The AMD 3rd Gen Ryzen Deep Dive Review: 3700X and 3900X Raising The Bar
by Andrei Frumusanu & Gavin Bonshor on July 7, 2019 9:00 AM EST** = Old results marked were performed with the original BIOS & boost behaviour as published on 7/7.
Benchmarking Performance: CPU Legacy Tests
3DPM v1: Naïve Code Variant of 3DPM v2.1
The first legacy test in the suite is the first version of our 3DPM benchmark. This is the ultimate naïve version of the code, as if it was written by scientist with no knowledge of how computer hardware, compilers, or optimization works (which in fact, it was at the start). This represents a large body of scientific simulation out in the wild, where getting the answer is more important than it being fast (getting a result in 4 days is acceptable if it’s correct, rather than sending someone away for a year to learn to code and getting the result in 5 minutes).
In this version, the only real optimization was in the compiler flags (-O2, -fp:fast), compiling it in release mode, and enabling OpenMP in the main compute loops. The loops were not configured for function size, and one of the key slowdowns is false sharing in the cache. It also has long dependency chains based on the random number generation, which leads to relatively poor performance on specific compute microarchitectures.
3DPM v1 can be downloaded with our 3DPM v2 code here: 3DPMv2.1.rar (13.0 MB)
x264 HD 3.0: Older Transcode Test
This transcoding test is super old, and was used by Anand back in the day of Pentium 4 and Athlon II processors. Here a standardized 720p video is transcoded with a two-pass conversion, with the benchmark showing the frames-per-second of each pass. This benchmark is single-threaded, and between some micro-architectures we seem to actually hit an instructions-per-clock wall.
CineBench 11.5 and 10
Cinebench is a widely known benchmarking tool for measuring performance relative to MAXON's animation software Cinema 4D. Cinebench has been optimized over a decade and focuses on purely CPU horsepower, meaning if there is a discrepancy in pure throughput characteristics, Cinebench is likely to show that discrepancy. Arguably other software doesn't make use of all the tools available, so the real world relevance might purely be academic, but given our large database of data for Cinebench it seems difficult to ignore a small five minute test. We run the modern version 15 in this test, as well as the older 11.5 due to our back data.
447 Comments
View All Comments
djayjp - Sunday, July 7, 2019 - link
So results for Intel chips are completely invalid then.futrtrubl - Sunday, July 7, 2019 - link
You will need to have to explain that then. Comparing Intel with mitigations vs AMD with mitigations.djayjp - Sunday, July 7, 2019 - link
No. Fallout/ZombieLoad does not affect AMD chips.djayjp - Sunday, July 7, 2019 - link
Intel performance will suffer whereas AMD's won't be affected.WaltC - Sunday, July 7, 2019 - link
Ha-ha...;) So because AMD has a newer architecture without most of the vulnerabilities that plague Intel's ancient CPU architectures--it should be held against AMD? Rubbish...;) Look, what is unfair about testing both architectures/cpus with all the mitigations that each *requires*? I can't see a thing wrong with it--it's perfect, in fact.extide - Sunday, July 7, 2019 - link
They tested Intel WITHOUT Fallout/ZombieLoad which would affect them. Probably not by much, though, honestly.RSAUser - Monday, July 8, 2019 - link
Well the results are close enough for a lot of tests to be error margin, that the mitigation would put AMD in the lead.The tests should reflect real world as of when the article is published, using old results without declaring that Intel doesn't have mitigation applied on every page is the equivalent of falsifying the results as people will buy based on these tests.
mkaibear - Monday, July 8, 2019 - link
"using old results without declaring that Intel doesn't have mitigation applied on every page is the equivalent of falsifying the results as people will buy based on these tests."Oh, that's just inane. They quite openly state the exact test specification on the "Test Bed and Setup" page, including which mitigations are applied. Arguing that not putting one particular piece of information on every page means it's the equivalent of falsifying the results is completely ridiculous.
RSAUser - Tuesday, July 9, 2019 - link
How many go through the test bed set up page?Meteor2 - Sunday, July 14, 2019 - link
Pretty much everyone reading such an in-depth review, I should think.