AMD Shares ZEN info and talks about based 32 core 64 thread product
AMD today shared a bit more info in which they talk about Zen, as well as Zen server solutions, yes a 32-core / 64 threads ZEN based CPU developed under code-name Naples is on target to be released in Q2 2017 and thus now confirmed as well. AMD also shares some details on the AMD AM4 motherboard platform.
As you guys know the The new “Zen” architecture is designed to scale across all of AMD’s CPU business, from enthusiast desktop platforms to enterprise-class servers and notebooks, to embedded and semi-custom products. The initial “Zen” CPU core is stated to deliver more than 40 percent improvement in instructions per clock cycle over the previous generation cores and will come to market first in an 8-core, 16-thread system-on-chip for desktops (=Summit Ridge). For the new architecture AMD focused on three key areas when designing this special architecture:
- Performance of the engine itself with completely new branch prediction, introduction of a micro-op cache and a much wider instruction window;
- Throughput, to keep that high-performance engine fed with data and instructions out of memory through pre-fetching and a completely new cache memory hierarchy with 8 MB of L3 cache; and finally,
- Efficiency having performance and throughput without increasing power, by leveraging a 14nm FinFET process and a wealth of power saving design techniques in the architecture.
Over at IDF AMD is showing ZEN performance demos where Summit Ridge (8c/16t) is competing with Broadwell-E.
In the video above you can see some impressive performance (video at 1:30), AMD Zen 8-core "Summit Ridge" versus 8-core Core i7-6900K with Blender 3D.
For us performance consumers Summit Ridge is likely going to be the processor that appeals the most with it's 8 cores and 16 threads. Bristol ridge will become the 4 core 8 threads processor series. Expectations are high and Summit Ridge may prove to be a make or break product for AMD. The Zen architecture will be built on a more efficient 14 nanometer FinFET process at GloFo, rather than the 32 nm and 28 nm processes of previous AMD FX CPUs and AMD APUs, respectively. Four variants of ES Zen are actually already spotted in the industry:
- AM4 8 cores with 95W TDP (Summit Ridge)
- AM4 4 cores with 65W TDP (Bristol Ridge)
- SP3 24 cores with 150W TDP
- SP3 32 cores with 180W TDP (Naples)
The "Summit Ridge" Zen family will feature a unified AM4 socket with its GPU-equipped "Bristol Ridge" APU counterparts, and feature DDR4 support and a an expected 95W TDP. Newer roadmaps don't confirm the TDP for desktop products, they suggest a range for low-power mobile products with up to two Zen cores from 5 to 15W and 15 to 35W for performance-oriented mobile products with up to four Zen cores. Though unconfirmed we expect each Zen core will have four integer units, two address generation units and four floating point units, and the decoder can decode four instructions per clock cycle. L1 data cache size is 32 KiB and L2 cache size 512 KiB per core.
Todays press release:
AMD Demonstrates Breakthrough Performance of Next-Generation “Zen” Processor Core
Performance preview of “Summit Ridge” desktop processor demonstrates AMD achieved 40% Instruction Per Clock improvement; AMD unveils first demonstration of 32-core “Naples” server processor —
SAN FRANCISCO — Aug. 18, 2016 — At an event last night in San Francisco, AMD (NASDAQ: AMD) provided additional architectural details and a first look at the performance of its next-generation, high-performance “Zen” processor core. AMD demonstrated the “Zen” core achieving a 40% generational improvement in instructions per clock, delivering a landmark increase in processor performance. During the event, AMD demonstrated an 8-core, 16-thread “Summit Ridge” desktop processor (featuring AMD’s “Zen” core) outperforming a similarly configured 8-core, 16-thread Intel “Broadwell-E” processor when running the multi-threaded Blender rendering software with both CPUs set to the same clock speed.
AMD also conducted the first public demonstration of its upcoming 32-core, 64-thread “Zen”-based server processor, codenamed “Naples,” in a dual processor server running the Windows Server operating system. “The performance and efficiency of our ‘Zen’ core showcases AMD at its best,” said Dr. Lisa Su, president and CEO of AMD. “Over the last four years we have made significant investments to develop a high-performance, multi-generation CPU roadmap that will power leadership products. Customer excitement for ‘Zen’ continues to grow as we make significant progress towards the launch of new products that will span from the datacenter to high-end PCs.” The “Zen” processor core features multiple architectural advances designed to increase the performance, throughput, and efficiency of AMD’s future products. “Zen” is based on a cleansheet design and features a new cache hierarchy, improved branch prediction and simultaneous multithreading (SMT). These advances will allow the “Zen” core to scale to meet the needs of a broad range of applications, including fanless 2-in-1s, embedded systems, high-performance computing, and the datacenter.
