Friday, January 29th 2021
Samsung Exynos SoC with AMD RDNA GPU Destroys Competition, Apple 14 Bionic SoC Kneels
Some time ago, Samsung and AMD announced that they will be building a mobile processor that utilizes AMD RDNA architecture for graphics processing. Samsung is readying its Exynos 2100 SoC and today we get to see its performance results in the first leaked benchmark. The new SoC design has been put through a series of GPU-only benchmarks that stress just the AMD RDNA GPU. Firstly there is Manhattan 3 benchmark where the Exynos SoC scored 181.8 FPS. Secondly, the GPU has scored 138.25 FPS in Aztek Normal and 58 FPS in Aztek High. If we compare those results to the Apple A14 Bionic chip, which scored 146.4 FPS in Manhattan 3, 79.8 FPS in Aztek Normal, and 30.5 FPS in Aztek High, the Exynos design is faster anywhere from 25% to 100%. Of course, given that this is only a leak, all information should be taken with a grain of salt.
Source:
Tom's Hardware
43 Comments on Samsung Exynos SoC with AMD RDNA GPU Destroys Competition, Apple 14 Bionic SoC Kneels
Now AMD needs to start using Samsung fabs.
Tegra has been out performed by multiple SoC.
Yes it was the best performer for few years, but it suffered from high-power consumption, it was one of the reasons the Tegra didn't succeed well in the mobile market (the other reasons are cost and lack of radio silicon).
Today, almost all high-end SoC's from even 2-3 years outperform Tegra, depending on which benchmark and what SoC (which vendor). For example, according to notebookcheck, the Snapdragon 865 is on average over 80% faster than Tegra X1 (both CPU and GPU sides). When Hector Ruiz was the CEO of AMD, he bought ATi, and after few months the financial crisis started so he sold the mobile and embedded centric part of ATi to Qualcomm to get cash (among other things), this includes the Adreno GPU (which is just an anagram of Radeon, eg.. same letters, different order) and IIRC the Geode processors intended for embedded systems like TV's and so on. Because he felt at that time that those embedded and mobile things were not the future. It was just few years before the exploding of smartphones.
An RX6900XT blows A14 out of the water...at 300W.
So when I'll see that this SoC beats A14 at the same power consumption, then I'll buy it.
But I seriously doubt it.
AMD been neglected iGPU for far too long, 3 generations with same config and "a clock bump" is lame to put it mildly. Intel has catch up with its 96EU Xe Graphics, as demonstrated by reviewer, its even surpass Vega 10 and close to MX350 in some test.
www.techpowerup.com/277974/amd-reportedly-in-plans-to-outsource-partial-chip-production-to-samsung
I think AMD will have to divide their nodes, so use the 7nm node for some processors, but use 6nm for others. In the GPU space they are likely to go with 6nm or 5nm because they sell less GPU's and they definitely need the power and speed boost to stay competitive with Nvidia.
So I'm thinking Epyc processors with 5nm or 6nm(whichever they choose), Ryzen 6000 and Threadripper still with 7nm. Monile cpu's on 7nm, custom apu's on 7nm, use Samsung's capacity for some GPU's, probably the low and mid range, and big Navi RX 7800 and above on 5nm or 6nm.