Sign-in / Sign-up
Your question

amd 64*2 3800+ or intelcore2duo which is better

Tags:
  • CPUs
  • Performance
  • AMD
  • Intel
Last response: in CPUs
December 18, 2006 1:16:25 PM

i wanted 2 buy pc which is going to be better
amd 64*2 3800+
or
intel core2duo
which has better performance

More about : amd 3800 intelcore2duo

December 18, 2006 1:19:32 PM

depends on which C2D, but they all destroy the 3800+, so the core 2 duo i guess.
Related resources
December 18, 2006 1:29:47 PM

Try using the [search] feature, before posting....

Did it ever occur to you, this may have been talked about before?

Sorry - I get flashbacks from my relatives bothering me, becasue they are to lazy to do a little research....
December 18, 2006 2:29:07 PM

As you've seen, the new 65nm X2 are more energy efficient than any thing intel has to offer (if you'd like to wait a little for retail ones). I'd go for the X2 3800 just for sake of having great 64-bit potential on windows Vista.
Right now the Core 2 suffers a ~16% performance penalty when running in 64-bit mode while the K8 suffer only a ~6% on average running in 64-bit mode. The greater drop in the Core 2 is because the Macro/Micro OP fusion is a 32-bit exclusive feature. This is something often overlooked by most tests because they only test 32-bit apps. However Vista will be the first true 64-bit OS by Microsoft (Server 2003 x64 and XP x64 were really just hacks) Microsoft also plans to allow people to upgrade to the 64-bit version of the OS during its life time meaning that Microsoft has great faith that some point in the near future (~2yrs) the majority of system will be sold as 64-bit by default. This would mean that suddenly older K8 and newer K8L will suddenly preform much better compared to the Core 2, because these chips do not have the same limitations as the Core 2. Intel might be able to hack something together to make the Macro/Micro OP fusion work in 64-bit mode, but from what I read on the design this is highly unlikely. Nothing short of a brand new core would work to fix this. But by this time AMD should be fully into a modular CPU (K10) design which will render convential benchmarks useless. The best that Intel will able to do is have the best "well rounded" chip which is an oxymoron.
December 18, 2006 3:07:08 PM

Quote:
As you've seen, the new 65nm X2 are more energy efficient than any thing intel has to offer (if you'd like to wait a little for retail ones). I'd go for the X2 3800 just for sake of having great 64-bit potential on windows Vista.
Right now the Core 2 suffers a ~16% performance penalty when running in 64-bit mode while the K8 suffer only a ~6% on average running in 64-bit mode. The greater drop in the Core 2 is because the Macro/Micro OP fusion is a 32-bit exclusive feature. This is something often overlooked by most tests because they only test 32-bit apps. However Vista will be the first true 64-bit OS by Microsoft (Server 2003 x64 and XP x64 were really just hacks) Microsoft also plans to allow people to upgrade to the 64-bit version of the OS during its life time meaning that Microsoft has great faith that some point in the near future (~2yrs) the majority of system will be sold as 64-bit by default. This would mean that suddenly older K8 and newer K8L will suddenly preform much better compared to the Core 2, because these chips do not have the same limitations as the Core 2. Intel might be able to hack something together to make the Macro/Micro OP fusion work in 64-bit mode, but from what I read on the design this is highly unlikely. Nothing short of a brand new core would work to fix this. But by this time AMD should be fully into a modular CPU (K10) design which will render convential benchmarks useless. The best that Intel will able to do is have the best "well rounded" chip which is an oxymoron.


Sharikou? Is that you?

I'm not even going to respond to this.
December 18, 2006 3:13:38 PM

Quote:
As you've seen, the new 65nm X2 are more energy efficient than any thing intel has to offer (if you'd like to wait a little for retail ones). I'd go for the X2 3800 just for sake of having great 64-bit potential on windows Vista.
Right now the Core 2 suffers a ~16% performance penalty when running in 64-bit mode while the K8 suffer only a ~6% on average running in 64-bit mode. The greater drop in the Core 2 is because the Macro/Micro OP fusion is a 32-bit exclusive feature. This is something often overlooked by most tests because they only test 32-bit apps. However Vista will be the first true 64-bit OS by Microsoft (Server 2003 x64 and XP x64 were really just hacks) Microsoft also plans to allow people to upgrade to the 64-bit version of the OS during its life time meaning that Microsoft has great faith that some point in the near future (~2yrs) the majority of system will be sold as 64-bit by default. This would mean that suddenly older K8 and newer K8L will suddenly preform much better compared to the Core 2, because these chips do not have the same limitations as the Core 2. Intel might be able to hack something together to make the Macro/Micro OP fusion work in 64-bit mode, but from what I read on the design this is highly unlikely. Nothing short of a brand new core would work to fix this. But by this time AMD should be fully into a modular CPU (K10) design which will render convential benchmarks useless. The best that Intel will able to do is have the best "well rounded" chip which is an oxymoron.

