AMD vs Intel: A Linux Bout

← Back to Stories (view on slashdot.org)

Posted by CmdrTaco on Monday September 20, 2004 @07:32AM from the battle-of-the-titans dept.

CrzyP writes "AnandTech puts the latest and greatest AMD and Intel CPUs, including 32-bit and 64-bit versions, to the test in their first ever "Linux Desktop CPU Roundup" to see which performs the best in various Linux applications including database, compiling, rendering, encryption, and more. They suggest the Athlon 64 3500+ over the P4 560 for "balancing price and performance". Very informative!"

26 of 227 comments (clear)

Min score:

Reason:

Sort:

Wintel vs AMDnux ? by mirko · 2004-09-20 07:35 · Score: 3, Interesting

So, isn't there some Bias here ?
We associated so much Intel and Windows in the past that it now seems obvious that AMD is better for Linux ?
Just a question, I have not checked the thoroughness of these tests.

--
Trolling using another account since 2005.
AMD by yonatanh · 2004-09-20 07:35 · Score: 2, Interesting

I've always been an AMD fan, this just confirms my beliefs and makes my next processor being an AMD processor decision final. Anyone know why an AMD 2400+ would be running at over 50 degrees Celsius? Check the fan and the case is at around 30 degrees Celsius.
Hyperthreading by grub · 2004-09-20 07:36 · Score: 5, Interesting

So this latest benchmark suggests that HyperThreading doesn't do a whole lot. Is this the case on all unixy systems (ie: is HT geared more to Windows?) or is lacklustre performance on Windows the case as well?

I'm leaning heavily to the AMD 64 stuff for my next home unixy machine, any arguments for the P4?

--
Trolling is a art,
1. Re:Hyperthreading by Nos. · 2004-09-20 07:43 · Score: 4, Interesting
  
  No hyperthreading is hyperthreading. How the OS is written will make some difference, but HT is not some supper technology that makes slower processors act like faster ones. It just makes them a bit more efficient. Your better off going dual processors or to a faster processor. Of course, I'd agree with your last statment, go for the AMD64, according to everything I've read and heard, you won't regret it.
2. Re:Hyperthreading by stratjakt · 2004-09-20 07:51 · Score: 1, Interesting
  
  IIRC, Windows makes much better use of HT than Linux. Windows understands that there are two "virtual" processors, while linux treats them as two physical processors.
  
  The best argument for the P4 I can think of is the wide assortment of platforms, chipsets and motherboard vendors. The choices in that arena have historically been skewed towards Intel.
  
  That and AMD have been having some QC issues which seem to be getting worse and not better.
  
  I've leaned towards Intel in cases where I absolutely do not want to deal with some obscure hardware headache, particularly in the server arena. AMD's much improved I'm sure, but I've been burned in the past.
  
  Just don't be a dope and buy into the "this company is good, that company is bad" tripe that slashdot is selling. Companies are companies and both these guys want to squeeze the other out so they can gouge you and I more deeply.
  
  --
  I don't need no instructions to know how to rock!!!!
3. Re:Hyperthreading by Anonymous Coward · 2004-09-20 08:05 · Score: 5, Interesting
  
  No input re: M$ on HT'd cpu's.
  
  Under Linux I have seen various effects due to HT in the CPU. In the bad old (2.4 kernel) days dual Xeons would often run slower under relatively high load because the scheduler was agnostic of the fact that virtual cpu's share cache. I have also seen some sections of code which a)caches up nicely and b) is VERY heavy on FP math, give me about 1.8x the throughput by enabling HT. Seems the more a feature is hyped by marketing, the more likely it is that your mileage will vary greatly. Now if we could just get an OS/compiler that could use all the freakin' MMU's on a Power4 instead of just one .....
4. Re:Hyperthreading by arendjr · 2004-09-20 08:12 · Score: 2, Interesting
  
  It suggests HyperThreading doesn't do a whole lot on a single process not designed for multi-threading. That's quite a big difference. HyperThreading will give you some nice speed-ups when running multiple processes together.
  Furthermore, Linux actually works better with HyperThreading when you run a single multi-threaded program, the program will actually be scheduled to run on both cores. On Windows, you will only see an advantage when running multiple processes.
I love AMD processors by rhsanborn · 2004-09-20 07:39 · Score: 2, Interesting

I don't have an enormous pile of money to shell out all over. AMD has always done everything I need for significantly less than anything Intel has offered.
1. Re:I love AMD processors by Anonymous Coward · 2004-09-20 16:45 · Score: 1, Interesting
  
  Do me a favor and compare AMD's revenue's to Intel's. AMD's stock is at the same price it was 20 years ago, Intel's has grown 5000%. Now tell me who has the better business strategy?
Re:Linux Users Prefer Underdog Company by Acrodizer · 2004-09-20 07:42 · Score: 1, Interesting

exactly. Intel is far from the evil empire that many folks think. It's not fair to put them in the same boat as Microsoft. They make good products, but folks seem to care more about "supporting the little guy" - no matter who it is or how big that little guy is compared to #1.
The Price/Performace of a VIA C5 (or C7)..... by Powertrip · 2004-09-20 07:56 · Score: 4, Interesting

