SoftBank is reportedly assessing spin-off options for its semiconductor firm, Arm Holdings. The Wall Street Journal reports from its sources that those options include having an initial public offering or a sale. The Japanese tech conglomerate picked up Arm back in 2016 for $32 billion and currently shares some ownership with investors in the SoftBank Vision Fund.
The moves are being considered as SoftBank fends off challenges from activist investment house Elliott Management over major losses for its Vision Fund, including WeWork’s attempted IPO. SoftBank is supposedly targeting $41 billion in immediate fundraising through share buybacks and divestitures. Depending on Arm’s current prospectus, a sale could be more likely to happen than an IPO.
I’m linking to the AndroidPolice item since the original article is stuck behind a paywall.
Whoever intends to acquire ARM better have a very good story to tell antitrust regulators, because I doubt Intel, Apple, Google, or any of the other major technology companies will be allowed to acquire it. I wonder who else could be a potential buyer – maybe another investment fund?
Potential acquisitions of big important hardware vendors makes me very nervous. (Oracle Sun Acqisition!!??!) Its always a crap shoot whether the new owner will mismanage it into the ground, or sell it off as spare parts. I guess nothing lasts forever, but it feels like ARM is just coming into its own in a big way, and it would be nice to see it truly go head to head with intel.
ARM’s biggest problem in the server space is a resurgent AMD and Nvidia.
AMD has lots of cores, which was kind of ARM’s thing.
Nvidia owns the GPU processing market, and they now own Mellanox. Their play seems to be GPU compute boxes attached via Ethernet, Infiniband, or the gen-z interconnect.
ARM’s server window might have already closed, and they should focus on vertically integrated client devices.
You would think so, but ARM is at the top of the top 500 right now.
So clearly with the new ARM CPU design (and help from Cray) it can be done.
How does that relate to the market in general?
HPC is interesting, but it’s pretty niche, and only tangently connected to the market in general.
For instance, Intel Phi was all the rage in HPC a couple of years ago, and it went nowhere. Nvidia GPUs, the incumbent technology then and now, took it to the woodshed.
Cell procs were also the big thing in HPC for a few years. Once again, they went nowhere.
As far as ARM goes, I’m not doubting the ISA’s ability to perform; I’m doubting they have a value proposition over incumbent technology in the server space.
I need to see Dell start selling ARM servers, I need to see ARM accelerate the software I use on a regular basis, and I need to see the ARM ecosystem not devolve into warring fiefdoms which lock me into a vendor increasing the stress in my life. Part of this is on the devs to tune the software, part of this is ARM vendors need to add accelerators to their chips, and part of this is probably going to happen because ARM doesn’t have control over what vendors put into their chips.
People can’t say “We’re going to standardize on ARM.” the way they can say “We standardize on x86.” x86 chips are all roughly symmetric (acknowledging the market segmentation games Intel likes to play), and it remains to be seen if the ARM ecosystem comes to a similar feature equilibrium.
I’ll probably buy some personal ARM server stuff to play with, but I’m not going to spend business money on ARM hardware right now. Definitely nothing above super cheap low end tertiary servers.
I’m still annoyed by ARM’s sale to Softbank in 2016 – it was the crown jewels of the British IT industry and there seemed to be little or no outcry at the time that it would become Japanese instead. I’m hard-pressed to think of another UK IT company with anywhere near the worldwide recognition of ARM.
As for this latest news, it doesn’t surprise me much after the WeWork fiasco and just reinforces my view that the origjnal ARM sale was ill-advised. Sadly, I don’t think ARM will return to UK hands – I don’t think any UK company is big enough to afford it.
Wasn’t that idiotic Brexit thing partly responsible for that?
ARM’s licensing model has always been founded on a level of independence from chip manufacturers and its customers. It’ll be interesting to see if a change in ownership causes the balance to shift, which could have the knock-on effect of increasing interest in RISC-V.
It would be a strange result if a bizarre failure in the co-working space causes a global shift in instructions set usage.
Tbh, I think Google or Amazon would be the best “giants” to take it over. They have the most interest in licensing designs to multiple manufacturers. Other companies would represent a challenge to ecosystem if they had sole/majority ownership – especially someone like Apple.
The key concern is ensuring that there are non-prejudicial terms for licensing the designs.
One other solution would be for some of the major players to come together and purchase it as a joint venture.
I was really thinking about this possibility a few days ago
I think ARM would be better off as an independent company, as it was until 2016.
I have doubts that Google/Apple would really profit from a direct acquisition, pretty sure it wouldn’t make any economical sense.
I don’t know the actual numbers, but I guess Apple pays at most 250 millions to ARM each year (probably much less than that), …. peanuts compared to its current revenues.
Shutting the other vendors out is priceless and may break some of them, due to costs of migrating.
That would be quite stupid to do.
That would be perfection, honestly. Apple started ARM with Acorn as processor for the Newton, and 30 years later, they are transitioning everything to ARM with their own in-house design team.
Apple can’t let ARM get bought by a competitor. They would, once again, become beholden to another company which doesn’t care about their interests, at best, hostile at worst, and they can’t announce another transition so soon. “We’re going ARM!” 6 months later, “We’re going MIPS! We’ll need everyone to send back the ARM Macs.”, RISC-V, x86, SPARC, POWER, or something else.
It’s also a golden opportunity to hamstring the competition for Apple, or diversify to increase revenue via licensing fees. It’s why companies should have a war chest full of cash to protect the supply lines.
