AMD Ryzen 5 1600X and 1500X – light and shadow in duet

1 - Einführung und Testsystem 2 - AotS: Escalation, Battlefield 1 3 - GTA V, Hitman (2016) 4 - Middle-Earth: Shadow of Mordor, Project Cars 5 - Rise of Tomb Raider, The Division, Time Spy 6 - Workstation-Benchmarks 7 - Temperaturen und viele Deltas 8 - Leistungsaufnahme im Detail 9 - Zusammenfassung und Fazit 10 -

The Ryzen 5 1600X and 1500X presented today (and tested) are actually no longer really new, because we find the same archtitecure in the whole CPUs of the Ryzen 7 family. If you want to refresh the knowledge of AMD's new architecture a little bit, please refer to the launch article "AMD's Ryzen 7 1800X in Test" and the first follow-up "The Ryzen-Family: Three Sevens with Eight Cores in the Second Comparison", where everything is already very accurate described.

According to the manufacturer, the advantage of AMD's scalable architecture should also be to be able to adapt the CPUs with their expansion almost at will, so that a wide range of constellations from consumer to server CPUs can be put on the market without much additional effort. can bring. One of the backgrounds is, of course, purely economic, because it also means that many chips can be reused, for which a part is not functional or only to a limited extent. But this is neither new nor offensive, but important for better yield and in the end also ensures survival.

The basis for this versatile denomination is AMD's CPU Complex (CCX for short), which we still know from the architectural description of the Ryzen launch. Each of these CCX contains four cores, so you might suspect a real halved eight-core behind the Ryzen 5 1500X, which would have limited the L3 cache to 8MB.

Instead, AMD uses two cores (2-2) of each CCX. The L3 cache remains at 16 MB, which is definitely not a disadvantage. In addition, each of the cores of both CPUs still has 64KB L1 cache (commands and data) and 512KB L2 cache available.

The Ryzen 5 1600X is a six-nucleus that also utilises both CCX, but with three cores (3-3). From a purely external lynot, however, they do not differ. The Ryzen 5 1400 exits different information about CCX usage and The L3 cache size. We have seen a boxed label with the indication of 10MB, which would correspond to the full cache of a CCX plus half of the second CCX, other sources write from a CCX and 8 MB L3 cache. AMD has nauseuly not responded so far, so we have to put a question mark on this CPU.

There is still the eternally young price question. Was e.g. the Ryzen 7 1700 as a real 8-core still a small price bomb, the R5 1500X with 209 euros (incl. VAT) and the R5 1600X with 279 euros in the store (both EIAs), which is now a much more optimistic and more researched price announcement. After all, an Intel Core i5-7600K currently costs "only" 238 euros and a Core i5 7500 starts at 190 euros. Although this cannot be overclocked and, like the Core i5 7600K, it must be set to SMT (or hyper-threading), but can also be operated worry-free on lower-cost motherboards. Here we will now have to take the actual test results as the actual yardstick for the final assessment.

Windows 10 Creators Update

We wavered back and forth like reeds in hurricane strength and then decided to re-benchmark everything completely after the final Creators Update for Windows 10 was still available last week. This shortens some versions somewhat (e.g. the number of screen resolutions in the gaming test), however, seemed so important to us that we were happy to make the extra effort and time-related compromises.

Test system and configuration

We are testing the two CPUs on a new platform by using the MSI B350 Tomahawk, a more cost-effective motherboard from the 100 Euro class. Of course, overclocking attempts can also be started with this motherboard – at least as far as the CPU as such really allows. In our later tests we will see that the Ryzen 5 1500X could be stable to 3.9 GHz and the Ryzen 5 1600X to 4 GHz overclocked and operated equally long-term stable.

As a special feature, the board offers its own temperature measurement by means of a sensor in the socket area, whereby we want to refer explicitly (without pre-pre-presetting in detail now) to our own measurements and explanations of AMD's Tctl values, which will follow later in the article. . At least the findings from MMI's sensors were the trigger for our somewhat more detailed paragraph on this topic (from page 7).

The AM4 motherboard is based on the AMD B350 chipset and has four DDR4 slots for up to 64 GB of memory, whereby we will only use two slots in our test with a total of 16GB (2x 8GB DDR4 3200). On board are a PCIe 3.0 x16 slot, a PCIe 2.0 x4 slot, two PCIe 2.0 x1 slots and two older PCI slots. The motherboard also features a 7.1 onboard sound chip, a Gigabit Ethernet interface, as well as four SATA3 ports, an M.2 port, as well as USB 3.0 Type-C and USB 3.0 ports.

As RAM we used two 8GB modules G.Skill Ripjaws DDR4 3200 (CL15-15-15-35) as well as two 8GB modules Geil EvoX DDR4 3200 (CL16-16-16-36), all of which ran smoothly.

Commissioning and technical data

The new test system and the methodology we have already described in great detail in the basic article "So we are testing graphics cards, as of February 2017" and so we now only refer to this detailed description for the sake of simplicity. So if you want to read everything again, you are welcome to do so.

In this case, only the hardware configuration with CPU, RAM, motherboard, as well as the new cooling is different, so that the summary in table form quickly gives a brief overview of the system used here and today: