mfenn
Elite Member
I understand the concept of racing to idle, but this wouldn't apply to an straight 8 hour gaming run as the processor would always be under load. Tomshardware did a test and the T processor used around 10 W less energy when under load for the same workload. Of course they also concluded that this saving is always mood because the processor has to run longer.
I just doubt I will feel the difference between both processors when gaming (though round turn times might take longer on Civ 5, but 10 sec is not going to make a big difference in the long run). So my question still remains: wouldn't a T processor be more energy efficient when I really don't look at finishing a workload but rather have it run under full load for the same amount of time? The question seems to be yes. What has to been seen though, if a T processor is powerful enough for gaming purpose.
Oh, I think I see the misunderstanding here. You've got to think about what happens on a very short timescale and apply it to the longer one. The race to idle principle applies over very short time scales (fractions of a second) just as it does for longer ones. Assuming that you are running at a constant framerate, the CPU will actually spend good chunk of its time between frames waiting idling (waiting on the GPU). The faster CPU will spend relatively more time idling, thus saving more power.
Also I'm not trying to run a 1,35V stick on 1,5V. I'm trying to figure out if a mainboard that says it supports 1,5V RAM modules can run a 1,35V at 1,35V.
I can't guarantee for a specific board, but every motherboard I've worked with will allow that.