Thursday, August 18th 2016

AMD Demos Breakthrough Performance of the ZEN CPU Core

At an event last night in San Francisco, 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 clean-sheet 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."

"AMD invested where it counts, with an x86 core that can scale from PCs to high-performance 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.
Add your own comment

187 Comments on AMD Demos Breakthrough Performance of the ZEN CPU Core

#76
TRWOV
Well, this was pleasantly unexpected :D

The only drawbacks I think Zen will have is that the CPUs will sport a higher TDP; I'm thinking about 165w for a 8C/16T part at 3.2Ghz or so. I'm pretty sure AMD is brute forcing all of this performance for the time being and will later be refined. Maybe I'll pick up the revised Zen.
Posted on Reply
#77
OneMoar
There is Always Moar
$ReaPeR$look, even marketing idiocy has its limits, there has to be some truth at least. i doubt that zen will be faster, but imo it will be within 10% of the corresponding intel counterpart. at this point that will have to do.
I hope so but don't count on it and certainly don't give anybody any money until you see numbers
Posted on Reply
#78
alucasa
It's simple really.

Make judgement - after - it's released, not before.
Posted on Reply
#79
Jism

"We are on par with our performance targets"

Sums about it. And if anyone has any doubts about that blender render, please count the time from the moment that button was pressed until that bench finished up. Compare results to other systems and do your own math.
Posted on Reply
#80
the54thvoid
Super Intoxicated Moderator
alexsubriAMD Stocks are rising up to ~5% today! Glad I bought my shares cheap ^___^
Sell them before the review sites do their independent analysis. It'll be the best AMD CPU for a long time, no doubts but I think the people (evil speculators and shysters) who make the main play with stocks will not like it's performance.
Posted on Reply
#81
alucasa
JismAnd if anyone has any doubts about that blender render, please count the time from the moment that button was pressed until that bench finished up. Compare results to other systems and do your own math.
As someone who uses Blender everyday, I can tell you that it's really easy to rig the rendering time. For a simple image like in the Youtube, 500 sample render will look identical to 5,000 sample render.
Posted on Reply
#82
evernessince
Kevin KingHmm, the Broadwell-e was clocked wayyy below it's normal clock speed, 3.0GHz versus what it's sold as, 3.7GHz. This tells me that Summit Ridge likely won't be clocked at a similar frequency to Broadwell-E, otherwise they would've shown that comparison, right? It's exciting, sure, but if it's not able to match the clock rate of Intel, I'm not exactly hyped...
We can't say for sure yet. I doubt AMD's current samples will be the final result. The last benchmark with Zen we saw was clocked at 2.8 Ghz so they obviously managed to increase the clock speed. Even without the same clock speed this would mean AMD is competitive again. All they need to do is release a successor to the quad core low end athlon and they win the biggest part of the market, the budget end. Intel's low end pentiums only have 2 cores and don't even have hyper-threading. An AMD quad-core of even remotely close IPC would make Intel's offerings look silly.
Posted on Reply
#83
rtwjunkie
PC Gaming Enthusiast
JismIt's always never enough with you people. They go core to core thread to thread clock to clock, on a bench that is HIGLY reliant on all cores / threads. Outcome is that AMD finishes up that test FASTER then Intel's counterchip. I dont think they would be clocking down that intel chip by putting the bclk down, slower memory speeds and all but simply set for a 30x MP and not 37x.

Bottom point is is that AMD actually offers something with raw power that COMPETES with intel's 8c/16t CPU worth 1000$.
What you're ignoring, is that at their real speeds, running in the wild, the AMD will not be able to keep up unless (A) They can speed the clocks up, and (B) demonstrate its single core supremacy. By slowing down the Intel for the demonstration, they may be showing that the Intel is slower at completing instructions...at that speed. But real life speeds, which is where consumers will actually use them, and what consumers will judge on, the Broadwell will win because it is clocked higher, and people will buy accordingly.

Don't get me wrong, I want Zen to be great and succeed! Just don't read too much into this.
Posted on Reply
#84
Supercrit
I really hope that will force i3 becoming quad cores or at least turbo enabled, with the whole line have less disabled features just because intel could with no competition.
Posted on Reply
#85
HD64G
dwadeCouldn't care less about clock for clock comparison since they are 2 different architectures. Only metric that matters is performance per watt.
If we suppose linear performance to clocks for the Intel CPU in this comparison and we assume that this 8C/16T Zen is the one that is going to be sold as the 95W, with 140W for Intel's Zen is 20% better in Perf/W. Simple assumptions there. Future reviews will prove if this benchmark is accurate or not.
Posted on Reply
#86
Dimi
I remember the AMD hype from back in the days of the FX processors.

I think i jumped from Intel to an AMD Athlon 64 4000 (San Diego) cpu while shortly after it got rendered useless compared to a cheaper intel core duo. I think i paid close to 400 euro for the AMD cpu while the i3 was like 120 ish.

