Ladies and gentlemen, good morning. Thank you for standing by, and welcome to the Arm Q2 Results Analyst Conference Call. At this time, all participants are in a listen-only mode. There will be a presentation followed by question and answer session, during which, if you wish to ask a question, you'll need to press star and one on your telephone keypad. I must also advise you that this conference is being recorded today, Wednesday the 27th of July, 2016. I'm now handing the call over to your first speaker today, Mr. Ian Thornton. Please go ahead, sir.
Thank you, Andre. Good morning, everybody. This is Ian Thornton, VP of Investor Relations at Arm. On today's Q2 results conference call, we have Simon Segars, Chief Executive Officer, and Chris Kennedy, Chief Financial Officer. On today's call, Simon and Chris will take us through the highlights and comments on the quarter's results. Then we'll open up the call to a Q&A session. As a reminder, the press release and some financial data can be found on Arm's investor relations website at www.arm.com/ir. I now have a few words with respect to this conference call and what we're about to discuss. The contents of this conference call are being directed only to those of you who have professional experience in matters related to investment. The information communicated on this call is being made available only to investment professionals.
Any persons present on this call who does not have professional experience in matters related to investments should not act on or rely on the contents of this call. The following conference call may include forward-looking statements, which will be statements which are other than statements of historical fact. The company's actual results for future periods may differ materially from these statements as they are based on current expectations and are subject to a number of risks and uncertainties. Clearly, this quarter is a little different from our normal quarterly results. As a reminder, on the 18th of July, 2016, SoftBank and Arm announced that we had reached agreement on the terms of a recommended all-cash acquisition of GBP 17.00 per share. Revenues up 12% year-on-year.
This quarter, more thought-leading companies made long-term commitments to Arm technology. Our semiconductor partners announced many new and exciting Arm-based products for a wide variety of applications, including sensors, supercomputers, and satellites. These design wins characterize the drivers of our long-term growth. More advanced Arm technology being chosen by more partners to be deployed in more markets. Putting some detail on this. Fujitsu and RIKEN announced that their next-generation supercomputer will be based on Armv8-A-based chips. When that machine goes into operation in 2020, RIKEN expect it to have 100 times the performance of their current supercomputer, which is ranked number five in the world. At the smaller end of the spectrum, we are seeing Arm technology being deployed in an increasingly wide range of devices.
As an example, NASA recently completed a three-year study into high-performance space computing requirements and selected the Cortex-A53 processor for its onboard computers. We saw dozens of Arm partners, including AppliedMicro, Broadcom, Cavium, Mellanox, and NXP all announce new chips for networking infrastructure and servers. Qualcomm, ST, Xilinx, and many others announced new embedded intelligence chips for smart sensors, medical equipment, automotive applications, and so on. Arm has been investing and developing technology for many years in these areas. It is great to see our partners being so successful. Arm is continuing to invest in new technologies. During the quarter, we announced the Cortex-A73 CPU and Mali-G71 graphics processor. Together, these processors will enable virtual reality and augmented reality in smart mobile devices.
We also announced the supporting technologies needed to integrate these processors into an SoC, including on-chip interconnect, physical IP libraries, software tools, and support from our ecosystem partners. 10 of our customers now have access to the Cortex-A73 and/or Mali-G71, and are building them into their next-generation mobile application chips. In May, Arm announced the acquisition of Apical for GBP 251 million. Apical is at the forefront of embedded computer vision technology, building on its leadership in imaging products. Together with our existing technology, this will help the Arm partnership to address some of the challenges of next generation products, such as driverless cars, robotics, and sophisticated security systems. At acquisition, the Apical team was about 100 people strong, the majority being R&D engineers. We are very pleased with the integration process so far. The companies have similar cultures, which is helping with integration of teams, processes, and roadmaps.
In June, we entered into a strategic partnership with HOPU Investment Management, a Chinese private equity firm, through which we are launching a fund that will be focused on the Internet of Things, smart devices, big data, and cloud computing. Also in June, we signed an architecture license with Guizhou Huaxintong, a joint venture between Qualcomm and China's Guizhou Province. Guizhou Huaxintong stated aim is to develop advanced server chipset technologies in the rapidly expanding Chinese server market, the second-largest data center market in the world. Now I'll discuss the revenue drivers in the different parts of the business in more detail, starting with technology licensing. We signed 25 licenses with 23 customers in the quarter. Just as in Q1, over half of our customers this quarter were new to Arm.
