Processador AMD Ryzen / Threadripper

DvdCst disse:
Falam em Zen cores mesmo. Falta saber o que um "Zen core" consegue fazer, o que vale em relação a um Intel e se os lançam a um preço enquadrado com a concorrência a nível de rendimento. Continua a ver-se números grandes, provavelmente para ajudar a vender/hype, mas o que importa verdadeiramente é o rendimento, e nessa parte a Intel parece andar mais do que descansada neste segmento. Era bom ver verdadeira concorrência no high-end para abanar um pouco o mercado, mas em 2016-17 já devem andar os Skylake a dominar o mercado de vendas e não vai ser fácil (a menos que sejam mesmo muito bons).
Clicar para expandir...

Eu disse que são CPU cores porque falam em "up to 16 AMD Zen x86 cores."
 
Muito dessas histórias são rumores e duvido que lançados pela AMD.
Existe meia dúzia de sites de reputação duvidosa que se está a entreter a inventar notícias dia sim dia não dos CPUs e GPUs da AMD para gerar cliques.

Essa noticia veio do Fudzilla caramba!

Este hype todo é contra-produtivo porque a AMD mal se consegue manter competitiva.
 
Acho que descobri a "origem" do devaneio do Fudzilla.

28gv61z.jpg

Esta imagem terá sido apresentada durante um evento no Japão, e refere-se a um hipotético APU destinado ao mercado HPC (High Performance Computing), e como se vê na imagem, surgem duas setas a ligar um Next Gen Server CPU e um Next Gen Firepro GPU.
 
The next generation Opteron has 32 Zen x86 cores

No APU, 16MB L2, 64MB L3 cache

We recently showed you a new 16 Zen core next generation processor with Greenland integrated graphics and DDR4 support.

This part definitely sounds interesting but we got an update on the 2016 Opteron server market parts. The next generation Opteron won't have an integrated graphics part but it will have up to 32 Zen x86 cores with 64-thread support. Unlike the highest end compute HSA part that comes with Greenland HBM graphics, the next generation Opteron doesn't have any integrated graphics. The Opteron needs all the silicon space for the L2, L3 cache as well as its Zen x86 cores.

Just like the 16 Zen core high performance market APU, each core has 512KB of L2 cache and four processors share 8MB L3 cache. The highest end part will come with eight clusters of 4 cores and if you do the math this server oriented CPU will come with 64GB of L2 cache and 16MB of L2 cache for its CPU cores.

A few other notable features for the next generation server parts include a new platform security processor that enables secure boot and crypto coprocessor. The next generation Opteron has eight DDR4 memory channels capable of handling 256GB per channel. The chipset supports PCIe Gen 3 SATA, 4x10GbE Gig Ethernet and Sever controller HUB. Of course, there will be a SMP, dual socket version.

The next generation Opteron will have 32 CPU cores in its highest end iteration, and we expect some Stock Keeping Units (SKUs) with fewer cores than that for inexpensive solutions.

In case AMD comes to market with this part on schedule, and if the Zen core ends up performing as expected, Intel might finally get some competition. Let's just hope for AMD's sake that this server CPU is coming in 2016, sooner rather than later.

We can only on possible Zen-based FX parts for high-end desktops, or the manufacturing process for Zen chips, but at this point we cannot confirm FX parts are coming, and whether or not they will be manufactured in 14nm.


http://www.fudzilla.com/news/processors/37564-the-next-generation-opteron-has-32-zen-x86-cores
 
Última edição:
Não se sabe se os 16 cores aparecem logo em 2016.
Se no próximo ano começarem por ex. nos 8-12 cores, e escalarem por ex. em 2017 para os 16 cores, nessa altura pode já não ser assim tão impressionante.
Eu não me acredito minimamente que em 2016 vamos ver um APU da AMD com 16 cores Zen + quad-channel DDR4 a 3200MHz + 16GB HBM2 a 512GB/s + iGPU enorme + 64 lanes PCI-Express. Seria um conjunto de especificações completamente redundante, caro, desnecessário com péssimos yields e o chip mesmo a 14nm ficaria praí a 800mm^2.
 
