Soldato
AMD Ryzen 1400 upwards now get a FREE copy of the yet to be released Quake Champions: https://forums.overclockers.co.uk/t...s-free-with-amd-rx-series-amd-ryzen.18790620/
Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.
Oh no you don't, it's too early for the Zen+ hype train to start up.
However it could be tweaked for better power efficiency and hopefully IF not linked to the memory speed, considering it also needs to link up the GPU and not 2 CPU complexes.
I don't buy it, it's too soon. Threadripper uses the same Zen cores so why wouldn't these APUs? I always assumed the 2xxx series would be "Zen+" (i.e. still 14 nm, maybe with a small bump in clock speeds and/or IPC), with the 3xxx series being "Zen 2" (i.e. 7 nm). But if the APUs will be 2xxx, either I'm wrong and they are indeed improved cores, or their naming structure makes no sense. I guess the third alternative is they're just going to bump the generation number up at every opportunity to catch up to Intel, in which case Zen+ refreshes next year will be 3xxx and next year's APUs will be 4xxx.It could be they use Zen+ cores??
I don't buy it, it's too soon. Threadripper uses the same Zen cores so why wouldn't these APUs? I always assumed the 2xxx series would be "Zen+" (i.e. still 14 nm, maybe with a small bump in clock speeds and/or IPC), with the 3xxx series being "Zen 2" (i.e. 7 nm). But if the APUs will be 2xxx, either I'm wrong and they are indeed improved cores, or their naming structure makes no sense. I guess the third alternative is they're just going to bump the generation number up at every opportunity to catch up to Intel, in which case Zen+ refreshes next year will be 3xxx and next year's APUs will be 4xxx.
It's certainly possible, I just won't believe it until I see it.It follows from all their APUs though. Llano had updated cores over the Phenom II,Trinity was demoed before the FX8350 and had Piledriver cores,and both Steamroller and Excavator were more mobile orientated. If you go by the original release date of Ryzen which was late 2016,this would be one year later,and I would probably think its a more power optimised core,so they can hit higher clockspeeds in a TDP limited environment.
That is OFC my guess,not saying I am right though!!
It's certainly possible, I just won't believe it until I see it.
Also quad cores don't interest me, even with IGPs, so personally I'll be waiting for the next batch of desktop CPUs (at least) regardless.
I will be interested to see what a native one CCX design will be capable off,as it won't have any CCX penalty. It actually might be quite a competent gaming CPU too,especially if the cache and clockspeeds are comparable to the desktop Ryzen CPUs.
I would like to know this as well. My guess though is that all the working quads go into the higher models, so it is unlikely unless they are getting exceptional yields.
Remember,the reason why Ryzen performs better with fast RAM is due to the CCX penalty,so could it be quite possible that the CPU section is less dependent on fast RAM too??
I think AMD have to focus on the mobile market. For 90% of people PC= laptop and making a big push in the mobile business is a great way to get the AMD and Radion brand better recognised.
The groundwork AMD have put in with APUs makes them ideal supplier for a market thats deaperate for high performance chips that offers everything in one package.
Advanced Micro Devices, Inc. (NASDAQ:AMD) announced the expansion of its global ecosystem of partners for its EPYC datacenter processors today at the China EPYC Technology Summit. New datacenter customers – Tencent and JD.com – along with OEM partners Lenovo and Sugon, joined AMD on-stage with new product announcements and to showcase a wide range of systems and performance demonstrations.
“Today we celebrate the AMD EPYC family of CPUs for the datacenter market in China and are excited to announce support from the leading cloud service providers, Tencent and JD.com, while continuing momentum from Baidu, Lenovo and Sugon,” said Forrest Norrod, senior vice president and general manager, Enterprise, Embedded and Semi-Custom products, AMD. “The AMD EPYC family of processors provides incredible performance and scalability for the workloads that matter today and in the future. By partnering with these market leaders, AMD is bringing choice and competition to one of the fastest growing technology markets in the world.”
Additional ecosystem participation at the Summit came from Acer, Asus, Dell, Fiberhome, Gigabyte, HPE, H3C, INVENTEC, Lenovo, Sugon, SuperCloud, SuperMicro, TYAN, and Wistron. Key server hardware and software ecosystem support was provided by partners including Mellanox, Redhat, Samsung and VMware.
Cloud Datacenters Choose EPYC
“Tencent Cloud provides public services to benefit everyone. With our commitment to offer users with more choice and a more convenient user experience, Tencent Cloud is continuously seeking more cores, more I/O interfaces, more secure hardware features and improved total cost of ownership for server hardware products,” said Sage Zou, senior director of Tencent Cloud. “To continue as a leading provider of high-performance and high-value cloud services, Tencent needs to adopt the most advanced infrastructure and the chip industry’s latest achievements. By the end of this year, Tencent Cloud will launch AMD EPYC-based 2P cloud servers, with up to 64 processor cores and superior single system computing capability, to provide the industry with a more diverse portfolio of cloud products and services.”
“China Internet and e-commerce companies need more compute cores and higher memory bandwidth. We saw AMD EPYC processors have up to 32 cores, providing competitive advantage over current 2P server systems, and the eight memory channels enable greater memory bandwidth, which are believed to better match domestic customers’ requirements,” said Andrew Wang, technology leader of hardware system department at JD.com. “AMD EPYC will help JD.com improve the total cost of ownership (TCO) of our server systems. JD.com will collaborate with AMD on Big Data, AI and Cloud Services based on AMD EPYC in the future.”
Featured OEM Platforms
“As a strategic partner, Sugon and AMD have been working together for more than 15 years, AMD’s new EPYC data center processors will bring a new value experience in the datacenter,” said Cao Zhennan, vice president of Sugon. “Sugon will introduce full AMD EPYC processor based product line with nine new products across workstation, rack, blade and super rack systems for high-performance computing, cloud computing, large data analysis and deep learning applications today.”
“AMD EPYC processors present unique opportunities for our customers to lower total cost of ownership via an unprecedented balance of cores, memory bandwidth, and I/O. We are excited to collaborate with AMD and several global Hyperscale customers to develop and deploy single socket and dual socket EPYC-based servers,” said Paul Ju, vice president and general manager, Lenovo Global Hyperscale Business.
Mmmm, maybe, but then fast RAM would still be important as the CPU will likely communicate with GPU via Infinity Fabric, no? Also the RAM will be used as VRAM, so arguably fast RAM JUST as important for the APU as the CPUs, just for slightly different reasons!
Yes but IGPs always rely on fast RAM speeds since they don't have much/any VRAM, so either way you need fast RAM to get the most out of it.Remember,the reason why Ryzen performs better with fast RAM is due to the CCX penalty,so could it be quite possible that the CPU section is less dependent on fast RAM too??
Yes but IGPs always rely on fast RAM speeds since they don't have much/any VRAM, so either way you need fast RAM to get the most out of it.