Dual Core Linux Performance: Two Penguins are Better than One
by Kristopher Kubicki on July 1, 2005 5:55 AM EST- Posted in
- Linux
Multitasking Scenario 1: DVD Transcoding
Ripping DVDs on Linux is one of the applications Linux really does well at - if you don't mind getting past the pretty GUI of DVD Shrink or a similar application. Ripping DVDs in general is fairly common for any desktop regardless of OS, and checking mail while ripping a disc can inflict havoc sometimes. The following test attempts to mirror Anand's Windows benchmarks on Linux and is very comparable to the Windows alternatives.
- Open FireFox 1.0.4 and load all 5 web pages. We chose 5 pre-downloaded pages from various sites like AnandTech, CNN and ESPN
- Open XMMS and start playing a Nine Inch Nails CD ripped to Ogg
- Open Thunderbird for news
- Login to our news server and start downloading headers for our subscribed news groups
- Open dvd::rip
- Start backup of Star Wars Episode VI - Return of the Jedi. All default settings
Again the emphasis is not particularly whether or not one system is capable of finishing the DVD backup faster than the other, but whether or not the system is still useable while doing so. We might be pushing the load of our system a bit far - but a machine that can rip DVDs fast but can't play music at the same time is effectively useless to us. Like Anand's test, the vob files were already placed on the hard drive to eliminate the DVD drive bottleneck. We have instructed dvd::rip to reduce our vobs from 7.5GB with transcoding to 4.5GB.
Although the load from our other application is not substantial, the combination of several low intensity applications and the introduction of a maximum load application (dvd::rip) end up bringing our single core platforms to a halt. The results of this benchmark were eerily similar to Anand's Windows benchmark, and the same culprits are pretty much to blame. Even if Linux might have a leg up on Windows' scheduler, no amount of clever preempting can beat a second processor core.
Facebook
Twitter
RSS
69 Comments
View All Comments
JGunther - Friday, July 1, 2005 - link
They're going for $660 on ZipZoomFly, which is sold out again, but they were in stock yesterday.All I'm saying is, look around. Other review sites have come to the same conclusions: the AMD dual cores are better than their Intel couterpoints. Better deals? No. But this is not a "budget Linux dual core system" article, so let's leave the "best bang for your buck" talk for another time. I'd only ask that it be a performance-oriented review, like it claims to be, instead of him choosing to compare CPUs at only the price-point where Intel comes out on top. Cutting one CPU out because it's just $60 too much seems a bit arbitrary and biased to me.
I tend not to get worked up about the whole AMD vs. Intel stuff-- my laptop at home is runnning a Pentium 4 in it. But when I clicked on this article, and saw who wrote it, I knew EXACTLY which CPU was going to be at the top of the charts. That should never happen. That's all.
Tegeril - Friday, July 1, 2005 - link
I think the compiling graphic is simply in error. Why would the 660 have a-j3 score better than the D 820. That doesn't make much sense. I'd bet the X2 is supposed to occupy that space and the 660 and 640 got shifted down by accident.mikellpp - Friday, July 1, 2005 - link
Now why do you suppose there are no -j3 compile times for any of the Athlons in the last compile test? The author sooo wants intel to look good, he just can't bear to see AMD's better times. Are we supposed to be stupid readers?Tegeril - Friday, July 1, 2005 - link
Ugh "None of his [dual core] CPUs tested..." is what I meant.Calin - Friday, July 1, 2005 - link
Why the Athlon64 X2 has no -j3 compilation benchmarks?Also, in short, one can compare the Intel D840 and the AMD X2 as they cost about the same. And not everyone will keep their mainboards, as not everyone upgrades every six months. I think most buyers will buy both a new mainboard and a new processor. I would have thought of buying myself an Athlon64, but I would like an Socket939, and I think their price (compared to the 754) are highway robbery.
I'll wait with the upgrade
Tegeril - Friday, July 1, 2005 - link
JGunther.None of his CPUs tested topped $600. Newegg has the 4400+ for $719. That is not in the same price range. He drew no unfavorable conclusions, finding AMD fantastic in gaming and showed its dual core option that competes with the D820 and D840 also performs well in multitasking environments.
Your high performance quote is meaningless when you look on the "the hardware" page and read: "These processors are a bit on the high end"
Stop creating controversy where there is none.
JGunther - Friday, July 1, 2005 - link
#19, If he wanted to do some equal price comparisons, there's also the 840 EE vs. the 4800+. Those should have been tested as well-- the article's talking about "high performance Linux machines", so the most prudent thing to do would be to show which CPU actually gives the best performance and let the reader know the respective prices, so that they can come to their own judgements.If Kris wants to nail home the point that doller for dollar, the Intel CPUs are the better deal by omitting benchmarks of competing, though higher priced, CPUs, let him do it on his own time, on a site that doesn't have the same integrity and non-bias standards as Anandtech should.
There is only one dual core AMD chip being compared to two dual core Intel chips, despite the fact that the 4400+ is only a marginal increase in price. I suspect this is to keep his benchmark graphs more Intel-heavy near the top, but if it's because he simply doesn't have a 4400+ (they're in stock at Newegg and ZipZoomFly), as #20 suggests, then he should have waited to post the article until he had a more balanced roundup of CPUs.
Yeah I know my words are harsh, and I'm sorry. But this isn't the first time Kris has done this. It's irresponsible 'journalism' and reflects poorly on the site, and users *should* speak out against it.
TheJet - Friday, July 1, 2005 - link
One thing to note, I usually run -j3 on my _single_ processor Celeron 466 *NIX workstation and get significantly better performance than a standard make. So really, performance numbers should have been shown for -j3 across the board, not just for the dual core parts.Furen - Friday, July 1, 2005 - link
He has mentioned (repeatedly) that he doesnt have an X2 4400 and that Anand is the one who has the X2 4800, so he had to make do with what he had. I, personally, like the article (though I hate pentium Ds ^^), just wish there was more data on some of the tests, heh.IKeelU - Friday, July 1, 2005 - link
#17, He's comparing simiarily-priced CPUs. Would it be more fair to compare a $1000 AMD CPU to intel's $558?