AMD EVP Henri Richard
This is Part I of two-part interview.
***
Q: If you look at the figures that have been given, the ATI acquisition is being funded by the US$2.5 billion loan, plus cash and cash equivalents and short-term investment balances of approximately US$3.0 billion. That’s total funding of US$5.5 billion. That covers a basic acquisition valued at US$5.4 billion. I think the suspicion, here, among the general public and among your shareholders, might be that AMD is taking too much of a risk with this acquisition, that you’re betting the farm on a deal that might be very questionable. A: I don’t think we have the feeling that we’re betting the farm. We’re very enthusiastic about this transaction because we think that it’s a necessary step in our quest to break the monopoly. Having the opportunity to innovate, in an equal way, at the platform level, more so than we could in the past, is certainly part of our strategy. As well, we have always been recognizing that there’s a certain amount of balance needed in system performance, between all the various components. Now we have two of the key components, two of the key technologies, the CPU and the GPU, and we’re going to be able to continue to move forward in terms of innovation to balance out, for the best interests of our end users and customers, these two key technologies. That’s something we couldn’t do before. So there’s no trepidation from our side, and certainly not from the management team, in terms of, “Is this a transaction we can’t afford?” If we had felt that way, we wouldn’t have done it.
Now it’s clear that recently the market has been taking a beating, and not only our stock. That said, to a certain degree, we’ll pass. We’ve always known this was a marathon. You don’t go and break a monopoly the size of Intel in a few short quarters, and Hector [AMD President and CEO Hector Ruiz] is keeping his eyes focused on the long-term strategy, which is to be able to offer two alternatives to the industry, and I think this takes us one more step in that direction. In fact, for the first time this positions AMD in the marketplace with solutions that Intel doesn’t have, because one of the very exciting aspects of the ATI product portfolio is the technology they have in the digital consumer space. So not only are we becoming a more potent PC player, and to a certain degree we’re doubling down on the investment for the PC marketplace, we’re also now adding a certain number of very key technologies in the consumer-electronics space, both in handsets and digital TV, and that’s pretty exciting. ATI’s got a very large market share in cell phones, and ATI is also a very important player in digital TV.
Q: So, obviously, part of the acquisition strategy is to leverage ATI’s technology in both those segments?
A: Well, frankly, these are segments where we’re not present, so for us they are new markets, and clearly, with the deep knowledge and respect that ATI has in those markets, it opens up new opportunities for AMD by essentially using some of our IP and getting some of it applied to some of those particular segments.
Q: So are you saying that handsets are a new segment for AMD?
A: Through this acquisition, absolutely. We had absolutely no presence in that market. Previously we had a presence in the handset market through Spansion and our flash business, but after the separation from Spansion, we basically had lost the relationship or the reason to establish a relationship in handsets.
Q: So what exactly is it that you hope to gain from ATI, in term of handsets?
A: Well, if you think about the evolution of the handset market, there’s an increasing body of evidence that new products, like the UMPC and portable devices that are neither a small notebook nor a smartphone, seem to be one of the hot topics of conversation these days, particularly here in Taiwan, and I think that for us, re-establishing a very strong partnership, through this ATI acquisition, with companies such as Nokia or Motorola, is an interesting element. If you start to think about the future of mobile computing, there’s now the possibility that some of those key industry market makers might decide to enter that space.
Q: Of course, digital TV might well become a feature of the handset market.
A: Well, I look at digital TV from a different angle. If you think about it, if you have a flat-panel TV, it doesn’t take much to turn it into a media center, so there again there could be some opportunities, as we leverage the very very deep and wide relationship that ATI has established in that marketplace, to listen to customers and see in what direction we could innovate to service their needs.
As you know, with our AMD Live! initiative, we took a very different approach to that of Intel. We’re not trying to force everyone into a single bottle. We’re trying to listen to customers and what they think they really need in the living room, and frankly, if it was the case that some large flat-panel manufacturers decided they should add PC functionality to their flat panels, that could be an interesting opportunity for AMD or the new company.
As well, I was talking with some customers today, and there are a lot of new ideas, where the phone could become more linked in the home, becoming a portable display, for example, so there are a lot of things going on, particularly as VoIP becomes more pervasive. In this kind of scenario, by default, phones will have some sort of IP based connection with your home network, and that in itself presents a lot of interesting opportunities for us.
Q: In the past AMD has always been closely associated with the name Nvidia, and it’s always been assumed that AMD has a very positive relationship with Nvidia, if not an actual partnership. We’ve heard terms such as SNAP, the strategic Nvidia-AMD partnership. What are the implications for your relationship with Nvidia, now that AMD has made this move to acquire a company that has always been a competitor with Nvidia?
A: Well, first, we have to remember the principle under which we’re going to continue to operate our joint business, which is one of open platforms and open standards. There’s nothing in our strategy that, unlike our competitor, is going to be aimed at hurting or impairing the ability of other folk to innovate on our platform. As such, I expect our relationship with Nvidia to continue to be one that is both a partnership and a strategic relationship because again we respect the innovation that Nvidia brings to the marketplace, and we certainly want them to feel that if they invest some of that capability for innovation on the AMD platform, there’s going to be a just return on that investment. So those principles that have helped build AMD are not going to change.
Q: So can we continue to anticipate AMD based motherboards that carry Nvidia graphics and Nvidia core logic?
A: I surely hope so, absolutely. I come as you know from a background at IBM, and one of the principles I learned there, and I think this is very important, is that if internal divisions are to remain competitive, you should not shield them from external competition. I think that’s one very important aspect of competition. So, although I would expect that our internal chipset division would provide us with very high quality, very high performance chipsets, I know for a fact that artificially shielding them from external competition would probably be bad business practice.
