The ARM Diaries, Part 1: How ARM’s Business Model Works
by Anand Lal Shimpi on June 28, 2013 12:06 AM ESTHow ARM Makes Money
AMD, Intel and NVIDIA all make money by ultimately selling someone a chip. ARM’s revenue comes entirely from IP licensing. It’s up to ARM’s licensees/partners/customers to actually build and sell the chip. ARM’s revenue structure is understandably very different than what we’re used to.
There are two amounts that all ARM licensees have to pay: an upfront license fee, and a royalty. There are a bunch of other adders with things like support, but for the purposes of our discussions we’ll focus on these big two.
Everyone pays an upfront license fee and everyone pays a royalty. The amount of these two is what varies depending on the type of license.
The upfront license fee depends on the complexity of the design you’re licensing. An older ARM11 will have a lower up front fee than a Cortex A57. The upfront fee generally ranges from $1M - $10M, although there are options lower or higher than that (I’ll get to that shortly).
The royalty is on a per chip basis. Every chip that contains ARM IP has a royalty associated with it. The royalty is typically 1 - 2% of the selling price of the chip. For chips that are sold externally that’s an easy figure to calculate, but if a company is building and selling a chip internally the royalty is based on what the market price would be for that chip.
Both the up front license fee and the royalty are negotiable. There are discounts for multiple ARM cores used in a single design. This is where things like support contracts come into play.
Buses/interfaces come for free, you really just pay for CPU/GPU licenses. ARM’s Mali GPU is typically going to be viewed as an adder and is currently viewed as demanding less of a royalty than ARM’s high-end CPU licenses. A rough breakdown is below:
| ARM Example Royalties | ||
| IP | Royalty (% of chip cost) | |
| ARM7/9/11 | 1.0% - 1.5% | |
| ARM Cortex A-series | 1.5% - 2.0% | |
| ARMv8 Based Cortex A-series | 2.0% and above | |
| Mali GPU | 0.75% - 1.25% adder | |
| Physical IP Package (POP) | 0.5% adder | |
In cases of a POP license, the royalty is actually paid by the foundry and not the customer. The royalty is calculated per wafer and it works out to roughly a 0.5% adder per chip sold.
It usually takes around 6 months to negotiate a contract with an ARM licensee. From license acquisition to first revenue shipments can often take around 3 - 4 years. Designs can then ship for up to 20 years depending on the market segment.
Of the 320 companies that license IP from ARM, over half are currently paying a royalty - the rest are currently in period between signing a license and shipping a product. ARM signs roughly 30 - 40 new licensees per year.
About 80% of the companies that sign a license end up building a chip that they can sell in the market. The remaining 20% either get acquired or fail for other reasons. Royalties make up roughly 50% of ARM’s total revenues, licensing fees are just over 33% and the remainder is equally distributed between software tools and technical support.
ARM's revenues are decent (and growing), but it's still a relatively small company. In 2012 ARM brought in $913.1M. Given how many ARM designs exist in the market (and the size of some of ARM's biggest customers), it almost seems like ARM should be raising its royalty rates a bit. Because of ARM's unique business model, gross margin can be north of 94%. Operating margin tends to be around 45% though.
Facebook
Twitter
RSS
64 Comments
View All Comments
Crono - Friday, June 28, 2013 - link
So basically ARM is like an author who choose to work with a bunch of different publishers?It amazes me how dominant they are, though, in the mobile processor industry, all without having to manufacture chips themselves. 45 billion chips... wow. And I'm guessing there is still plenty of room to grow with more embedded chips in more devices.
airmanchairman - Friday, June 28, 2013 - link
"It amazes me how dominant they are, though, in the mobile processor industry, all without having to manufacture chips themselves. 45 billion chips... wow."Their philosophy has always been centred on the needs of their client industries, which extend far beyo
airmanchairman - Friday, June 28, 2013 - link
...nd the mobile processor industry into the vast all-purpose market (street lights, lifts, automatic turnstiles etc). These industries strictly specified the maximum power output the chips could be capable of (used to be 700mW) and ARM designed within those parameters. As such their entire R&D/Logistics/Marketing focus has been perfectly suited to the battery-power-constrained mobile industry.Intel, on the other hand, after decades of dictating to the booming desktop industry with its faster/more cores/larger plants philosophy, is only just coming to grips with the power efficiency and discipline required to compete with ARM in the rapidly growing mobile sector.
