Intel has recently been facing fierce competition from their rival AMD in the processor department. It all started with Zen and continues with Zen and while there seems to be no end to it, it looks like Intel is very actively accessing AMD's advances in the processor and process industry and trying to find ways to counter them.
Intel's "AMD Competitive Profile" Internal Employee-Only Post Revealed- Talks On "Where we (Intel) go toe-to-toe, why they (AMD) are resurgent, which chips of ours beat theirs"
In the article which is posted on Reddit, Intel's Circuit News Managing Editor, Walden Kirsch, talks about AMD in general and how they both have been engaged in the tech industry's single longest-running business rivalry. It's absolutely true that both companies have been competing for many years but a few years ago, Intel didn't even saw AMD as their main competition anymore. But this post drills down on why that perception changed on the blue side.
Intel says that they are 10 times the size of AMD, in terms of revenue, but they also face the largest and most competitive threat from AMD. The article starts off with the main heading "AMD is now a formidable competitor". Intel is looking into every aspect of AMD's business and looking at the key indicators for their success story. Intel also believes that the main competitive threats from AMD are their high-end products. Following are some of the competitive threats as summarized by the Intel Performance, Power, Competitive Analysis Team.
Key AMD Competitive Threats To Intel Are From High-End Products
- AMD offers high-performance CPUs, posing direct competition to Intel in both our client and datacenter CPU businesses. With our announced ambitions to bring new discrete graphics to market, we are bringing new competition to both AMD's and NVIDIA graphics businesses.
- AMD has recently been gaining some traction in winning public cloud offerings. And competition from AMD is shaping up to be especially tough in high-performance computing. HPC performance is usually driven by the number of cores and the number of memory channels (or memory bandwidth). Intel is challenged on both fronts.
- AMD's upcoming next-generation Zen core products, codenamed Rome for servers and Mattise for desktop, will intensify our desktop and especially server competition. The latter is likely to be the most intense in about a decade.
- Outside of desktop and servers, Intel's competitive position in notebooks and business PCs is stronger as customers value specific aspects such as productivity performance, battery life, and overall manageability where Intel has clear advantages versus the competition.
- By leveraging TSMC's 7nm manufacturing, AMD no longer manufactures its own chips. AMD can drive higher core counts and higher performance than it could previously with Global Foundries as its in-house manufacturer. These 7nm products will amplify the near-term competitive challenge from AMD.
Now I would agree with most of the points that Intel has shared here and it's an odd sight that Intel is being so clear about their competitions successes in each respective field. AMD's Ryzen Threadripper and Ryzen processors have gained huge ground for team red in the two years since their release. They have managed to increase market share in both desktop & notebook markets and considering that the next best series is on the verge of launch, AMD can expect even more share to rise in their favor. At the same time, Intel's own processors are faced with supply issues which their CEO, Bob Swan, stated won't get better till Q4 of 2019.
Sure Intel has the fastest CPUs for mainstream and HEDT users but with Ryzen 3000 series launch just two weeks from now, we can expect even that lead to disappear. The Ryzen Threadripper CPU lineup has almost completely decimated Intel's HEDT offerings by offering insanely high core count at disruptive prices. Intel even agrees with this much as they say "The latter (Mattise) is likely to be the most intense in about a decade".
Then comes the server department which has also seen major share gain since EPYC launched. AMD's EPYC servers are gaining huge demand in the HPC and cloud/data center market. Recent deals such as Azure (Microsoft), AWS (Amazon) Frontier Supercomputer (ORNL), SHASTA (Cray), XH2000 (AtoS), Oracle have put AMD's EPYC chips in the limelight.
In addition to that, AMD's Rome server lineup which is being highlighted by Intel as a major threat to their Data Center Group offerings which includes Xeon class products. AMD is bringing the chip competition both at number of cores and memory bandwidth that they offer.
AMD EPYC Rome Server Processors – Here’s What To Expect
As for the EPYC Rome processors, AMD has confirmed that they are aiming a launch in Q3 2019 which should be a few months apart from the Ryzen and the Ryzen Threadripper processors. The AMD EPYC Rome processor family is expected to lift AMD's server CPU market share to 10% by 2020 which is a great deal considering Intel's ex-CEO, Brian Krzanich, had stated that they don't want AMD capturing 15% market share but given the demand and adoption of EPYC processors in major server platforms, 15% shouldn't be too far from now.