Got links? Proof? Benchmarks? Empirical evidence that proves this? If not, than this is just empty banter.
December 18, 2006 3:14:19 PM

Core2 Duo
December 18, 2006 3:31:50 PM

Quote:
Sharikou? Is that you?

I'm not even going to respond to this.

Nope, it is LameNoobMike.
December 18, 2006 3:44:29 PM

Quote:
As you've seen, the new 65nm X2 are more energy efficient than any thing intel has to offer (if you'd like to wait a little for retail ones). I'd go for the X2 3800 just for sake of having great 64-bit potential on windows Vista.
Right now the Core 2 suffers a ~16% performance penalty when running in 64-bit mode while the K8 suffer only a ~6% on average running in 64-bit mode. The greater drop in the Core 2 is because the Macro/Micro OP fusion is a 32-bit exclusive feature. This is something often overlooked by most tests because they only test 32-bit apps. However Vista will be the first true 64-bit OS by Microsoft (Server 2003 x64 and XP x64 were really just hacks) Microsoft also plans to allow people to upgrade to the 64-bit version of the OS during its life time meaning that Microsoft has great faith that some point in the near future (~2yrs) the majority of system will be sold as 64-bit by default. This would mean that suddenly older K8 and newer K8L will suddenly preform much better compared to the Core 2, because these chips do not have the same limitations as the Core 2. Intel might be able to hack something together to make the Macro/Micro OP fusion work in 64-bit mode, but from what I read on the design this is highly unlikely. Nothing short of a brand new core would work to fix this. But by this time AMD should be fully into a modular CPU (K10) design which will render convential benchmarks useless. The best that Intel will able to do is have the best "well rounded" chip which is an oxymoron.

Got links? Proof? Benchmarks? Empirical evidence that proves this? If not, than this is just empty banter.

Anandtech, Techreport and many other review sites have benched C2D under 64-bit Windows and it's main advertised feature Macro/Micro OP fusion does not work in 64-bit. You can read Intel's tech sheet of Cd2 where they even aknowledge this!
December 18, 2006 3:44:39 PM

Quote:
Sharikou? Is that you?

I'm not even going to respond to this.

Nope, it is LameNoobMike.Not entirely sure....He didn't mention "the dark side". :wink:
December 18, 2006 3:46:11 PM

AMD SHILL!

AMDroid, go troll another forum.
December 18, 2006 3:47:09 PM

Quote:
I'd go for the X2 3800 just for sake of having great 64-bit potential on windows Vista.
Right now the Core 2 suffers a ~16% performance penalty when running in 64-bit mode while the K8 suffer only a ~6% on average running in 64-bit mode. The greater drop in the Core 2 is because the Macro/Micro OP fusion is a 32-bit exclusive feature. This is something often overlooked by most tests because they only test 32-bit apps. However Vista will be the first true 64-bit OS by Microsoft (Server 2003 x64 and XP x64 were really just hacks) Microsoft also plans to allow people to upgrade to the 64-bit version of the OS during its life time meaning that Microsoft has great faith that some point in the near future (~2yrs) the majority of system will be sold as 64-bit by default. This would mean that suddenly older K8 and newer K8L will suddenly preform much better compared to the Core 2, because these chips do not have the same limitations as the Core 2. Intel might be able to hack something together to make the Macro/Micro OP fusion work in 64-bit mode, but from what I read on the design this is highly unlikely. Nothing short of a brand new core would work to fix this.

This is interesting.
I must admit, I don't use 64-bit OS (who does?)
But I recall, I have seen some discussion about this. I remember, Core2 didn't work as well in WinXP x64 as AMD did...
Yes, of course this should become more 'noticable' next year :o 
Regards
December 18, 2006 3:48:25 PM

Quote:

Anandtech, Techreport and many other review sites have benched C2D under 64-bit Windows and it's main advertised feature Macro/Micro OP fusion does not work in 64-bit. You can read Intel's tech sheet of Cd2 where they even aknowledge this!


Congratulations, your post has earned the 8x8 FUD award.


Your post is misinformation. Do some more research before you spread FUD.

