• We’re currently investigating an issue related to the forum theme and styling that is impacting page layout and visual formatting. The problem has been identified, and we are actively working on a resolution. There is no impact to user data or functionality, this is strictly a front-end display issue. We’ll post an update once the fix has been deployed. Thanks for your patience while we get this sorted.

Discussion NV Re-Enter ARM PC market in Q2-2026!

Page 12 - Seeking answers? Join the AnandTech community: where nearly half-a-million members share solutions and discuss the latest tech.
coercitiv said:
I was in a hurry, as evidence by the fact that I missed the video segment being discussed in this thread last week by the exact person I replied to. The news was about the Windows driver.
Click to expand...
Isn't this a CPU feature that W11 recently made mandatory? I don't get why nvidia would back off its requirement at this point. Has to be a pretty niche application somewhere.
 
blckgrffn said:
Isn't this a CPU feature that W11 recently made mandatory? I don't get why nvidia would back off its requirement at this point. Has to be a pretty niche application somewhere.
Click to expand...
Is it mandatory for WoA too or only for the x86-64 versions ?
 
I was replying to @marees. WoA can't require popcnt because that is an x86 instruction in the SSE 4.2 group. ARM has a broadly similar instruction called CNT that may be required for WoA, but I don't know either way.

Windows 11 added the requirement recently, likely having to do with encryption/cryptography processing.
 
I get it now, sorry, I didn’t follow the flow well.

I would imagine MS would “require” every modern extension known to man for W11 as that’s baseline for WoA going forward. I would imagine they would try to set the bar high now. Pure speculation on my part.
 
Thibsie said:
First attempt, ok by me but competition is always good.
Click to expand...
I mean, it is their 1st collab with Mediatek, but beyond that whether we are talking off the shelf or custom based it is by no means their first ARM SoC rodeo with a decade of DRIVE and even longer with Tegra (which was basically what DRIVE built from anyway).
 
adroc_thurston said:
Drive stuff is still a Tegra.
Grace is also a Tegra.
Click to expand...
I assumed as much.

The only thing I can guess that is tripping them up would be MTK's modem integration, as it's not like they had a whole lot of luck with cellular modems with their Icera acquisition and haven't made any similar acquisitions since.
 
soresu said:
The only thing I can guess that is tripping them up would be MTK's modem integration, as it's not like they had a whole lot of luck with cellular modems with their Icera acquisition and haven't made any similar acquisitions since.
Click to expand...
Oh but they're laptop parts. These have no modem.
 
Specifically, NVIDIA doesn't have the experience going for performance in the deep power management scheme required for laptops. Everything else they've done has been more about managing thermals while pushing for peak performance, or just throwing out a bunch of resources for the user to just figure out how to use. Mediatek, on the other hand, has no excuses. They've been making (crummy) processors for low end Chromebooks for a long time on top of their mobile phone, tablet and work for SoCs for Amazon fire devices.
 
Gideon said:
First Geekbench leak of N1x AFAIK (unfortunately only GB 6.2.2)

NVIDIA N1x - Geekbench

Benchmark results for a NVIDIA N1x with an ARM ARMv8 processor.
browser.geekbench.com browser.geekbench.com

Actually not terrible performance. 93-95% of a 9950X in both ST and MT
Click to expand...
N1x = GB10 ?

20 Threads total APU definitely refer to GB10 with 10 x X925 and 10 x A725. ST score of 3096 seems higher than D9400 X925 3.62GHz, no sure cause by Ubuntu or SME???
 
Last edited:
You must log in or register to reply here.

TRENDING THREADS

Back
Top