Mr. Krzanich was very matter-of-fact in saying that Intel would lose server share to AMD in the second half of the year (2018). This wasn’t new news, but we thought it was interesting that Mr. Krzanich did not draw a firm line in the sand as it relates to AMD’s potential gains in servers; he only indicated that it was Intel’s job to not let AMD capture 15-20% market share.” – Romit Shah, Nomura Instinet
Just for number's sake, Dell EMC has announced that they will be tripling their AMD server offering by adopting more of the EPYC range of processors.
"Out of, let's say, 50 or so platforms that we have today," he said, "three of them are AMD - we'll probably triple that by the end of this year."
He also confirmed that Dell EMC will be launching servers powered by AMD's newest architecture - a 7nm architecture codenamed 'Rome' - in the second half of 2019.
- Dominique Vanhamme (DELL EMEA vice president and general manager for storage and compute)
Based on such strong growth figures and adoption rate, we can expect AMD to give major blows to Intel's Xeon efforts and their server side of operations. We should expect up to 64 cores and 128 threads along with impressive PCIe Gen 4 connectivity with up to 162 lanes as summarized here.
It should also be pointed out that when AMD was designing their 7nm Zen 2 based EPYC Rome processors, they had internally estimated what the performance of Intel's next-gen server part would be like.
The next-gen 10nm part known as Ice Lake-SP is scheduled to launch for 2020 with Cascade Lake-SP and Cooper Lake-SP being offered as an intermediary solution based on 14nm (++) while the Cascade Lake-AP and Cooper Lake-AP would be designed as a multi-core HPC part.
“Rome was designed to compete favorably with “Ice Lake” Xeons, but it is not going to be competing against that chip. We are incredibly excited, and it is all coming together at one point.” – Forrest Norrod.
“Our plan for the Naples-Rome-Milan roadmap was based on assumptions around Intel’s roadmap and our estimation of what would we do if we were Intel,” Norrod continues.
“We thought deeply about what they are like, what they are not like, what their culture is and what their likely reactions are, and we planned against a very aggressive Intel roadmap, and I really Rome and Milan and what is after them against what we thought Intel could do. And then, we come to find out that they can’t do what we thought they might be able to. And so, we have an incredible opportunity
AMD confirmed that their EPYC Rome processors have been designed to compete favorably against Intel's Ice Lake-SP parts. This only means that AMD would have an even greater edge versus the Intel 14nm++ server parts arriving this year.
One of the biggest advantage that EPYC Rome processors will have over Intel parts is that they will be socket compatible with EPYC Naples so all of those vendors who have been using Naples would get drop-in compatibility for AMD's next-gen 7nm EPYC Rome processors on day one.
AMD looks to be in a really good position with their EPYC server processors, even more so than their desktop and mobility portfolios. If everything runs smoothly for AMD and their long-term Zen roadmap in the years to come, we can see them dominating all sectors of the CPU market again.
A Challenging Period Ahead For Intel And A Dominating Position For AMD
The article further goes on into a one-on-one Q/A session with the director of Intel's Data-centric & AMD Competitive Assessment Group. Steve says that "Today and into the future, we will be facing tough competitive challenges". There are a few key points on how Intel's products compare to the AMD lineup as mentioned by Steve and listed below:
- Intel 9th Generation Core processors are likely to lead AMD's Ryzen-based products on lightly threaded productivity benchmarks as well as many gaming benchmarks. For multi-threaded workloads, such as heavy content creation workloads, AMD's Mattise is expected to lead.
- In the longstanding industry debate over benchmarks, whose to use? Cinebench is often used by AMD since it favors high core/thread count and represents one of the best-case benchmarks for AMD. Intel believes that Cinebench is not a representative benchmark for general platform task evaluations and real-life workloads. Intel continues to work with the press on using real applications for evaluation performance.
- In general, Intel's mainstream Xeon server products will be challenged on throughput-oriented benchmarks that scale well with core count. Architecturally, AMD's Rome product for a server is improved over 1st generation EPYC, but Xeon is still expected to have cache and memory latency advantages. For this reason, Intel sill expects Xeon to be competitive on applications that require fast response times and are sensitive to memory latencies like database, analytics, web serving and so on.
