Test Bed and Setup

As per our processor testing policy, we take a premium category motherboard suitable for the socket, and equip the system with a suitable amount of memory running at the manufacturer's maximum supported frequency. This is also typically run at JEDEC subtimings where possible. It is noted that some users are not keen on this policy, stating that sometimes the maximum supported frequency is quite low, or faster memory is available at a similar price, or that the JEDEC speeds can be prohibitive for performance. While these comments make sense, ultimately very few users apply memory profiles (either XMP or other) as they require interaction with the BIOS, and most users will fall back on JEDEC supported speeds - this includes home users as well as industry who might want to shave off a cent or two from the cost or stay within the margins set by the manufacturer. Where possible, we will extend out testing to include faster memory modules either at the same time as the review or a later date.

Test Setup
AMD Ryzen 3000 AMD Ryzen 9 3950X
AMD Ryzen 9 3900X
Motherboard ASRock X570 Taichi 2.50 (AGESA 1004B)
CPU Cooler Kraken X62
DRAM Corsair Vengeance RGB 4x8 GB DDR4-3200
GPU Sapphire RX 460 2GB (CPU Tests)
MSI GTX 1080 Gaming 8G (Gaming Tests)
PSU Corsair AX860i
SSD Crucial MX500 2TB
OS Windows 10 1909

We must thank the following companies for kindly providing hardware for our multiple test beds. Some of this hardware is not in this test bed specifically, but is used in other testing.

Hardware Providers
Sapphire RX 460 Nitro MSI GTX 1080 Gaming X OC Crucial MX200 +
MX500 SSDs
Corsair AX860i +
AX1200i PSUs
G.Skill RipjawsV,
SniperX, FlareX
Crucial Ballistix
DDR4
Silverstone
Coolers
Silverstone
Fans
Going For Power: Is 105W TDP Accurate? CPU Performance: System Tests
POST A COMMENT

204 Comments

View All Comments

  • halfflat - Wednesday, November 27, 2019 - link

    For Brownian motion? That seems weird. Nonetheless, it can't alone explain the speed up.

    Most favourable scenario: code consists only of floating point mul and add pairs, together with 64-bit integer multiplication. The floating point operations could become 4x faster in AVX2 (twice as wide as SSE, and using FMAs); to see the observed 2x speed up, that means the floating point operations constituted 2/3 of the execution time in the SSE version.

    The AVX512 version, ignoring any consequent downclocking, could make those floating point operations 8x faster than the SSE case, and the 64-bit integer multiplies also 8x faster. That's still not 10x, it ignores the lower throughput of 8-wide i64 muls compared to scalar muls, and also discounts the slower clock speed.
    Reply
  • halfflat - Thursday, November 28, 2019 - link

    Just an update: ran a simple test (square eight times all the 64-bit ints in a 1024-long array) wrapped in google benchmark on a Skylake Xeon with gcc-8.2 -O3. The kernel is almost entirely multiplications, and ultimately saw a roughly 2x speed up with AVX512 compared to AVX2, and a 2.5x speed up with AVX512 compared with a 'no architecture specified' compilation. Reply
  • w1p30ut3r - Friday, November 22, 2019 - link

    Its very, very simples. If you gaming lonly buy an intel... If you work and gaming buy a 3950x... If you only work buy a threadripper or a xeon... Reply
  • philipwebdev - Monday, February 17, 2020 - link

    Thanks for your fabulous post! I enjoy while reading your post. If you required digital marketing services then contact us or see us at https://philipwebdev.com Reply

Log in

Don't have an account? Sign up now