Deutsch
Español
Français
Italiano
Nederlands
Polski
Português
Русский
Türkçe
Svenska
Chinese
MagyarFirst AMD Ryzen 7 6800U benchmarks: Closer to Core i7-1260P in multi-core, Radeon 680M impresses in synthetics but lags in gaming
AMD first announced the new Ryzen 6000 series back during CES 2022 followed by the Ryzen Pro 6000 series in April. Devices with Ryzen 6000 processors have now started to become available.
We now have access to the new Asus ZenBook S 13 OLED laptop, which is based on the AMD Ryzen 7 6800U and the integrated Radeon 680M GPU. Asus's implementation allows the Ryzen 7 6800U to work at a sustained PL1 of 17 W with short PL2 bursts up to 30 W in the Performance power profile.
In this article, we take a preliminary look at the performance of the Ryzen 7 6800U and the Radeon 680M in the Asus ZenBook S 13 OLED. A holistic review of the device will follow shortly. Since these are still initial performance data, we have also included reference scores from Asus's reviewer's guide for the ZenBook S 13 wherever available.
Ryzen 7 6800U CPU performance: Targets Core i7-1260P in multi-core, equals Ryzen 9 5900HS in single-core
Despite being housed in a compact 13-inch form factor, the Ryzen 7 6800U puts up an impressive show in CPU benchmarks. Improvements to single-core performance finally see the Ryzen 7 6800U equaling the likes of an average 35 W Ryzen 9 5900HS in Cinebench R23. That being said, the actual performance delta between an average Ryzen 7 5800U and an average Ryzen 9 5900HS isn't really significant to begin with.
The average Core i5-1240P lags behind the Ryzen 7 6800U by 22% in Cinebench R23 single-core while an average Core i7-1260P manages a 20% lead. It helps to note here that the Core i7-1260P in the Lenovo Yoga 9i 14 actually runs at more than 2x the TDP under load (38 W). A more complete picture will be available as we get to test additional Alder Lake-P devices.
Cinebench R23 multi-core sees the Ryzen 7 6800U faring 20% better compared to an average Ryzen 7 5800U, which is not a bad improvement for an interim refresh. The advantage with the Ryzen 9 5900HS diminishes here owing to the higher wattage of the Cezanne chip. Still, the Ryzen 7 6800U nearly ties with an average Core i7-1260P, which in turn is about 18% faster than an average Ryzen 7 5800U.
In Blender CPU, we see the Ryzen 7 6800U to be about 9% faster at completing the render compared to an average Ryzen 7 5800U. The new Rembrandt-U APU nearly ties with the Core i7-1260P and is 24% faster than a Core i5-1240P.
Overall system performance in PCMark 10 is about 5% better with the Ryzen 7 6800U than an average Ryzen 7 5800U and a decent 10% higher compared to the Core i7-1260P-powered laptops.
* ... smaller is better
Ryzen 7 6800U: Sustained performance affected by chassis constraints
The Ryzen 7 6800U does seem to be capable of decent sustained performance in a Cinebench R15 loop. However, in this test, we see the chassis limitations of the Asus ZenBook S 13 OLED affecting the CPU's ability to fully stretch its muscles. The HP ProBook x360 435 with a Ryzen 7 5800U is able to offer higher scores and a better sustained performance comparatively.
During the run, we observed that the laptop isn't quite capable of sustaining the 30 W peak TDP. It gradually settles down to about 17 W with the core clocks eventually levelling off near the 2.5 GHz mark and temperatures settling around 70 °C.
Compared to the Lenovo Yoga 9i 14 with a Core i7-1260P, the Asus ZenBook S 13 OLED is at a disadvantage. The Core i7-1260P is able to raise the PL2 to as much as 64 W, which affords much higher boosts while the Ryzen 7 6800U can only attain 30 W PL2 in the ZenBook S 13's Performance mode.
Radeon 680M iGPU: Beats the MX450 in synthetics but not in gaming, requires FSR
One of the main USPs of the Ryzen 6000 generation is the introduction of an RDNA 2-based iGPU that should finally negate the need for an entry-level dGPU like the GeForce MX450.
In Geekbench 5.3 compute tests, the Radeon 680M is able to handily beat an average MX450 in both OpenCL and Vulkan tests with comparable performance to that of an average GTX 1650 Mobile. In 3DMark, the Radeon 680M is able to lead an average MX450 by as much as 37% in Fire Strike Graphics and 25% in Time Spy Graphics. However, it loses out to the GTX 1650 Mobile and the GTX 1650 Max-Q in this test.
Gaming presents a strange conundrum. We have scores only from Dota 2: Reborn for the moment, but the Radeon 680M trails the average MX450 and the average GTX 1650 Max-Q by 21% and 28%, respectively in 1080p High. This performance gap between the GPUs further increases by up to 7% at 1080p Ultra. The deltas are somewhat lower at lower resolutions and settings.
We will be testing more games with this GPU, but first impressions indicate that performance can be potentially further enhanced with FidelityFX Super Resolution (FSR) 2.0 upscaling in supported games or via Radeon Super Resolution (RSR) in the driver.
Overall, the Ryzen 7 6800U does impress on several counts even though the actual processor itself isn't too different from previous Cezanne offerings. Sustained performance definitely has room for improvement, but it is more likely to be a limitation of the Asus ZenBook S 13 OLED's compact chassis than the processor itself.
We now have an iGPU that performs comparably to entry-level dGPUs thanks to the RDNA 2-based Radeon 680M. Though we found the Radeon 680M to perform well in compute and synthetics tests, it looks like it does require some additional impetus in gaming via FSR or RSR.
We will be evaluating other metrics of the Asus ZenBook S 13 OLED in our full review that is coming up soon and will also be testing more Ryzen 6000 and Alder Lake offerings in the days to come for a more holistic picture, so stay tuned.
Source(s)
Own
Pricecompare
More articles related to this device
Asus ZenBook 13 OLED UM5302T (ZenBook 13 Series)Related Articles
