSD utils), it would be real nice if they could have thrown pf in instead of developing their own (potentially less secure / less configurable) firewall.

Re:OK then.

[Waffle Iron]

My system becomes very sluggish if I do something like md5sum an ISO. However, my CPU utilization is not maxed, and lowering the process priority doesn't seem to help. If I let it run by itself and then come back to the machine when it's done, I see all of the apps and window manager have been totally swapped out and will take many seconds just to repaint the first time I clear the screensaver.

This tells me that the sluggishness is due to the interaction between virtual memory and heavy I/O activity, not CPU load. Another CPU probably isn't going to help at all in this situation.

Re:OK then.

[Jim_Maryland]

I have to admit, XP doesn't crash very often compared to previous MS releases. I have had it crash/lockup though due to applications and the only thing to do is power it down. I can't recall seeing a BSOD with it yet. On the other hand, my Solaris systems for the last 7 years and the DEC Ultrix systems for 3 years prior to that have had better reliability with zero OS freezes/crashes. I have had applications hang, but they haven't hung the OS in the least.

Microsoft has to overcome years poor stability reputation among IT professionals. Linux will likely face the same issue. The major difference I see though between the MS and Linux systems is that when I pay for the OS, I'd expect better quality, just because I'm paying for it. I'm probably a bit more forgiving to the Linux system crashing than a Microsoft OS.

As for the hardware comments, vendor support for the dominant desktop OS makes sense. Unless the hardware vendor sees a particular reason to release drivers simultaneously for multiple operating systems, I would expect MS to be the first driver released.

If you're looking for hardware to install into your PC, then how about showing me where I can get MS Win32 drivers for Sun GigaSwift PCI Ethernet Card. Likely this piece of hardware will only provide Solaris drivers. Note: I'm only showing this as an example to disprove your statement. Generally speaking though, you are correct about the hardware.

Re:one step closer to the day

[advocate_one]

Pah... I can do that already... I've got 4 linux boxes networked and three of them are being driven using X over ethernet... so basically, yes, I am doing all four tasks at once on the same display... however, they are being done on four separate computers...