Hint: it's MACRO fusion that doesn't work under 64-bit, but MICRO fusion sure as hell works.
December 18, 2006 3:50:32 PM

Whatever...

The important thing is that you got my message. 8)
December 18, 2006 3:54:08 PM

Quote:
i wanted 2 buy pc which is going to be better
amd 64*2 3800+
or
intel core2duo
which has better performance

In a MrsBytch fashion:

Quote:
Any C2D performs better than the Athlon64 X2 3800+.
December 18, 2006 3:59:21 PM

Quote:
Whatever...

The important thing is that you got my message. 8)


And until you provide a post with links to *all* of your statements, meaning information that soundly backs it up, then you're full of crap and no one here will believe you with the exception of maybe BaronMatrix.
December 18, 2006 4:01:41 PM

Wait, let me get this straight... there's a performance PENALTY for going to 64-bit... for both K8 and C2D?

Don't know what you're smoking...
December 18, 2006 4:06:38 PM

Quote:
Wait, let me get this straight... there's a performance PENALTY for going to 64-bit... for both K8 and C2D?

Don't know what you're smoking...


What I'm trying to say is that K8s are better suited for 64-bit computing than C2Ds. Both will work fine but C2D does suffer due to not having macro OP fusion in it and it uses the same old EMT64 instruction set from previous Netburst architecture + SSE4.
December 18, 2006 4:06:44 PM

Quote:
Wait, let me get this straight... there's a performance PENALTY for going to 64-bit... for both K8 and C2D?

Don't know what you're smoking...


Awesome
Magical
Dope

:D 
December 18, 2006 4:33:52 PM

Quote:
Wait, let me get this straight... there's a performance PENALTY for going to 64-bit... for both K8 and C2D?

Don't know what you're smoking...


What I'm trying to say is that K8s are better suited for 64-bit computing than C2Ds. Both will work fine but C2D does suffer due to not having macro OP fusion in it and it uses the same old EMT64 instruction set from previous Netburst architecture + SSE4.

That still doesn't make sense. Last I checked, x86-64 doubled the number of GP registers. That alone should result in a ~10% performance boost for the register-starved x86-32 ISA.
December 18, 2006 4:39:46 PM

Quote:
Wait, let me get this straight... there's a performance PENALTY for going to 64-bit... for both K8 and C2D?

Don't know what you're smoking...


What I'm trying to say is that K8s are better suited for 64-bit computing than C2Ds. Both will work fine but C2D does suffer due to not having macro OP fusion in it and it uses the same old EMT64 instruction set from previous Netburst architecture + SSE4.

That still doesn't make sense. Last I checked, x86-64 doubled the number of GP registers. That alone should result in a ~10% performance boost for the register-starved x86-32 ISA.

Sorry for the misinformation if I didn't make myself clear. In conclusion, AMD's K8 and intels C2D bring good performance in 64-bit BUT AMD's procesors scale better which translates in better overall performane than intels.
December 18, 2006 4:47:37 PM

I don't think anyone would want to move to 64-bit if there was any type of performance penalty.

Just assuming *wink* that there is a performance boost when moving to x86-64:
I won't argue that K8 gains more PER CLOCK compared to C2D in x86-64 code, but C2D gains more performance OVERALL compared to K8.

It's pretty simple:
Assume K8 has a performace index of 1.0 in 32 bit code and it has a speedup of 10% in 64-bit code. Then, K8's 64bit performance would be 1.1.
Assume C2D has a performance index of 1.3 in 32-bit code and it has a speedup of 8% in 64-bit code. Then, C2D 64-bit performance would be 1.404. By virtue of higher base performance, the C2D's 64-bit speedup is greater than that of K8, though it's less than that of K8's clock for clock.

Jack has the data,
December 18, 2006 5:02:46 PM

Quote:
I don't think anyone would want to move to 64-bit if there was any type of performance penalty.

Just assuming *wink* that there is a performance boost when moving to x86-64:
I won't argue that K8 gains more PER CLOCK compared to C2D in x86-64 code, but C2D gains more performance OVERALL compared to K8.

It's pretty simple:
Assume K8 has a performace index of 1.0 in 32 bit code and it has a speedup of 10% in 64-bit code. Then, K8's 64bit performance would be 1.1.
Assume C2D has a performance index of 1.3 in 32-bit code and it has a speedup of 8% in 64-bit code. Then, C2D 64-bit performance would be 1.404. By virtue of higher base performance, the C2D's 64-bit speedup is greater than that of K8, though it's less than that of K8's clock for clock.

