[ad_1]
Nvidia showcased RTX graphics running on ARM processors to highlight the potential performance non-x86 processors can offer consumers – and to emphasize that games don’t require an x86 microprocessor. There are a few caveats to the presentation (we’ll get to them), but make no mistake: this is as much a warning to Intel and AMD as it is an opportunity to brag about Nvidia’s graphics prowess.
Nvidia showed a game, Wolfenstein: Youngblood, and a tech demo (The Bistro, from the Open Research Content Archive (ORCA). Video posted by Nvidia for GDC (embedded below) indicates that the benchmark is performing an RTX 3060 on a MediaTek Kompanio 1200, which is based on the MT8195 SoC MediaTek calls these parts “chipsets,” but this word is generally used to refer to the motherboard control logic, not an SoC.
MediaTek is generally known as a low budget device developer, but the MT8195 seems to be carrying some weight. The SoC is built on TSMC’s 6nm process, which offers higher logic density compared to its 7nm node. TSMC built the 6nm node for companies that wanted to take advantage of its logical density but didn’t want to jump ship for 5nm and its new design rules. The processor contains four Cortex-A78 cores and four Cortex-A55 cores with an AI processing unit, which MediaTek confusingly calls an “APU”. The chip uses four-channel LPDDR4x 2133, but we don’t know how wide the channels are. It’s a modest chip overall, however – the MT8195 only achieves a maximum default clock speed of 2.2 GHz on Cortex-A78 cores. SoC is typically used in Chromebooks.
We suspect that’s part of Nvidia’s point of view. The MT8195 is not the type of processor we would expect to see in an actual ARM-powered gaming system. While it can run RTX games well when paired with a single RTX 3060, that implies good things about the products that might appear in the future.
Nvidia ported five Linux-specific capabilities: DLSS (presumably used above to improve performance), RTX Direct Illumination, RTX Global Illumination, Nvidia Real-Time Denoisers, and RTX Memory Utility. DLSS is used to improve image quality by increasing lower resolution output at higher resolutions, direct lighting supports dynamic lighting, and global lighting is used to calculate ray bounces. Real-time denoising is self-explanatory, and the RTX memory utility is supposed to optimize GPU memory usage.
The fact that the system is running Linux (Arch Linux) throws an interesting wrinkle in all of this. Windows support on ARM exists, but Nvidia chose not to demonstrate a system running Windows. Linux isn’t normally considered a gaming operating system, but Valve has announced that its next Steam Deck will ship with SteamOS and run a number of Windows games through Proton, its fork of Wine. The Steam Deck relies on an AMD APU, not an Nvidia solution, but it’s still an interesting demonstration of gaming performance.
Nvidia isn’t going all out with a major product announcement; GTC is more of a developer conference than a consumer event. The message here is unequivocal, albeit understated: neither x86 nor Windows are inherently required for high-performance gaming.
Nvidia has more reasons than most companies to hammer out this message. If the company’s attempt to purchase ARM is approved by regulators, it will have the opportunity to play a role in becoming the leading concurrent GPU vendor. and CPU solutions for games. Intel is making its own game for this title with its Xe HPG product line, and AMD already holds that position if you’re a console gamer. But no manufacturer has ever claimed a dominant position in the processor and GPU market when it comes to supplying the PC gaming market. Showing ray tracing on Linux is good for Linux as a gaming operating system. Showing it on a MediaTek MT8195 is good for Nvidia. There are no plans to release this version of Youngblood to market, according to the video, but it probably won’t be the last time we see an Nvidia CPU ARM + GPU sharing space in a laptop or desktop.
Now read:
[ad_2]
Source link