Navigate: Advice Beginners BIOS Guide CPUs Links Mainboards Memory Network Storage Video/Sound Cards Contact Forum SiteMap Sponsors WebNews Home	.	.
Prices: Mainboards ABIT ASUS Chaintech Shuttle Soyo Tyan CPU Intel P4 2.4C-800 P4 2.6C-800 P4 2.8C-800 P4 3.0-800 P4 3.2-800 AMD AthlonXP XP 1700+ XP 2000+ XP 2400+ XP 2500+ XP 2700+ XP 3000+ XP 3200+ Athlon64 Athlon64 3200+ Athlon64 FX-51 Opteron Opteron 240 Opteron 242 Opteron 244 Opteron 246 Memory Corsair Crucial Kingston Mushkin OCZ Search Prices:

After the mental exercises of the last few pages, the rest should be fairly easy. We ran a number of benchmarks and they all can be categorized into two classes, that is, either, they show an advantage of HT, or they don't. In the extreme case, there is a performance hit but it is usually small enough to be able to live with. One example for the latter is Content Creation Winstone2002:

ContentCreation Winstone2002

Even though some of the applications used for CCWS2002 are capable of multithreading, it always depends on the respective workload whether HT can actually take advantage of the capabilities of the program. So far, there is some evidence that CCWS can take advantage of SMP but it is only marginal anyway and, moreover, the multitasking nature of CCWS, even though multitasking is one of the fortes of HT also has the tendency to really hammer the CPU resources. Add the kernel overhead of the OS for HT or MP and CCWS is a good candidate for a performance hit in an HT environment.

ContentCreation Winstone2002
The HT disabled P4 3.06 outperforms the 2.8 GHz counterpart while HT enabling causes a drop-off to the last place in this chart.

MainConcept

At the beginning of this review we mentioned that the optimal environment to benchmark the effect of HT would be a multithreaded application which, by nature of its own would encompass identical runtimes for the individual threads and at the same time combine lighter with heavier workloads, e.g. the swiss cheese and the holes therein. One such application is MainConcept MPEG encoder 1.3 which can be used similarly to Ulead MediaStudioPro or Adobe Premiere to convert AVI files into MPG or DVD formats by simultaneously encoding audio and video streams. Granted that this application was recommended by Intel, it still shows some rather hefty performance benefits for HT.

Runtime in seconds, lower is better: We used a 50 MB AVI file that we converted into DVD format. The application is capable of taking advantage of SMP as well as HT and shows a 22% performance gain from enabling HyperThreading.

As always, we could add tons of additional benchmarks but who would really care, besides, our goal has been to dissect the performance profiles of HT in the various conditions any modern PC can encounter.

next page:    => Conclusions =>