Jack has the data,


I've been reading these forums a while ago and to be sincere, I just don't trust or believe whatever jack has to say. he's just a die-hard intel fan (just a personal thought).

I get my facts and numbers from other review sites and from personal experience.

I don't and never denied C2Ds great design and performance, but the hard and true fact is that intel optimized this chip for 32-bit performance instead of making it a true 64-bit killer. You, as a user, expect that your processor perform the same in 32-bit as in 64-bit, that's why you've spend your hard earned dollars. It's ridiculous to think that intel doesn't support Macro OPs fusion in 64-bit and intel do knows that EMT64 sucks compared to AMD64. Things will get even nastier when AMD releases K8L and to add insult to injury, they (AMD) will be adding more instructions to the AMD64 instruction set like LZCNT, POPCNT, EXTRQ/INSERTQ, MOVNTSD/MOVNTSS and 48-bit memory addressing.

Let's give credit where it's due. Intel came up with an astounding uArch like conroe, but AMD also did a great job with the K8 architecture and people in this forum seems to suffer from amnesia and all they have to say is crap about it.
December 18, 2006 5:12:09 PM

Quote:

I've been reading these forums a while ago and to be sincere, I just don't trust or believe whatever jack has to say. he's just a die-hard intel fan (just a personal thought).

I get my facts and numbers from other review sites and from personal experience.

I don't and never denied C2Ds great design and performance, but the hard and true fact is that intel optimized this chip for 32-bit performance instead of making it a true 64-bit killer. You, as a user, expect that your processor perform the same in 32-bit as in 64-bit, that's why you've spend your hard earned dollars. It's ridiculous to think that intel doesn't support Macro OPs fusion in 64-bit and intel do knows that EMT64 sucks compared to AMD64. Things will get even nastier when AMD releases K8L and to add insult to injury, they (AMD) will be adding more instructions to the AMD64 instruction set like LZCNT, POPCNT, EXTRQ/INSERTQ, MOVNTSD/MOVNTSS and 48-bit memory addressing.

Let's give credit where it's due. Intel came up with an astounding uArch like conroe, but AMD also did a great job with the K8 architecture and people in this forum seems to suffer from amnesia and all they have to say is crap about it.


O RLY, you just joined today. But that's not the point.

Jack used benchmark data all over the internet.
Data is data is data. Data is fact.
Interpretation is opinion. I don't think there's anything to do with bias here. Data doesn't lie... benchmarks don't lie.

Quote:
I get my facts and numbers from other review sites and from personal experience.

Why don't you enlighten us with your sources, then?
December 18, 2006 5:13:02 PM

You can call Jack whatever you want, but he's a more reliable and believable source than you will ever be, since you have that "AMD is still a better deal, no matter what else is out" attitude.

You can give credit to whomever you want, but right now, you're spreading misinformation. What other review sites? Link them. You linked to Scientia's Blog as a source that AMD is better at 64-bit than Intel's CPUs, but
Quote:
The Windows Media Encoder shows a significant drop with 4 threads, but it drops about the same for Intel and AMD. With the original code Intel had a 23% lead, however with the newer 64 bit code this lead drops to only 15%. This is true for both 2 and 4 threads.

Well, it still says Intel lead with 23% in 32, but dropped to a 15% LEAD in 64-bit. How is that slower than AMD in 64-bit apps?

And frankly, in another post, you used Scientia, as a source again, to claim that benchmarks were "fixed" in some way. So, every review site is being paid off by Intel to run the same tests, on the same configuration, and no one will be the wiser? Right. You say, "What about cryptology?" WHO FREAKING CARES ABOUT IT? I play games, and do some encoding. I don't need to encrypt my files.

K8 was a good thing for AMD, what you are forgetting is that AMD is trying to milk it for what it's worth, and until they revise or create a technology, it might not be able to compete in what is now an even more competitive CPU environment.
December 18, 2006 5:28:55 PM