Five of the licenses signed were for Arm's Cortex-A processors, mainly for use in smartphones and networking infrastructure. 13 of the licenses were for Cortex-M processors for use in embedded devices and in the Internet of Things. Two of the licenses were for Cortex-R class processors for use in advanced memory controllers and wireless communications. We also signed two licenses for our Mali multimedia processors for smartphone and digital TVs. Physical IP licensing was down 37% compared to the strong number reported in Q2 2015. This was a result of the timing of engineering milestones and a higher proportion of semiconductor companies opting to take single-use licenses rather than multi-use licenses for our POP IP technology. We also signed a major deal in the quarter to develop a 10 nanometer library of physical IP for a foundry's most advanced manufacturing process, which was backlog building.
Now I'll switch to the royalty side of the business. Arm's royalty revenues are reported one quarter in arrears, so our royalty for Q2 was generated from chips sold by our licensees in Q1. Processor royalty revenue was up 11% year-on-year. By comparison, the semiconductor market declined 3% in the relevant period, we outperformed the industry by 14 percentage points. In Q1, the smartphone market was down around 10% sequentially, there was also a widely reported inventory correction in the smartphone supply chain. We saw an impact from both of these factors in Q2. The sale of application processors in smartphones also declined by about 10% sequentially, and the sales of peripheral chips for use in a smartphone declined to a much greater extent. During the quarter, approximately 250 million chips contained Armv8-A processors, and 110 million of these chips were octa-core.
Around 65% of all smartphones sold in Q1 contained an Armv8-A processor, around 30% contained an octa-core chip, up from around 25% in the prior quarter. Physical IP royalty revenue grew 17% year-on-year due to the increase in shipments of wafers using Arm's Physical IP at advanced nodes. I'll now hand over to Chris, who'll provide some further details on the numbers.
Thank you, Simon. Good morning, everybody. Hopefully many of you will have had a chance to have a quick look at Arm's Q2 2016 earnings release. The quarterly results and financial data are on our website as usual. Q2 dollar revenues of $388 million were up 9% year-on-year, with 7% growth in technology licensing and 12% growth in technology royalties. In Q2 2015, the sterling to dollar effective exchange rate was 156. In Q2 this year, it was 145. That's an 8% weakening of sterling versus the dollar. Q2 sterling revenues were up 17%. We also had the release of a tax provision following the receipt of correspondence from a tax authority. Combining these together resulted in EPS growth of 18% year-on-year. Without the tax provision release, EPS would have increased around 7%.
For those wondering about the impact of Brexit and the subsequent weakening of sterling, as it happened right at the end of the quarter, there was very little impact on our Q2 results. As we have often discussed on these calls, quarterly license revenues can be lumpy. In Q2, processor licensing was particularly strong, up 14% year-on-year. This was driven by 25 processor licenses, including a high-value architecture license. In contrast, Physical IP license revenue was down 37% year-on-year to $14 million for the reasons that Simon has already covered. About 50% of our processor licensing revenue came from backlog in the quarter, which is within the typical range of between 40%-60%. Group backlog was down around 10% sequentially. Normalized OPEX in Q1 was GBP 130.7 million, up 32% year-on-year and down 2% sequentially.
As you will recall, we announced a step-up in R&D investments in Q3 last year. Most of that step-up was achieved in H2 last year, and we're now in a period of more stable headcount and cost growth. Excluding the acquisition of Apical, headcount was up about 1.5% sequentially. The group's effective normalized tax rate was 6% in Q2 2016. This was much lower than prior quarters due to the release of a tax provision of GBP 11.5 million. Excluding the tax provision release, our tax rate would have been around 15%. Normally at this point, I would talk about outlook. However, given the restrictions based on forward-looking statements as a result of the SoftBank offer for Arm, it's not appropriate for us to provide the usual guidance for full-year 2016 revenues or next quarter's cost range or the full year's tax rates. I'll now come back to Simon.
