• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

AMD vs Intel

Page 2 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
Originally posted by: lenjack
It may have changed, but I recall amd is 9 instructions per clock and intel is 6.
you'r talking about "Operations Per Clock" not "Instructions Per Cycle". northwood C /prescott E 6x2 HT enabled.


 
I might be a little off here, but I was under the impression that AMD's engineers were working with slightly better branch predicition.

Not that much attention is paid to such things, and you can't really attribute a CHUNK of extra "speed" to good branch prediction, but on a design like Intel's (with more in between each transaction) bad precictions can hit up performance (since you'd need to place the instruction back at the start while flushing everything out of your working buffer).

But then again, I'm gathering all of this from these lovely oveprice textbooks I'm forced to read, and much may have changed since those books were written.

Is this still (not sure it ever was) an issue with Intel systems?
 
Originally posted by: NittanyLAncer
I might be a little off here, but I was under the impression that AMD's engineers were working with slightly better branch predicition...

...Is this still (not sure it ever was) an issue with Intel systems?

No, your not off. Right on topic I think. And you're right too. I briefly touched it, included in the
"The reason for P4's lower IPC is partially because long pipe doesn't really work so well. There are problems with dependencies and other *balls* to keep in the air, for so many instructions. So the goal, one instruction off the end of the pipe each cycle, is not achieved. Lots of results must be discarded, and be redone. "

Prescott feature improved branchprediction, which sometimes results in benchmark characteristics to lean towards AMD flavor. But mainly, the improvement, as well as the larger L2 cache, only compensates for the penalties of an even longer pipe.
 
Originally posted by: dug777
its not remotely on the topic, but arguably mhz is a lot more accurate/truthful than the amd system- which is related 2 whatever fairydust the amd marketing folk r smoking & the intel ratings 😛

It _IS_ on topic. And it's complete nonsens. MHz is irrelevant. Nowhere near "accurate/truthful".


 
Originally posted by: dug777
its not remotely on the topic, but arguably mhz is a lot more accurate/truthful than the amd system- which is related 2 whatever fairydust the amd marketing folk r smoking & the intel ratings 😛

Then what do we need review sites for?

Go and get yourself a X600XT videocard clocked at 500MHz which most cerainly beats 6800ultras lame 400Mhz. A 3.8P4 which lays the smack down on FX-55's little 2.6Ghz. And some expensive DDR2 600Mhz which beats DDR's 200 LL.
:roll:
 
Originally posted by: dug777
its not remotely on the topic, but arguably mhz is a lot more accurate/truthful than the amd system- which is related 2 whatever fairydust the amd marketing folk r smoking & the intel ratings 😛


That whole statement was nor accurate or truthful. No offense but you have no idea what your talking about on this subject. Frequency is one of the things that measures a CPU's performance, you have to consider how many Instructions per second a CPU can calculate, lantency, cache prefetch, branch prediction and so so so many other things that only the AMD fab workers know about. Only newby's (General Public) or overclockers buy Intel's CPU's, and if you dont want to be a part of that General Public then I suggest the following.

Do your homework!
Listen to people on these forums!
 
Originally posted by: Shenkoa
Originally posted by: dug777
its not remotely on the topic, but arguably mhz is a lot more accurate/truthful than the amd system- which is related 2 whatever fairydust the amd marketing folk r smoking & the intel ratings 😛


That whole statement was nor accurate or truthful. No offense but you have no idea what your talking about on this subject. Frequency is one of the things that measures a CPU's performance, you have to consider how many Instructions per second a CPU can calculate, lantency, cache prefetch and so so so many other things that only the AMD fab workers know about. Only newby's (General Public) or overclockers buy Intel's CPU's, and if you dont want to be a part of that General Public then I suggest the following.