Quote:
As you've seen, the new 65nm X2 are more energy efficient than any thing intel has to offer (if you'd like to wait a little for retail ones). I'd go for the X2 3800 just for sake of having great 64-bit potential on windows Vista.
Right now the Core 2 suffers a ~16% performance penalty when running in 64-bit mode while the K8 suffer only a ~6% on average running in 64-bit mode. The greater drop in the Core 2 is because the Macro/Micro OP fusion is a 32-bit exclusive feature. This is something often overlooked by most tests because they only test 32-bit apps. However Vista will be the first true 64-bit OS by Microsoft (Server 2003 x64 and XP x64 were really just hacks) Microsoft also plans to allow people to upgrade to the 64-bit version of the OS during its life time meaning that Microsoft has great faith that some point in the near future (~2yrs) the majority of system will be sold as 64-bit by default. This would mean that suddenly older K8 and newer K8L will suddenly preform much better compared to the Core 2, because these chips do not have the same limitations as the Core 2. Intel might be able to hack something together to make the Macro/Micro OP fusion work in 64-bit mode, but from what I read on the design this is highly unlikely. Nothing short of a brand new core would work to fix this. But by this time AMD should be fully into a modular CPU (K10) design which will render convential benchmarks useless. The best that Intel will able to do is have the best "well rounded" chip which is an oxymoron.



In some ways you're right but X64 was not a hack. I was at MS when it was being developed and it is true 64bit with 32bit functionality.

MS has already said that all server apps REQUIRE X64 chips. I figure next year they will move toward 64bit on the desktop. AMD64 is easy to do since it mostly needs just a recompile ( though boundary layers have to be moved to 8bytes).

BTW, you have committed balsphemy. You will hear about it.
December 18, 2006 5:32:40 PM

Quote:
Whatever...

The important thing is that you got my message. 8)


And until you provide a post with links to *all* of your statements, meaning information that soundly backs it up, then you're full of crap and no one here will believe you with the exception of maybe BaronMatrix.

More insecure stalkers. Is there some kind of sign up page for people like you?

Thsi is not your CPU class. This is supposed to be just fun, not somewhere people show off that they can read links and regurgitate the info.
December 18, 2006 5:33:01 PM

Quote:

Anandtech, Techreport and many other review sites have benched C2D under 64-bit Windows and it's main advertised feature Macro/Micro OP fusion does not work in 64-bit. You can read Intel's tech sheet of Cd2 where they even aknowledge this!


Congratulations, your post has earned the 8x8 FUD award.


Your post is misinformation. Do some more research before you spread FUD.

Hint: it's MACRO fusion that doesn't work under 64-bit, but MICRO fusion sure as hell works.


HOLY CR*P. Where Can I get that motherboard... does anyone know which mobo that is, and which vid cards those are? I really want to buy that cause it must be the greatest system ever ever created.
December 18, 2006 5:57:59 PM

Quote:
Whatever...

The important thing is that you got my message. 8)


And until you provide a post with links to *all* of your statements, meaning information that soundly backs it up, then you're full of crap and no one here will believe you with the exception of maybe BaronMatrix.

More insecure stalkers. Is there some kind of sign up page for people like you?

Thsi is not your CPU class. This is supposed to be just fun, not somewhere people show off that they can read links and regurgitate the info.

Fun is one thing.

Saying mistruths and spreading misinformation is not fun. What's fun about claiming that one thing is better because you say it is, but it isn't. Yeah, really fun when someone comes in here and asks for some advice, and all they get are facts that aren't even close to the truth. That's why benchmarks, review sites, etc. are asked. To prove the point of the writer. Hell, I could say that my system can run a server farm with just it's single CPU, but where's my proof? Would anyone believe it? Maybe, and there in lies the danger of spreading misinformation.

You can have fun in here BM. I do. I just don't think saying misinformation is fun. And no, this isn't a CPU class, but neither is this place a forum where pure BS is gonna be accepted as fact, either.
December 18, 2006 6:05:14 PM

Quote:
Whatever...

The important thing is that you got my message. 8)


And until you provide a post with links to *all* of your statements, meaning information that soundly backs it up, then you're full of crap and no one here will believe you with the exception of maybe BaronMatrix.

More insecure stalkers. Is there some kind of sign up page for people like you?

Thsi is not your CPU class. This is supposed to be just fun, not somewhere people show off that they can read links and regurgitate the info.

Fun is one thing.

Saying mistruths and spreading misinformation is not fun. What's fun about claiming that one thing is better because you say it is, but it isn't. Yeah, really fun when someone comes in here and asks for some advice, and all they get are facts that aren't even close to the truth. That's why benchmarks, review sites, etc. are asked. To prove the point of the writer. Hell, I could say that my system can run a server farm with just it's single CPU, but where's my proof? Would anyone believe it? Maybe, and there in lies the danger of spreading misinformation.

You can have fun in here BM. I do. I just don't think saying misinformation is fun. And no, this isn't a CPU class, but neither is this place a forum where pure BS is gonna be accepted as fact, either.


You guys really need hobbies. Even if someone purposely misleads people the worse that could happen is that their PC may only get 75fps rather than 90fps.

