Intel Dual Core Performance Preview Part II: A Deeper Look
by Anand Lal Shimpi on April 6, 2005 12:23 PM EST- Posted in
- CPUs
Multitasking Performance
As we discovered in the first article, multitasking performance requires a slightly different approach to benchmarking methodology. While for single application performance in which we test with a system that's in a very clean state with nothing but the benchmark and drivers loaded, for our multitasking tests, we have the system configured as what a real system would be. That means tons of programs and lot's of tasks running in the background. If you missed Part I, here's a quick recap of what our system configuration is like for our multitasking tests; the following applications were installed:
Daemon Tools
Norton AntiVirus 2004 (with latest updates)
Firefox 1.02
DVD Shrink 3.2
Microsoft AntiSpyware Beta 1.0
Newsleecher 2.0
Visual Studio .NET 2003
Macromedia Flash Player 7
Adobe Photoshop CS
Microsoft Office 2003
3ds max 7
iTunes 4.7.1
Trillian 3.1
DivX 5.2.1
AutoGK 1.60
Norton Ghost 2003
Adobe Reader 7
What's important about that list is that a handful of those programs were running in the background at all times, primarily Microsoft's AntiSpyware Beta and Norton AntiVirus 2004. Both the AntiSpyware Beta and NAV 2004 were running with their real-time protection modes enabled, to make things even more real world.
Multitasking Scenario 1: DVD Shrink
For this test, we used DVD Shrink, one of the simplest applications available to compress and re-encode a DVD to fit on a single 4.5GB disc. We ran DVD Decrypt on the Star Wars Episode VI DVD so that we had a local copy of the DVD on our test bed hard drive (in a future version of the test, we may try to include DVD Decrypt performance in our benchmark as well). All of the DVD Shrink settings were left at default, including telling the program to assume a low priority, a setting that many users check in order to be able to do other things while DVD Shrink is working.
Next, we did the following:
1) Open Firefox and load the following web pages in tabs (we used local copies of all of the web pages):
We kept the browser on the AT front page.
2) Open iTunes and start playing the latest album of avid AnandTech reader 50 Cent on repeat all.
3) Open Newsleecher.
4) Open DVD Shrink.
5) Login to our news server and start downloading headers for our subscribed news groups.
6) Start backup of Star Wars Episode VI - Return of the Jedi. All default settings, including low priority.
DVD Shrink was the application in focus. This matters because by default, Windows gives special scheduling priority to the application currently in the foreground (we will test what happens when it's not in the foreground later in this article). We waited until the DVD Shrink operation was complete and recorded its completion time. Below are the results:
The results here aren't too surprising. With dual core, you can get a lot more done at once, so the Pentium D 2.8 cuts the DVD Shrink encode time by about half when compared to the Athlon 64 3500+.
There is one element that caught us off guard, however. When looking at these numbers, we noticed that they were unusually high compared to the numbers from our first article. Yet, we ran and re-ran the numbers and had fairly consistent results. Even running the CPUs at the same speeds as in our first article yielded lower performance than what we saw in that piece. Comparatively, the processors all performed the same with reference to each other, but the DVD Shrink times were all noticeably higher. So, we started digging, and what we uncovered was truly interesting.
Facebook
Twitter
RSS
106 Comments
View All Comments
Umbra55 - Thursday, April 7, 2005 - link
Why do you still use the nForce 4 reference board for AMD tests? The real nForce 4 boards exist since over a quarter now and perform 10% better than the reference board. You also pretend to compare CPUs with similar price. Wouldn't it make more sense to compare combinations of mobo/CPU/memory of similar price?This is not a fair comparison (Or is it the intention?)
Umbra55 - Thursday, April 7, 2005 - link
yde - Thursday, April 7, 2005 - link
In this atricle, I was unable to find cache information about the AMD chip (usually equipped with a 512 KB L2 cache).To my point of view, in multiple threads scenario, the cache size may have dramatic influence and may explain several handicaps in the benchmarks. It would be nice to know what happens with a 1MB L2 cache Athlon 64 to keep things equal.
AMD chip has shorter branch prediction lines and seem quite well equipped for multitasking in theory, so why is it appearing so weak?
snorre - Thursday, April 7, 2005 - link
We're still waiting for proper benchmarks comparing dual core Smithfield with dual Opteron/Xeon. When will we see this?Comparing dual core CPUs with single core CPUs is like comparing apples and oranges, totally meaningless.
Calin - Thursday, April 7, 2005 - link
Thanks for the minimum frame rate comparison! And maybe you should not use so many flash-heavy pages, especially considering that you already told us that Athlon64 is much slower in Flash than Pentium4...Looks like Pentium D is a better choice in more ways than the Prescott is
xsilver - Thursday, April 7, 2005 - link
The multitasking gaming analysis combined with HT and NCQ was very insightful, almost pioneering...and with dual core AMD's people have to remember that the whole architecture is designed differently, and hence the possible suprises in tasks that the pentium D may not perform well on
but the power consumption advantage on amd is mighty tempting as amd will be the cpu that just keeps on saving -- with power bills that is
Azsen - Thursday, April 7, 2005 - link
Hi Anand,I was thinking, seeing most games are single-threaded you might like to try and benchmark this scenario on the dual core machine:
Set all the processes including operating system processes and other background processes to run on CPU 0 (first core). Then set the particular game to run on CPU 1 (second core). Have nothing else but the game running on CPU 1. This should dedicate a whole CPU core to the game for maximum performance in theory. I believe you can set the affinity in the task manager or use a batch file to do it.
Then run a Doom3 benchmark or HL2 benchmarks to see if the gaming performance is increased by letting the single-threaded game have a whole CPU core with no interuptions.
Would this be feasible to test?
Cheers. :)
AnnoyedGrunt - Wednesday, April 6, 2005 - link
The AMD not competing with the PD 2.8 will be due to price.Remember that AMD will need to sell a 2 core 2.2 GHz processor for about double the price of the single core version (maybe even more if yields are a problem). Therefore, you should expect a dual core 2.2 proc to be closer to $500-600 instead of the ~$300 processors tested here.
This also explains why AMD is focusing on the workstation market first. It's an area where price is typically not as much a factor, and where they are competing with Xeon prices, so it will be much easier for them to sell procs @ the higher prices.
I'm guessing that AMD's desktop dual core procs will start @ 1.6 or 1.8 GHz, and be priced on par with the Intel offerings. I think the overall performance will be similar to the current single core behavior, but now you can have more stuff running.
I currently have an AMD 939 3200+, so I am looking forward to their dual core offerings in a couple years, and hoping my mobo will be compatible. That would be very cool.
Also, I would like to add vote for a WoW test with Teamspeak running in the background as well as an istance if IE or Firefox with tthotbot.
Thanks,
D'oh!
SLIM - Wednesday, April 6, 2005 - link
Hi Anand,Any chance you could give us a little peak at what dual opterons on the nforce pro chipset can do in the same benches that you did in this article.
A configuration something like this:
Opteron 252s clocked at 1.8GHz, 2.2GHz, 2.6GHz
1 GB ECC DDR400
NVIDIA nForce pro motherboard
ATI Radeon X850 XT PCI Express
NCQ enabled HDD (maxline III, 7200.8, etc)
I looked through all the old articles I could find, but most seemed to contain only server oriented benches. I think that kind of article would be very enlightening as to the future performance of dual core athlons (probably within a couple percent)... maybe help put to rest some of the questions a lot of us have about waiting for amd dualcore, intel dualcore or just overclocking the heck out of a venice a64.
SLIM
Jep4444 - Wednesday, April 6, 2005 - link
SSE3 does almost nothing for the Venice which is barely faster than the Winchester(Xbit Labs benchmarked it already)While Anandtech said the dual cored A64s wouldn't compete with the Pentium D in encoding, unless they actually have dual cored A64s which they can't show us, i'd be willing to argue with that. Encoding is one of those things thats largely affected by Hyper-Threading and the Pentium D loses hyper threading. When the A64 goes dual, its performance increase will be larger than Intels.