Nvidia abandons deal to buy ARM *spawn*

Apple is irrelevant.
The larger Apple's marketshare is, the less devices can be made with NVIDIA GPUs and Apple marketshare is growing in important markets.

If ARM gets bought by a vulture capitalist which invests fuck all (which is IMO the most likely alternative to NVIDIA) the consumer market will get worse for NVIDIA.
 
The larger Apple's marketshare is, the less devices can be made with NVIDIA GPUs and Apple marketshare is growing in important markets.

If ARM gets bought by a vulture capitalist which invests fuck all (which is IMO the most likely alternative to NVIDIA) the consumer market will get worse for NVIDIA.
Samsung & other big players at least planned to make offer as co-owning it among many Arm using companies (if NVIDIA doesn't get it)
 
The larger Apple's marketshare is, the less devices can be made with NVIDIA GPUs and Apple marketshare is growing in important markets.
Apple's market share in there most important markets is pretty much at a plateau. Markets - which are relevant here - are growing?

If ARM gets bought by a vulture capitalist which invests fuck all (which is IMO the most likely alternative to NVIDIA) the consumer market will get worse for NVIDIA.

Are you suggesting the regulators would be smart enough to reject Nvidia acquiring ARM but stupid enough to approve an organisation who just want to loot it? Do you have any evidence of this every having happened before?
 
Are you suggesting the regulators would be smart enough to reject Nvidia acquiring ARM but stupid enough to approve an organisation who just want to loot it? Do you have any evidence of this every having happened before?
It has been mentioned that companies like Qualcomm (who expressed an interest in ARM shares) would have an overall negative effect. Not sure if regulators would have any say in a case like this.
While not the same as an acquisition, Qualcomm purchasing a share of ARM does raise cause for concern when considering the business practices of Qualcomm. From integrating their technology into global standards to charging customers for technology licenses that they do not use, Qualcomm is far from perfect in this situation. It could be argued that Qualcomm themselves are being highly hypocritical.

While Qualcomm having a portion of ARM would not be able to control it outright, it would give Qualcomm a foot in the door to ARM and its decision-making process. Furthermore, Qualcomm’s somewhat immoral businesses practices could easily spread into ARM management and see a push towards unfair licensing terms. Thus, a company that complains about how NVIDIA will harm ARMs reputation as an independent business could end doing the exact same thing.
https://forum.beyond3d.com/posts/2212960/
 
It has been mentioned that companies like Qualcomm (who expressed an interest in ARM shares) would have an overall negative effect. Not sure if regulators would have any say in a case like this.
The UK absolutely does. In addition to traditional monopolies and mergers considerations, the economic impact and national security implications of mergers and acquisitions are considered. If I own a company that created sensitive technology and I decided to sell the company to a foreign owner, I could be prevented from doing so under variation legislation that is intended to preserve the economy and not introduce risks to national security. It's happened recently.

And the UK is far from the only county in Europe with such powers. This is very much an emerging legislatory measure designed to counter foreign investors who may not share the interests of your company/country.
 
It would be an interesting idea to grant 20% of ARM's shares to QCOM, 20% to Apple, 20% to NVIDIA, 20% to Samsung and the remaining 20% to all others interested. The fact that it might take more than a decade for each hw generation to appear because the owners wouldn't easily agree to anything, is besides the point *cough*

While the original sale of ARM to Softbank is done and old news I'd still like to hear what the exact reasoning of the British government was to allow the sale of such a large AND profitable technology powerhouse as ARM. Usually governments or even nations are proud of such achievements.....which brings me to the point that whatever legislation has appeared ever since within the EU or in countries that used to belong to the EU is simply too little and too late.
 
It could be an interesting idea for x86 and RISC-V companies to do that to Arm, yeah.
Apple and Qualcomm would likely also be very okay with that (for some time at least) since they design most of their Arm h/w in-house anyway.
But for a wide market such option would be nothing short of a disaster.
 
