Arm Seeks To Raise Prices Ahead of Hotly Anticipated IPO (ft.com) 29
Arm is seeking to raise prices for its chip designs as the SoftBank-owned group aims to boost revenues ahead of a hotly anticipated initial public offering in New York this year. From a report: The UK-based group, which designs blueprints for semiconductors found in more than 95 per cent of all smartphones, has recently informed several of its biggest customers of a radical shift to its business model, according to several industry executives and former employees. These people said Arm planned to stop charging chipmakers royalties for using its designs based on a chip's value and instead charge device makers based on the value of the device. This should mean the company earns several times more for each design it sells, as the average smartphone is vastly more expensive than a chip.
Headlines from six months in the future... (Score:4, Insightful)
Apple, Samsung switch to RISC-V
ARM holdings files for bankruptcy
Re: (Score:2)
Just what I thought reading the summary.
Hubris comes before the fall, as the saying goes.
Comment removed (Score:5, Informative)
Re: (Score:2)
TSMC make whatever chips their customers want. I expect they make the majority of both ARM and RISC-V chips excluding those made by Samsung.
Re: (Score:1)
Very true. Not in 6 months but at some point.
The problem is the culture. Arm are assholes. Sniveling little weasels trying to protect their "secrets" and pulling dollars out of your pockets.
Why do I have to sign over my first born just to get the free SDK they provide?
Re: (Score:2)
There's a lot of stickiness to an architecture when there's an existing software ecosystem. Apple has a huge investment in ARM technology, so they won't be switching anytime soon. Android won't, either, at least until everything on the Google Play store is available for both.
There's also a stickiness from the design investment. Companies have done tons of engineering to make really good ARM processors. That investment probably exceeds the licensing costs being paid to ARM, so they aren't excited about s
Re: (Score:2)
There's no incentive for Qualcomm, Samsung or MediaTek to switch ISA
But expect the low end market to be flooded with Android tv boxes and tablets with Allwinner riscv chips within 3 years. NB performance will still be adequate at best but that's true of $50 tablets today.
Re: (Score:3)
The thing with ARM is the design isn't the important part, the testing is. Most of ARM's employees do not work on the chip designs, they work on testing the chip designs with extensive test suites so if you implement an ARM chip, you can take the test suite and run it against your design to make sure your chip will work as expected.
It's how Apple and everyone else tests their designs to make sure they conform to the ARM ISA, and how licensees can test to make sure their chip was fabricated properly. It's al
Re: Headlines from six months in the future... (Score:2)
Why yes, I'd like to license that octa-core arm CPU and cache cluster tuned for and preverified on the silicon process vs implementing and verifying my own. It has extensive debug capabilities, with rich performance measurements you say? Delightful!
The best thing for an ARM IPO (Score:1)
Is for the company to stop suing its largest customers. That behavior creates uncertainty in the marketplace, and accelerates the development of alternative architectures like RISC-V. Investors want a steady business with growth and great returns, not the asinine behavior ARM has been showing lately.
Re: (Score:2)
For the last year or more, you can see that ARM has been caught with their parent company, Softbank, after making many risky bets that failed, trying to squeeze all the blood they can out of their successful investment before throwing away its dried husk. It's obvious they don't care what happens to it as long as it balances their books in the end. After the UK government stepped in to prevent the sale to NVIDIA, they've been doing everything in their power to use it for a quick turnaround for cash, which
Re: (Score:2)
It's obvious that ARM's days are numbered. ARM doesn't make chips, other companies figure out how to make performant processors that implement their architecture. ARM produces reference designs, and then the chip makers produce actual designs. That means they can do the same thing for RISC-V instead of ARM, and then ARM ceases to exist. Linux proved OSS could dominate in software, now it's time for it to [begin to] happen to hardware.
Re: The best thing for an ARM IPO (Score:2)
That's much easier said than done. If you create a custom chip design, you're going to pay through the nose to get it produced. Starting at the low, low price of something to the tune of say...1 million? I think if it were going to happen, it would have already happened a while ago with FPGAs. Anything more than that is a lot more involved than simply compiling an HDL and rigging it to a breadboard. Oh, you aren't an electrical engineer in addition to being a software engineer? Ah...well...(though I somewha
Re: (Score:2)
That's much easier said than done.
Of course it is. So? There's a lot of doing going on.
If you create a custom chip design, you're going to pay through the nose to get it produced.
ALL SoCs are "custom chip designs"
Bold Move (Score:2)
Can you imagine if food was priced this way? "This food costs 20% of your annual salary, since that's what we decided the value of providing you calories is."
I'm really curious about the economic theory behind charging percents of value for a good or service instead of charging absolute price.
I mean if you can get someone to pay you a percent of their revenue instead of a fixed price, I guess that's good for you? But what are the long-term effects - it's not possible for every good or service to be priced t
ARM should become a public utility (Score:1)
ARM should become a public utility
FU-Greedy. (Score:2)
Apple already went through this once with Qualcomm. They switched away to their own modem/cellular chipsets if I remember right.
Qualcomm was trying to charge based on the value of the iphone, not the chip. (just as an example of principle, not actual numbers:) : So a basic iphone 8 with 64GB of storage that goes for $400 will net them $5 for the cellular modem chip... An iPhone 14max pro with the exact same chip that sells for $1500 will net them $12-20
I don't see how this is different from a lumberyard
Bean counter understand numbers, not values (Score:2)
Re: Bean counter understand numbers, not values (Score:2)
Nah, it's more a case of companies looking at their customers making bank and then demanding a bigger cut because they aren't. Much like Bernie bros that think that just because their employer (aka their customer) makes too much, so they're entitled to a bigger cut. Just like them, ARM doesn't realize that they just might price themselves out of a job, particularly when it gives their customers an incentive to look for substitutions.
Re: (Score:2)
Engineers generally get a company up and running. It is management who understand the market and their customers that keep it successful and it is management who only listen to bean counters that destroy companies.
Re: Bean counter understand numbers, not values (Score:2)
Bean counter to me means accountants and financial types, usually the ones more concerned with cutting their costs, reducing spending, etc. What you're talking about are strategic decisions. Stereotypical bean counters generally aren't known to strategize, poorly or otherwise, and are basically bureaucrats.
Re: (Score:2)
ARM likely can't do this across the board (Score:2)
There are many IP licensees that have contracts. ARM can't just unilaterally come in and raise the price on fixed-price deals. If they do - I guarantee there are license terms that protect the purchaser from such things and it would cost ARM more than they would recover from the deal.
There are also architecture licensees like Apple and Faraday that have designed their own silicon version so the Arm CPUs that likely can't be affected by this price rise.