Do your homework!
Listen to people on these forums!


true, but overclockers tend to go amd now. 2.6oc on a 3200+ is easy and the chip is a hella lot cheaper than a comparable intel p4 that can oc and achieve similar performance with the stock cooler.
 
for fvuck's sake u crazy buggers 😛

Had it crossed your minds (obviously not or u wouldnt have comedown on me like a sack of shiat 🙁 ) that all i'm saying is that the mhz intel speed rating is just that- a rating of a chip based on it's raw, measurable clockspeed- i'm not trying to say that it is an accurate rating of performance accross the board (or god forbid across architectures, i am well aware that "Frequency is one of the things that measures a CPU's performance, you have to consider how many Instructions per second a CPU can calculate, lantency, cache prefetch and so so so many other things that only the AMD fab workers know about." 😛 )

In a moment of foolishness i posted a comment (half-jokingly as it happens) because the IMO the AMD pr rating system is not quantifiable in any way,and whatever they like to say was heavily based on a comparison with intel processors...(certainly used to be before A64 ne way)

When i said that it wasn't on topic, i meant MY post, as i can't easily see what relevance pr ratings have to answering "How do AMDs manage to keep up with a much slower clock?", and so i thought what i was saying was rather irrelevant...

ne way- what i was trying to say is simply that the mhz ratings given by intel are an actual, quantifiable characteristic, while the relation b/w a 2.2gz 512kb l2cache & the rating of 3200+ is somewhat harder to see, especially given the different variants given the same name...ie. what i was saying was off topic & rather pointless, but also kinda true.

If u aren't still in flame mode u should b able 2see what i'm getting at, tho' badly expressed, and i've learnt my lesson when it comes 2posting comments like that 😛
 
Originally posted by: sandorski
Originally posted by: CheesePoofs
Originally posted by: InseName
well y would u hate the pr system, it's not really amd's fault that intel has chosen to max their frequencies but not perfrom

because the pr rating does not accurately reflect actual performance, and also its confusing. For example, there are 3 different Athlong 64 3200+. one is 2.4 ghz, 512K L2 cache, socket 754; one is 2.2 ghz, 1MB L2 cache, socket 754; and the last is 2.2ghz, 512K L2 cache, socket 939. And they aren't even equal in perfromance.

It is as accurate as using mhz.

I would say that AMD's PR system is much better than just using mhz. I personally don't really mind it myself, i was just stating why some people don't.
 
correct me if i'm wrong folks (tho u will ne way 🙂, i don't need 2give u permission), but the OP was lookin 4 replies xplaining the architectural diff b/w AMD & intel that means that AMD can have higher performing CPUs at massively lower clockspeeds ..not a biatch-fight bout pr ratings

Vee & Zebo seem 2 have covered the basics of it on page1 of the thread...
 
Originally posted by: dug777
for fvuck's sake u crazy buggers 😛

Had it crossed your minds (obviously not or u wouldnt have comedown on me like a sack of shiat 🙁 ) that all i'm saying is that the mhz intel speed rating is just that- a rating of a chip based on it's raw, measurable clockspeed- i'm not trying to say that it is an accurate rating of performance accross the board (or god forbid across architectures, i am well aware that "Frequency is one of the things that measures a CPU's performance, you have to consider how many Instructions per second a CPU can calculate, lantency, cache prefetch and so so so many other things that only the AMD fab workers know about." 😛 )

In a moment of foolishness i posted a comment (half-jokingly as it happens) because the IMO the AMD pr rating system is not quantifiable in any way,and whatever they like to say was heavily based on a comparison with intel processors...(certainly used to be before A64 ne way)

When i said that it wasn't on topic, i meant MY post, as i can't easily see what relevance pr ratings have to answering "How do AMDs manage to keep up with a much slower clock?", and so i thought what i was saying was rather irrelevant...

ne way- what i was trying to say is simply that the mhz ratings given by intel are an actual, quantifiable characteristic, while the relation b/w a 2.2gz 512kb l2cache & the rating of 3200+ is somewhat harder to see, especially given the different variants given the same name...ie. what i was saying was off topic & rather pointless, but also kinda true.