The Price/Performace of a VIA C5 (or C7)would be interesting to see here. No, I don't expect it to come even close in 'horsepower' to the players, but it would be of great interest for low-cost server appliances of sorts.
1. Re:The Price/Performace of a VIA C5 (or C7)..... by Anonymous Coward · 2004-09-20 08:40 · Score: 1, Interesting
  
  The advantage to the Via setup has very little to do with cpu performance. For what you'd use them for, enough is enough.
  
  They have a built in media decoder for your mpeg2/4 movies and such. Also they have a hardware-based encryption setup that stomps what even the highest end AMD cpu can do.
  
  So for small media players, and network routers that have to do VPN work (especially with the dual nic motherboards) the Via offers best bang for the buck.
  
  Low power, low cost, low heat. Low performance.
2. Re:The Price/Performace of a VIA C5 (or C7)..... by imroy · 2004-09-20 13:53 · Score: 2, Interesting
  
  I reckon what would be really interesting is to factor in the power requirements (including cooling) of these systems, especially in a cluster configuration. Sure, a bunch of Xeons or Opterons (or G5's) will give you great peformance. But how much are you paying to power AND cool those hogs? A rack-full will pump out enough heat to require a good air-conditioning unit. How does a cluster of mini-itx's compare then? Inquiring minds want to know, and I'm available to do the testing for anyone who wants to donate the hardware! :)
Re:AMD is the cheaper but which one is faster? by Anonymous Coward · 2004-09-20 08:04 · Score: 1, Interesting

I don't know, but I know the fact that lots of linux stuff (Wine comes to mind) won't compile under AMD64 right now don't affect the results.

Anandtech really has AMD's cock buried firmly in their ass, dont they?
Where's the 754s? by ameoba · 2004-09-20 08:08 · Score: 5, Interesting

Any article that bases its conclusions on price/performance ratios that leaves out AMD's socket 754 Athlon64s is overlooking a major contender. Socket 754 chips generally cost far less than the s939 ones at comparable speeds and with the current generation of chips the dual-channel memory that s939 offers doesn't provide that much of a performance boost.

Some might say that the s754 chips are an upgradability dead end but most people aren't upgrading CPUs without replacing the motherboard & RAM anyways. A s939 chip doesn't really get you much more upgrade headroom since there are no 939 boards with PCI-Express and DDR2 on them anyways...

--
my sig's at the bottom of the page.
Price/Performance by Slightly+Askew · 2004-09-20 08:19 · Score: 3, Interesting

They suggest the Athlon 64 3500+ over the P4 560 for "balancing price and performance".

Naturally, I didn't RTFA, but doesn't this suggest that I, as a geek who doesn't care about the value of my money, would get better performance with the Intel? Otherwise, they would just come right out and say that the AMD is the fastest of all processors, wouldn't they? I mean, I know that I would choose a DLP HDTV for "balancing price and performance", but that LCD is just so damn cool. Hell, I don't know, maybe I'll go read the article, but this sounds like some of that marketing speak we were recently warned about.

--
Public use of any portable music system is a virtually guaranteed indicator of sociopathic tendencies. -- Zoso
1. Re:Price/Performance by catenos · 2004-09-20 12:10 · Score: 2, Interesting
  
  They suggest the Athlon 64 3500+ over the P4 560 for "balancing price and performance".
  Naturally, I didn't RTFA
  
  I don't see anything natural about that.
  
  Yes, I am aware that you are referring to typical behaviour of the Slashdot croud, but the natural behaviour would have been to read the article, if you wonder about that statement.
  
  but doesn't this suggest that I, as a geek who doesn't care about the value of my money, would get better performance with the Intel?
  
  No. Just because an Intel processor isn't the winner in the performance/price class, it doesn't mean it's the winner in the performance class. The reasonable assumption you can make from the given statement, considering a performance/price statement was made, is that it isn't the Athlon 64 3500+ who was best in performance.
  
  And that's obviously true: Even ignoring how the Intel procs did, the other AMD Athlon's in the test, the FX-53 and the 3800+ repeatedly won against the 3500+, which is to be expected.
  
  As for AMD vs. Intel, they both won a share of the tests (IMHO the Athlons leading a bit) and it is not easy to declare a clear winner (for the 32-bit performance).
  
  Otherwise, they would just come right out and say that the AMD is the fastest of all processors, wouldn't they?
  
  You realize that it was the submitter's choice to emphasize a performance/price statement, do you?
  
  Hell, I don't know, maybe I'll go read the article, but this sounds like some of that marketing speak we were recently warned about.
  
  I am not sure what you are referring to.
  