While the original sale of ARM to Softbank is done and old news I'd still like to hear what the exact reasoning of the British government was to allow the sale of such a large AND profitable technology powerhouse as ARM. Usually governments or even nations are proud of such achievements.....which brings me to the point that whatever legislation has appeared ever since within the EU or in countries that used to belong to the EU is simply too little and too late.

ARM's sale to Softbank in 2016 came with a number of regulatory strings, including the UK entity's retainment of the IP and continuance of the UK HQ operation. The UK's underlying legislation, the Enterprise Act, stems from 2002 so is not new. There are evolving considerations which are new but the UK's ability to regulate and prevent acquisitions and investments is not new.

Softbank's acquisition of ARM was declared as being a financial investment with no intention to meddle with the business. This is perhaps not the case with Nvidia's intended acquisition. That's probably the biggest perceived different between the two acquisitions and why they are panning out differently.

As for the EU, whilst the EU has it's own regulatory approval process, this is focussed on the impact of mergers and acquisitions on the market regarding competition and consumer choice. It has always been up to individual member states to apply their own considerations and approval processes where their national interests are at stake.

And again, this is not restricted to just the UK.
 
Last edited by a moderator:
It could be an interesting idea for x86 and RISC-V companies to do that to Arm, yeah.
Apple and Qualcomm would likely also be very okay with that (for some time at least) since they design most of their Arm h/w in-house anyway.
But for a wide market such option would be nothing short of a disaster.

I consider that scenario so unrealistic, that I thought the sarcasm was obvious ;) (OT: nice to read you again).
 
ARM's sale to Softbank in 2016 came with a number of regulatory strings, including the UK entity's retainment of the IP and continuance of the UK HQ operation. The UK's underlying legislation, the Enterprise Act, stems from 2002 so is not new. There are evolving considerations which are new but the UK's ability to regulate and prevent acquisitions and investments is not new.

Agreed.

Softbank's acquisition of ARM was declared as being a financial investment with no intention to meddle with the business. This is perhaps not the case with Nvidia's intended acquisition. That's probably the biggest perceived different between the two acquisitions and why they are panning out differently.

While they didn't disrupt any of ARM's business, I don't think they made any serious investments in the end either. Again ARM wasn't in depth, au contraire. What companies like that would need from any government is strong assistance in order for their business to further florish. Softbanks true intentions of the past show today.

As for the EU, whilst the EU has it's own regulatory approval process, this is focussed on the impact of mergers and acquisitions on the market regarding competition and consumer choice. It has always been up to individual member states to apply their own considerations and approval processes where their national interests are at stake.

And again, this is not restricted to just the UK.

This is obviously not the place and time to go that far OT, but I never thought much of decorative organizations since their birth.
 
While they didn't disrupt any of ARM's business, I don't think they made any serious investments in the end either. Again ARM wasn't in depth, au contraire. What companies like that would need from any government is strong assistance in order for their business to further florish. Softbanks true intentions of the past show today.
I couldn't find any definite articles about Softbank's investments inARM between 2016 and 2020, but Softbank themselves talk about rapid growth in the engineering departments. ARM don't manufacture so it probably doesn't take significant investment to expand certain areas.

This is obviously not the place and time to go that far OT, but I never thought much of decorative organizations since their birth.
What is a 'decorative organisation'? :???:
 
I couldn't find any definite articles about Softbank's investments inARM between 2016 and 2020, but Softbank themselves talk about rapid growth in the engineering departments. ARM don't manufacture so it probably doesn't take significant investment to expand certain areas.

Engineers are IMHO the highest percentage of R&D related investments. Question would be (as just one example) if their Mali/GPU IP department wasn't/isn't understaffed why they never catered for higher end GPU IP and why they lost their one and only major GPU IP client to AMD. With Samsung soon out of the picture, their last big GPU IP customer is Mediatek.

Next best question would be why NVIDIA is again promising high investments after the hypothetical takeover, which in this case I would tend to actually believe them (knowing how they usually operate) irrelevant which purpose the investment would actually serve.

