AMD Ryzen 5 2600X + Ryzen 7 2700X Linux Benchmarks

[char]

Ian is now an AMD shill,

I don't think so, we'll see...

Ian Cutress wrote:
"Update: A number of comments have noted that some of our gaming numbers are different to other publications. To clarify, we used the latest ASUS 0508 BIOS (on X470), full Windows RS3 + updates, Spectre/Meltdown patches, and updated gaming titles. We are reviewing the data."

[austin754]

There are several reasons why I wonder if the Phoronix test configuration for redis is particularly appropriate to use in comparing these systems. After looking at the redis benchmark page and timing the application/workload, I'll note:

1. Redis is a single-threaded server not designed to benefit from multiple CPU cores. It also appears Phoronix runs this instance without pinning on a NUMA topology, which adds uncertainty. The entire workload run is very short, completing in less than 1/2 second on all your test systems. A single test itself runs for slightly over 10 seconds but the vast majority is a 10 second sleep(1) to make sure the database is ready.

2. The Redis benchmark page linked below, says "Redis is, mostly, a single-threaded server from the POV of commands execution (actually modern versions of Redis use threads for different things). It is not designed to benefit from multiple CPU cores. People are supposed to launch several Redis instances to scale out on several cores if needed. It is not really fair to compare one single Redis instance to a multi-threaded data store." Wouldn't it make more sense to run many instances of Redis when comparing servers with many cores?

3. The Redis benchmark page also says, "on multi CPU sockets servers, Redis performance becomes dependent on the NUMA configuration and process location. The most visible effect is that redis-benchmark results seem non-deterministic because client and server processes are distributed randomly on the cores. To get deterministic results, it is required to use process placement tools (on Linux: taskset or numactl)." When the benchmark doesn't pin, in addition to being non-deterministic, it might be difficult to separate out the effects of bad process placement vs. performance of the network sockets that are pushing along 2-byte data sets.

What exactly are you using redis to measure and compare?

--mev

[char]

About anandtech's test:

-----------------------------------
TechRadar & the wccftech preview has the same results. If you have been following Spectre as I have, you would've seen even users find this result. See the top comments here. https://np.reddit.com/r/pcmasterrace...atch_and_bios/

AT, TR & WCCF's results are accurate. Many reasons for this.
- Many reviewers used the old Ryzen balanced power setting which cripples the 2700X
- Disallowed the motherboard settings that push the chip over TDP
- Fully patched as possible for Spectre v1 & v2, which cripples Intel up to 50% in IO heavy tasks (streaming textures for games that do so).
-----------------------------------

Summary of various reviews: (anandtech' is not the only one with poor intel)

https://www.3dcenter.org/news/ryzen-2000-launchreviews-die-testresultate-zur-anwendungs-performance-im-ueberblick