The Stock and Overclock Tests
To make our performance data as useful as possible, we decided to run the stock 4200+ and the overclocked CPU through our standard motherboard test suite. As Anand has already shown in the launch article for AMD Dual-Core these are not the most revealing tests you can run on dual core. However, the test suite is a fair representation of the types of applications you run today, and it is a set of benchmarks that are very familiar to regular readers. This makes it a revealing set of tests for dual-core and overclocked dual-core performance.Performance tests were run at stock speed with the 4200+ and at the highest stable overclocked speed that could be achieved with basic air cooling on our test bed. For comparison the same suite of benchmarks were run using the single core 4000+ CPU. In looking at comparisons in the graphs, keep in mind that the 4000+ runs at 2.4GHz versus 2.2GHz for the 4200+. The 4000+ also features 1MB L2 cache compared to 512KB cache on each core in the 4200+.
The test configuration is our most recent DDR memory test bed built around the DFI LANParty nF4.
| Performance Test Configuration | |
| Processor(s) | AMD Athlon64 x2 4200+ (2.2GHz 512KB cache each) 939 AMD Athlon64 4000+ (2.4GHz 1MB Cache) Socket 939 |
| RAM | 2 x 512MB OCZ PC3200 Platinum Rev. 2 |
| Power Supply | OCZ 520 watt PowerStream |
| CPU Cooling | Thermaltake Silent Boost K8 HeatSink/Fan |
| Hard Drive | Seagate 120GB 7200 RPM SATA (8MB Buffer) |
| Video AGP & IDE Bus Master Drivers | nVidia nForce 6.39 |
| Video Card | nVidia 6800 Ultra (PCIe) |
| Video Driver | nVidia nForce 71.89 |
| Operating System | Windows XP Professional SP2; Direct X 9.0c |
| Motherboard | DFI LANParty nF4 SLI-DR |
For CPU cooling we used the same Thermaltake Silent Boost K8 HSF fan we have used for recent overclocking tests in motherboard reviews. In looking at our overclock numbers, keep in mind that cooling is just a decent stock Heatsink/Fan. Higher overclock will be achieved with more aggressive cooling like liquid cooling or phase-change.
Facebook
Twitter
RSS
53 Comments
View All Comments
val - Thursday, June 23, 2005 - link
also consider another thing. Many games and applications are never alocating all available memory, usually when they are well programmed, they are alocating for example 75-80 of available main memory and managing post loading of levels by own alghorytms. Testing now on 1 GB makes in many games no sense. Those games (for example battlefield 2, GTA,...) takes about 600 MB of RAM when you have only 1 GB but they takes with same settings up to 1 GB when you have two. Result? Less loading, no lags, smoother gameplay, higher performance.Here i am not calculating the case that windows can manage file cache so nicely when there is enough memory. In CS:S is loading another level than damn fast - result? You are first on to choose the side to play and you dont look half minute to loading screen.
Gamer should never have less than 2000MB of ram when it costs damn <200€!
Gatak - Thursday, June 23, 2005 - link
I feel something is very strange with the 3D benchmarks. I thought that Maya, 3D Studio and SPECViewperf were made to support dual CPU systems - yet we see little improvement here.Is really dual CPU enabled in those applications?
ElFenix - Thursday, June 23, 2005 - link
what core is the 4000+?val - Thursday, June 23, 2005 - link
Wesley, it is not truth that playing games are single threaded. Problem is that you are using only demo plays and not real gameplay.Internet play is hardly multitasking and can slow down single cpu lot, same for high quality sound card. Audigy drivers are nice multitasking and needs lot of CPU.
You do not calculate with time needed for system managing the swap file and dynamic memory alocations, also very often and also nice multitasking.
In real world, everybody is able to see difference. I am now not talking only about best frames per second (who cares them) but about minimal, about that lags, short drops, loading level times and so on!!!
Who ever had dual cpu + at best dual scsi hdd machine, newer wants to go back. Fact that AMD is not able to produce at reasonable price and only trying us to think that we are buing PCs for benchmarks instead of hard gaming only shows they PR as BS.
Many of us even likes to have some applications running while gaming, ever heard about VoiceIP tools, email applications?
Thanks to intel to understanding that two slower cores are much better than useless single core PCs at any speed. At least in windows.
Wesley Fink - Thursday, June 23, 2005 - link
Houdani - We really want to do an article exactly as you describe, but we also have extremely limited Dual-Core samples. We have asked AMD for a 4400+ and 4800+ for comparison, but until supplies ease up a bit we don't have the processors for such an article.On page 1 the table specifies the speed, cache, and expected price of the x2 CPUs. We realize vendors are charging more in this early going, but that is nothing unusual. Our p.1 prices are where prices will settle after the first adopter flurry.
yacoub - Thursday, June 23, 2005 - link
#20 - Totally agree. For maybe $50 more you can get double the cache but does it overclock as well and what sort of performance gains does that net at equal overclocks?Houdani - Thursday, June 23, 2005 - link
This article almost started out as a "which X2 is the best value" piece, but turned into an overclocking article for the 4200 without any real comparison for the performance of the 4200 vs 4400 vs 4600 vs 4800 vs Cost.I suppose we could probably fill in the blanks by looking at the previous article by Anand and compare it to this article by Wesley, but we'd still be left with trying to extrapolate the performance of the 4600 and 4800 at stock settings.
Any chance we can get an article which puts all four of the current X2 processors on the same set of graphs with a performance per dollar blurb thrown in for good measure? Something like that would help the folks who are on the fence about spending the extra money for the 4400 over the 4200.
Yeah, yeah, the easy answer it to spend the extra money for the extra cache of the 4400 -- after all, you're spending $550 to $600 already so another $50 likely won't break your bank. Even so, it'd still be nice to see the performance numbers to make our wallet feel better about itself. :)
KristopherKubicki - Thursday, June 23, 2005 - link
dumbnewbie: Yonah launches Q1'06, but Intel has plans for small factor motherboars one quarter later. So I would optimistically say Q2'06.Kristopher
fishbits - Thursday, June 23, 2005 - link
"$500 would get you started on an Intel i955 motherboard and cpu, so why would you invest in the end-of-the-road on AMD socket 939 and old DDR1? It's crazy."Because it outperforms the Intel option and doesn't necessitate a new motherboard like the Intel does?
Why shouldn't folks take advantage of the upgrade capacity of 939 and its compatibility with X2? Or this i955 you speak of, you mean to say no one should ever upgrade the CPU they put in it, no option but 1 cpu per mobo ever? All chips/mobos are by definition "end-of-the-road?" If so, you'd be equally compelled to say Intel dual core and i955 is a futile path. If not, then you wouldn't see any problem with people upgrading their CPUs to X2.
Tell you what, just buy your Intel dual core and (try to) be happy with it. In fact, their low-end DCs are pretty nice for the price... if you didn't have to buy a new mobo just to get in the door and can put up with relatively anemic performance in some areas. But even buying a new 939 board to get to X2 today, you're going to get an across-the-board powerhouse that'll last anyone 2 years and beyond depending on their upgrade strategies, especially with the OC results we've seen today. What's not to love if you can or will need the power and it fits in your budget?
Viditor - Thursday, June 23, 2005 - link
There are a few reviews out there already that compare a dual cpu vs dual core...http://tinyurl.com/8dt7r