Conroe 64-bit numbers?

[LstBrunnenG]

Hey, I'm just curious if any reviewers did any extensive 64-bit testing of Conroe vs the K8, seeing as these numbers are left out of even [H]'s expansive set of reviews. I saw one site with some numbers, but it simply compared it with other Conroes and earlier Xeons, so that wasn't mch help.

Some really stalwart AMD zealots have been saying all along that EMT64 was inherently inferior and just a clone of AMD64, that EMT64 on Core 2 would suck as much as it did on Netburst because it was based on Yonah, which was a 32-bit notebook chip, and so on and so forth. Thus, I was wondering whether they had been proven right or wrong.

If I may throw out something for any potential debate, it would matter less to me whether Conroe "gains" going from 32-bit to 64-bit mode as much as whether or not it was faster than the other guy's chip. I know little about the specifics of such things, but couldn't the fact that AMD64 gains by going from 32-bit mode to 64-bit mode mean that AMD's 32-bit mode is less efficient? If I'm wrong on that point, feel free to educate me.

 

[Serge84]

Hey, I'm just curious if any reviewers did any extensive 64-bit testing of Conroe vs the K8, seeing as these numbers are left out of even [H]'s expansive set of reviews. I saw one site with some numbers, but it simply compared it with other Conroes and earlier Xeons, so that wasn't mch help.

Some really stalwart AMD zealots have been saying all along that EMT64 was inherently inferior and just a clone of AMD64, that EMT64 on Core 2 would suck as much as it did on Netburst because it was based on Yonah, which was a 32-bit notebook chip, and so on and so forth. Thus, I was wondering whether they had been proven right or wrong.

If I may throw out something for any potential debate, it would matter less to me whether Conroe "gains" going from 32-bit to 64-bit mode as much as whether or not it was faster than the other guy's chip. I know little about the specifics of such things, but couldn't the fact that AMD64 gains by going from 32-bit mode to 64-bit mode mean that AMD's 32-bit mode is less efficient? If I'm wrong on that point, feel free to educate me.

Conroe? Whaaa... Whats that... whatever it is we will never get one...

 

[brucedeluxe169]

http://www.xbitlabs.com/articles/cpu/display/core2duo-64bit.html

Core 2 would suck as much as it did on Netburst because it was based on Yonah, which was a 32-bit notebook chip, and so on and so forth. Thus, I was wondering whether they had been proven right or wrong.

K8 is based on K7.... which was also a 32-bit chip, so that arguement is invalid.

Conroe's 64-bit performance is just as good as AMDs

 

[Donnie27]

Hey, I'm just curious if any reviewers did any extensive 64-bit testing of Conroe vs the K8, seeing as these numbers are left out of even [H]'s expansive set of reviews. I saw one site with some numbers, but it simply compared it with other Conroes and earlier Xeons, so that wasn't mch help.

Some really stalwart AMD zealots have been saying all along that EMT64 was inherently inferior and just a clone of AMD64, that EMT64 on Core 2 would suck as much as it did on Netburst because it was based on Yonah, which was a 32-bit notebook chip, and so on and so forth. Thus, I was wondering whether they had been proven right or wrong.

If I may throw out something for any potential debate, it would matter less to me whether Conroe "gains" going from 32-bit to 64-bit mode as much as whether or not it was faster than the other guy's chip. I know little about the specifics of such things, but couldn't the fact that AMD64 gains by going from 32-bit mode to 64-bit mode mean that AMD's 32-bit mode is less efficient? If I'm wrong on that point, feel free to educate me.

http://techreport.com/reviews/2006q3/core2/index.x?pg=1

OS
Windows XP Professional x64 Edition
Windows XP Professional with Service Pack 2 (WorldBench only)

All benchmarks except Worldbench were performed with WinXP 64bit. Conroe gains about 2% and A64 gains about 5%. That just means instead of getting beat by 30% in 32bit mode, it will get by 27% instead

 

[brucedeluxe169]

shut up donnie

*what a troll*

 

[Donnie27]

shut up donnie

*what a troll*

Hahaha!

 

[drizzt81]

http://techreport.com/reviews/2006q3/core2/index.x?pg=1


All benchmarks except Worldbench were performed with WinXP 64bit. Conroe gains about 2% and A64 gains about 5%. That just means instead of getting beat by 30% in 32bit mode, it will get by 27% instead

Your math is wrong.:

1.30 * 1.02 = 1.326
1.00 * 1.05 = 1.05

(1.326 - 1.05 )/1.05 = .263 -> 26.3%

Let's just remember that most of the population sucks at percentage calculations. I know that you knew better than to do simple subtraction, which is fine for a quick estimate, but we don't want the general population to think that that is the correct way.

 

[Donnie27]

Your math is wrong.:

1.30 * 1.02 = 1.326
1.00 * 1.05 = 1.05

(1.326 - 1.05 )/1.05 = .263 -> 26.3%

Let's just remember that most of the population sucks at percentage calculations. I know that you knew better than to do simple subtraction, which is fine for a quick estimate, but we don't want the general population to think that that is the correct way.

Thank you for that correction Now that will make a great deal of difference.