Thanks, Chris. Now we'll go to Q&A. As usual, if I could ask you just to ask one question at a time, we will be able to get everyone's question in, and if there's time at the end, we can come back for a second round. Also as a reminder, we will not be discussing the details of the offer from SoftBank to Arm, nor will we be giving any forward-looking statements. If I can hand back to the operator to manage the Q&A, please.
Certainly. Thank you. Ladies and gentlemen, as a reminder, if you wish to ask a question, please press star and one on your telephone keypad and wait for your name to be announced. If you wish to cancel your request, please press the hash key. Your first question comes from the line of Francois Meunier from Morgan Stanley. Please go ahead.
Morning, gentlemen.
Hey, Francois.
Hey. Yeah. Just a quick question on Q2. I think on Q1 you disclosed the growth which you had in networks, which was +10%. I don't know if you could disclose or give some flavor around network infrastructure revenue growth in Q2, if you can.
Not off the top of my head. I don't have that in front of me. In terms of proportion of units, we did see a bit of sequential growth in that split, up about a percentage point sequentially and year-on-year into that enterprise segment.
Okay. Thanks for the clarification. Thank you.
It's all right. Just bear with me. It was also up 10% in Q2 as well as Q1.
Okay. Thanks again.
Thank you. Your following question comes from the line of Matt Ramsay from Canaccord Genuity. Please go ahead.
Thank you very much for taking my questions. Good morning, guys. Obviously, congratulations on the deal. It's been a pleasure to work with you guys over the years. I guess, Simon, I don't know how much you can comment on this, but I'd be interested in just any feedback you've gotten from licensees around the SoftBank announcement. There's been lots of debate of that potentially freeing you guys up to invest more heavily. One of the things I've always admired about the team is the ability to balance the needs of all the different licensees as you guys have moved forward with the business. Any comments there? Chris, if you could break out at all for us any contributions, either costs or revenue from Apical in the quarter, that would really help. Thanks, guys.
Okay. Thanks, Matt. As I said, I can't comment on structure of the deal. We spent a lot of time communicating with all constituencies. A key message that we said previously is that there would be no change to our business model, no change to the way that we go about working with our partners. Beyond that, I don't want to go into any specifics.
On Apical, I appreciate that you haven't had too much time to look at the statement. Bottom of page 18, we break out the Apical results. In Q2, it added about GBP 1 million in revenue, and GBP 1 million that is in revenue and pre-tax profit of GBP 100,000. We also give the sort of pro forma had it been consolidated for six months, which is GBP 6.9 million in revenue and GBP 0.4 million pre-tax profit.
All right. Thanks, guys, appreciate it.
Thank you. Your next question comes from the line of Sandeep Deshpande from JP Morgan. Please go ahead.
Yeah. Hi. Thanks for letting me on. Simon, you've announced a few wins in the server market in the quarter. Do you see yourself, with what you've already, in terms of the wins you've announced and what you've not announced publicly, that you are on track to do the 25% share in units by 2020 at this point in the server market? Secondly, do you see that you're already seeing the impact of the smartphone slowdown in your numbers, that this is now actually already in the number. From here onwards, you will not see a further impact? Thank you.
Well, I'll just answer those questions in the context of Q2.
Yeah.
If you look sequentially, units are down. If you look at the split of where those units have gone, you'll see that mobile and connectivity for Q2 sits at 40%, versus 45% in Q1, which obviously relates to chips that were sold in Q4. That's a seasonality that we would typically see, Q1 being slower for smartphone shipments than off of Q4. Obviously during Q1, there was a lot of pessimism and discussion about supply chain inventory. I think it's reasonable to say that that effect is seen in Q2. Can't say anything about the future because that would be a forward-looking statement, as we said, we won't be making any of those today. Similarly with servers. Very pleased with the progress that we've demonstrated to date. Very pleased with the deals that we've closed and announced in Q2.
Beyond that, can't go into any more detail.
Thanks, all the best.
Thank you. Your following question comes from the line of Andrew Gardiner from Barclays. Please go ahead.