From time to time it may happen that we will have the best chipset for a particular application, and from time to time it may happen that Nvidia has the best chipset. And if that’s the case, there’s absolutely no reason why they shouldn’t reap the benefit of their innovation. We’re certainly going to maintain an open platform strategy, so that they can innovate on the AMD platform. Also, unlike the competition, I don’t foresee putting in place artificial marketing barriers to force our customers to make certain choices that are not necessarily good for their business.
Q: In the same context, AMD has always had a strong relationship with VIA Technologies. SiS and several other companies could probably be mentioned as well. Certainly VIA has always been regarded as a key provider of core logic for AMD, so do you have any comments on AMD’s future relationships with these third-party providers?
A: Well, all that I’ve just said about Nvidia would also be true for VIA, SiS and any other partner. It’s interesting to note that those three companies have been providing chipsets for the Intel eco-system for years, and I think that everyone would agree with me that working under the Intel umbrella is a lot more difficult than working with AMD. So these companies have been able to produce Intel based chipsets for years and finding a way to survive and thrive in that environment, so I don’t see why they wouldn’t be able to do it in the AMD eco-system.
Again, I can’t make comments for them. Some of them may decide that instead of making a broad offering, they’re going to focus on specific segments. I’m sure they’ll be looking at their strategies. But across the board, we’re going to maintain a very open platform and an open-standard approach to partnership with those three companies.
Q: Can you say what you hope to gain from the acquisition in terms of some specific components and functionality; for example, standalone graphics processors, an integrated graphics capability, gaming platforms and system-on-chip (SoC) solutions?
A: From time to time, over the past few years, we were in certain situations where a customer would have an idea, or would propose a level of collaboration or integration between the various components – for example, a chipset integrated with a CPU – and although we have great quality partnerships in the industry, issues can always arise. That’s not necessarily having an opposite view on a particular solution, but barriers can get in the way of implementing the most efficient solutions, simply because different companies have different priorities at different times. So one of the things that the ATI acquisition is going to allow us do, now that we have all the components of the puzzle, is listen to customers and really figure out how we can bring them innovation that is really meaningful to their business and that will allow them to differentiate and extend their business.
I think that this lack of differentiation, and this lack of expansion of their business might well be what has been hurting the industry the most, mainly because, again, our competitor has a tendency to want to normalize everything, rather than allow differentiation. So clearly we expect to continue to push the envelope on GPUs and provide ever increasing performance and quality and visualization of graphics. We believe there are opportunities to bring very high quality integrated graphics solutions to the marketplace, and I think that it’s particularly important to have a high quality cost-competitive solution in integrated UMA chipsets and gaming platforms for sure.
AMD has always been near and dear to the hearts of gamers. We can now think about innovating at that level, probably with our 4x4 enthusiast platforms. You’ll see some interesting and potential development around some of the IP that came with the ATI acquisition, and so forth. ATI has a relationship with Microsoft around the Xbox 360, and there are some interesting things there, I’m sure, for the future. There’s a vast array of opportunities for us to go and innovate and figure out how to balance performance and cost for the best interests of the end user. Basically, that means innovating without the artificial barriers that exist when you have separate companies, across the various elements.
In the long run, you can debate whether it’s at 45 nanometers or 32 nanometers or 22 nanometers, but at some point we’re going to have on-die a transistor budget that could probably support a further level of integration. In a multicore system on a chip, perhaps the CPU and GPU could be on the same die, for example. That’s not for the short-term. That requires a lot of R&D and development, but it’s clearly a possibility. That is of course something that a lot of customers will be asking us about, and that we’ll be starting to think about.
As you know, we’ve also just announced our Torrenza initiative, which is a multi-socket open architecture. I haven’t looked yet at what this could mean in terms of some of the IP that comes with the ATI acquisition, but clearly we are moving as quickly as we can in trying to offer this open-platform, open-socket, architecture, potentially licensing Coherent HyperTransport. Who knows what’s coming with all the talent and all the IP that’s part of the ATI acquisition? Maybe there are opportunities there for us to accelerate the Torrenza initiative and do some interesting things with some partners.
Q: How about systems-on-chip, as such, which would imply not only a processor and graphics capability but also memory, and so on.
A: Again there, we’re going to go back to the UMPC type of space. This is a very interesting space. With the 50x15 initiative, we’re very very close to all those new types of devices that require a system-on-chip capability, whether they’re for the ultra-mobility space, whether they’re for the ultra-low-cost space, and there again we believe that by having the complete set of tools and building blocks, we’ll be able to bring a level of innovation at the complete system level that we couldn’t when we were just AMD as a CPU company. It’s too early for me to disclose anything. We’ve got to go and do some hard work now, but certainly we know there are a lot of great opportunities to take some of the great technology that ATI develops for markets that are adjacent to the PC market and see if by combining some of those technologies with what we know about the PC market, we can create a new paradigm or a new level of performance at power and cost envelopes that have not been achieved so far.
Q: Could you concretize a little more how that might work out in terms of specific devices. You mentioned 50x15, for example.
A: Well, if you look at the context of 50x15, and if you look today, for example, at a very high-end multimedia phone, one could argue that one approach is a UMPC device that could provide Internet access to 50% of the population in 2015. To achieve that, you could start from a PC and try to cost reduce it, reduce its size down to that level of device, or you could start from a multimedia phone and try to up-size it, slightly, to arrive at the level of functionality you want in that device.
That’s one of the areas, clearly, where we’re going to ask our engineers to go think about how they can take the best technologies from both parties and innovate and provide a device that will attract new customers.
***
sa: http://www.digitimes.com/news/a20060725VL204.html
Linear Mode
21 OCZ Fans!
