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
xtknight - Friday, July 1, 2005 - link
Why was FireFox 1.0.2 used on Win32 and 1.0.4 used on Linux? Just wondered.KristopherKubicki - Friday, July 1, 2005 - link
Something is wrong with the graph on the Compilation test (the -j3 for the dual core Athlon was in the Pentium 660 slot). We only ran -j3 on the dual core chips. I redid the graph and it should be rendering correctly now.Kristopher
suryad - Friday, July 1, 2005 - link
#34 great question. I was about to ask that but you beat me to it.smn198 - Friday, July 1, 2005 - link
#22 Give him a break - The removal of the 2nd xbox vs ps article meant that this had to come out a day early.Kocur - Friday, July 1, 2005 - link
Well, Kris, do you have any theory of why single core Athlons seem to perform better with regard to single core P4s than X4200+ with regard to dual P4s?
In my opinion there are two possibilities.
1. HT slows both 640 and 660.
2. X4200+ does not stretch its wings.
In my opinion you should have adressed this problem in the article as we have learnt to expect the opposite.
Kocur.
n yusef - Friday, July 1, 2005 - link
I tink tis article is fair. The Pentium D 820 is good for the price, if you don't already have a S939 mobo and RAM. It's not as ceap as you would think, becuase if the extra ~$100 for more expensive RAM, motherboard, and aftermarket HSF, but it's still cheaper for a whole system. For someone like myself who already has a S939 mobo, and DDR1 RAM, the X2s are a better bargain. The make -j3 tag should have been on all of te CPUs (it even makes non-HT single core CPUs faster), especially the 4200+.When you get a chance, please get a 4400+, or lower your 4800+s (Anand's actually) multi to x11 so we can see how much cache affects a dual-core CPU (each core is sharing the same memory bandwidth that only one had to itself before, so even an A64 might be bandwidth limited).
TheMatt - Friday, July 1, 2005 - link
One comment I always have with these tests is why do you never see pbzip2? On *any* dual, quad, etc. machine I use, I always pester the admin to install pbzip2: http://compression.ca/pbzip2/ .I was hoping I'd see it here to see if it gives the same speed-up with dual-core as it does with my SMP machine. I'd suggest to anyone here to try it out, it's a great program.
fishbits - Friday, July 1, 2005 - link
"If you can't admit that the D820 is a good performer at its price, then I don't know what's going on with you, but nowhere did he slam AMD along the way."The D820 is really good for the price, especially if ram/mobo weren't issues. But saying you "pay through the nose" for the AMD chip and not the D840 at virtually the same price was pretty stunning.
"In our opinion, the Pentium D 820 is really an underdog in this roundup"
What the heck does that mean? The 820 looks like a really good multi-tasking performer for the dollar. How does that make it an "underdog?"
We've gotta give the guys a little slack on various errors, but in "Final Thoughts?" Should the concluding paragraphs where judgement and recommendations are handed out be read two or three times before releasing them? Or maybe by at least a second set of eyes?
JGunther - Friday, July 1, 2005 - link
#29Gaming bechmarks are the ONLY benchmarks in this article where the Pentium D doesn't occupy the top spot. Find me another review site that shows this to be the case.
The D820 is definitely a good performer for its price, I'd never say otherwise. But there's no denying that Kris and his articles are slanted. There are no slams, but little statements like this:
"At $558 you pay through the nose for the additional performance of the Athlon 64 X2 4200+."
Attempt to portray AMD as the culprit, but the fact is, you pay through the nose for that additional performance whether you're looking at the 4200+ or the D840 (the latter of which is is actually the more expensive of the two).
Anand once said that visitors to his site, he found, were about 50/50 Intel or AMD users. Though statistically, it could go either way, there is no doubt in my mind that Kris' primary computer at home runs an Intel chip. It's written all over his articles.
Tegeril - Friday, July 1, 2005 - link
I don't see his analysis showing the Pentium CPUs "on top" at all. He specifically states that the AMD offering beats the pants off of Intel as usual in gaming, that the memory costs on the Intel boards are higher, and that the D820 and D840 don't suck quite as much as people want it to. He simply says that at the price point of $252, the D820 becomes a tempting option.To sacrifice minimal overall performance and save $290 - memory costs is probably a very good suggestion and is warranted by the analysis. It's pretty impartial from here.
If you can't admit that the D820 is a good performer at its price, then I don't know what's going on with you, but nowhere did he slam AMD along the way.