Synthetic and Legacy Results (15W)

The realm of synthetic testing is a tricky one, given that there are plenty of benchmarks in the wild that provide a number, but aren’t actually based on real workloads, or are very limited in what they actually test. The issue here is that this software tries to emulate real-world, but it isn’t immersed in the harnesses or matrix of what a user might actually experience. For that reason, we only tend to use these benchmarks based on reader requests.

Legacy benchmarks are included for similar reasons, but can help to get a historical perspective.

GeekBench ST

GeekBench MT

x264 HD 3.0 Pass 1

x264 HD 3.0 Pass 2

System Results (15W) Gaming Results (15W and 25W)
Comments Locked

261 Comments

View All Comments

  • Xyler94 - Sunday, August 4, 2019 - link

    Slower in games, faster in almost everything else (when the R7 1800X came out, the i7 at the time was the i7 7700k)

    Just like today, the Ryzen 9 3900x is slightly slower in most games (beating it in others) while smashing the i9 9900k in almost any other task. Funny how now we have two CPUs for two different things, huh? Competition is good~
  • Thunder 57 - Sunday, August 4, 2019 - link

    No point in trying to talk sense into a grade A troll. You would think AMD beat him with a belt as a kid or something.
  • Korguz - Friday, August 2, 2019 - link

    um HStewart, " Big difference is power difference 0 series are 9V while U is 15/28 and also 28." 9 volts??? again.. do you NOT not know the difference between WATTS and VOLTS ?? i guess by YOUR logic, U uses 15 and 28 VOLTS ????
  • quadibloc - Friday, August 2, 2019 - link

    It's easy to focus on AMD and say that the reason Intel didn't do a pure "tick" release this time was because it couldn't afford to pass up any opportunity for improvement in a more competitive environment. But this forgets another factor: during the years of the "tick-tock" strategy, Dennard Scaling was operative, so a process shrink of itself brought a major performance improvement.

    Now, without Dennard Scaling, even without serious competition from AMD, if Intel had brought out a process shrink without architectural improvements, people would have said "What's the point?".
  • 0ldman79 - Friday, August 2, 2019 - link

    Ignore the whiners.

    Good tech preview.

    Looking forward to a full on review.

    The SPEC numbers vs the 3900x are pretty insane. Is that single thread or what? Are the benches short enough to run within the max turbo power envelope?

    I'm trying to understand how a 15W Ice Lake and 15W Whiskey Lake are comparable to a 9900K or 3900x.
  • Alistair - Friday, August 2, 2019 - link

    Of course it is single threaded (Hence Ian saying words like IPC in that section). There are 0 reasons you'd pick this CPU to compete against the 3900x in multi core workloads.
  • Phynaz - Friday, August 2, 2019 - link

    Show me a laptop with a 3900x.

    Thought so.

    Idiot.
  • Alistair - Saturday, August 3, 2019 - link

    Are you dumb? Single core performance is equal, not multicore (it is 1/3 the speed).
  • Xyler94 - Sunday, August 4, 2019 - link

    Show me a laptop with an i9 9900ks which can handle the insane tdp it needs.

    Thought so,

    Idiot
  • R6E7980XE - Saturday, August 3, 2019 - link

    “For Sunny Cove this has increased, but it gets a bit more complicated. In one clock Sunny cove can perform 2x64-byte reads and either 1x64-byte write, or 1x32-byte write, or 2x16-byte writes.”

    @Ian Cuttress Please correct me if I’m remember wrong.

    Sunny Cove Microarchitect consisted 2 x 64 Byte / cycle Load and 2 x 64 Byte / cycle Store.
    according to https://en.wikichip.org/wiki/File:sunny_cove_block...

Log in

Don't have an account? Sign up now