AMD is yet again pushing its Average CPU Power (ACP) ahead of Thermal Design Power (TDP) measurements and certainly doesn't consider its own TDP ratings to be comparable to Intel.
AMD's, Nigel Dessau, Senior Vice President and Chief Marketing Officer, tweeted yesterday to another hack: "No two people measure TDP the same way. How valid is the compare?"
According to his official blog, Dessau tries to state AMD's position as one of being green and of energy efficiency. He claims AMD is going down this route in order to better educate IT managers into making informed decisions about power budgets and all the rest of it.
AMD is measuring ACP with "typical data centre workloads" - which means it's designed for Opteron CPUs predominantly. We can accept this probably a good thing for servers and high power data centres where electricity and cooling bills can be rather large. But it's clear that anyone taking the figures still needs to grab a fist full of salt: are AMD's methods of measurement applicable to your environment and workloads? And worse yet - trying to work out whether you want Intel, AMD, IBM or whatever else and having only a marketing number, opposed to a scientific one, doesn't really help anybody.
AMD also claims it "provide[s] TDPs for system designers who need to know what worst-case thermal limits to use when designing a system, which is of course a practical application of that metric."
Now, are we talking worst case thermal loads for short bursts or hours at a time? For example, for years now Intel has used a "typical" TDP rather than TDP-Max because, again, it looks better on paper.
We were curious how far its ACP rating was being pushed so we asked AMD, who confirmed that was limited to server products only - desktop and notebook parts are not included. The problem this industry has is there's no official, quantifiable rating system so if AMD chooses to try and suggest something new for its own product, even if there is some technical reasoning behind it to aid certain customers, there will be an inevitable claim that it's a marketing push - especially if the numbers are conveniently lower.
What Nigel Dessau says is right - there's direct no way to compare the two without knowing that they are measured identically, however is AMD effectively forced to use the TDP nomenclature just because "that's how it's always been" or is one/are both companies simply playing hard and fast with those three letters in whatever way it suits them? Does an arbitrary TDP value even matter as long as the CPUs continually strive for performance efficiency, the motherboards support everything thrown at them and they keep making CPU coolers big enough? Let us know your thoughts in the forums.
AMD's, Nigel Dessau, Senior Vice President and Chief Marketing Officer, tweeted yesterday to another hack: "No two people measure TDP the same way. How valid is the compare?"
According to his official blog, Dessau tries to state AMD's position as one of being green and of energy efficiency. He claims AMD is going down this route in order to better educate IT managers into making informed decisions about power budgets and all the rest of it.
AMD is measuring ACP with "typical data centre workloads" - which means it's designed for Opteron CPUs predominantly. We can accept this probably a good thing for servers and high power data centres where electricity and cooling bills can be rather large. But it's clear that anyone taking the figures still needs to grab a fist full of salt: are AMD's methods of measurement applicable to your environment and workloads? And worse yet - trying to work out whether you want Intel, AMD, IBM or whatever else and having only a marketing number, opposed to a scientific one, doesn't really help anybody.
AMD also claims it "provide[s] TDPs for system designers who need to know what worst-case thermal limits to use when designing a system, which is of course a practical application of that metric."
Now, are we talking worst case thermal loads for short bursts or hours at a time? For example, for years now Intel has used a "typical" TDP rather than TDP-Max because, again, it looks better on paper.
We were curious how far its ACP rating was being pushed so we asked AMD, who confirmed that was limited to server products only - desktop and notebook parts are not included. The problem this industry has is there's no official, quantifiable rating system so if AMD chooses to try and suggest something new for its own product, even if there is some technical reasoning behind it to aid certain customers, there will be an inevitable claim that it's a marketing push - especially if the numbers are conveniently lower.
What Nigel Dessau says is right - there's direct no way to compare the two without knowing that they are measured identically, however is AMD effectively forced to use the TDP nomenclature just because "that's how it's always been" or is one/are both companies simply playing hard and fast with those three letters in whatever way it suits them? Does an arbitrary TDP value even matter as long as the CPUs continually strive for performance efficiency, the motherboards support everything thrown at them and they keep making CPU coolers big enough? Let us know your thoughts in the forums.
RELATED ARTICLES
MSI MPG Velox 100R Chassis Review
October 14 2021 | 15:04
Want to comment? Please log in.