If u aren't still in flame mode u should b able 2see what i'm getting at, tho' badly expressed, and i've learnt my lesson when it comes 2posting comments like that 😛


Hey I think AMD's "ratings" helped them with joe six pack. Those of us in the know, see the PR ratings for what they are, a marketing scam just like the P4 is.😛
 
Originally posted by: CheesePoofs
Originally posted by: sandorski
Originally posted by: CheesePoofs
Originally posted by: InseName
well y would u hate the pr system, it's not really amd's fault that intel has chosen to max their frequencies but not perfrom

because the pr rating does not accurately reflect actual performance, and also its confusing. For example, there are 3 different Athlong 64 3200+. one is 2.4 ghz, 512K L2 cache, socket 754; one is 2.2 ghz, 1MB L2 cache, socket 754; and the last is 2.2ghz, 512K L2 cache, socket 939. And they aren't even equal in perfromance.

It is as accurate as using mhz.

I would say that AMD's PR system is much better than just using mhz. I personally don't really mind it myself, i was just stating why some people don't.

"Barton" athlonxp. 3200+ my ass.
 
Originally posted by: mwmorph
Originally posted by: Shenkoa
Originally posted by: dug777
its not remotely on the topic, but arguably mhz is a lot more accurate/truthful than the amd system- which is related 2 whatever fairydust the amd marketing folk r smoking & the intel ratings 😛


That whole statement was nor accurate or truthful. No offense but you have no idea what your talking about on this subject. Frequency is one of the things that measures a CPU's performance, you have to consider how many Instructions per second a CPU can calculate, lantency, cache prefetch and so so so many other things that only the AMD fab workers know about. Only newby's (General Public) or overclockers buy Intel's CPU's, and if you dont want to be a part of that General Public then I suggest the following.

Do your homework!
Listen to people on these forums!


true, but overclockers tend to go amd now. 2.6oc on a 3200+ is easy and the chip is a hella lot cheaper than a comparable intel p4 that can oc and achieve similar performance with the stock cooler.

Well not so much now, but with Northy's man, I would to have a 2.4C, a good 865 or 875 mobo and danger den water cooling at my disposal = fun stuff!

Overclockers buy Intel but thats not to say they dont AMD as well.

I would say that AMD's PR system is much better than just using mhz. I personally don't really mind it myself, i was just stating why some people don't.


I agree Intel's rating is more accurate because its based on a more concrete system, AMD cant measure their performance in raw MHz because of marketing. The PR is a good system and I dont see what else AMD could use. S
 
Originally posted by: dug777
for fvuck's sake u crazy buggers 😛

Had it crossed your minds (obviously not or u wouldnt have comedown on me like a sack of shiat 🙁 ) that all i'm saying is that the mhz intel speed rating is just that- a rating of a chip based on it's raw, measurable clockspeed- i'm not trying to say that it is an accurate rating of performance accross the board (or god forbid across architectures, i am well aware that "Frequency is one of the things that measures a CPU's performance, you have to consider how many Instructions per second a CPU can calculate, lantency, cache prefetch and so so so many other things that only the AMD fab workers know about." 😛 )

In a moment of foolishness i posted a comment (half-jokingly as it happens) because the IMO the AMD pr rating system is not quantifiable in any way,and whatever they like to say was heavily based on a comparison with intel processors...(certainly used to be before A64 ne way)

When i said that it wasn't on topic, i meant MY post, as i can't easily see what relevance pr ratings have to answering "How do AMDs manage to keep up with a much slower clock?", and so i thought what i was saying was rather irrelevant...

ne way- what i was trying to say is simply that the mhz ratings given by intel are an actual, quantifiable characteristic, while the relation b/w a 2.2gz 512kb l2cache & the rating of 3200+ is somewhat harder to see, especially given the different variants given the same name...ie. what i was saying was off topic & rather pointless, but also kinda true.

If u aren't still in flame mode u should b able 2see what i'm getting at, tho' badly expressed, and i've learnt my lesson when it comes 2posting comments like that 😛

Oops! DP

Think of how many ways it can be taken.

 
Back
Top