Analyzing Intel Core M Performance: How 5Y10 can beat 5Y71 & the OEMs' Dilemma
by Brett Howse & Ian Cutress on April 8, 2015 8:00 AM EST3DMark Sky Diver Results
Most of the previous benchmarks were day to day tasks. Some involved the GPU, but it was never the focus. We will now move on to benchmarks which focus on GPU performance to see what kind of an effect this can have. Remember, the TDP of Core M is 4.5 watts including the integrated graphics, so any thermal room needed for graphics is going to come at the expense of the CPU. 3DMark Sky Diver is aimed for gaming laptops and mid-range PCs, so it is a bit too much load for integrated graphics. But it does feature DirectX 11 and includes both graphics and physics tests. The benchmark is around five minutes long.
We can see that the Core i5 continues shrugging off these tests. While the SoC did heat up, the GPU frequency was flat throughout the results. This is quite a bit different than all of the Core M processors, which had to throttle both the CPU and GPU as needed. It is very interesting especially in the UX305 results to see that the GPU is throttled on high CPU workloads to give more headroom for the CPU, which you can see on the third heat spike in its graph. This would be the physics test, which relies heavily on the CPU. The Dell Venue 11 and Yoga 3 Pro had very different temperature curves, and the Yoga 3 Pro had to throttle the GPU quite a bit to stay at its target SoC temperature.
Looking at the average CPU frequencies reaffirms what we have seen in previous results. The ASUS, despite having the lowest turbo frequency, has the highest average for the Core M devices. But it is the GPU frequencies which are the most important in this test.
All of the Core M devices had to throttle the GPU to some extent, but the ASUS did the least. The Yoga 3 Pro and Dell Venue 11 Pro were basically tied in average GPU frequency for the duration of this test. GPU workloads can pull a lot of power into the SoC, which can raise temperatures as we will see in the next graph.
Looking at the SoC temperatures explains the results. The Yoga 3 Pro has an average of 65.2°C, which is the target temperature for the Yoga. This means it was not able to leverage the breaks in workloads to ramp up its higher turbo frequencies when needed. The Dell Venue 11 is at almost 90°C for the benchmark, and that is also its limit. The ASUS, with its better cooling, manages to basically mirror the Core i5 for SoC temperature.
The excellent cooling of the ASUS form factor shines in the GPU tests. For the overall score, it comes very close to the Core i5. Both of the 5Y71 devices struggle under sustained GPU workloads, as the scores confirm.
Facebook
Twitter
RSS
110 Comments
View All Comments
maxxbot - Wednesday, April 8, 2015 - link
If the device buyer's choice is between the Core M and an ARM or Atom they're going to go with the Core M because it's faster in every aspect, especially burst performance. If the Core M in unacceptable slow for you then there aren't any other options at the 4.5W TDP level to turn to, it's the best currently available.name99 - Wednesday, April 8, 2015 - link
That ("Maybe Intel made too many compromises") seems like the wrong lesson.I think a better lesson is that the Clayton Christensen wheel of reincarnation has turned yet again.
There was a time more than 40 years ago when creating a computer was a demanding enough exercise that the only companies that could do it well were integrated top to bottom, forced to do everything from designing the CPU to the OS to the languages that ran on it.
The PC exploded this model as standardized interfaces allowed different vendors to supply the BIOS, the OS, the CPU, the motherboard, the storage, etc.
BUT as we push harder and harder against fundamental physics and what we want the devices to do, the abstractions of these "interfaces" start to impose serious costs. It's no longer good enough to just slap parts together and assume that the whole will work acceptably. We have seen this in mobile, with a gradual thinning out of the field there; but we're poised to see the same thing in PCs (at least in very mobile PCs which, sadly for the OEMs, is the most dynamic part of the business).
This also suggests that Apple's advantage is just going to keep climbing. Even as they use Intel chips like everyone else, they have a lot more control over the whole package, from precisely tweaked OS dynamics to exquisitely machined bodies that are that much more effective in heat dissipation. (And it gets even worse if they decide to switch to their own CPU+GPU SoC for OSX.)
It's interesting, in this context, where the higher frequency 1.2GHz part is difficult for some vendors to handle, to realize that Apple is offering a (Apple-only?) 1.3/2.9GHz option which, presumably, they believe they have embodied in a case that can handle its peak thermals and get useful work out of the extra speed boost.
HakkaH - Friday, April 10, 2015 - link
Device buyers don't even see beyond the price tag, brand name and looks. 90% of the people who buy tech are pretty oblivious on what they are buying. So they wouldn't even know if a device would throttle the speed at all.Secondly I'd rather have a device that throttles good which processors are doing the last couple of years than have a steady pace at which it just crawls along and maybe after 5 minutes decides... hey maybe I can add 200 MHz and still be okay. If that is your case I bet you still have the first generation smartphone in your pocket instead of a more recent model because they all aggressively throttle the CPU and GPU in order to keep you from throwing your phone out of your hands ;)
HP - Saturday, August 8, 2015 - link
Your description doesn't follow the usage paradigm of most computing tasks. As the user is actively using their device what they do on the machine roughly tracks the user's thought patterns which largely takes place in series. He doesn't batch the tasks in his head first and then execute them. So race to sleep is where it's at.milkod2001 - Wednesday, April 8, 2015 - link
What about Intel's native 4 core mobile CPUs. Are any in the works?Core M,Y, U(2 core) etc might be OK for bloggers, content consumers etc but if one wants/needs real performance on the go, there's not that much new to offer, right?
nathanddrews - Wednesday, April 8, 2015 - link
I think we'll have to settle for the i7-4700 until Skylake. Not a bad place to settle.kpkp - Wednesday, April 8, 2015 - link
"Atom competed against high powered ARM SoCs and fit in that mini-PC/tablet to sub 10-inch 2-in-1 area either running Android, Windows RT or the full Windows 8.1 in many of the devices on the market."Atom in Windows RT? Wasn't RT ARM only?
Essence_of_War - Wednesday, April 8, 2015 - link
Very impressed by the Zenbook, especially at its price point.boblozano - Wednesday, April 8, 2015 - link
Thanks for the detailed article.In this space it's clear that the top design consideration is cooling - do that well, and everything else follows. Performance will be delivered by the SoC's ability to turbo as needed, power consumption by the SoC and the rest of the design.
Of course materials, size, the question of passive vs. active cooling ... all that also factors decisively into the success of a design, whether the target market actually buys the devices.
But the effectiveness of the cooling will largely determine performance.
Refuge - Wednesday, April 8, 2015 - link
The efficiency of the cooling too. Can't have it take up too much space or too much power (If active and not passive)otherwise you leave either no room for your battery, or you drain it too fast keeping the thing cool (In the case of active)