Heavily-threaded rendering applications remain Ryzen's stomping ground due to the hefty allotment of threads and cores that easily outstrip Intel's Core i3-9350KF. AMD's first-gen Zen architecture (3400G) often leads the 9350KF in threaded tests, while the third-gen Ryzen 5 3600X and 3600 models take convincing wins. That picture changes when we switch to single-threaded rendering tests, with the Core i3-9350KF taking the lead by virtue of its higher clock rates.
The Intel/Netflix-designed SVT-AV1 video encoder leverages Ryzen's threaded horsepower to great effect, giving the third-gen Ryzen processors a commanding lead, but the Ryzen 5 3400G doesn't fare as well in this series of tests. Flipping through to the single-threaded LAME encoder results reveal that Intel still holds the advantage with encoders that prize per-core performance. Given the emergence of threaded encoders, that victory is a bit shallow.
AMD has made tremendous progress with its AVX processing performance in threaded workloads, which is evident in the HandBrake tests in both x264 and x265 flavors. The x265 test uses a heavier distribution of AVX instructions, and we can see the massive difference between the Zen 2 Ryzen 5 3600/X and the Ryzen 5 3400G that comes with the first-gen Zen architecture.
Browsers tend to be impacted more by the recent security mitigations than other types of applications, so Intel has generally taken a haircut in these benchmarks of fully-patched systems, though that handicap doesn't stop Intel from sweeping the competition.
The ARES-6 web browser benchmark focuses on forward neural networks used for machine learning tasks, along with overall browser responsiveness. The Ryzen 5 3400G suffers during this test as it falls to the bottom of the char, but the Ryzen 5 3600/X models grapple with Intel's stock Core i3-9350K. Tuning the Intel processor affords it the pole position as it barely edges its previous-gen counterpart our for the lead.
Office and Productivity
Microsoft's office suite runs via PCMark 10's new application test. This benchmark tests with real Microsoft Office applications. The Ryzen 3000 series processors are competitive, but the Core i5-9350KF takes the overall win after tuning. Much of that strength stems from its higher clock rate, but the third-gen Ryzen models nearly match the 9350KF at stock settings.
The application start-up metric measures load time snappiness in word processors, GIMP, and Web browsers. Other platform-level considerations affect this test as well, including the storage subsystem. The Core processors trail at stock settings, but take a big lead after overclocking. Adding a PCIe 4.0 SSD to our Ryzen test system would swing this benchmark in favor of the Ryzen 3000-series processors.
Flipping over to the heavily-threaded Timed LLVM compilation test finds the Ryzen processors back in a workload that can leverage their threaded heft, granting them a large win over the Intel competitors. The Ryzen 5 3400G isn't nearly as agile as its counterparts in this test, but it still challenges the stock Core i3-9350KF.
Compression, Decompression, Encryption, AVX
The 7zip and Zlib compression/decompression benchmarks rely heavily upon threading and work directly from system memory, thus avoiding the traditional storage bottleneck in these types of tasks. Ryzen dominates these tests without storage throughput restrictions, but you also have access to the PCIe 4.0 interface with AMD's X570 platform. That extra throughput means these tremendous gains will largely transfer over to real-world application performance when you pair the Ryzen chips with a speedy PCIe 4.0 SSD.
The heavily-threaded y-cruncher benchmark, which computes pi using the taxing AVX instruction set, finds the third-gen Ryzen processors offering a massive performance improvement over the Core i3-9350KF, partly due to their extra threads and AMD's AVX improvements.
MORE: Best CPUs
MORE: Intel & AMD Processor Hierarchy
MORE: All CPUs Content
I imagine that most those buying the lower end processors will be doing routine computer work like web browsing, word processing ect... I see that as the perfect use case for a very fast low core count processor. Feels fast in normal home/office use at a low cost.
Yeah especially when you can get a 2nd gen Ryzen CPU for ridiculously cheap right now. Even the 3rd gen CPUs are still a relative bargain compared to their Intel counterparts.
I've been under the assumption that games are being optimized for 8 threads, because of the consoles.
I'll need to look into if 4c/8t outperforms 6/6 at gaming to see if that's true, even though 4 hyperthreaded cores are usually going to be worse than 6 physical cores at overall number crunching.
I wonder how console-comparable pre-ryzen 8-weak-core AMD processors hold up with modern games compared to 4 fast cores.
Anecdotally, I'm seeing a lot of "FX-8350 or better" in the 'recommended' listing of $50-60 games on Steam, so I'm guessing those older 8 cores are working better now that things are optimized for 8 cores.
Games are starting to use more threads, because of Ryzen, really. Hence Intel finally pushing core counts higher, across all product lines. I would expect a 6c/6t to do better, than a 4c/8, but still suffer from frametime issues, a 6c/12t CPU like an intel 8700k, or a Ryzen 5 3600, would not.
This processor should be $59 , and a questionable purchase even then, but, it is NEVER worth $180+....(Intel, trim it's price by 60-70% , pronto..!)
While the are not taking advantage of the bandwidth, AMD's NAVI cards are PCI-E 4.0. X570 is not compatible with 1st gen Ryzen. Only a 3rd gen CPU, with an x570, gets you PCI-E 4.0. I do agree that PCI-E 4.0, is not really necessary, for the average user, at this time.