About the benchmarks, Intel tried to assert that the benchmarks that most reviewers are using right now are artificial and not indicative of real-world use cases. In their pre-Computex session, Intel called Cinebench an artificial benchmark which no one really uses aside from comparison purposes and that reviewers should running applications such as games and general office tasks that are more important to the user.
I would say it's true to one extent but reviewers right now are already including those in the tests and benchmark software such as Cinebench, 3DMark, etc are widely popular amongst the DIY builders which make up a large majority of PC users. Even Intel themselves had occasionally been using the so-called *artificial* benchmarks to show off the world-class performance of their top-tier chips and suddenly because the competition had eclipsed them in performance, the benchmarks don't matter anymore. I found PCWorld's discussion on the top really interesting on this and you should check it below:
Similarly, in a Q/A session with AMD's CEO, Lisa Su, PCWorld's Senior Editor, Mark Hachman asked if *artificial* benchmarks still matter. Following was Lisa Su's response:
Su: We also believe that real-world applications are what’s important, no doubt about it. But at some point, you’ve got to compare X to Y. So we will use benchmarks, we do. You might have noticed that we switched from Cinebench R15 to R20. We did that on purpose because we thought it was a harder test, frankly than R15.
When we look at gaming performance, we do our very best to benchmark. All of our stuff is apples to apples, and we’ll continue to do that.
Performance Per Dollar, TSMC or Renewed Focus on High-End Products, What Is The Key To AMD's Success?
All of them to be very honest. Moving from GloFo to TSMC has shown great results for AMD. They have managed to improve their performance per core and in terms of price, they are positioned highly competitive. AMD will be offering a 12 core, 24 thread chip with 4.6 GHz clock speeds for a price less than Intel's 8 core, 16 thread Core i9-9900K processor.
"AMD made improvements in their 2nd generation Zen core and their disaggregated chiplet-based architecture scales cores efficiently. Therefore, on workloads that are heavily threaded, including heavy content creation and most server workloads, they'll get great performance results. And on price, we expect their pricing to be significantly below ours. So they'll likely get good performance-per-dollar. That's what they're going to compete on, and that's the risk to Intel."
Steve also points out that Intel offers a much larger selection of processors to choose from compared to AMD. They start off with processors aiming the entry tier and go all the way to flagship tiers. But at the same time, he mentions that people don't buy a chip, they buy a complete system and that's where Intel says they have the lead, in offering a complete system with unmatched validation, software, and security (although the recent security vulnerabilities have taken a huge toll on Intel CPUs).
Also, saying that only OEM/ODM buys a chip, Intel may be dismissing the entire DIY market who do buy a single chip and the components for that single chip to build their PCs. At the end of the day, the DIY PC market has performed consistently well compared to the entire PC market.
AMD's Success Story According To Intel Themselves:
From 2006 to 2017, AMD had positive net income only three of the twelve years. I'm not sure we can point to a single thing that turned AMD around. But I do think it was absolutely rooted in the strategic changes AMD initiated in 2015/2016 that narrowed and simplified their focus. AMD shifted to focus on higher margin in premium segments, specifically high-end client, datacenter, graphics for gaming. And the continued their investment in their semi-custom and console business.
Rather than going after lower-margin, low-end products, they refocused on how to win higher-margin business. AMD added much-needed clarity since they were previously distracted by a market that didn't align with their strengths. They simplified their investments and roadmap and started leveraging best in class foundries. Most importantly, they executed that strategy. Having a clear focus and direction helps enable great execution.
I also believe AMD's comeback was a result of being very product-centric. A top priority for AMD was building great products, higher performance Compute and graphics solutions, from definition to development to delivery. - Steve Collins (Director of Intel's Data-centric & AMD Competitive Assessment Group)
In the end, this article is very to-the-point on where the industry is heading and the challenges that Intel faces. Intel is very reasonable in their assessment on the threats it faces and how it plans to tackle them in years to come. Just like in the past, Intel has overcome the challenges it faced by offering top-tier products and they have a solid foundation in the form of their six pillars (process, architecture, memory, interconnect, security and software). AMD rose from the ashes to once again become a dominant force in the industry and Intel could do the same if we take a look at their past, that's the beauty of competition!