Good morning, guys. Yes, sort of congrats from my side as well. Certainly Arm will be missed within the coverage. Just within the what you can comment in terms of the quarter and how you're seeing the smartphone space, I'm just interested in hearing your thoughts on the level of content gain or feature gain in terms of v8 and multi-core and Mali. Certainly if you look at v8, it seems to have flattened out at least quarter-on-quarter. I presume it's sort of seasonal, but I'm just interested in your thoughts on how the licensees are progressing there and whether you're seeing any slowing in terms of Apple's processors through mid and low end of the smartphone market. Thanks.
Well, again, focusing on all of that around Q2. If we look back at the growth of V8A into smartphones, that was about 30%, or that was 30% a year ago. It grew to 50% in Q4. Q1 was at 65%. Q2, as we just reported, 65%. Seeing a growth in octa-core in that time period. Q2 2015, octa-core was at 5% penetration, up to 30% in Q2. Now, as I just said, Q1 was a slower quarter for smartphones. I don't think you can read too much into the sequential changes in those numbers from Q4. Well, our Q1 to our Q2, obviously market Q4 to market Q1. Time will tell, again to make any further comment on that, I'd be talking about the future.
In terms of general appetites to upgrade, I think the Q2 licensing reflects continued demand for our technologies across the different categories of our product lineup. We're very pleased with the breadth of licensing in Q2, and that would seem to continue our historic trend.
Okay. Thank you. Best wishes for the future.
Thanks, Andrew.
Thank you. The following question comes from the line of Kai Korschelt from Merrill Lynch. Please go ahead.
Yes, good morning, gents. Again, congratulations on getting a good deal for the company. My first question was really a bit broader, and it was particularly around IoT. We've been talking about it for some years. What do you think so far has been holding back this sort of breakthrough adoption, and what do you think can or could be a catalyst for that sort of product category to take off? Then just a quick follow-up on the tax rate, it was very low. Were there any sort of unusual effects in the second quarter? Thank you.
As you say, we've been talking about IoT for a long time. We believe building or creating the building blocks that will be required for IoT systems. Cortex-M is a big part of that, and you've seen in the quarter very strong licensing of Cortex-M again. Our Mbed platform that we've discussed for some time has continued to make good progress in terms of delivering against the objectives that we set out. We believe that these are key technologies that will help IoT become a reality. What we've seen up till now is people developing IoT platforms in a very vertically oriented way. I think that's an understandable step for the industry to be making at this stage as people really work out what the benefits of IoT are going to be. Up till now, I think we have made good progress.
How this proceeds from here is again, a projection of the future. I'm not going to make one of those. I think the work that Arm has done has been the right things for us to have done over the last couple of years to prepare for this market. On the tax rate, it is unusually low. It's 6%. As I said, underlying we'd expect that to be around 15%. That's because in the quarter we released a tax provision which actually goes back over 10 years. We've concluded we no longer need it following some recent correspondence with the tax authorities.
Okay. Thank you. Best of luck.
Thank you.
Yep. Your next question comes from the line of Lee Simpson from Stifel. Please go ahead.
Great. Thanks for letting me on, and congratulations to all on such a strong deal.
Thank you.
Couple of questions if I could. Just trying to understand, maybe you can or can't answer this one. Could you maybe outline again for us what is the nature of the relationship with ThunderSoft? I mean, is this an arm's length investment, or do you see something more strategic developing from that? Maybe alongside that, a couple of quick ones. 1,695, the number of U.K. employees, does that now include Apical? Could you just remind me when the scheme documents actually release to investors?
Multiple questions in there. Well done for not following the rules. Very quickly, 1,695 U.K. employees includes Apical. The question about ThunderSoft, we made a completely arm's length investment in them a long time ago. They went public, I think it was last year. We still hold shares in ThunderSoft. From an investment point of view, it's just completely arm's length. There's no overlap in the way that they're managing their business or we're managing ours. They are a partner of ours in our accelerator program in China. We have a joint venture with ThunderSoft that has set up a number of accelerators or incubators around China, which is bringing in entrepreneurial startup companies into the Arm ecosystem. We're very pleased about that going. Schedule document, can't put a concrete date on that. Sometime over the next couple of weeks, that will be published.
Great. Thanks, guys, and all the best.
Thank you.