Oh the horror. The world will end with a few of those. Also, please come up with a new word. BS is getting old. Especially considering that I'n rarely wrong.
December 18, 2006 6:18:05 PM

Right. So paying $2k for a system that runs slower/equal to a system built for $1k is fine by you.

That's why people here are not sheep, and accept everything written as fact, especially by people who seriously have a love, yes, LOVE, for a company that doesn't care about them in a personal sense.

I don't care if someone is willing to pay $4-5k for a system, it's their money. But if someone comes in asking for a system that runs specific tasks, and someone misleads them into believing that a certain BRAND does better, cause they are morally better, the underdog, a monopoly, have more factories, etc., instead of just linking benchmarks, tests, and other sources of FACT, then that's pure and utter CRAP.

You could be rarely wrong, BM, but you're hardly right, either.
December 18, 2006 7:47:28 PM

Quote:
intel do knows that EMT64 sucks compared to AMD64.


Ummm. . . if I remember correctly, Intel's EMT64 is just a rebranded AMD64 that Intel licensed from AMD. Sure wish I had the link for this claim, I remember reading about it- probably on Tom's- when the first EMT64 Pentium came out.
December 18, 2006 8:30:51 PM

Quote:

I've been reading these forums a while ago and to be sincere, I just don't trust or believe whatever jack has to say. he's just a die-hard intel fan (just a personal thought).

I get my facts and numbers from other review sites and from personal experience.

I don't and never denied C2Ds great design and performance, but the hard and true fact is that intel optimized this chip for 32-bit performance instead of making it a true 64-bit killer. You, as a user, expect that your processor perform the same in 32-bit as in 64-bit, that's why you've spend your hard earned dollars. It's ridiculous to think that intel doesn't support Macro OPs fusion in 64-bit and intel do knows that EMT64 sucks compared to AMD64. Things will get even nastier when AMD releases K8L and to add insult to injury, they (AMD) will be adding more instructions to the AMD64 instruction set like LZCNT, POPCNT, EXTRQ/INSERTQ, MOVNTSD/MOVNTSS and 48-bit memory addressing.

Let's give credit where it's due. Intel came up with an astounding uArch like conroe, but AMD also did a great job with the K8 architecture and people in this forum seems to suffer from amnesia and all they have to say is crap about it.


O RLY, you just joined today. But that's not the point.

Jack used benchmark data all over the internet.
Data is data is data. Data is fact.
Interpretation is opinion. I don't think there's anything to do with bias here. Data doesn't lie... benchmarks don't lie.

Quote:
I get my facts and numbers from other review sites and from personal experience.

Why don't you enlighten us with your sources, then?

Then, I'll also say that C2D does suffer once running 64-bit code because Data is data is data. Data is fact: :) 

Xbitlabs
XtremeSystems
XtremeSystems2
December 18, 2006 8:36:17 PM

Quote:
You can call Jack whatever you want, but he's a more reliable and believable source than you will ever be, since you have that "AMD is still a better deal, no matter what else is out" attitude.

You can give credit to whomever you want, but right now, you're spreading misinformation. What other review sites? Link them. You linked to Scientia's Blog as a source that AMD is better at 64-bit than Intel's CPUs, but
The Windows Media Encoder shows a significant drop with 4 threads, but it drops about the same for Intel and AMD. With the original code Intel had a 23% lead, however with the newer 64 bit code this lead drops to only 15%. This is true for both 2 and 4 threads.

Well, it still says Intel lead with 23% in 32, but dropped to a 15% LEAD in 64-bit. How is that slower than AMD in 64-bit apps?

And frankly, in another post, you used Scientia, as a source again, to claim that benchmarks were "fixed" in some way. So, every review site is being paid off by Intel to run the same tests, on the same configuration, and no one will be the wiser? Right. You say, "What about cryptology?" WHO FREAKING CARES ABOUT IT? I play games, and do some encoding. I don't need to encrypt my files.

K8 was a good thing for AMD, what you are forgetting is that AMD is trying to milk it for what it's worth, and until they revise or create a technology, it might not be able to compete in what is now an even more competitive CPU environment.

Well, that's fine, but I do apologize if I don't and won't believe in your Master, is just that I don't want to join his "sect" of blind worshippers.
December 18, 2006 8:39:16 PM

Quote:
You can call Jack whatever you want, but he's a more reliable and believable source than you will ever be, since you have that "AMD is still a better deal, no matter what else is out" attitude.

