I get more insulted when companies disable features on their "enthusiast" parts and then charge a premium for the privilege...
I get more insulted when companies disable features on their "enthusiast" parts and then charge a premium for the privilege...
And how much slower than an i7 3930/60/70 is the FX-9590
GF process can endure slightly higher voltage than intels.
My i5 is uses a $50 air cooler, I'm not sure what point you were attempting to make.
I was only alluding to the fact that this seems to simply be a factory overclocked 8350, nothing more.
Both the rumored 220w TDP and $1000 price tag are clearly jokes.
I expect this to be true because it is a 32nm SOI node compared to a 22nm finfet node, but is there any actual proof or evidence that supports this statement?
So far as I am aware, neither Intel nor AMD have published a max allowed Vcore for the respective products under discussion here.
I am surprised about how many people care about energy efficiency for high end chips...
You should have been here in 2005, you couldn't get through ten posts without a pro-AMD member talking down how foolish it was to buy a Prescott P4 because the energy efficiency was so poor compared to their fancy SOI-based X2 chips.
The pro-Intel members would respond with something more or less identical to your post above.
What you see today in these forums is nothing new for those of us who've been around a while. The only thing new is that the shoe is now on the other foot now and it is the pro-AMD members who are now arguing that energy efficiency is a red herring.
I expect this to be true because it is a 32nm SOI node compared to a 22nm finfet node, but is there any actual proof or evidence that supports this statement?
So far as I am aware, neither Intel nor AMD have published a max allowed Vcore for the respective products under discussion here.
So far as I am aware, neither Intel nor AMD have published a max allowed Vcore for the respective products under discussion here.
Because Intel has it's chips made at Global Foundries. Please don't pull any more of this out of you know where. TY.
For thoses who dont waste time on semantics this means
GF process can endure slightly higher voltage than intels.
In that context it's fair. But the reaction seems more to be disgust at the CPU drawing 220W, when in reality any overclocked CPU has fairly horrific power consumption. TDP in performance desktops is not as important as CPUs in mobile form factors.
My point is, if TDP on the desktop REALLY mattered to people, they wouldn't overclock. The act of overclocking is contradictory to the pursuit of efficiency.
everything matters at some point. 220 watts for out of the box TDP is beyond laughable and is of no interest to most people here even if it was a $400 cpu. you can get the same performance with less than half the power usage. my 2500k oced to 4.4 uses less than 100 watts which is a pretty good balance.In that context it's fair. But the reaction seems more to be disgust at the CPU drawing 220W, when in reality any overclocked CPU has fairly horrific power consumption. TDP in performance desktops is not as important as CPUs in mobile form factors.
My point is, if TDP on the desktop REALLY mattered to people, they wouldn't overclock. The act of overclocking is contradictory to the pursuit of efficiency.
my 2500k oced to 4.4 uses less than 100 watts which is a pretty good balance.
I just ran Fritz again at 4.8GHz, idle is 40w, load was 140w at the wall, so the cpu couldn't have been drawing more than 100w for 15579 nodes per second.
The joke with 220w TDP is that its not going to be 220w power draw, much closer to 300w actually if all the review samples need 1.55v :|
I just ran Fritz again at 4.8GHz, idle is 40w, load was 140w at the wall, so the cpu couldn't have been drawing more than 100w for 15579 nodes per second.
The joke with 220w TDP is that its not going to be 220w power draw, much closer to 300w actually if all the review samples need 1.55v :|
Likewise I find it hard to imagine anyone with $1000 to blow on a CPU would even look at this chip.
We're already seeing the 8350 draw around 334w at 4.8GHz 1.5v, of course that's total system though. But it's hard for me to budget too much of that for the rest of the system considering the idle is very similar to Haswell.
If you look Haswell is around 80w idle for them, while the 8350 is around 90, this is stock, and once loaded up for them Haswell "jumps" to 132w giving it about a 50w delta, whereas the 8350 jumps from 90w to 283, or basically 193w accounted for the cpu.
Likewise if we look at Haswell overclocked, 80 -> 162 shows a 82w delta, on the other hand 8350 90 -> 334 (4.8GHz 1.5v) has a 244w delta. So it's not hard for me to stretch this new $1000 chip of AMD out to 260-270w at stock.
Likewise I find it hard to imagine anyone with $1000 to blow on a CPU would even look at this chip.
We're already seeing the 8350 draw around 334w at 4.8GHz 1.5v, of course that's total system though. But it's hard for me to budget too much of that for the rest of the system considering the idle is very similar to Haswell.
If you look Haswell is around 80w idle for them, while the 8350 is around 90, this is stock, and once loaded up for them Haswell "jumps" to 132w giving it about a 50w delta, whereas the 8350 jumps from 90w to 283, or basically 193w accounted for the cpu.
Likewise if we look at Haswell overclocked, 80 -> 162 shows a 82w delta, on the other hand 8350 90 -> 334 (4.8GHz 1.5v) has a 244w delta. So it's not hard for me to stretch this new $1000 chip of AMD out to 260-270w at stock.
In that context it's fair. But the reaction seems more to be disgust at the CPU drawing 220W, when in reality any overclocked CPU has fairly horrific power consumption. TDP in performance desktops is not as important as CPUs in mobile form factors.
My point is, if TDP on the desktop REALLY mattered to people, they wouldn't overclock. The act of overclocking is contradictory to the pursuit of efficiency.
This chip drawing 220W as opposed to the 244W of that 8350 can be explained by silicon quality. I don't know what Piledriver is like, but I remember with the 8150, there were absolutely massive differences in leakage between individual samples.