System Performance

For the new model, Lenovo has taken the step of moving from the 15 watt Haswell-U series processors to the 4.5 watt Broadwell-Y based Core M. Their first run of Yoga 3 Pros came with the Core M-5Y70, which has a base frequency of 1.1 GHz, and a boost of 2.6 GHz. Lenovo has refreshed the Yoga 3 Pro to include the 5Y71 processor, which increases the base clock to 1.2 GHz and boost to 2.9 GHz. It will be interesting to see the performance delta between the Yoga 3 Pro and the Yoga 2 Pro.

The Broadwell-Y processors have two cores and hyperthreading. With a thermal envelope of just 4.5 watts though, boost clocks with heavily multithreaded workloads will be reduced as compared to the U series processors. During a multi-threaded run of Cinebench R15, the Core M processor would hover between 1700 and 1800 MHz, but the Core i5-5200U would do the same workload at 2400 MHz and not deviate at all. On single-threaded workloads, the Core M has more headroom to allow the one core to clock higher, and on a single-threaded run of Cinebench R15 it would clock as high as 2600 MHz. However there was a lot of variance during the run, and the Core M-5Y71 would fall as low as 1200 Mhz, with the majority of the time spent between 2000 Mhz and 2200 Mhz. Broadwell-U on the other hand would stay exactly at 2400 Mhz again during the same workload, with very little change in frequency during the run. As compared to Broadwell-U with a 15 watt TPD, the Broadwell-Y limits clock speed quite a bit more. Let’s go through some benchmarks to see how much this affects the Yoga 3 Pro, and especially if the IPC improvements with Broadwell can make up for the lower TDP.

For our benchmarks, I have selected a sample of similar devices in which to compare against. I have also included the Dell XPS 15, which has a quad-core CPU, GT750 GPU, and 91 Wh battery to bracket these scores on the high end. To compare against the low end, I have included the HP Stream 11, which has the Atom N2840 processor. It is interesting to note that the Atom (rebranded as Celeron) has a slightly higher TDP (7.5 watts vs 4.5 watts of the Core M) and it will be a good comparison to see how much more performance Core has than Atom. To compare the Yoga 3 Pro against any other devices, please check out our Notebook Bench.

PCMark 8 - Home

PCMark 8 - Creative

PCMark 8 - Work

PCMark 8 - Storage

PCMark 7 (2013)

PCMark 8 has four workloads which are Home, Creative, Work, and Storage. Each has a different set of workloads which Futuremark believes represent a good sampling of tasks for each category. Here the Yoga 3 Pro does very well, with scores that are more often than not well above the Yoga 2 Pro’s Core i5-4200U. The Broadwell Core i5s in the XPS 13 still pull a good lead here. The 7.5 watt Bay Trail-M in the HP Stream is sorely outclassed by the 4.5 watt Core M.

TouchXPRT 2014 Overall Score

TouchXPRT 2014 Beautify Photos

TouchXPRT 2014 Blend Photos

TouchXPRT 2014 Convert Videos for Sharing

TouchXPRT 2014 Create Music Podcast

TouchXPRT 2014 Create Slideshow from Photos

Our next benchmark is TouchXPRT 2014, which is a Windows Store app. It performs several common tasks for a typical home user, and converts the time taken for each task into an overall score. Once again, the 4.5 watt Core M compares very favorably to the 15 watt Yoga 2 Pro processor. In almost all of the tests, Core M is ahead of the Haswell Core i5-4200U of the Yoga 2 Pro. In order to achieve these scores, the GPU driver was updated to the latest version on the Intel site which fixes a problem with the Video Conversion score. We had the same issue with the XPS 13, so if you are the owner of a Broadwell laptop, you may want to update your drivers. The Atom core once again is decimated by Core M.

Cinebench R15 - Single-Threaded Benchmark

Cinebench R15 - Multi-Threaded Benchmark

Cinebench R11.5 - Single-Threaded Benchmark

Cinebench R11.5 - Multi-Threaded Benchmark

x264 HD 5.x

x264 HD 5.x

Cinebench is a good long benchmark which shows us how the CPU will perform when maxed out for the duration of the benchmark. Here we can see where Core M starts to falter. On long sustained workloads, it has to cut the CPU frequency to keep within its thermal envelope, whereas the Broadwell-U and Haswell-U CPUs can keep the clocks a lot higher. While running the multithreaded R15, the Core M would hover around 1800-1900 MHz with all cores in use, and the Broadwell-U would lock in at 2400 MHz and stay there for the duration. x264 is similar, and is a very long benchmark. The higher power envelope of the U series processors allow for more work to be done. You can see the quad-core Dell XPS 15 really pulls ahead on these benchmarks, and the HP Stream continues to fall well behind.