Depende dos I7 que estiverem a falar... Os da plataforma x99 8-core têm 16 threads.

Acho que o que todos querem saber é a performance por core da plataforma Zen.
 
ToTTenTranz disse:
Não se sabe se os 16 cores aparecem logo em 2016.
Se no próximo ano começarem por ex. nos 8-12 cores, e escalarem por ex. em 2017 para os 16 cores, nessa altura pode já não ser assim tão impressionante.
Eu não me acredito minimamente que em 2016 vamos ver um APU da AMD com 16 cores Zen + quad-channel DDR4 a 3200MHz + 16GB HBM2 a 512GB/s + iGPU enorme + 64 lanes PCI-Express. Seria um conjunto de especificações completamente redundante, caro, desnecessário com péssimos yields e o chip mesmo a 14nm ficaria praí a 800mm^2.
Clicar para expandir...

Faz sentido para um processador para o mercado HPC, mas concordo contigo que para o mercado desktop não vejo como seja possível.

Vejamos: o Haswell-E com 18 cores tem um die size de 661 mm2 nos mesmos 14 nm. Não tem GPU nem HBM e "só" 40 lanes Pci-Ex. Custa mais de 4000$.

Não vejo como seja possivel meter 16 cores Zen + quad-channel DDR4 a 3200MHz + 16GB HBM2 a 512GB/s + iGPU enorme + 64 lanes PCI-Express no mercado consumidor.

EDIT: O Haswell-E é a 22 nm e não 14. Mas mesmo assim este Zen do artigo não faz sentido para o mercado desktop.
 
Última edição:
AMD Zen CPU Core Block Diagram Leaked – Features 512bit Wide Floating Point Unit And A Wider Integer Pipeline
HardwareLeak 4 hours ago by Khalid Moammer
An unbelievable leak showed up at our doorstep today, one that’s pertaining to AMD’s upcoming high performance Zen CPU core. A slide showcasing the block diagram for AMD’s upcoming CPU architecture has found its way to the internet.



Before we dive into what this slide shows we should make sure to mention that at the present moment we absolutely cannot verify the authenticity of the slide in question. However it won’t take long before we can find out for certain if this slide is truly legitimate or not. As it’s supposed to be part of AMD’s upcoming roadmap unveiling at the company’s Financial Analyst Day event on May 6th.

A First Glimpse Into AMD’s Next Generation High Performance x86 Zen CPU Core
We first broke the news about AMD’s next generation high performance core back in September of last year. At which point AMD’s then CEO Rory Read revealed the code name for the company’s upcoming high performance x86 CPU architecture. Prior to then we only had knowledge of its sister ARMv8 core code named K12.

However recently we’ve witnessed a flood of leaks pertaining to AMD’s brand new inception. Three months ago we heard that AMD was preparing an entirely new line-up of CPUs on a brand new platform. We learned that the platform was code named Summit Ridge and would feature an entirely new socket and an updated feature set including DDR4 memory support. And more importantly we learned that the new platform would feature mainstream CPUs with up to eight Zen cores.

Two months later we learned that AMD was also working on a monstrous High Performance Computing APU with 16 Zen cores and a huge integrated GPU in addition to stacked High Bandwidth Memory. Finally last week we learned that AMD would also introduce high performance server CPUs with up to 32 Zen CPU cores. Hearing about all of those different SKUs is jolly exciting but is also quite frustrating as we had no idea what to expect from Zen. That is until today, because for the first time we are being given a glimpse into the CPU core itself rather than the different SKUs it will show up in.

We don’t have a die shot of the core but we have the next best thing, a block diagram. Below you can see Zen on the right, compared to AMD’s upcoming and last Clustered Multithreading / CMP CPU core code named Excavator. Excavator is the fourth and as mentioned above the last of AMD’s Bulldozer family of cores. It will debut with AMD’s upcoming Carrizo APU, which the company hails as the most power efficient mainstream APU the company has ever made.



The first thing we can spot is that there is only one integer cluster in a Zen core rather than two as in the Excavator module on the left. These two integer clusters are what forms the two separate CPU cores / threads in each Excavator module. Zen takes on a more traditional AMD CPU layout resembling that of Phenom and Athlon K series cores. Featuring a single large Integer cluster and one equally large floating point unit.

