http://ultrabooknews.com/2014/10/16...erformance-tests-reveal-throttling-fan-noise/
this does not seem promising.
this does not seem promising.
I'm curious as to how Skylake turns out performance wise, and when it actually launches.Skylake seems to be on track for a launch in 2015.
http://seekingalpha.com/article/2563875-intel-skylake-development-appears-to-be-on-schedule
It performs as well as processors with 3x the TDP. I think that's very promising.http://ultrabooknews.com/2014/10/16...erformance-tests-reveal-throttling-fan-noise/
this does not seem promising.
I'm curious as to what changes Intel is going to make to improve performance. The bag of tricks has run pretty empty...I'm curious as to how Skylake turns out performance wise, and when it actually launches.
Nevertheless, I actually just bought a Z97-Pro and a Pentium AE to hold me over until Broadwell. Figured I would maximize my DDR3 and SATA 6Gbps drives (HDD and SSD) before doing a complete overhaul sometime in ~3ish years.
Best thing about haswell is that it did (re)introduce easy undervolting via Intel XTU. That why i bought into Haswell cause other than that, no reasons really to buy over Ivy. Undervolting is making such a huge difference in laptops...You can run at full turbo forever or at least at higher clocks. So many tend to throttle at stock voltage. If undervolting is still on with Broadwell, np.
It performs as well as processors with 3x the TDP. I think that's very promising.
Yes since it start the bench at 3 x its nominal TDP...at least since the peak TDP is 15W if there s enough thermal headroom.
Intel has learned well from Qualcomm and Apple.
It works well for bursty workloads, which is often what these types of devices handle.What is a waste is that to get apparently good numbers everyone is stretching the designs, people should understand once and for all that DT class performance is not possible in thermaly constrained devices other than for ultra short periods, and it s not a free lunch, quite the contrary, it s extremely counterproductive in matter of battery life; hence the slow or even absence of significant progres in this matter, batteries technology is not the only thing to blame.
What is a waste is that to get apparently good numbers everyone is stretching the designs, people should understand once and for all that DT class performance is not possible in thermaly constrained devices other than for ultra short periods, and it s not a free lunch, quite the contrary, it s extremely counterproductive in matter of battery life; hence the slow or even absence of significant progres in this matter, batteries technology is not the only thing to blame.
At least Qualcomm and Apple chips are decent enough to throttle in silenceIntel has learned well from Qualcomm and Apple.
Yes since it start the bench at 3 x its nominal TDP...at least since the peak TDP is 15W if there s enough thermal headroom.
At 1300 also...
So are you saying that a shrink and other improvements did nothing at all?
It seems rather weird that in a fanless tablet the score were so high and now in a notebook with fan they're less... either it's made of plastic and throttling a lot or this doesn't make sense.
Could be a windows setting? Maybe it's in eco mode... no sure it is!
Skylake seems to be on track for a launch in 2015.
http://seekingalpha.com/article/2563875-intel-skylake-development-appears-to-be-on-schedule
IIRC Intel demos were done with TDP set at 6.5W.
You think that this heatsink would have been necessary for 6.5W.?.
TDP was 6.5W BUT peak TDP was 15W, the bench start at 15W, as long as the CPU doesnt reach it max temp it will stay at 15W, once max temp is almost reached it will begin to throttle progressively down to 6.5W, on the big CB scores it s about sure that the CPU was running at full peak TDP, to add insult to injury if the YOGA is benched several times using CB you ll see the consecutive scores collapsing down to the level at wich the device thermal resistance is enough to sustain 6.5W or eventualy more, it all depend of the dissipation capability of the notebook.
The way Intel does this is to briefly spike the CPU up to 15W for a couple of microseconds in order for it to hit its top Turbo speed as quickly as possible.
'If I want to raise my voltage very fast and I want to go from 500MHz to 2.6GHz,' explained Piednoel. 'I better crank up the wattage just for a few micro-seconds just to get my voltage up. If you measure very closely you will see a peak, but that's not the TDP.'
So now you're sugesting that 14nm offers worse performance at 15W than previous gen (Haswell-U is quite a bit faster than your imaginary 15W TDP Broadwell-Y CPU-wise @ Intel reference platform).
Unless you have power consumption measurements for Core M during the entire benchmark please stop with this nonsense and deal with the fact that this is a very interesting Intel product.
TDP was 6.5W BUT peak TDP was 15W,
as long as the CPU doesnt reach it max temp it will stay at 15W, once max temp is almost reached it will begin to throttle progressively down to 6.5W,
You have a misunderstanding. PL3 is only reachable for 10ms. That won't affect much longer running benchmarks like Cinebench, although it might help for response rates.
Even for Haswell it mentions that bit and the duration is same at 10ms.
PL2 is what is reachable for above TDP levels. It can be set to tens of seconds, and its a 25% above TDP. That's the new one that came with Sandy Bridge chips. Which is what you are saying here:
The difference is the huge casted heatsink used in their demo by Intel, i did point that the back cover was a piece of work impossible to duplicate in mainstream designs and that thermal storage would be very limited, heck it did cost several hundreds thousands $ to cast the back cover, machining is marginal at this level...
That seems kinda a lot, look at what tablet BTs use as a heatsink in comparison.
don't mind his nonsense. That was the rear metal chassis of the tablet. Obviously they dont really cost thousands of dollars, as many retail tablet use the same setup.
The back plate works as a heat-sinkThis is a Broadwell based Core M powered tablet; it doesn't have a codename, however, it is based on a modified Llama Mountain platform. It carries a Core M 5Y70, which was benchmarked right in front of us, to display how much faster it was than Atom. What is really amazing about this processor is the amount of processing power it packs into a CPU with a TDP of only 4.5W. This device is actually cooled through its back plate.