I smells a apple fanboy. Apple is one of the few companys, if not the only one, who would close arm down
“Apple started ARM with Acorn as processor for the Newton”
You need to read up on your history. ARM was created by Acorn, for Acorn, to power their RiscPC and Acorn Archemedes range of computers. Apple only used ARM in the Newton because it was small, cheap, and power efficient. Apple tailored the ARM architecture for use in the Newton, but much of the architecture and designs were developed way before Apple got involved
https://en.wikipedia.org/wiki/Acorn_Archimedes
The comment system won’t allow me to reply to @The123king (nesting too deep?), but hopefully my meaning will be clear anyway.
While this is true if we’re talking about the ISA, if we’re talking about the company then @Flatland_Spider’s comment isn’t unreasonable. ARM was originally spun out of Acorn as a joint venture: 30% Apple and the rest split between Acorn and VLSI.
If Apple have the same sense they did back in 1990, they’ll realise the importance of ARM being independent. I speak as someone who discovered computing through Acorn, but could never get excited about Apple kit, just so my bias is clear.
@th22
You’re smelling your butt. Take a shower.
Intel, Nvidia, and IBM are others I can think of off the top of my head. Possibly, Alphabet/Google since they close down projects they acquire.
@The123king, @flyingpig
Yes, ARM the company.
https://en.wikipedia.org/wiki/ARM_architecture
https://en.wikipedia.org/wiki/Arm_Holdings
I’m sure they do see the value of keeping ARM independent, but at the same time, they have a lot more at stake now. They’ve announced the ARM ISA is going to be the ISA for their entire lineup going forward, and if I had the money, I would buy ARM to jack up the rates on Apple. I’m probably not the first person to think of that, so there have to be other jackals circling.
Yep. This is the worst scenario for everyone. The only good spots ARM could land is AMD or Apple.
AMD designs and sells chips. They have highend GPU tech with FOSS drivers which would be a boon to the ARM ecosystem, and they have experience selling custom designs.
Apple needs ARM to maintain control of their destiny. They can’t let Qualcomm, Samsung, or Intel buy ARM. They would, more then likely, close ARM down since they don’t like to share, and it would be kind of fun to see the chip vendors scatter.
Is MIPS still on the market? Is that even an option, or does the future belong to Power and RISC-V?
As for now I would tell that future belongs to ARM. There is godzillon units sold every year with it, Meanwhile I don`t see anything with Risc-V. And Power? Oh…
WD uses RISC-V as the controller on their hard drives. Nvidia cards have RISC-V chips on them.
Power shows up in servers. They are really good at throughput.
Apple had an opportunity 4 years ago and let it go.
And Thom is right on this one: a corporation like Apple or Google cannot buy ARM without having many many untitrust problems around the globe.
RISC-V has only just started.
The good thing about it could be: one of the big companies buys ARM and then the competitors think: euh… I think we’ll invest some more money in RISC-V as well so we can get away from ARM in the long run.
Some companies have started replacing low end ARM cores with RISC-V cores. Nvidia and Western Digital are two that come to mind that use RISC-V cores to shave some cents off of the bill of materials.
I have a soft spot for Power, so I’d like to see it make a comeback.
I bet Apple will make a play for it regardless. It fits their CPU adventure and can kill off Adroid in theory.. or at least make all Android phone makers pay a tax to APple. All that’s needed is for aApple to spin off another company to do the dirty work. M$ did it for SCO in funding them the needed money to fight against NOVELL for Linux and SYSV. It also would work out for AMD, as they’re second place in ECO Chips to ARM.
Apple would probably buy it just to stick it to Qualcomm, or at least Jobs would have. Sure Qualcomm has cell patents, but they don’t own their own ISA. >:)
I can see a consortium buying ARM, and then donating the ISA to a neutral foundation. Put Jim Keller at the helm as a semi-retirement gig.
ARM Foundation: 60% Apple, 20% AMD, 20% Samsung
Let’s handicap the companies interested.
# Apple:
They really want to control their own fate, and they need an ISA they can control. ARM really fits their certical integration strategy quite well.
# Nvidia:
Wants to break out of the box, and they can only do that if they have a basic CPU to take care of maintenance tasks. They’re also petty enough to see an ARM acquisition as payback to Apple and Intel.
# Samsung:
One of the top mobile handset producers, and they have their own fabs.
# Intel:
Could either milk ARM or shutdown a competitor sending the tech world into disarray.
# Marvell:
They bought Cavium, so why not buy the ISA.
# Broadcom:
Acquiring chip companies seems to be what they do.
# Qualcomm:
They don’t own their own ISA, and they would love to bring more competitors to heel.
# TSMC/Global Foundries:
A fab company with it’s own chip design services seems to work pretty well for Intel.
# Nokia:
Why not?
Regulators in the the current anti-trust climate wouldn’t even look at it as long as the buyer is a western company, at least in the US.
Yeah, I wouldn’t count on US regulators in this political climate to even bat an eye, no matter who wants to buy ARM…
Would there be any impedence to a Chinese company buying ARM?
I REALLY don’t want to see Qualcomm grab it. I’d certainly like to see a company who will open source the ISA.
Data protection is very important right now
If antitrust issues are such a problem, I could ease SoftBank burden.
I just bought a car but still have a few coins in my wallet. I offer 43.26EUR