Intel Core 2 Duo E4400 processor is a low-end product equipped, like E4300, with an "old" 800 MHz FSB, 2MB shared L2 cache, and it doesn't support virtualization. But! Its clock rate is 200 MHz higher than that of E4300. Because of this... It's also higher than the clock rate of Core 2 Duo E6300, the junior product of E6xxx series! Therefore, E6300 merely offers questionable (from the end-user point) virtualization support and 1066 MHz FSB. And E4400 has "just" 140 MHz higher clock rate. Actually, we can draw conclusions already now, but let's still see what test results are. Hardware and softwareTestbeds
* - "2 x ..." means per core Software
BenchmarkingEssential foreword to chartsOur test method has two peculiarities of data representation: (1) all data types are reduced to one - integer relative score (performance of a given processor relative to that of Intel Core 2 Duo E4300, given its performance is 100 points), and (2) detailed results are published in this Microsoft Excel table, while the article contains only summary charts by benchmark classes. We will nevertheless focus your attention on detailed results, when needed. 3D ModellingIn this article we are naturally interested in the results of E4300/E4400/E6300 processors, while other products are provided for reference (we have just added the results of E4400/6300 to the previous article charts.) And in the very first benchmark we see that faster bus loses to higher clock rate. CAD/CAEThis is just as in previous benchmark, and even the absolute points are the same. Digital Image ProcessingE4400 and E6300 earned 1 more point each, but the general picture remained the same: Core 2 Duo E4400 is still a confident leader among the lower-end trio. CompilingAt compiling E4400 even increased its result comparing to previous benchmark results. While E6300 still sticked to its favourite 104 points... Web ServerThis benchmark was a hope for E6300. In principle, given there are multiple simultaneously processed requests, a faster bus should have offered advantage. However, our hopes were not justified and the clock rate remained the critical argument. Synthetic testsIn CPU RightMark Core 2 Duo E4400 showcases 100% clock rate scalability: it's actually 11% higher than that of E4300. ArchivingAt last. We were already afraid that E6300 would make no score. However, there's a little fly even in this teaspoonful of ointment: in WinRAR E4400 still outperformed E6300 and only the considerable advantage in 7-Zip allowed the latter to score. It's very indicative: 7-Zip seems to be very critical about memory performance. OCRSame old, same old... Audio EncodingAn "olde" test group which nearly lost its importance today due to high result predictability. Video EncodingAnd again E6300 scores 104 points. It just can't break away from this value... GamesBoth processors were not perfectly scalable here. Even the E4400 rolled back from its usual 108-110 range down to 106 points. Well, games are more sensitive to GPU than to CPU, as we know. Total scoreThe total score doesn't tell us any news. Essentially, in all benchmarks except one the game was the same, so the final result was rather predictable. Supposed power consumptionAnd this is more interesting already. It seems that Core 2 Duo E6300 didn't get away with VT support and faster bus: its idle power consumption is much higher than that of E4300/4400. Speaking of 12W of the faster E4400 vs. 13W of the slower E4300, we shouldn't forget there are measurement errors. And it's not exactly CPU power consumption that we measure, as well... ConclusionsAs we have mentioned in the beginning, the competition of faster bus and higher clock rate is rather predictable, and within a single architecture the latter wins. But we still had some hopes that the lag would be slight... Alas, we can't say 4% is slight in the value CPU price segment, where buyers count money well, and the choice is dictated by price. If Core 2 Duo E4400 becomes more cheaper (and, judging by the product, this is actually possible), then E6300 would only be recommended for those in dire need of virtualization. For all other customers there would be no sense in overpaying for a 1066 MHz FSB, which is de facto useless here. There's actually a rather straightforward way out of this situation: to avoid competition between its own products, one of which is surpringly better than the other despite their market positioning, Intel can just quietly remove E6300 from the market. And leave E6320 instead, for example, with its doubled L2 cache. Will E6320 outperform E4400 armed with both faster bus and larger cache? An interesting question it is, and the answer is not that predictable already. We shall conduct the necessary benchmarking in the nearest future... Memory for testbeds provided by
Corsair Memory Russia Stanislav Garmatyuk (nawhi@ixbt.com) May 3, 2007 Write a comment below. No registration needed!
|
Platform · Video · Multimedia · Mobile · Other || About us & Privacy policy · Twitter · Facebook Copyright © Byrds Research & Publishing, Ltd., 1997–2011. All rights reserved. |