  Note that despite the quotes signs, that "balancing price and performance" isn't even an actual quote (it's "balancing price with performance") and taken out of context, too: First they look at the prices and availablity and come to the conclusion:
  
  "Realistically, the Pentium 4 560 and the Athlon 64 3500+ are the best contenders in this match up. In six months when we run this shootout again we will likely be saying the same things about the Athlon 64 3800+. For now, however, the Athlon 64 3500+ does an excellent job of balancing price with performance."
  
  They could have chosen the Pentiums for the "in six month" example as well. In other words, it was more a statement about new, high-priced processors and current ones (and how in 6 month the sweet spot will have shifted to the next generation), than about an Intel vs. AMD preference.
  
  They state an AMD preference afterwards, but that's for their compelling performance with binaries compiled for 64-bit.
  
  --
  Keep an eye on which arguments are silently dropped in replies. Not always, but often times it's very telling.
Anandtech, AMD isnt the only 64bit x86 player by cwcpetech · 2004-09-20 08:20 · Score: 3, Interesting

I'd allow them to skip the Itanium processor line, but to skip the 64bit EM64T is quite odd for such a review. Handing the 64bit categories with no competition like that is almost a PR exercise. You might as well be throwing G5 performance numbers at x86 machines instead of consulting the POWER5 equivalents (closest competition).
Hyperthreading trivia and scheduling domains by j1m+5n0w · 2004-09-20 09:25 · Score: 2, Interesting

How the OS is written will make some difference
Slightly offtopic, but there was an article at linux weekly news awhile ago about a new scheduling algorithm for NUMA machines. They made an interesting point that load balancing between hyperthread cores (by migrating processes) is much cheaper than load balancing between separate physical cpus, since both hyperthread cores share the same L2 cache, and thus the process doesn't have to start over with an empty cache.
-jim
Re:publish your methodology by Wytter · 2004-09-20 09:40 · Score: 3, Interesting

Well, all of the benchmarks are available. To make it easier for people to run benchmarks on their computers and make more tech sites to GNU/Linux benchmarking, I am currently developing a LiveCD for benchmarking purposes (and of course the benchmark scripts will be usable by everyone as well as the programs for configuration etc.). http://linbench.sf.net
Still, remember the Amiga. by Spy+der+Mann · 2004-09-20 10:01 · Score: 3, Interesting

AMD is doing something else. "Word of mouth" can be VERY powerful today. But I bet they'd be doing LOTSA publicity if the internet hadn't arrived yet.

Remember what the lack of marketing did to the Commodore Amiga (with its powerful Video Toaster), when IBM only made bleeps and creeps.

Think about it.
Would be nice to see... by Warhaven · 2004-09-20 10:05 · Score: 2, Interesting

...a slightly larger roundup. xServe is supposedly the best bang for the buck when it comes to big cluster servers. How about: P4-, AMD-, G5-Linux roundup?
Re:AMD is the cheaper but which one is faster? by TrekCycling · 2004-09-20 10:24 · Score: 2, Interesting

I don't know about that last part, but I would agree regarding the overlooking of the obvious. i.e. That by and large the P4 just works whereas the AMD64 is more of a headache. I went with a P4 last week. Worked great. Changed my mind because of noise and heat and returned it for an Athlon 64. The system ran slow, filesystems were corrupted when I could get a clean install, which took numerous attempts and required me to pull the ram. In the end I went back to the P4. It just works.
Double bias test by Anonymous Coward · 2004-09-20 12:11 · Score: 2, Interesting

Reson Amd chips run better when kernel is built for amd. Intel chips are the same. Benchmark total screwed up. But no more bias than windows vs linux ones it is always the wrong kernel.

Rebuilding the kernel to match process can give upto 40% speed boast depending on the processor.

Poor windows users never see there processor chip work to its best. Hypertheading disabled ment that intel did not stand a chance since Hyperthreading is required to make up for there lack of general performace(linux kernel does support Hyperthreading if it is build right).

Ever wondered why you have to add drivers to windows to use hyperthreading yep windows does not support it at all.
Re:Linux: Intel vs. AMD vs. IBM by wobblie · 2004-09-20 12:56 · Score: 2, Interesting

I'd like to see something like this as well, with two incarnations:

1. a test based on what you could get for a certain amount of money

2. an all out test comparing the top of the line in each class
Re:64 bit compiling? by MrNemesis · 2004-09-21 05:16 · Score: 2, Interesting

Depending on the application, anything from nothing to alot ;)

Unlike the EMT64 Intel x86-64 processors, the AMD64's actually contain extra registers which only work on 64bit code, so be running 64bit code you get extra registers.

This really makes a difference with some CPU-heavy apps; a couple of examples that spring to mind are LAME and MySQL, which show performacne increases of ~30-50% vs. 32bit code on the same hardware. Not bad for a "free" upgrade :)

--
Moderation Total: -1 Troll, +3 Goat