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:

LOSTCIRCUITS	SHORTCUTS: A Higher Bus Frequency Test Configurations Memory Subsystem Power Consumption 3D Rendering Energy Efficiency Cinebench AV Encoding: DVD-Shrink, DrDivX, MainConcept AV Encoding: Virtualdub / DivX and SSE4 Gaming: 3DMark '06 FarCry F.E.A.R. The Down and Dirty Dis-Illusions Give Us Some Feedback on this Review
Intel's Yorkfield QX9770 at 3.2 GHz I have the Power!!
(Review by Michael Schuette, November 25, 2007)

Memory Performance

We used SiSoft Sandra to measure memory bandwidth available to the CPU. Increasing the bus speed to 1600 MHz with a concomittant increase in memory frequency is guaranteed to increase the available bandwidth, as long as everything else remains equal. In this case, all we did to accomodate the QX9770 was to physically swap CPUs and manually configure the bus to a 400 MHz clock with the memory in 1:1 mode for 1600 MHz data rate on both buses.

Buffered Memory Bandwidth. Compared to the QX9650, the QX9770 shows a significant increase in memory bandwidth. In fact, it is the highest memory bandwidth result we have ever seen on any Intel architecture system. Buffering was enabled to allow "intelligent prefetching as it is enabled by Intel's smart memory management strategies.

As mentioned, the QX9770 - even on an "overclocked" P35 chipset-based platform achieves record benchmark results for any Intel platform. Using the QX9650 at the same settings gave the same results within the margins of error.

L2 Cache Latencies

Although it may appear trivial since at superficial glance the QX9650 and the QX9770 should be identical, we thought it was necessary to make sure that this is really the case, at least on the level of cache access times or latencies - whichever metric is preferred by the reader

L2 Cache access latency cycles depending on stride length: lower is better! Series 1-3: 2048 KB block size, series 4-6: 4096 KB block size. We show the access latency cycles for the QX6850, the 9650 and the 9770. To make a long story short, the 9650 and the 9770 are identical, wich is clearly overt by the merging of the yellow and red plot lines to a combined orange throughout most of the data range.

(BX80557E6300)

next page: => The Power Paradox =>