You can give credit to whomever you want, but right now, you're spreading misinformation. What other review sites? Link them. You linked to Scientia's Blog as a source that AMD is better at 64-bit than Intel's CPUs, but
The Windows Media Encoder shows a significant drop with 4 threads, but it drops about the same for Intel and AMD. With the original code Intel had a 23% lead, however with the newer 64 bit code this lead drops to only 15%. This is true for both 2 and 4 threads.

Well, it still says Intel lead with 23% in 32, but dropped to a 15% LEAD in 64-bit. How is that slower than AMD in 64-bit apps?

And frankly, in another post, you used Scientia, as a source again, to claim that benchmarks were "fixed" in some way. So, every review site is being paid off by Intel to run the same tests, on the same configuration, and no one will be the wiser? Right. You say, "What about cryptology?" WHO FREAKING CARES ABOUT IT? I play games, and do some encoding. I don't need to encrypt my files.

K8 was a good thing for AMD, what you are forgetting is that AMD is trying to milk it for what it's worth, and until they revise or create a technology, it might not be able to compete in what is now an even more competitive CPU environment.

Well, that's fine, but I do apologize if I don't and won't believe in your Master, is just that I don't want to join his "sect" of blind worshippers.Classic MMM line.
December 18, 2006 8:41:37 PM

Quote:
Whatever...

The important thing is that you got my message. 8)


And until you provide a post with links to *all* of your statements, meaning information that soundly backs it up, then you're full of crap and no one here will believe you with the exception of maybe BaronMatrix.

More insecure stalkers. Is there some kind of sign up page for people like you?

Thsi is not your CPU class. This is supposed to be just fun, not somewhere people show off that they can read links and regurgitate the info.

Lol. I have much better things to do with my time than stalk you my challenged little friend. That you consider the random idle comment stalking just points out how insecure you are, not I. More of that patented logic I guess....

Have a great day.
December 18, 2006 8:48:03 PM

Quote:
You can call Jack whatever you want, but he's a more reliable and believable source than you will ever be, since you have that "AMD is still a better deal, no matter what else is out" attitude.

You can give credit to whomever you want, but right now, you're spreading misinformation. What other review sites? Link them. You linked to Scientia's Blog as a source that AMD is better at 64-bit than Intel's CPUs, but
The Windows Media Encoder shows a significant drop with 4 threads, but it drops about the same for Intel and AMD. With the original code Intel had a 23% lead, however with the newer 64 bit code this lead drops to only 15%. This is true for both 2 and 4 threads.

Well, it still says Intel lead with 23% in 32, but dropped to a 15% LEAD in 64-bit. How is that slower than AMD in 64-bit apps?

And frankly, in another post, you used Scientia, as a source again, to claim that benchmarks were "fixed" in some way. So, every review site is being paid off by Intel to run the same tests, on the same configuration, and no one will be the wiser? Right. You say, "What about cryptology?" WHO FREAKING CARES ABOUT IT? I play games, and do some encoding. I don't need to encrypt my files.

K8 was a good thing for AMD, what you are forgetting is that AMD is trying to milk it for what it's worth, and until they revise or create a technology, it might not be able to compete in what is now an even more competitive CPU environment.

Well, that's fine, but I do apologize if I don't and won't believe in your Master, is just that I don't want to join his "sect" of blind worshippers.

Isn't this sortta like the pot calling the kettle black? Sheesh. This thread is retarded much like your logic. Have fun with it.

/exit thread
December 18, 2006 9:01:34 PM

I understand what you are trying to say, but you can't compare percentage increase/decrease in speed over 32-bit since you don't have an even baseline to measure the 2 platforms. This is "fuzzy math".

If one is slower than the other, it could even have less of a performance gain and wind up with a bigger percentage.

For example:
If I make $2/hour and get a 10% raise, and you make $1/hour and get a 16% raise - who got the bigger raise?
December 18, 2006 9:04:43 PM

Quote:
I understand what you are trying to say, but you can't compare percentage increase/decrease in speed over 32-bit since you don't have an even baseline to measure the 2 platforms. This is "fuzzy math".

If one is slower than the other, it could even have less of a performance gain and wind up with a bigger percentage.

For example:
If I make $2/hour and get a 10% raise, and you make $1/hour and get a 16% raise - who got the bigger raise?


Thanks for your support. Maybe, I haven't made myself clear from the first.
December 18, 2006 9:13:23 PM

Quote:
I understand what you are trying to say, but you can't compare percentage increase/decrease in speed over 32-bit since you don't have an even baseline to measure the 2 platforms. This is "fuzzy math".