Thank you. Next question comes from the line of Jagheer Bajwa from Arete Research. Please go ahead.
Hi, guys. Thanks for taking my question. The first one, can I get a clarification on that terms number that you gave during the prepared remarks? Is that okay?
Yes. 50% of the license revenue came from backlog and 50% from terms.
Okay, great. Then just for the question, despite the big architectural deal you did in the quarter, backlog down, and the maturity of your licensing now, did you recognize more from the architectural deal in the P&L this quarter than you kind of usually have done historically? Maybe could you give us a rough estimate of how much of that architectural deal went into the P&L this quarter? Thanks.
We don't disclose details about individual licenses, but it's a rules-driven process. The amount we recognize in the quarter is to do with the IP that that license involves. If it's mature IP that the customer's capable of using, then we take that to revenue. If it's for future delivery, then that goes into backlog and is released when that IP is delivered. There is no difference in the handling of that license than any other similar license.
Okay. Thank you very much. Cheers.
Thank you.
Yep.
Your next question comes from the line of Douglas Smith from Agency Partners. Please go ahead.
Yeah, good morning. I was wondering if you could talk a little bit about what you're seeing in the automobile market. I know it's one you've always said you had a lot of hopes for. Many other people seem to think it's very strong right now. Could you comment particularly, are you seeing strength, and is it navigation, infotainment, self-driving capabilities?
Well, Arm has historically played a role in automotive electronics. Our processors have been designed into many different parts of the car over the years. We've been in airbags and ABS systems and infotainment systems within cars. Over the last few years, as interest around autonomous driving has grown and as the sophistication of driver assistance systems has grown, we have seen increased demand for embedded processing. Many of our partners have been working on products based on our technology to go and address that space. Clearly, the trends are towards a greater silicon content in cars. We've talked about that before. How it plays out, how much market share Arm gains, not in a position to talk about today, as that's a discussion about the future.
As I said, we've just seen over the last few years, an increasing amount of semiconductor content, an increasing amount of processor-based semiconductor content going into cars.
Okay. Thanks a lot for all that, and it's been a pleasure working with the management and the investor relations team as well.
Thank you very much.
Thank you. Ladies and gentlemen, as a reminder, if you wish to ask a question, please press star and one on your telephone keypad and wait for your name to be announced. If you wish to cancel your request, please press the hash key. Your next question comes from the line of Johannes Schaller from Deutsche Bank. Please go ahead.
Thanks, congratulations on such an amazing deal from my side as well. Just two questions around IoT, given that we heard about that so much. I know you can't talk about things going forward, but could you give us a sense just where, in your view, IoT related royalty revenue stood in the second quarter and how that grew? I know it's probably not a lot, and there's some definition to that, but maybe you can give us a bit of a feeling. Then also, if you just look at the deals you have so far, would you imagine most of your IoT related deals to just get you a percentage royalty rate on the silicon, like most of your other deals? Or is there maybe also something where you can derive more of the value of the services, et cetera, that are related to IoT?
If you can give us a bit of a feeling on that would be great. Thank you.
We don't break out royalty dollars between the different segments. Again, if you look at the table that's on page seven, you can see that 42% of the volume in Q2 went into embedded. That's a very broad category. That can be all manner of devices. Some of those are connected IoT-like systems, and some of them are more, well, unconnected control applications, controlling the motor in a dishwasher, for example, a standalone black box type application. Embedded, very broad space, 42% of 3.6 billion chips. I can't do that math in my head, but that's a large number of chips. How many of those are connected, how many of those aren't? We don't have the breakdown here.
In terms of today's business and today's business model, the IoT opportunity, or, sorry, not IoT opportunity, revenues from the IoT space are about the percentage royalty on the chips that contain Cortex-Ms that are going into these connected devices. Again, whether or not that changes over time, we'll be talking about the future, so I better not speculate on that.
That's helpful. Thank you very much, Simon. All the best.
Thank you.
Thank you very much. There are no further questions. Please continue. Thank you.
Well, if there are no further questions, then thank you very much for joining us today. We'll call that a day. Thank you very much. Goodbye.
Thank you. That does conclude the conference for today. Thank you all for participating. You may all disconnect. Have a good rest of the day.