4 ms·
The best play would be to take the Intel approach: run ARM independently, and just build best-in-class integration with Nvidia GPUs / features. Substantial ant
by ethbr0 6y ago
The best play would be to take the Intel approach: run ARM independently, and just build best-in-class integration with Nvidia GPUs / features.
Substantial anti-trust concerns are avoided, while building exactly the moat you want: no competitor can afford to compete with the level of integration investment.
- someperson 6y agoWhy can't NVidia achieve this without spending tens of billions buying ARM?
- ddingus 6y agoThey can, as can others. Having some advantage may be worth the purchase.
- ethbr0 6y agoYou have a lot of options available when you own a company, that you don't have when they're independent. Example: Nvidia would prefer certain specifications on an internal bus they'd use, but nobody else uses as a standard.
- justaguy88 6y agoThey can already do that
- ethbr0 6y agoNot exclusively. ARM's roadmap is driven by what all of their customers want, not just Nvidia. There doesn't appear to be sufficient profit margin in the industry for too many greenfield, non-reference custom implementations. So if Nvidia-ARM sticks out a reference implementation or architecture choice that's beneficial to Nvidia, there's limited capability for others (Apple, Cavium/Marvell, Qualcomm, & Samsung?) to strike their own path, and they'll certainly be paying steeply to do so.
- someperson 6y agoSounds like https://en.wikipedia.org/wiki/NVLink https://en.wikipedia.org/wiki/NVLink and NVidia doesn't own AMD, Intel or motherboard manufacturers
- numpad0 6y agoThat's the Tegra, every luxury sedans has them, plus Nintendo Switch use one of them, as they're GPU focused and power management don't matter.
- dathinab 6y agoBuying ARM + their manufacturer improvements with Samsung might give Nvidia another real chance to enter the mobile marked! Thats if they keep ARM alive and mostly like it is for now at least.