If one is slower than the other, it could even have less of a performance gain and wind up with a bigger percentage.

For example:
If I make $2/hour and get a 10% raise, and you make $1/hour and get a 16% raise - who got the bigger raise?

Exactly what I tried to explain earlier.
December 18, 2006 9:23:48 PM

Wasnt the point though that the raise is higher and not the endresult speed.
December 18, 2006 9:24:06 PM

Quote:

Well, that's fine, but I do apologize if I don't and won't believe in your Master, is just that I don't want to join his "sect" of blind worshippers.


Is that all you can say, since I refuted your claim? "your Master"? Jack has DATA to back up his statements, though. That I believe more than someone's opinion.

I am not as blind as you are in your undaunted belief in a product that is not matching anything that a competitor is offering.

You link an article to X-bit labs that concluded with:

Quote:
This way, there is a certain difference: AMD K8 turns out 6% mode efficient in 64-bit mode than Intel Core. However, this difference cannot compensate for the 20% performance advantage of the Intel Core 2 Duo over the Athlon 64 X2 working at the same clock speed, which we have pointed out in our previous articles. Therefore, we will not change our conclusions about the performance of the new Intel processors even keeping in mind the upcoming launch of 64-bit Windows Vista OS family.


Oh, but that's not important, is it. Just look at the tasks that it did fail on. I mean, really, did you even look at the testing results of that article? :roll: Yes, in 64-bit, the C2D was lower than it's 32-bit test results in some tests, but so was the FX-62. So, does that mean that the AMD CPUs aren't any better in 64-bit than the C2Ds? Hell, the C2D lost 1 benchmark in that all those tests, in either 64-bit or 32-bit.

You go believe what you want. You can say Jack is my master, if you want. I suppose it's easier trying to get under someone's skin and fling insults, than try to show some real proof.
December 18, 2006 9:28:25 PM

Quote:
As you've seen, the new 65nm X2 are more energy efficient than any thing intel has to offer (if you'd like to wait a little for retail ones). I'd go for the X2 3800 just for sake of having great 64-bit potential on windows Vista.
Right now the Core 2 suffers a ~16% performance penalty when running in 64-bit mode while the K8 suffer only a ~6% on average running in 64-bit mode. The greater drop in the Core 2 is because the Macro/Micro OP fusion is a 32-bit exclusive feature. This is something often overlooked by most tests because they only test 32-bit apps. However Vista will be the first true 64-bit OS by Microsoft (Server 2003 x64 and XP x64 were really just hacks) Microsoft also plans to allow people to upgrade to the 64-bit version of the OS during its life time meaning that Microsoft has great faith that some point in the near future (~2yrs) the majority of system will be sold as 64-bit by default. This would mean that suddenly older K8 and newer K8L will suddenly preform much better compared to the Core 2, because these chips do not have the same limitations as the Core 2. Intel might be able to hack something together to make the Macro/Micro OP fusion work in 64-bit mode, but from what I read on the design this is highly unlikely. Nothing short of a brand new core would work to fix this. But by this time AMD should be fully into a modular CPU (K10) design which will render convential benchmarks useless. The best that Intel will able to do is have the best "well rounded" chip which is an oxymoron.


Sharikou? Is that you?

I'm not even going to respond to this.
Haha... nice. Seems pretty Shariou-ish, but he would have had to mention something about how if you use an Intel CPU your laptop will burn up and hurt your children. :lol: 
December 18, 2006 9:32:19 PM

Actually, the point is that if they don't share the same baseline, then percenatges are meaningless (unless one is positive and the other is zero or negative).
December 18, 2006 9:36:33 PM

64bits is a myth... we are all using 8bit programming to its fullest.
December 18, 2006 9:44:34 PM

*Grabs a soldering iron. Prepares to spend all night "piggybacking" RAM chips to get 128k on his C-64* :wink:
December 18, 2006 9:46:27 PM

Quote:
This way, there is a certain difference: AMD K8 turns out 6% mode efficient in 64-bit mode than Intel Core. However, this difference cannot compensate for the 20% performance advantage of the Intel Core 2 Duo over the Athlon 64 X2 working at the same clock speed, which we have pointed out in our previous articles. Therefore, we will not change our conclusions about the performance of the new Intel processors even keeping in mind the upcoming launch of 64-bit Windows Vista OS family.



What a dumb***, Barcelona Xtreme Noob should read the article before posting. :roll: Kinda reminds me of 9inch aka 9nm, posted anything that sounded like it supported his argument.
      • 1 / 2
      • 2
      • Newest