A Surface Pro X Benchmark

I went to a local BestBuy to try out the Surface Pro X, and I ran a browser benchmark on the X as well as some of the other devices to compare.

As you already know, benchmarks can be misleading, and I wasn’t able to create a perfect lab benchmark in the demo environment; however, I thought some people would be interested before more official benchmarks and reviews come out.

Methodology

·        I used Speedometer 2.0: https://browserbench.org/Speedometer2.0/

·        Benchmark was only run once

·        All tests were run in Edge with only one tab (I was unable to confirm all browsers had the same edge version since the demo units did not allow me to view edge settings)

·        Task Manager was open and sorted by CPU

·        All apps were closed (e.g. the demo full screen app)

·        Power was plugged in

·        If power management allowed it, I set the battery slider to “Best Performance”

Results

·        13.5in Surface Laptop 3 (i5) – 57.6

·        Surface Pro 7 (i5) – 46.9

·        Surface Pro X – 44.3

·        Surface Book 2 (i7) – 42.8

·        15in Surface Laptop 3 (R5) – 41.8

·        Lenovo Yogo C630 (Snapdragon 850) – 25.7

·        Surface Go – 17.3 (Note: This was the 4GB version, and memory was reported as greater than 80%. This means disk swapping was occurring when the other device did not have that issue due to having 8GB ram)

After running these benchmarks, I felt the back of these devices for heat. The Surface Pro 7 was extremely hot. I could not feel any heat on the Surface Laptops. The Surface Pro X was not hot at all. You could feel the temperature difference where the processor is located on the device, but it was comfortable to hold. I would be unable to tell the difference if I wasn’t looking for it.

My assumption is that thermals are the reason why the 13.5in Laptop and Pro 7 had such a different score with the same specs.

Conversation 12 comments

  • jimchamplin

    Premium Member
    30 October, 2019 - 5:38 pm

    <p>That’s pretty solid. Was it classic Edge or Edge Insider? As far as I know Edge Insider doesn’t yet have an ARM native version, which is odd to me.</p>

    • Eagle

      31 October, 2019 - 4:13 pm

      <blockquote><em><a href="#484867">In reply to jimchamplin:</a></em></blockquote><p><br></p><p>It was stock Edge on the device, so I would assume classic Edge.</p>

      • jimchamplin

        Premium Member
        31 October, 2019 - 11:34 pm

        <blockquote><em><a href="#485316">In reply to Eagle:</a></em></blockquote><p>Thanks for the clarity. It’s pretty impressive what they did with the custom silicon in that thing!</p>

  • smidgerine

    Premium Member
    31 October, 2019 - 8:53 pm

    <p>Thanks for posting this. I preordered an X after taking back an iPad Pro. I hope they will hurry up with the native ARM ChrEdge.</p>

  • briantlewis

    01 November, 2019 - 9:34 am

    <p>What's damning about this is an iPhone 11 Pro scores 155.</p>

    • thejoefin

      Premium Member
      01 November, 2019 - 11:00 am

      <blockquote><em><a href="#485514">In reply to briantlewis:</a></em></blockquote><p><br></p><p>if the iPhone could run the programs I need then that score would be relevant. Windows on ARM vs Windows on Intel is a fair comparison. Windows vs iOS doesn't really make sense.</p>

    • KPixel

      01 November, 2019 - 11:38 am

      <blockquote><em><a href="#485514">In reply to briantlewis:</a></em></blockquote><p><br></p><p>As I wrote above, this is not a CPU benchmark.</p><p>It tests a very specific scenario (user interaction) that a mobile browser would implement very differently than a desktop browser.</p>

  • KPixel

    01 November, 2019 - 11:36 am

    <p>This Speedometer is very browser-dependent:</p><p>On my i7 8700K, I get a score of 117 with Chrome and 64 with Edge (Classic).</p><p><br></p><p>So, I think it is barely useful as a CPU benchmark.</p><p><br></p><p>Its About page says:</p><p>"Speedometer tests a browser's Web app responsiveness by timing simulated user interactions."</p><p>So, depending on how each browser defines "user interaction", the results will vary widely.</p>

    • Daishi

      Premium Member
      01 November, 2019 - 4:18 pm

      <blockquote><em><a href="#485537">In reply to KPixel:</a></em></blockquote><p>Sure, but at least in these examples they are, more or less, all using the same browser on the same OS. So while I wouldn’t want to use them as a definitive source on the performance of the Pro X it is at least an interesting data point. </p>

  • Vladimir Carli

    Premium Member
    05 November, 2019 - 1:14 pm

    <p>How can a surface book 2 i7 Be slower than a surface pro i5?</p>

    • digiguy

      Premium Member
      05 November, 2019 - 4:54 pm

      <blockquote><em><a href="#486618">In reply to Vladimir:</a></em></blockquote><p>because of different generations. SB2 is on 8th gen, while the i5 in the pro 7 is 10th gen and it's also faster than the i7 in the pro 6. 10th gen Ice Lake is the 10nm jump so it's also a jump in performance</p>

Windows Intelligence In Your Inbox

Sign up for our new free newsletter to get three time-saving tips each Friday

"*" indicates required fields

This field is for validation purposes and should be left unchanged.

Thurrott © 2024 Thurrott LLC