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: SMP and the New Intel Kentsfield Key Features Power Management and Cache Questions Test Setup, Benchmark Overview Power Consumption Memory System 3D Rendering: TrueSpace POV-Ray 3.7 Cinebench 2003 DVD Shrink, Dr DivX & Mainconcept H.264 Futuremarks FarCry, DOOM3 F.E.A.R. Prey, Call of Duty2 Final Thoughts Give Us Some Feedback on this Review
Intel's Core 2 Quad Extreme Edition QX6700 Codename "Kentsfield"
(Review by MS, November 1, 2006)

Memory Performance

Raw memory bandwidth has not been the strongest point of the latest Intel chipsets and processors. Where the P4 ExtremeEdition 3.73 scored an impressive 7.1 GB/sec bandwidth using prefetch and buffering in SiSoft Sandra, no subsequent Intel Netburst processor got even close to this number and the Core2 family did not improve in this respect either. On the other hand, it is not scrolling through the pages that counts, on the contrary, memory disambiguation in combination with overlapping commands a.k.a. Intel Fast Memory Access is the secret weapon that - unfortunately - escapes the metrics of essentially all current memory bandwidth benchmarks. It is, therefore, not surprising that the QX6700 does not shine in Sandra.

SiSoft Sandra Buffered Memory Bandwidth

Memory bandwidth: higher is better

Memory and Cache Access Latencies

Memory access latencies: lower is better. We compared the Core2 Duo E6700 (dark edges) with the Core 2 Quad QX6700 (white edges). The single die version is slightly faster across the bench. Note that the cache size for both CPUs comes out the same, despite the 8MB spec, 8 MB blocks cannot be cached. As mentioned, the reason is that the 8MB total L2 cache refers to two independent and separate "non-shared" blocks of 4 MB cache per die that are only communicating via the system bus.

(BX80557E6300)

next page: => 3D Rendering Performance =>