Conclusion Conroe spanks AMD FX-62. Link to Hexus article
Just get the chip released. I want to build a new computer!
Just get the chip released. I want to build a new computer!
Follow along with the video below to see how to install our site as a web app on your home screen.
Note: This feature may not be available in some browsers.
MrGuvernment said:http://www.hardforum.com/showthread.php?t=1057259
Foir thoughts on the other review of the FX60
i love how they didnt trust Intel's tests and now their own are proving otherwise - Intel isn't stupid, they know people would get these chips in their hands and find out how they perform, so why lie.
ON that review, things seem rather evened out, conroe wins in Gaming the Am2 pretty much breaks even or just ahead in all else, and almost double in memory banwidth is sandra is coded right to read those results.
duby229 said:I realize this is the Intel forum, so I'll make this one post and shutup....
Those benches seem... wrong... FX60 pounds the 955... In those tests, it doesnt.... SOOOO I'm not quite sure whether to trust them or not.
It just seems to be a little off to me.
I think thats English slang, bring on Conroenobi125 said:Pretty sad that they can't even spell booty............."Botty" are you kidding me?
pay attention to the test like the gaming ones. This is why they are all over the place. the last two gaming tests would be probably video card limited.visaris said:- 38.02% ScienceMark 2.0 memory bandwidth
- 02.16% ScienceMark 2.0 memory latency
+ 20.89% HEXUS Pifast
- 12.57% HEXUS Cryptography
+ 19.09% Realstorm Raytracing 2004
+ 23.78% DivX encode - multithreaded
+ 24.50% WAV conversion - multi-threaded
+ 15.77% CINEBENCH multi-CPU render
+ 51.32% KribiBench v1.1 - Jetshadow model
+ 39.47% Far Cry - 1024x768 - speed
+ 03.67% Quake 4 - 1600x1200 - 4x AA 16x AF
+ 00.88% Splinter Cell: CT - 1600x1200 - 4x AA 8x AF
Those scores are all over the place. I really would like to see more testing.
When you average those results one could come to the conclusion that the Conroe E6700 is around 12.22% faster than the FX-62 over all.
Assuming linear performance increase with clockspead, the FX-64 will be around 7.14% faster than the FX-62.
This puts the Conroe E6700 at around 5.08% faster than the FX-64.
----
So, in summary, Cornoe is a very impressive chip. No doubt about it. Still, will it be enough to keep Intel in the lead for long? Intel says they have moved from a 4 year design cycle to a 2 year cycle. Can conroe last them 2 whole years agains AMD's 65nm (and K8L in around 9 months)? Discuss!
NulloModo said:For some reason they also gave the FX-62 system an extra gig of ram (running 2 gigs while the rest were all at 1) so that could have swayed things a bit more in that direction as well.
Also fun to note, this is the $500 E6700 trouncing a 1,000+ FX chip, just imagine what the Core 2 EE could do to the FX-62.
duby229 said:I realize this is the Intel forum, so I'll make this one post and shutup....
Those benches seem... wrong... FX60 pounds the 955... In those tests, it doesnt.... SOOOO I'm not quite sure whether to trust them or not.
It just seems to be a little off to me.
Hexus said:Conroe - we got two!
The formal release is at least a month off but we, being an inquisitive and resourceful bunch of hacks, have managed to procure a couple of Conroe samples to test in our own (unbiased) lab.
Least said:It's definitely impressive. Then again, a comparison between these two isn't exactly a "fair fight." The primary similarities are they are the flagship processors. One is a "new" design, the other is... not.
In any case, it'll be nice when AMD comes out with something genuinely new in their line to compete with the Conroe because well, the fx-62 ain't cutting it.
Yay for competition
That is very true.Donnie27 said:The fight [. . .] is between what ever is on hand.
These shouldn't be used in your average. These are the characteristics of a particular platform. Otherwise, we can throw in Conroe's L1 and L2 cache bandwidth and latencies too.visaris said:- 38.02% ScienceMark 2.0 memory bandwidth
- 02.16% ScienceMark 2.0 memory latency
These are clearly GPU limited.+ 03.67% Quake 4 - 1600x1200 - 4x AA 16x AF
+ 00.88% Splinter Cell: CT - 1600x1200 - 4x AA 8x AF
Using only the relevant benchmarks, it's over 20%.When you average those results one could come to the conclusion that the Conroe E6700 is around 12.22% faster than the FX-62 over all.
100% scaling with clockspeed is unlikely in most applicationsAssuming linear performance increase with clockspead, the FX-64 will be around 7.14% faster than the FX-62.
SLee said:These shouldn't be used in your average. These are the characteristics of a particular platform. Otherwise, we can throw in Conroe's L1 and L2 cache bandwidth and latencies too.
These are clearly GPU limited.
Using only the relevant benchmarks, it's over 20%.
100% scaling with clockspeed is unlikely in most applications
SLee said:These shouldn't be used in your average. These are the characteristics of a particular platform. Otherwise, we can throw in Conroe's L1 and L2 cache bandwidth and latencies too.
These are clearly GPU limited.
Using only the relevant benchmarks, it's over 20%.
100% scaling with clockspeed is unlikely in most applications
thecoldanddarkone said:I agree with everything this guy said
Well, based on AMD's and Intel's numerbing system, this is Intel's 8th generation x86 core beating AMD's 8th generation core.krameriffic said:And wait, what's this? A new generation, new architecture chip beating a new revision of an older chip? That's unpossible!!
krameriffic said:And wait, what's this? A new generation, new architecture chip beating a new revision of an older chip? That's unpossible!!
SLee said:Well, based on AMD's and Intel's numerbing system, this is Intel's 8th generation x86 core beating AMD's 8th generation core.
chrisf6969 said:Very true. Especially the memory bandwidth part. You can have 50% more bandwidth but often that only translates to a small improvement in overall performance. Thats why I've always gone for the highest CPU overclock and used memory ratios when necessary. Memory bandwidth is often overrated.
krameriffic said:And wait, what's this? A new generation, new architecture chip beating a new revision of an older chip? That's unpossible!!
freeloader1969 said:It amazes me how people are still comparing a new architecture to one that's three years old now. For Intel's sake, the Conroe should be 15 to 20% faster than any current AMD chip; they've had a few years to work on it.