CPU Performance: Synthetic Tests

As with most benchmark suites, there are tests that don’t necessarily fit into most categories because their role is just to find the peak throughput in very particular coding scenarios. For this we rely on some of the industry standard tests, like Geekbench and Cinebench.

GeekBench4: Synthetics

A common tool for cross-platform testing between mobile, PC, and Mac, GeekBench 4 is an ultimate exercise in synthetic testing across a range of algorithms looking for peak throughput. Tests include encryption, compression, fast Fourier transform, memory operations, n-body physics, matrix operations, histogram manipulation, and HTML parsing.

I’m including this test due to popular demand, although the results do come across as overly synthetic, and a lot of users often put a lot of weight behind the test due to the fact that it is compiled across different platforms (although with different compilers).

We record the main subtest scores (Crypto, Integer, Floating Point, Memory) in our benchmark database, but for the review we post the overall single and multi-threaded results.

Geekbench 4 - ST OverallGeekbench 4 - MT Overall

LinX: LINPACK

The main tool for ordering the TOP500 computer list involves running a variant of an accelerated matrix multiply algorithm typically found from the LINPACK suite. Here we use a tool called LinX to do the same thing on our CPUs. We scale our test based on the number of cores present in order to not run out of scaling but still keeping the test time consistent.

This is another of our new tests for 2020. Data will be added as we start regression testing older CPUs.

LinX 0.9.5 LINPACK

 

Cinebench R20

The Cinebench line of tests is very well known among technology enthusiasts, with the software implementing a variant of the popular Cinema4D engine to render through the CPU a complex scene. The latest version of Cinebench comes with a number of upgrades, including support for >64 threads, as well as offering a much longer test in order to stop the big server systems completing it in seconds. Not soon after R20 was launched, we ended up with 256 thread servers that completed the test in about two seconds. While we wait for the next version of Cinebench, we run the test on our systems in single thread and multithread modes, running for a minimum of 10 minutes each.

Cinebench R20 Single ThreadedCinebench R20 Multi-Threaded

CPU Performance: Web and Legacy Tests CPU Performance: SPEC 1T
POST A COMMENT

220 Comments

View All Comments

  • Gastec - Friday, May 22, 2020 - link

    Basically you just have to type "allyourbasearebelongtous +$50/surprisemechanic" and you get all the framerate you want in your favorite multiplayer FPShooter. Reply
  • Boshum - Wednesday, May 20, 2020 - link

    I think it's a viable alternative to Ryzen 3000, so it's not pointless. It's about equal in performance for most people. A little more expensive and power hungry core for core, but it's more of a flavor thing now. It's still better for certain gaming and application scenarios. Hyperthreading makes the low to midrange a much more reasonable option too, with heat and power being no big deal there. The only place it can't compete with Ryzen is at the very high end for power users doing heavy multi-core work. Reply
  • Dribble - Wednesday, May 20, 2020 - link

    I'd be the sort of person to look at a 10700K but power usage is just too high. I want to be able to stick a high end air cooler on it, o/c and still have it run pretty quiet. I'd have to go water with one of these and I can't be bothered with that. Not worth it for the small performance increment over more efficient chips. Reply
  • IBM760XL - Wednesday, May 20, 2020 - link

    Agreed. The 10700K and 10900K use more power per core than my ancient-but-trusty 2500K, at least with stock settings. Sure, the new chips get somewhat better IPC, but I can't justify switching from a Sandy Bridge that's nice and quiet even at 100% load, to a Comet Lake that will require Serious Cooling to have an outside chance of being as quiet.

    I could look at lower-end hex-core Comet Lake chips instead, but why would I do that when I could just as well get an octo-core Ryzen 7 3700, or a Ryzen 5 3600 that will have better performance than an i5-10500?
    Reply
  • Boshum - Wednesday, May 20, 2020 - link

    I should think the 10500 and 3600 would be pretty close at stock, though you have more overclocking options with the 3600. It's the future Rocket Lake vs Ryzen 4000 options that is more interesting. Reply
  • warrenk81 - Wednesday, May 20, 2020 - link

    typo in the dropdown for the final page, move/more. Reply
  • colonelclaw - Wednesday, May 20, 2020 - link

    Grammar error, too. Less/fewer. Reply
  • Flunk - Wednesday, May 20, 2020 - link

    Well, Intel's back on top for gaming, by a small marging, with chips that can fry an egg. Maybe it'll force AMD to lower their prices on their high-end chips. I don't really fancy a 250+ Watt CPU. Reply
  • DrKlahn - Wednesday, May 20, 2020 - link

    You can already get the 3900x for $410 on Amazon. Unless you have a use case that heavily favors Intel that would seem to be a pretty good value already. A good B450 board capable of handling it could be had for not much more than the difference in chip cost (provided that fits your needs). Reply
  • Irata - Wednesday, May 20, 2020 - link

    Yup, and like the article says that includes an HSF that will do the job.

    Contrast that with the 10900k which retails for $530 on Newegg (not available) and which requires you to spend $ 200+ for a proper cooling set up and you are looking at $ 410 vs. $ 730, i.e. paying 56% more for the 10900k. And that does not even include case fans, mainboard, PSU.

    If gaming is what one is after, the 9700k looks much more attractive than the 10900k.
    Reply

Log in

Don't have an account? Sign up now