Originally, SoftBank appeared to have approached Apple about selling the tech company, but that seems to have fallen through. Now the possibility that SoftBank might be selling Arm to Nvidia raises some interesting questions about its future.
Once described as the jewel in the crown of British tech Arm, which licenses its IP to the likes of Apple, Qualcomm and Nvidia itself, has always been seen as a neutral supplier, one not in direct competition with its customers.
That could be where the proposed Nvidia acquisition becomes problematic, because although some argue that Arm’s IP is simply a ‘commodity’ and that it’s the other design IP that surrounds the Arm core that holds the key differentiators, Nvidia could end up influencing Arm’s future R&D direction ensuring that Arm provides it with priority access to new and future designs – which wouldn’t be surprising when they could be paying over $30bn for the company.
It would also end Arm’s ‘neutrality’. If it is acquired by Nvidia will it be able to serve the likes of Qualcomm or Apple? How will they feel about licensing IP from a unit that’s now owned by a significant rival?
Arm co-founder, Hermann Hauser, speaking to the BBC said that by allowing Nvidia to buy Arm it would compromise that neutrality and the ability of the company to address the needs of multiple disparate customers and there would be a good chance that Arm would lose many of them – in short it would be a ‘disaster’.
In response, perhaps to these concerns, reports have been circulated suggesting that Nvidia may simply look to take a stake in Arm, with SoftBank retaining a share. The idea of joint ownership has also been floated, according to reports.
So if this is true and Nvidia is looking to buy Arm, what are the advantages?
Nvidia licenses technology from Arm for its Tegra chips, but these devices don’t have a significant share of the mobile market so in terms of competition it seems unlikely it could be stopped on competition grounds – although any acquisition would face intense regulatory scrutiny.
Nvidia would probably use Arm’s capabilities to develop more powerful chips driving higher performance levels. It’s also ‘unlikely’ that Arm’s existing customers would drop them, when there aren’t really any other suitable architectures available to them.
How many of Arm’s customers does Nvidia actually compete against?
Could a tie up between these two tech companies actually be beneficial, bringing together Nvidia’s AI processing capabilities with Arm’s power saving techniques?
There are, however, some big questions that need to be answered about this possible deal.
Could Nvidia decide not to keep R&D focused in Cambridge and control everything from the US, and if Nvidia intends to buy ARM, how and where will it look to compete?
It’s a big deal, with significant consequences for the industry.
