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
seapeople - Thursday, April 9, 2015 - link
Won't an over-aggressive turbo actually decrease performance? Processors are generally less power efficient at higher clock speeds, i.e., running at 3GHz is twice as fast as 1.5GHz but generally uses more than 2x the power, and thus more than 2x the heat.In this case, therefore, a processor that races to 3GHz will quickly (and less efficiently) use up its thermal headroom and have to throttle back moreso than a processor that stayed at 2GHz.
It's like a footrace - if the race is 100m long, you're going to finish fastest if you go all out. However, if the race is a mile long, then the guy who starts off sprinting is going to be sputtering along a quarter of the way into the race as the joggers pass him up.
MrSpadge - Friday, April 10, 2015 - link
You are right that with agressive Turbo the chip is running in a less power efficient state initially and will have to throttle a bit earlier than a slower, steadily running chip. but if we're talking about low performance under sustained loads, this doesn't matter: it affects the first few seconds, or 10's of seconds at most, whereas in the following minutes both systems are running at the same power efficient throttled speed, which is basically determined by the system cooling. It's not like the sprinter who's completely exhausted and can't recover.retrospooty - Wednesday, April 8, 2015 - link
I dont think its really all that complicated... If you are looking for raw performance, Core M isnt for you. It is really for low power devices that do basic stuff like browsing, email etc. For that purpose, its one hell of a CPU. That performance level at 4.5 watts is a hefty accomplishment IMOYuLeven - Wednesday, April 8, 2015 - link
I do development on a Core M machine. Instead of carrying 4 pounds of computing power on my back, I let a cloud based development box do the heavy lifting. The plume light Core M notebook is used basically to write the code and give orders to the Dev box. IMHO opinion a far better setup than having scoliosis for the sake of running code locally.mkozakewich - Wednesday, April 8, 2015 - link
It's not for web browsing. That's what Atom is for. A Core-M device is good for all regular core tasks except sustained graphics tasks. I wouldn't get one to game, but it'll be great for anything else.retrospooty - Thursday, April 9, 2015 - link
That is pretty much exactly what am saying. Basic use, core M is fine. Not for high performance requirements.nathanddrews - Wednesday, April 8, 2015 - link
They have taken the exact opposite approach to their SSD design, where they try very hard to offer constant and consistent performance.xthetenth - Wednesday, April 8, 2015 - link
Both make sense from the perspective of increasing perceived speed. With storage, it hanging and being slow is the biggest way it can impact the feel of the device, while processors that trade finishing short tasks much faster for a tiny decrease in how fast they complete long tasks do a lot to achieve a responsive feel.xthetenth - Wednesday, April 8, 2015 - link
Device buyers don't buy devices to get a higher average frequency, they buy things to do what they want without the device holding them up. Look at the benchmarks where the ASUS holds higher average frequencies but the Yoga's higher maximum frequency means it completes tasks faster, and it performs better in the benchmark. That sort of responsiveness is what turbo is for. The time to complete long tasks isn't going to be materially changed but the time to complete short tasks is going to be reduced significantly if the processor can use a quick burst like turbo allows.I'm also pretty sure that most users consider not getting burned by their device a good thing that should continue, incidentally.
StormyParis - Wednesday, April 8, 2015 - link
That's not a real use case though. Real use case is load a page (low CPU), render page (high CPU) read page (low CPU). I don't care how fast my CPU is idling while I'm reading the page, I do care how fast the page renders. It'd be different if I were running simulations.. that's what desktop CPUs are for.