SPEC - Single-Threaded Performance

Single-thread performance of server CPUs usually isn’t the most important metric for most scale-out workloads, but there are use-cases such as EDA tools which are pretty much single-thread performance bound.

Power envelopes here usually don’t matter, and what is actually the performance factor that comes at play here is simply the boost clocks of the CPUs as well as the IPC improvement, and memory latency of the cores. 

The one hiccup for the Xeon 8380 this generation is the fact that although there’s plenty of IPC gains to be had compared to previous microarchitectures, the new SKU is only boosting up to 3.4GHz, whereas the 8280 was able to boost up to 4GHz, which is a 15% deficit.

SPECint2017 Rate-1 Estimated Scores

Even with the clock frequency disadvantage, thanks to the IPC gains, much improved memory bandwidth, as well as the much larger L3 cache, the new Ice Lake part to most of the time beat the Cascade Lake part, with only a couple of compute-bound core workloads where it falls behind.

SPECfp2017 Rate-1 Estimated Scores

The floating-point figures are more favourable to the ICX architecture due to the stronger memory performance.

SPEC2017 Rate-1 Estimated Total

Overall, the new Xeon 8380 at least manages to post slight single-threaded performance increases this generation, with larger gains in memory-bound workloads. The 8380 is essentially on par with AMD’s 7763, and loses out to the higher frequency optimised parts.

Intel has a few SKUs which offers slightly higher ST boost clocks of up to 3.7GHz – 300Mhz / 8.8% higher than the 8380, however that part is only 8-core and features only 18MB of cache. Other SKUS offer 3.5-3.6GHz boosts, but again less cache. So while the ST figures here could improve a bit on those parts, it’s unlikely to be significant.

SPEC - Multi-Threaded Performance SPEC - Per-Core Performance under Load
POST A COMMENT

170 Comments

View All Comments

  • rahvin - Sunday, April 18, 2021 - link

    The bigger the silicon ingot the more expensive it is to produce. Though the problems with EUV may have delayed development my guess would be the additional cost of the larger ingots negates the cost savings on the other side, particularly with the EUV tools being 10X more expensive. Reply
  • Lukasz Nowak - Thursday, April 8, 2021 - link

    There's another curious thing about the wafer. There are a lot of dies with just clipped corners. If they shifted the entire pattern to the left or right by a quarter of a die, they would get 6 more good ones. That's 7% more dies for free (90 instead of 84).

    Wouldn't that be worth doing?
    Reply
  • Smell This - Thursday, April 8, 2021 - link


    Maybe Chipzillah can glue them together . . . HA!

    Speaking of which __ it this aN *MCM* multi-chip module ?
    Reply
  • Chaitanya - Tuesday, April 6, 2021 - link

    Even that upgrade is falling short of catching up. Reply
  • Hifihedgehog - Tuesday, April 6, 2021 - link

    Exactly. What's with these nonsense comments, anyway? It is like bragging about how I can now run a 10-minute mile instead of a 20-minute mile while the star players are breaking world records and running 4-minute miles. *facepalm* Reply
  • Wilco1 - Tuesday, April 6, 2021 - link

    No kidding. It is not even matching Graviton 2! A $8k CPU beaten by an Arm CPU from 2019... Reply
  • Hifihedgehog - Tuesday, April 6, 2021 - link

    and yet that's still a lot of falling short for Intel 🤷‍♂️ Reply
  • fallaha56 - Tuesday, April 6, 2021 - link

    not really, the 38 and 40 core parts won't be available in any amounts (see Semiaccurate)

    and as can be seen in the lower spec parts, suddenly Ice Lake is barely beating Cascake Lake never mind AMD
    Reply
  • fallaha56 - Tuesday, April 6, 2021 - link

    https://semiaccurate.com/2021/04/06/intels-ice-lak... Reply
  • Gondalf - Wednesday, April 7, 2021 - link

    LOL semi-accurate. 40-38 cores parts already for sale. Reply

Log in

Don't have an account? Sign up now