Let's see:
https://group.softbank/en/ir/financials/annual_reports/2021/message/segars

In fiscal 2020, Arm increased its R&D headcount by 11.0% year-on-year, as it continues to develop more advanced technology and to further expand its product portfolio.

I wouldn't call 11% a significant investment, but I'd gladly stand corrected.

What is a 'decorative organisation'? :???:

One that exists only to verify its existence. If you want a close next example there's another one that starts with a N and ends with an O or another that starts with a U and ends with an N. [/end of OT] ;)
 
Last edited:
Engineers are IMHO the highest percentage of R&D related investments. Question would be (as just one example) if their Mali/GPU IP department wasn't/isn't understaffed why they never catered for higher end GPU IP and why they lost their one and only major GPU IP client to AMD. With Samsung soon out of the picture, their last big GPU IP customer is Mediatek.

In the semiconductor industry, people are cheap - it is manufacturing that is expensive. Look at ARM's running costs compared to TSMC. This is what I'm talking about. You probably couldn't spent a billion dollar pounds on engineers because you'd have too many to integrate in your business model. For TSMC, a billion dollars doesn't build a quarter of a plant. Now look at ARM's profits versus TSMC. Crazy huh.
 
In the semiconductor industry, people are cheap - it is manufacturing that is expensive. Look at ARM's running costs compared to TSMC. This is what I'm talking about. You probably couldn't spent a billion dollar pounds on engineers because you'd have too many to integrate in your business model. For TSMC, a billion dollars doesn't build a quarter of a plant. Now look at ARM's profits versus TSMC. Crazy huh.

R&D stands for research and development and not manufacturing. I noted clearly R&D related investments, as we're talking about an IP provider. Softbank had promised before acquiring ARM that they intend to make strong investments in the company, but I fail to see them; Softbank obviously bought ARM in order to resell it at their earliest convenience.
 
Last edited:
R&D stands for research and development and not manufacturing. I noted clearly R&D related investments, as we're talking about an IP provider. Softbank had promised before acquiring ARM that they intend to make strong investments in the company, but I fail to see them; Softbank obviously bought ARM in order to resell it at their earliest convenience.
The development of R&D in the semiconductor industry is running a fuckton of test runs to test and qualify processes, wafer compositions and validate deigns. This is staggeringly expensive. Semiconductor manufacture is only efficient when undertaken on a massive scale. If you're producing just 1,000 dies for qualification, the manufacturing cost for this aspect of the development process is astronomical.

Nobody goes from simulated design to fully-working mass-produced IC when you have billions of transistors. To develop your process you have to test them. A lot. It's fucking expensive. If you think TSMC are doing it wrong, you should email them. :yep2:
 
Last edited by a moderator:
This has absolutely zilk, nada, nothing to do with the claim that Softbank did not heavily invest in ARM.
 
Last edited:
This has absolutely zilk, nada, nothing to do with the claim that Softbank did not heavily invest in ARM.

Your claim there wasn't serious investment. I am clarifying that because ARM are a design box, and do not build or manufacture, they can have expand sizeably without significant investment because there isn't an associated expensive capital investment as well.

I do not see what relevance any of this has to be the regulatory hoops that need to be jumped through. SoftBank bought ARM, later sold a bunch of shares, and now want to sell their remaining interests. ¯\_(ツ)_/¯
 
“The FTC is suing to block the largest semiconductor chip merger in history to prevent a chip conglomerate from stifling the innovation pipeline for next-generation technologies,” said FTC Bureau of Competition Director Holly Vedova. “Tomorrow’s technologies depend on preserving today’s competitive, cutting-edge chip markets. This proposed deal would distort Arm’s incentives in chip markets and allow the combined firm to unfairly undermine Nvidia’s rivals. The FTC’s lawsuit should send a strong signal that we will act aggressively to protect our critical infrastructure markets from illegal vertical mergers that have far-reaching and damaging effects on future innovations.”

https://www.ftc.gov/news-events/pre...es-block-40-billion-semiconductor-chip-merger
 
Back
Top