Surface Book Discussion Thread

Discussion in 'Microsoft' started by DRTigerlilly, Oct 6, 2015.

  1. Utan-chan

    Utan-chan Pen Pal - Newbie

    Messages:
    37
    Likes Received:
    8
    Trophy Points:
    16
    Guys, here's a hands-on with the new Surface Pen and the Pen Tip/Nib Kit:
    Looks like there's still some problems with palm rejections. Hopefully Microsoft will fix it soon...
     
  2. excalibur1814

    excalibur1814 Scribbler - Standard Member Senior Member

    Messages:
    2,386
    Likes Received:
    383
    Trophy Points:
    101
    Tried to pre-order one for my boss, in the U.K., but i couldn't even find it in the store (Didn't look for long). Grumble.
     
  3. ATIVQ

    ATIVQ V⅁O⅄ Senior Member

    Messages:
    1,043
    Likes Received:
    588
    Trophy Points:
    131
    Geekbench has three very important advantages:
    • Basic benchmarking is free
    • There is a single benchmark for all devices
    • It has an extensive benchmark result library
    The benchmarking software you suggested has 10 different products (which one should I use?) that can't be compared between different devices and their online result comparison library is sparse. Oh and their Windows benchmarking software is 2.4GB.

    So even if the results skew towards Apple, I'll stick to Geekbench.
     
  4. dstrauss

    dstrauss Comic Relief Senior Member

    Messages:
    11,143
    Likes Received:
    9,753
    Trophy Points:
    331
    ATIVQ - not to question your motives, but your criteria for choosing Geekbench is that it is free, one result, with lots of results - yet it is skews towards one platform - Apple? That doesn't exactly exactly respond to Sonic's argument, but in fact solidifies his objection.

    Kind of like saying my Prius beats all other cars because it averages 40 mpg (which it does); despite the fact I CAN'T carry more than 3 people (without all their luggage); off-road with it; or carry a 4x8 sheet of plywood.
     
  5. dstrauss

    dstrauss Comic Relief Senior Member

    Messages:
    11,143
    Likes Received:
    9,753
    Trophy Points:
    331
    Vaga, Utan-chan and Mesosphere like this.
  6. ATIVQ

    ATIVQ V⅁O⅄ Senior Member

    Messages:
    1,043
    Likes Received:
    588
    Trophy Points:
    131
    I wasn't contradicting him, I was stating why Geekbench is preferable. If he thinks he needs 2.4GB worth of benchmarking software to roughly judge the performance of a device, great, but I doubt the average forum-goer is going to go through all that for a simple benchmark.
     
  7. dstrauss

    dstrauss Comic Relief Senior Member

    Messages:
    11,143
    Likes Received:
    9,753
    Trophy Points:
    331
    But if that "simple" solution is skewed towards one competing vendor, and is only used because it is free and a single score, how is that a service to forum users?
     
  8. ATIVQ

    ATIVQ V⅁O⅄ Senior Member

    Messages:
    1,043
    Likes Received:
    588
    Trophy Points:
    131
    People just want to know roughly how their device stacks up. When someone comes here and says "I think my processor is capped, how can I be sure?" I'm not going to point them to a 2 GB download. I'm going to point them to Geekbench and within three minutes they'll have a rough estimate of their device's performance.

    People who come to the forum don't need a complex benchmark. They just need to know their device's performance is within the ballpark of its expected performance.

    Nobody is ever going to go, "well the iPad Air 2 scores 1800 in Geekbench and the Dell Venue 11 Pro scores 3400, I'm going to get the Dell!" That's simply not what people use these benchmarks for, so it's irrelevant how inaccurate it is, as long as it's accurate enough to tell you on what playing field the device is performing. So the A8X is heavily optimized towards SHA1 multicore algorithms and that skews its Geekbench multicore performance up by 50%? So what. People who understand the last sentence know how to benchmark using their own software, and people who don't understand it just need a tool that spits out a number that gives them a rough idea of the device's performance.
     
  9. dstrauss

    dstrauss Comic Relief Senior Member

    Messages:
    11,143
    Likes Received:
    9,753
    Trophy Points:
    331
    There's the fallacy in a nutshell - when it "spits out a number that gives them a rough idea of the device's performance" AND that test "skews its Geekbench multicore performance up by 50%" towards one vendor who either: (a) optimizes for that test, or (b) the test maker bends towards that vendor, it is not even a rough idea of relative performance.

    It's like saying one of two identical cars is faster than the other, without disclosing the winner had a turbocharger and the loser did not...
     
  10. JoeS

    JoeS I'm all ears Senior Member

    Messages:
    5,730
    Likes Received:
    4,048
    Trophy Points:
    331
    Plus all those poor saps who break their clipboard screen will put those sweet GPU holding keyboards on eBay.
     
    Last edited: Oct 21, 2015
    Vaga and dstrauss like this.
Loading...

Share This Page