Intel Core 2 Extreme QX6800: The Fastest Desktop CPU, now with more cores
by Anand Lal Shimpi on April 9, 2007 12:59 PM EST- Posted in
- CPUs
General Performance
Although our usual general performance suites aren't available under Vista, something we often use in our motherboard reviews is: PCMark '05. PCMark is a somewhat unusual benchmark as it is a synthetic performance test that tries really hard to be applicable to the real world.
The tests used by PCMark '05 and the applications they are based off of are listed in the table below:
We focused on two of the PCMark tests: the system test suite and the CPU test suite. The system test suite is composed of the following tests:
And the CPU test suite consists of the following tests:
We reported the overall PCMark and CPU test scores.
PCMark has a handful of heavily multithreaded tests which obviously favor more cores, hence the three quad-core offerings dominating the charts here. Moving down the list however we see that the X2 6000+ manages to come in just under the E6600 in performance, but remains ahead of the E6400. The 5600+ remains ahead of both the E6400 and E6300, its two chief competitors based on price.
Isolating the CPU score itself doesn't change things much, although AMD looks a tad better in some of the results.
34 Comments
View All Comments
Sc4freak - Sunday, April 15, 2007 - link
It's one of the few games out there that benefit greatly from multi-core. It would have been interesting to see how this new CPU benefitted one of the most CPU-bound games out there right now.SilverMirage - Wednesday, April 11, 2007 - link
AT fails to be completely honest with the situation:1. AT conveniently publishes this on the exact day AMD's price cuts come into effect. That's interesting. Although AT mentions this, they could have mentioned that AMD's previous prices were not able to compete.
2. Benchmarking the 5000+ against the e6300 is inherantly biased since it will be the e6320 which is contending with the 5000+
"5000+ will have to contend with the E6300"
3. Now this depends a lot on the mobo, but I'd say that the conclusion from these benchmarks is that the E6320 and E6420 will be better for their price in a week or two.
he Athlon 64 X2 6000+ is a realistic alternative to the E6600/E6400, the 5600+ competes well with the E6400/E6300 and the 5000+ can hold its own against the E6300/E4300
(AT fails to mention again that the E6300 is an unfair comparison)
duploxxx - Monday, April 16, 2007 - link
In the beginning of C2D launch we saw many reviews reducing the multiplier to see what 4mb of cache could do against 2MB of cache... it was only a few % depending on the type of apps... so stop the crap that a 6320 will outperform a 5000, same with the e6400 vs 5600. because for sure it will not!As for power consumption, yes a K8 consumes more power at load, It also consumes A LOT LESS in idle, how long is you're system idle a day? And buy a normal ATI chipset like the asus M2R32-MVP and the total power consumption at load will be less than the C2D system...
any system can be oc'ed.... one bether than the other. you are talking about 5% of users maximum. the allendale tend to oc worse these days......
nice review but start using the ATI chipsets also, they are equal performers against NVidia and cunsume a lot less and also cheap these days.
yyrkoon - Wednesday, April 11, 2007 - link
It is funny, you come here saying the Anandtech crew is full of BS, yet you do not bring any proof with you, so excuse me if I call BS on you. Things do not magically work one way, instead of another, JUST BECAUSE *you* say so.DeepThought86 - Tuesday, April 10, 2007 - link
I don't understyand why only the CPU prices are considered?? Shouldn't the overall cost including a motherboard be a much more realistic measure? What about a performance/overall (CPU+MB) cost metric be very usefulGriswold - Tuesday, April 10, 2007 - link
I just skimmed over the article, but where are the numbers for power usage?RedWolf - Tuesday, April 10, 2007 - link
One thing that is in AMD's favor is that Dell is now selling AMD machines. The college I work for is buying all AMD machines this year. Even the slowest C2D machines are a few hundred dollars more than AMD machines. All of our machines this year, including laptops, are Athlon X2 powered machines because the price was so attractive. I simply could not configure a C2D machine that came close. That price difference allowed us to go to 2 gb of ram and still be under C2D pricing for the same machine. Granted we aren't building enthusiast machines or buying for business but we are buying AMD and getting them at great prices.dm - Monday, April 9, 2007 - link
It is important to note that Intel is also coming up with a better Intel® Core™ 2 Duo E6300 and Intel® Core™ 2 Duo E6400, which are Intel® Core™ 2 Duo E6320 and Intel® Core™ 2 Duo E6420. Both have full 4MB L2 cache and will be a lot better performer. I have done quite a few tests with them here (and it includes Intel® Core™ 2 Duo E4400 as well):
http://fanboyreview.blogspot.com/2007/03/brag-fanb...">http://fanboyreview.blogspot.com/2007/0...g-fanboy...
You missed a wonderful processor, which is the quad core Intel® Xeon® X3210 (2.13GHz/8MB L2/1066MHz) which is an LGA775-socket compatible CPU and would appear to be binned to worked at a lower voltage. And according to guru3D (http://www.guru3d.com/newsitem.php?id=4949)">http://www.guru3d.com/newsitem.php?id=4949) the price will be hovering the $430 range. I have done some testing with this Intel® Xeon® X3210 here:
Part I (Stock Benchmark): http://fanboyreview.blogspot.com/2007/04/article-l...">http://fanboyreview.blogspot.com/2007/0...icle-lit...
Part II (Overclocked up to 63%): http://fanboyreview.blogspot.com/2007/04/article-l...">http://fanboyreview.blogspot.com/2007/0...e-little...
Anyway, overall, nice article!!!
skrewler2 - Tuesday, April 10, 2007 - link
Doing a google search, I see the price is around $750-800.. Too bad, you got me excited tooyacoub - Tuesday, April 10, 2007 - link
DM I didn't see any head-to-head comparison of 6300 to 6320 and 6400 to 6420 in your review of them. Am I reading your graphs wrong?