I'm not falling for this hype again.
Posted on Reply
#87
cadaveca
My name is Dave
ROFL.

choo-choo! All aboard the hype train!

:clap:
Posted on Reply
#88
the54thvoid
Super Intoxicated Moderator
cadavecaROFL.

choo-choo! All aboard the hype train!

:clap:
What do you know about CPU's? /sarcasm.
Posted on Reply
#89
alexsubri
cadavecaROFL.

choo-choo! All aboard the hype train!

:clap:
That's right! You BETTER hop on that Zen train! :p

Posted on Reply
#90
cadaveca
My name is Dave
the54thvoidWhat do you know about CPU's? /sarcasm.
This is the sort of launch I can't miss. But I dunno... might be a bit late. :P
Posted on Reply
#91
OneMoar
There is Always Moar
I think TPU should just refuse to cover anything AMD
I mean we are totally bias here right its not like we have been lied to time and time again or anything
Posted on Reply
#92
cadaveca
My name is Dave
OneMoarI think TPU should just refuse to cover anything AMD
I mean we are totally bias here right its not like we have been lied to time and time again or anything
Meh. Put it in my hands and let me play with it, and I'll be honest. I wasn't one to buy into the hype in the past; you can find my posts denying it from the get go and even explaining why. :P
Posted on Reply
#93
Fx
cadavecaROFL.

choo-choo! All aboard the hype train!

:clap:
It isn't just hype that has some of us aboard the train. I am one of AMD's customers that continues to purchase their products to support them. Their processors have plenty of power needed for gaming. I use Xeons for my servers.

If Zen can improve its generational performance leap by 20-30% compared to the 40% it is claiming, I will be more than happy.
Posted on Reply
#94
cadaveca
My name is Dave
FxIt isn't just hype that has some of us aboard the train. I am one of AMD's customers that continues to purchase their products to support them. Their processors have plenty of power needed for gaming. I use Xeons for my servers.

If Zen can improve its generational performance leap by 20-30% compared to the 40% it is claiming, I will be more than happy.
AMD does a great job of pricing their product accordingly for the most part, so they'll do well. But they are a tiny CPU company, so I just don't expect much. It just doesn't matter how fast Zen is; it'll be priced according to its true value anyway. Ultimately THAT is what matters.
Posted on Reply
#96
Frick
Fishfaced Nincompoop
MelvisErm no, hate to tell you and about 10 000 other people on this forum will tell you just the same thing that AMD was cheaper on the 939 socket, the FX57 and 60 was around $1000-$1200 and the Pentium EE was around $1500 AUS, its just a fact :)

Anyone can post a pic like you did that is from god knows what source, and from even before launch date where in actual fact I never saw any CPU from AMD (price to performance comparison) be more then Intel.
Different continents. I am too quite sure that AMD was just as expensive as Intel when they were competitive. They often, iirc, offered a bit more punch for your money and often overclocked decently, but they weren't magical people giving stuff away if they could charge for it and get away with it, which they did.

Annywaay, real world test is nice. But I don't expect it to last if they perform different tests. Remember the goal was Haswell-ish performance, iirc.
Posted on Reply
#97
D007
Lol and so the war of the fanboys ensues.. Like always.. Why can't people just be happy, that AMD may finally make a processor worth a shit? lol.
I'm happy for them.. I go with performance.. Idc about cost. If AMD can beat intel, I'll be looking at it. Anyone with sense would do the same thing.
But don't make huge assumptions based on half tests..
Posted on Reply
#98
dozenfury
I'm excited and looking forward to Zen too. *However* AMD has also shown a strategy of embellishing in their marketing pretty extremely. Their problem is that real performance gains generally take big $ R&D, money which Intel has and AMD doesn't. It's far cheaper to market something as faster than to actually make a faster chip.

So I hope maybe AMD gets lucky and has a surprise performance breakthrough despite fighting an underdog battle without the huge R+D spend Intel has, because I would love to be able to go back to AMD and the competition is badly needed. But...I'll believe it when I see it, as in third-party real released product benchmarks. We've been hearing about your secret weapon Piledriver/Steamroller, etc. cpu for the past 10 years and have been let down every time, prove the doubters wrong this time AMD.
Posted on Reply
#99
slozomby
I'm confused. so the zen chip at 3k outperforms an underclocked 6900k. I wonder if they turned off boostclock as well.
Posted on Reply
#100
cadaveca
My name is Dave
slozombyI'm confused. so the zen chip at 3k outperforms an underclocked 6900k. I wonder if they turned off boostclock as well.
It doesn't matter really, does it? Being even within a few % of Haswell-E on a clock-per-clock basis is better than I could have hoped for.
Posted on Reply
Add your own comment
Jan 23rd, 2025 20:31 EST change timezone

New Forum Posts

Popular Reviews

Controversial News Posts