Compute & Synthetics

Shifting gears, we'll look at the compute and synthetic aspects of the RX 580. As this is the 3rd time we're seeing 'big' Polaris, there's nothing new to discuss, and with no other design changes we only expect to see the impact of increased clockspeeds.

Beginning with CompuBench 2.0, the latest iteration of Kishonti's GPU compute benchmark suite offers a wide array of different practical compute workloads, and we’ve decided to focus on level set segmentation, optical flow modeling, and N-Body physics simulations.

Compute: CompuBench 2.0 - Level Set Segmentation 256

Compute: CompuBench 2.0 - N-Body Simulation 1024K

Compute: CompuBench 2.0 - Optical Flow

Moving on, we'll also look at single precision floating point performance with FAHBench, the official Folding @ Home benchmark. Folding @ Home is the popular Stanford-backed research and distributed computing initiative that has work distributed to millions of volunteer computers over the internet, each of which is responsible for a tiny slice of a protein folding simulation. FAHBench can test both single precision and double precision floating point performance, with single precision being the most useful metric for most consumer cards due to their low double precision performance.

Compute: Folding @ Home Single Precision

Next is Geekbench 4's GPU compute suite. A multi-faceted test suite, Geekbench 4 runs seven different GPU sub-tests, ranging from face detection to FFTs, and then averages out their scores via their geometric mean. As a result Geekbench 4 isn't testing any one workload, but rather is an average of many different basic workloads.

Compute: Geekbench 4 - GPU Compute - Total Score

We'll also take a quick look at tessellation performance.

Synthetic: TessMark, Image Set 4, 64x Tessellation

Finally, for looking at texel and pixel fillrate, we have the Beyond3D Test Suite. This test offers a slew of additional tests – many of which use behind the scenes or in our earlier architectural analysis – but for now we’ll stick to simple pixel and texel fillrates.

Synthetic: Beyond3D Suite - Pixel Fillrate

Synthetic: Beyond3D Suite - Integer Texture Fillrate (INT8)

Total War: Warhammer II Power, Temperature, and Noise
Comments Locked

136 Comments

View All Comments

  • Opencg - Thursday, November 15, 2018 - link

    Its just a gap filler. It will help them make money from people wanting to get good performance at that price but... we're still waiting for the serious contenders. Vega refresh will be a little more toward the high end but... navi is where amd could potentially shift the market. RTX means you pay more for less performance outside rtx games. Navi is the opposite you pay less and get more. We need navi to function well and I believe amd can do it.
  • neblogai - Thursday, November 15, 2018 - link

    Is is not a gap filler. It is only there, because overproduction of P20 needs to be sold out first. Naturally- P30 should be a replacement of P20, at same price, with minimal investment by MAD or partners.
  • eva02langley - Thursday, November 15, 2018 - link

    Polaris is selling well, Vega, not so much beside mining. In Canada, the price are still having a huge price tag because the supply is just so low.
  • Allan_Hundeboll - Thursday, November 15, 2018 - link

    I am surprised a boost frequency increase of 15% nets 12% overall performance. Maybe Polaris isn't as starved for memory bandwidth, as people seem to think. Or AMD made other improvements besides the increased frequency...?
  • mapesdhs - Thursday, November 15, 2018 - link

    Frequency increases rarely produce linear performance gains, not when there are so many other aspects of a system that impact what's happening (main CPU, API, game engine, etc.)
  • Allan_Hundeboll - Thursday, November 15, 2018 - link

    Yes that is why I'm surprised 15% increase on the core frequency -without any increase of the Vram frequency, would result in 12% overall performance increase. Especially considering the "original Polaris" rx480 was configured with 1266Mhz core boost frequency and 2000Mhz Vram.
    If AMD figured 1266/2000MHz was near the sweet spot balance between core/mem speed then 1545/2000 would be severly starved for memory bandwidth.
    So I'm wondering if 2000MHz memory speed was kind of wasted on the rx480? Or did AMD tighten the Vram timings on this new 590? Or maybe the near linear increase in performance is due to the higher tdp allowing the rx590 to sustain the boost clock while the 480 would throttle to base core frequency of 1120MHz?
  • SlowSpyder - Thursday, November 15, 2018 - link

    Where is 7nm Vega??
  • mapesdhs - Thursday, November 15, 2018 - link

    Never going to happen, because gamers won't buy it.

    https://www.youtube.com/watch?v=guK2XoFbPFw
    https://www.youtube.com/watch?v=USMlET3L7mA
  • Cyborg997 - Thursday, November 15, 2018 - link

    Summer 2019 or late 2019. Definitely soon rtx fighter.
  • schujj07 - Thursday, November 15, 2018 - link

    The most surprising thing to me is just how good RX480 is when you look at the benchmarks. When it launched it was as fast as the GTX970, now it is faster than then GTX980 just due to drivers and newer games being able to use its resources better.

Log in

Don't have an account? Sign up now