This is an important distinction because in contrast, the Bulldozer family of cores achieved very high integer throughput but also sacrificed floating point performance. That’s because each pair of cores shared one floating point unit. Although the floating point unit itself was larger and more capable than the one found in AMD’s previous K10 CPU core in the Phenom II line of chips. Floating point performance was still lacking compared to integer, merely because the design was heavily weighted towards integer as can be seen above.

Obviously because Zen forgoes the CMT design of the bulldozer family we can see that AMD has returned to a single fetch and single decode unit on the front end. As opposed to the double decoders that were introduced with Steamroller, Excavator’s predecessor found in the 7000 series Kaveri APUs.

Right off the bat just by looking at the block diagram we can tell that Zen will have a substantially higher single threaded performance compared to Excavator and the Bulldozer family. Both in integer and floating point workloads. The bulldozer family will likely maintain a higher total throughput on integer if you compare a single Excavator module with two cores vs a single Zen core. But this is a sacrifice that has to be made for Zen to achieve better per thread performance. And once more, Bulldozer’s Integer throughput was already quite phenomenal as it rivaled Intel’s extreme i7 parts. So the overall throughput once you add all the threads was never a problem, the per thread / single threaded performance was the issue.

Comparing both floating point units of Excavator and Zen we can see that AMD has introduced a floating point that’s twice as wide as that of Excavator. Featuring two FMAC 256-bit units that in all probability will be able to fuse together to process 512-bit AVX flaoting point instructions. This is compared to the two 128-bit FMAC units found in AMD’s Bulldozer family, which can either process one 128-bit SIMD instruction each per clock or fuse together to process a single 256-bit AVX instruction per cycle. Hence the assumption above that we could see a similar behavior with Zen’s FPU which would allow both FMACs to cooperate and process 512bit AVX instructions.



Besides enabling 512bit AVX support, the wider floating point unit also means that Zen will be able to process less complex instructions at double the rate of Excavator. A massive boost in floating point performance. An area where AMD had historically excelled in but was put aside with Bulldozer.
I should mention that AVX-512 support was not listed for Zen in the official Linux patch which revealed the new instruction set extensions that the upcoming processor will support. This is slightly odd but could be explained by a possible lack of 512bit integer support in Zen, which is required for the AVX-512 extension.

Moving on to the integer cluster, based on the diagram above Zen will feature a 50% wider integer pipeline vs a single Excavator core. Which will also dramatically improve the single threaded / per core performance of Zen vs the Bulldozer family of cores.

In summary based on the diagram above we’re looking at a significantly beefed up integer cluster, a doubling in floating point capability and a streamlined front end. The net result should be a significantly faster, leaner, smaller and more power efficient CPU core.

Products based on the new core are scheduled to come out next year. And if this is anything to go by we should hear a lot more about the new core in two weeks time. In the meantime stay tuned so we can keep you updated.


Read more: http://wccftech.com/amd-zen-cpu-core-block/#ixzz3YYFdFQ6a
Clicar para expandir...

http://wccftech.com/amd-zen-cpu-core-block/
 
É interessante o diagrama, mas é uma visão ainda muito geral.

A ideia de terem uma unidade FP por cada dois cores INT, foi atirada ao rio. Aqui por cada core há uma unidade FP e à primeira vista, mais forte, com duas unidades 256 bit. Se isso quer dizer que suporta AVX-512 ainda é cedo para dizer.
Do lado INT, tem 6 pipelines, mas não sabemos o que é cada um. Por isso também é cedo para dizer que é melhor.

Parece que vai ter 512 KB de L2 por cada core e 8 MB de L3 por cada grupo de 4 cores. Parece-me uma boa base para um APU quad core para o mercado consumidor. Não se sabe nada da L1 e de muitos outros pormenores.

Parece que vai haver uma versão com 32 cores para servidores e 16+GPU para HPC. Parece também que o mercado servidor é o primeiro alvo da AMD.

Até agora os detalhes ainda são poucos, mas pode ser que este cpu volte a colocar a AMD no mapa.
 
Inicie sessão ou registe-se para poder participar.
Back
Topo