01:51
@qasdfdsaq Likewise, I just said they can't sustain burst.
@qasdfdsaq You're the one that brought a DOTA 2 measurement into this.
@qasdfdsaq Oh? Which one?
@qasdfdsaq ...as far as that, without hard data it's kinda hard to say. I made a pretty big assumption ("if they have a TDP around that of an Atom").
I didn't say that was the case. I'll say now that that's probably not the case.
@qasdfdsaq There are always situations where you can achieve max performance if you want. Those situations might even be common.
However, you're then operating above specified TDP.
You're operating above the manufacturer-recommended/required cooling specs.
More importantly, you end up playing with different efficiency (and power consumption) numbers than the spec sheet would suggest.
If you're getting more performance, but you spend more time at, say, 6 W power dissipation (or even 12 W at peak, reportedly) -- sure, your cooling solution might handle it but that doesn't automatically make you more efficient than a lower-power (both energy consumption and performance) chip.
On the other hand, you might be more efficient per watt but the lower-power chip will be more suitable to different conditions.
I don't know about you but I don't consider a (passively cooled or not) tablet at 70 degrees Celsius to be comfortable to hold or use.
If I threw more cooling at an Atom chip, I could probably overclock it significantly.
Probably not performance on par with a Core M, to be sure, (mostly due to architectural differences) but you'll see a significant boost.
There are two comparisons here: by specced TDP and by real-world performance in real devices.
I go by specced TDP because that's what the manufacturer recommends. That's what it's theoretically designed for. It's the easiest way to directly compare competing chips, if not necessarily the most accurate.
Real-world performance is more interesting when shopping for devices. You might also see what a chip is capable of at max when given all the resources it wants -- traditionally that's been the realm of overclockers.
> The processor TDP is the maximum sustained power that should be used for design of
the processor thermal solution. TDP is a power dissipation and junction temperature
operating condition limit, specified in this document, that is validated during
manufacturing for the base configuration when executing a near worst case
commercially available workload as specified by Intel for the SKU segment. TDP may
be exceeded for short periods of time or if running a "power virus" workload.
So, it's possible that the specced TDP actually assumes a sustained clock higher than base.
But it's hard to be sure.