Another reason for me not to buy AMD anymore

[varikonniemi]

The same reason you don't advertise the turbo clock as standard, you don't advertise the standard clock as standard in this case; it cannot sustain it in all workloads in manufacturer specified operating conditions. It is a lie. It is a 3.4ghz processor, not 3.8 ghz.

[bridgman]

Since the Phenom II X6 has a turbo function it better knows about its power envelope. The point is not that CPUs do that, but that a CPU that is advertized as 3.8 GHz CPU is not able to sustain that speed under load. While Kano thinks that such a CPU is a joke I would go a step further and say that such a CPU is a fraud.

I really don't understand the point you are trying to make. Ability to sustain clocks is always going to be a function of ambient temperature, cooling system effectiveness/cleanliness, case design re: airflow and turbulence etc... The only thing that has changed is that every year CPUs and GPUs are able to adjust clocks more finely and more quickly, and to do this with less (or zero) OS intervention and control.

I think you're saying "I want to go back to the good old days when hardware overheated under certain stress tests unless you had a big-ass aftermarket heat sink", and I sympathise with that (I hate it when my hardware makes its own decisions without consulting me ) but I don't think that is going to happen.

If you're just saying that processor specs should reflect 99.999999% of use cases rather than 99.99999% (or whatever the numbers are) I think that's fair, although I'm not sure it's a good idea. Again, look around a bit more. There have been complaints and questions about thermal throttling ever since the first implementations appeared, and for a while the most common question was "how do I turn this ^#$%$# off ?".

The problem is that fine-grained thermal management makes things better for users, although it makes spec'ing and benchmarking *much* more complicated. I suspect it's one of the reasons that different sites are increasingly showing significantly different results for the same products in their benchmarks and reviews.

[varikonniemi]

I think it's a case of consumer vs server workload. Consumer cpus (or gpus, see furmark) are not assumed to run at 100% for hours, so they're not thermally specced to withstand that, so that marketing can claim higher default numbers.

Server cpus typically are specced for such workloads.

Whether this is false advertising for consumer items is up for debate.

You must be kidding. Right?

It is a very serious flaw if the product is not 100% stable in the most serious workload imaginable, in continuous operation. The problem with furmark was that it was a harder workload than engineers could imagine. No natural workload is as demanding.

[bridgman]

Those older CPUs are able to downclock themselves if they reach their maximum temperature, but they never do that under default conditions.

Maybe not on Linux, because the OS support wasn't there, but if you search around Windows-oriented forums you'll find all kinds of questions about clock reduction over the last few years, particularly when running stress tests.

Newer CPUs make more clock decisions on-chip without requiring OS intervention.

[Vim_User]

Those older cpus simply use other methods of throttling than downclocking. (assuming they did reach thermal limits; depends on cooling, ambient temp etc.)

The most common way I remember was to delay instructions (start inserting nops every now and then).

Those older CPUs are able to downclock themselves if they reach their maximum temperature, but they never do that under default conditions.

[Vim_User]

AFAIK thermal throttling to stay within a power envelope is a pretty new feature for CPUs (ours and others). I'm not sure the other CPUs you mentioned are even able to do that.

Since the Phenom II X6 has a turbo function it better knows about its power envelope. The point is not that CPUs do that, but that a CPU that is advertized as 3.8 GHz CPU is not able to sustain that speed under load. While Kano thinks that such a CPU is a joke I would go a step further and say that such a CPU is a fraud.

[curaga]

My consumer Phenom II X6 (125W version) can run under 100% load for hours without downclocking itself, the same is true for my laptop's Athlon QL-66 and was true for my former Core 2 Quads (Q6600 and Q9550) and Athlon X2 5200+ and any CPU I used before that.
I can't see any reason why I should expect it to be different with newer CPUs, consumer or not. If a CPU is advertised as 3.8GHz model it has to run 3.8GHz, not 3.4GHz.

Those older cpus simply use other methods of throttling than downclocking. (assuming they did reach thermal limits; depends on cooling, ambient temp etc.)

The most common way I remember was to delay instructions (start inserting nops every now and then).

[Kano]

I don't think it is a good hardware design when the cpu consumes by default too much power that it has to lower clockspeeds on certain loads. Intel has power saving cpus as well, there the default clock is lower by default - the rest is done using turbo steps (for i5+). Turbo boost has of course a power usage limit (can be set in firmware for oc). So basically AMD should rebrand the cpus and use a lower default and more turbo steps. A cpu that does not run a specified speed under all loads is a joke.

[bridgman]

AFAIK thermal throttling to stay within a power envelope is a pretty new feature for CPUs (ours and others). I'm not sure the other CPUs you mentioned are even able to do that.

[Vim_User]

I think it's a case of consumer vs server workload. Consumer cpus (or gpus, see furmark) are not assumed to run at 100% for hours, so they're not thermally specced to withstand that, so that marketing can claim higher default numbers.

Server cpus typically are specced for such workloads.

Whether this is false advertising for consumer items is up for debate.

My consumer Phenom II X6 (125W version) can run under 100% load for hours without downclocking itself, the same is true for my laptop's Athlon QL-66 and was true for my former Core 2 Quads (Q6600 and Q9550) and Athlon X2 5200+ and any CPU I used before that.
I can't see any reason why I should expect it to be different with newer CPUs, consumer or not. If a CPU is advertised as 3.8GHz model it has to run 3.8GHz, not 3.4GHz.