Mozilla Kraken 1.1


Once again, in web use, the Core M processor is very similar to the outgoing Haswell U based Yoga 2 Pro. Just to put the numbers in a bit more context, I also ran the benchmarks on my Core i7-860 based Desktop (running Chrome, as were the Yogas) and it is pretty clear just how far we have come. The i7-860 is a four core, eight thread 45 nm processor with a 2.8 GHz base clock and 3.46 GHz boost, all in a 95 watt TDP. It was launched in late 2009. Five years later, we have higher performance in a 4.5 watt TDP for many tasks. It really is staggering.

There seems to be a general concensus that Core M equates slow, but clearly this is not the case. For some sustained workloads, yes, the 4.5 watt TDP limits how much performance you are going to get from the CPU, but for many tasks, especially short burst loads, the performance of the 5Y71 is very competitive, often outperforming the Haswell Core i5-4200U from last year’s Yoga 2 Pro. This says a tremendous amount about the Intel 14 nm process, because the IPC improvements of Broadwell vs Haswell are fairly limited. Clearly the CPU has quite a bit of headroom on the 14 nm process to keep the clock speeds up.


Display GPU Performance


View All Comments

  • name99 - Friday, March 13, 2015 - link

    Good luck with that. Intel puts more care into the design of the glue those stickers use than practically anything else they ship.
    After you do finally manage to pull the sticker off, you're left with a nasty glob of gunk on your machine, and it's not clear quite what will remove it while not hurting the finish of the machine. Acetone? Alcohol? Vinegar?
  • fokka - Friday, March 13, 2015 - link

    try baby oil. i agree though, those pesky stickers should simply use the glue they use on post-its, that would be nice for a change. or, you know, get rid of those useless things all together. Reply
  • jabber - Saturday, March 14, 2015 - link

    Actually I've found gently prising them off slowly lifts most of the glue with the sticker, especially when they are new. Then use the sticker to dab over any glue left gets 99% of it. Not difficult. Reply
  • mkozakewich - Saturday, March 14, 2015 - link

    Oils destabilize rubbers. Try accidentally leaving a bit of turkey stock in the rubber part of a baster, or try eating peanuts while chewing gum. Reply
  • Alexvrb - Sunday, March 15, 2015 - link

    I would probably use Goo Gone, if I was prone to caring about removing factory stickers. Reply
  • Manch - Sunday, March 15, 2015 - link

    wd40. put a spot on a rag and wipe. then clean with alcohol. Reply
  • limitedaccess - Friday, March 13, 2015 - link

    Can verify that it is actually 4.5w TDP and not set for 3.5cTDP? The original was version was found to use the lower cTDP setting which is why performance was low (especially in gaming).

    Was the performance actually sustainable? For instance the SP3 i3 did not show performance regression in gaming benchmarks until after the first pass of Dota 2.

    I'm also wondering whether or not mobile reviews from now on should start looking into sustained performance and perhaps monitoring what clock rates actually are when in use.

    What were the temperatures?
  • limitedaccess - Friday, March 13, 2015 - link

    Just to add in general I'm wondering in mobile reviews are they actually being down to account for performance differences under extended use?

    We now see the market move towards more variable and burst based performance I think this does need to be looked at more. A 10 minute gaming benchmark for instance may not represent an actual extended game play session if this isn't accounted for.
  • Brett Howse - Friday, March 13, 2015 - link

    CPU Z shows it as a 4.5w part, and the clock speeds agree. I actually was hoping it would be cTDP up to 6w since it has a fan but it does not appear to be.

    Sustained performance did not appear to be a big factor. I ran Cinebench R15 six times in a row (which took about an hour total) and all of the scores were within a fraction of a percentage of each other.
  • fokka - Friday, March 13, 2015 - link

    i think a cTDP of 6w would be a better fit for an actively cooled 13-inch machine, hell, i could even get behind a cTDP of 10 watts. the gap between 4.5 and 15w just seems so big otherwise. but with power states and turbo clocks diluting the whole TDP-game anyways, i don't even know what i want in a cpu anymore. o.O Reply

Log in

Don't have an account? Sign up now