AMD Ryzen 7 5800X3D On Linux: Not For Gaming, But Very Exciting For Other Workloads

[ResponseWriter]

Thanks, Michael. I've been waiting for Linux-relevant benchmarks. I still hope to see DXVK/wine gaming performance compared with 5900X and 3800X given the former seems to now be priced between the 5800X and 5800X3D, while the 3800X is the previous 8-core chip.

Zstd as well. Like Michael points out in the article, that probably had good benefits for file systems using Zstd for compression. I wonder if LZ4, XZ, and other codecs get performance improvements as well.

zram with zstd as well, I'd imagine.

[brucethemoose]

The only games I'm interested in buying such a CPU for are logic/simulation bound ones: Rimworld, modded Minecraft, Distant Worlds 2, Stellaris, Satisfactory, Starsector and so on. Games that need every bit of single core performance they can to scale up.



Hence I don't get this fascination with benching GPU bound AAAs on low at low res, especially with a focus on FPS instead of frametimes... and I'm not just talking about Phoronix.

[fuzz]

The only games I'm interested in buying such a CPU for are logic/simulation bound ones: Rimworld, modded Minecraft, Distant Worlds 2, Stellaris, Satisfactory, Starsector and so on. Games that need every bit of single core performance they can to scale up.

I'm going to try Dwarf Fortress on it, which typically runs into limits for map size/population pretty easily due to these reasons exactly. I think Rimworld is the same

[domih]

Thanks for this article as well as the one on Milan X!

I'm not a gamer so I'm mostly interested in the possible performance increase brought by 3D V-cache in development related tools and servers. What about (beyond ML/DL):
- JSON parsing (in various languages),
- XML parsing and other XML operations (in various languages),
- MySQL, PostgreSQL,
- Cassandra,
- Large Python or PHP list and dict handling,
- JIT compilation in Java, Python, PHP,
- Crypto (AES, RSA),
- JavaScript,
- Web servers (Apache, Nginx).

[pete910]

Not surprised as to see little gaming increase compared to windows with that collection of games.

The workload tests was a surprise as that looked to be the opposite of windows. Having said that Phoronix does better/broader benchys compared to the windows guys.

[Michael]

Thanks for this article as well as the one on Milan X!

I'm not a gamer so I'm mostly interested in the possible performance increase brought by 3D V-cache in development related tools and servers. What about (beyond ML/DL):
- JSON parsing (in various languages),
- XML parsing and other XML operations (in various languages),
- MySQL, PostgreSQL,
- Cassandra,
- Large Python or PHP list and dict handling,
- JIT compilation in Java, Python, PHP,
- Crypto (AES, RSA),
- JavaScript,
- Web servers (Apache, Nginx).

Click the openbenchmarking.org link on the last page of the article, some of those are covered. Others are coming.

Also navigating from this page will also yield in-progress metrics for other tests - https://openbenchmarking.org/s/AMD+R...5800X3D+8-Core

[nicalandia]

Would be interesting to test to see if there is any Performance boost on Windows Games running on Linux + Proton/Wine(SteamOS) vs Windows 10/11 with 3D V-Cache.

[chocolate]

Great companion to AM4 motherboards until their end of life!

I think it's just as great for Linux gaming as it is for Windows gaming in all modern, more "bloated" AAA games.
It's kind of a wash with the 5800X at stock frequencies for older games because those supposedly already "fit" in the constraints of abysmally poor Intel caches from the early Core days, and benefit from higher clocks in CPU-bound scenarios, but their absolute performance is probably already above v-sync for any reasonable monitor.

Michael, if I'm not mistaken, your own testing in the past showed how Rise/Shadow of the Tomb Raider scaled well with RAM frequencies. I fully expected it to perform better with a larger cache; the game was basically begging for data non-stop (as for the cause, we can only speculate, but that's how it seems to be, and it's not an isolated case in the industry nowadays).
Therefore, I'm not convinced it's appropriate to reiterate how poor the 5800X3D performs for "Linux gaming" (in such a general sense) so many times in the article, given the set of games benchmarked.
Using it for the same non-native titles that other reviewers have benchmarked on Windows would surely bring the same benefits. Unfortunately, those titles probably cannot be automated.

Given that the results here already hint at industry trends quite accurately, perhaps it would be best to remain cautious instead of disregarding the 5800X3D for a gaming usecase.
For example, Strange Brigade has always been praised for its performance relative to its graphics, and apart from using Vulkan (the main, simplistic explanation some reviewers love to resort to), cache-friendliness may very well be a part of that, hence its slight decrease under the 5800X3D compared to the 5800X.

Cheers.

[willmore]

LZ4 Does not appear to take advantage of the 3D V-Cache

LZ4.jpg

LZ4 is very cache friendly. It reads through its input buffer and coppies that to the output buffer and occasionally reads back a little bit in the output buffer and coppies that to the current end of the output buffer. This makes it very tolerant of cache eviction, etc. It's very easy for simple fetch predictors to keep up with the way LZ4 accesses memory. Other compression programs use different methods with much larger cache footprints, but LZ4 is not one of them. It's suitable for microcontrollers, etc. If you have enough memory for the input buffer and the output buffer, you can do LZ4 without extra storage. You can't do that with BZ2 or other transform based systems.

Being a bit pedantic here but the app don't "take advantage" of larger cache.
It is more like bloated apps that require larger caches.

If LZ4 was well written, then you won't see much of a boost.

You're speaking in the context of compression programs here and you are dead wrong. Saying that a compression program is using too much memory is like saying that an algorythm is lazy because it didn't find a way to solve the Traveling Salesman problem in polynomial time. I don't want to quote the Cat in the Hat, but he's right.

To address your overly pedantic opinion, a program is well written if it takes advantage of the hardware on which it runs. LZ4 doesn't happen to be able to do that as it's already fully satisfied by even a basic processor, but that's because it's designed for a little processor. Honestly, it's a bit silly to be using it as a benchmark in this way. It's about as good of a benchmark as 'grep /proc/cpuinfo "cpu MHz"'

[Templar82]

I would be interested to see how it compares to the 5900X on the professional workloads, given they are the same price.