“An engineer may get one chance in their career to work on a project of this size and scope, and maybe never one with as much potential to impact the future as much as ‘Zen,’” said Mark Papermaster, senior vice president and chief technology officer at AMD. “With ‘Zen’ we aim to do what many never thought possible – deliver a 40 percent generational improvement in instructions per clock while maintaining power requirements in line with our previous generation technology.2”
“AMD invested where it counts, with an x86 core that can scale from PCs to highperformance servers,” said Linley Gwennap, principal analyst, Linley Group. “Consumers today expect to get the most out of their systems to create transformative experiences. The versatile design of ‘Zen’ delivers highly-efficient performance that should provide increased computing capabilities across the spectrum – from devices to cloud computing.”
Expected to launch first, the “Zen”-based “Summit Ridge” desktops will utilize the AMD AM4 socket, a new unified socket infrastructure that is compatible with 7th Generation AMD A-Series desktop processors – previously codenamed “Bristol Ridge” – for exceptional performance and connectivity scalability required by AMD partners and customers. The first desktop systems featuring 7th Generation AMD A-Series processors and new AM4 sockets are scheduled to ship in the second half of 2016 in OEM PC designs.
With dedicated PCIe lanes for cutting-edge USB, graphics, data and other I/O, the AMD AM4 platform will not steal lanes from other devices and components. This allows users to enjoy systems with improved responsiveness and benefit from future-ready technologies that the AM4 platform provides with a powerful, scalable and reliable computing solution.
AMD AM4 platform key technology features include:
- DDR4 Memory
- PCIe Gen 3
- USB 3.1 Gen2 10Gbps
- NVMe
- SATA Express
Additional “Zen” architectural features will be detailed next week in a presentation at Hot Chips 28. I have made a selection of slides on this presentation that you can find below (click thumbnails).
AMD shares details on 8-core Seattle ARM architecture - 08/13/2014 08:23 AM
AMD released some details about their upcoming eight-core ARM Cortex-A57 based server processor developped under codename Seattle. Seattle's eight 64-bit Cortex-A57 cores are arranged in four dual-...
AMD shareholders approve spinoff operations - 02/19/2009 12:12 PM
A good step forward for AMD as shareholders of chipmaker Advanced Micro Devices (AMD) on Wednesday approved a spinoff of the company's manufacturing operations. The move, approved by 94 percent of the...
Senior Member
Posts: 10562
Joined: 2006-02-14
Since we don't know the OC capabilities of these chips, how can you figure that it's the opposite? If an Intel ES sample is stock low, but was somehow gotten by someone who would OC it and publish those numbers, how does that make it any different from what we are at now, other then someone not willing to do the latter?
I thought I saw later engineering samples of Haswell clocked at the same thing as the final product (4770K) at 3.5GHz. I didn't even hear of this 2.8GHz sample. Hope that clarifies what I was saying.
Senior Member
Posts: 6070
Joined: 2011-01-02
If AMD gonna show only software where AMD fully utilized like 480 crossfire beating 1080 in AoS, yeah, thats not false. Are we to expect a good product?
AMD shows people number and make people hype about it, but once proper benchmarking on common software is done, its suddenly calm.
GPU marketing was full of ****, aimed at wrong targets with wrong hardware. Now we got CPU marketing shooting air?
I highly recommend to ignore all AMD bull**** their marketing is telling and read proper benchmarks once released. Nothing but disappointment.
Senior Member
Posts: 2770
Joined: 2010-01-03
what could go wrong... amd just needs to match SB ipc and they will be a short OC from the latest intel cpus, priced at same price with twice the cores, dont see how that can be impossible really, i think they will have a win this time
essentially, intel is either waiting for them or they really hit a roadblock, which wont go away just because amd finally came next to them, so thats actually cool if they are both stopped there, will have to compete with more and more cores
Senior Member
Posts: 7435
Joined: 2006-09-24
Even haswell-e had 3ghz es. 4770k was indeed 2.8ghz.
But yeah engineering samples are rather low clocked dies and the tests against them are usually done by underclocking the existing parts.
Senior Member
Posts: 8408
Joined: 2008-07-31
Honestly with engineering samples coming in at 3GHz I'm not hopeful that AMD that pump out a final product with high enough clocks to compete.
This makes no sense. I'm, like, wtf dude is wrong with everyone?
It's an engineering sample
Intel haswell ES: 2.8Ghz
Intel broadwell ES: 2.2Ghz
Intel Ivy Bridge ES: 2.0GHz
Among more and more and more and more, and unless it was an engineering sample of an exact model number with higher frequency rather then an engineering model of the architecture itself, it's almost always low. So seriously people, wtf is wrong with you all that because AMD does the SAME thing as Intel, that somehow means AMD is set up to fail, or is likely to fail? What kind of logic is that? That is no logic. To derive a feeling or "fact" or "i knew it" from something like this, is illogical to the highest degree.
They all OC'd so high in every review that they had everyone thinking even a bad chip would hit 4.6GHz.
Since we don't know the OC capabilities of these chips, how can you figure that it's the opposite? If an Intel ES sample is stock low, but was somehow gotten by someone who would OC it and publish those numbers, how does that make it any different from what we are at now, other then someone not willing to do the latter?