The new Haswell architecture shows the promise and potential that Intel may
airmanchairman - Friday, June 28, 2013 - link
... bring to the battle against ARM.nadim.kahwaji - Friday, June 28, 2013 - link
veryyyy nice post, anand i know that you are so busy, but we miss the podcast !!!!!!blanarahul - Friday, July 26, 2013 - link
Umm... Does Qualcomm also pay royalty for the chips that use Krait?blanarahul - Friday, July 26, 2013 - link
Is Qualcomm automatically eligible to use any of ARM designs since they are above the "Subscription Licence" level in the pyramid?dealcorn - Friday, June 28, 2013 - link
I could not follow the sense of your statement "From Intel’s perspective, it made the mistake of licensing the x86 ISA early on in its life, but quickly retreated from that business." I thought IBM required that Intel license X86 to competitors as a pre condition of IBM's selection of Intel's 8086 for the IBM PC. Did Intel made a boo boo letting IBM use the 8086 in the IBM PC? It was reported that Intel's selection of x86 license partners was driven by the dual criteria that the partner must be acceptable to IBM and also likely incompetent to exploit the benefits of the license. In retrospect, their selection of partners achieved these goals. What was Intel's licensing mistake?The aphorism, "Knitter, stick to your knitting" has long been appreciated as a useful business strategy. ARM is properly commended for it's strict adherence to this truism in it's attempted optimization of architecture design. Intel appears focused on SoC level optimization. Intel's holistic approach is more time consuming and capital intensive. Consumers will ultimately vote with their wallets which approach is better in the contested ultra mobile space.
SleepyFE - Friday, June 28, 2013 - link
Intel licensed it's IP to AMD, but AMD didn't throtle it's CPU's so it would get more business with constant iteration of the same architecture with marginal improvements to it (Intel's tic toc development model). AMD made the best use of what they got from Intel. At that time their (AMD's) CPU's were faster and they were eating away Intel's business. Intel had since then made a new architecture and did not license it to AMD. Consequently AMD fell behind and had not recovered since.Wolfpup - Friday, June 28, 2013 - link
AMD and Intel cross license x86. It's not fair to claim x86 is Intel's thing along, given how much stuff Intel has to license from AMD for it-heck, the most obvious is 64-bit. The updated ISA is AMD's design, and Intel started using it a few years later."At that time their (AMD's) CPU's were faster and they were eating away Intel's business. Intel had since then made a new architecture and did not license it to AMD. Consequently AMD fell behind and had not recovered since."
I can't remember if AMD ever actually licensed chip designs-if they did, it wasn't recent, would have been like the 486. Everything since then that AMD sells has been AMD's design. When AMD was faster, that was with entirely AMD designs. And no, Intel doesn't license Core, but that's not new like you're saying, that's been the case since the 486 days (if those were even licensed), and it's not why AMD's fallen a bit behind. Intel was making horrible decisions with their CPUs, and AMD was designing better ones. AMD's stuff is still really strong/good, it's just Intel's bigger, has thrown more resources at it, usually has a process node advantage, and since getting serious again has been able to have more powerful chips at the high end (though of course the reality is AMD's chips keep getting more powerful too, and are only "bad" in comparison to Intel's newest, and sometimes even then AMD looks better, like that platform comparison using integrated graphics where AMD's new $140 CPU was stomping on one of Intel's best i7s, often 50% better).