Ryzen coming in 2017
AMD reveals the brand for Zen, a clock speed of 3.4 GHz and some new technical features.
As much as we might want it to be, today is not the day that AMD launches its new Zen processors to the world. We’ve been teased with it for years now, with trickles of information at event after event…but we are going to have to wait a little bit longer with one more tease at least. Today’s AMD is announcing the official branding of the consumer processors based on Zen, previously code named Summit Ridge, along with a clock speed data point and a preview of five technology that will help it be competitive with the Intel Core lineup.
The future consumer desktop processor from AMD will now officially be known as Ryzen. That’s pronounced “RISE-IN” not “RIS-IN”, just so we are all on the same page. CEO Lisa Su was on stage during the reveal at a media event last week and claimed that while media, fans and AMD fell in love with the Zen name, it needed a differentiation from the architecture itself. The name is solid – not earth shattering though I foresee a long life of mispronunciation ahead of it.
Now that we have the official branding behind us, let’s get to the rest of the disclosed information we can reveal today.
We already knew that Summit Ridge would ship with an 8 core, 16 thread version (with lower core counts at lower prices very likely) but now we know a frequency and a cache size. AMD tells us that there will be a processor (the flagship) that will have a base clock of 3.4 GHz with boost clocks above that. How much above that is still a mystery – AMD is likely still tweaking its implementation of boost to get as much performance as possible for launch. This should help put those clock speed rumors to rest for now.
The 20MB of cache matches the Core i7-6900K, though obviously with some dramatic architecture differences between Broadwell and Zen, the effect and utilization of that cache will be interesting measure next year.
We already knew that Ryzen will be utilizing the AM4 platform, but it’s nice to see it reiterated a modern feature set and expandability. DDR4 memory, PCI Express Gen3, native USB 3.1 and NVMe support – there are all necessary building blocks for a modern consumer and enthusiast PC. We still should see how many of these ports the chipset offers and how aggressive motherboard companies like ASUS, MSI and Gigabyte are in their designs. I am hoping there are as many options as would see for an X99/Z170 platform, including budget boards in the $100 space as well as “anything and everything” options for those types of buyers that want to adopt AMD’s new CPU.
What about this SenseMI Technology? This is the brand that AMD has assigned to a collection of five different features that improve the performance and efficiency of the Zen architecture as it exists in the Ryzen consumer product. I’ll be upfront – I don’t know if these items needed an encompassing brand umbrella though the features themselves are worth noting.
First up is Pure Power, an integration of a temperature and voltage monitoring system that helps optimize the clock speeds and voltages the CPU will run at in real world conditions. By keeping track of die temperatures and voltages running through it, Zen will utilize adaptive control schemes to lower power usages while maintaining performance and clock speeds at expected levels. There is a lot of jargon and terms on this slide above including “infinity control fabric” and “distributed embedded sensors” but this is technology we have seen before in Intel’s processor systems, both major vendors GPUs and in mobile devices. It helps keep performance levels at a static state but improves efficiency to keep power and thermals at lower levels.
Precision Boost is the other side of the same coin, using the same sensors and the same logic but for a different purpose. This is AMD’s answer to Intel Turbo Boost technology, dynamically adjusting the clock speed of the processor cores to improve performance for as long as the die can remain within the preset thermal constraints AMD (or a user) puts on it. We don’t have much detail on this feature yet but I would assume it would behave very similarly to Intel’s version – running at higher clocks during lower threaded workloads but lower clocks at highly threaded workloads. One interesting improvement with AMD Precision Boost is that it can operate in 25 MHz increments. Intel Turbo Boost only moves in steps of 100 MHz. While this will add more complexity to the system, in theory it should allow AMD to stretch that last little bit higher to get additional perf.
Extended Frequency Range has the potential to be very interesting. If you have better cooling, your processor will be able to run at higher clock speeds. AMD says this will scale with air cooling, water cooling and even liquid nitrogen! NVIDIA and Intel have both claimed that this capability exists for their products (NVIDIA GPU Boost), but in practice, we usually find that process technology creates a different limit than thermals for processor performance. For example, on NVIDIA GPUs, artificial voltage limits keep the GPU clock down well before thermals when running extreme cooling solutions.
The amount that XFR impacts consumers will be measured once we have samples in hand. If the difference between a basic air cooler and water cooling is ~100 MHz of usable, repeatable clock speed, then the value will be…debatable. If we can get an extra 300-400 MHz by just utilizing high end enthusiast coolers, that’s a great feature to add.
Now things getting a little more interesting. What AMD is calling Neural Net Prediction, and what they dub as a “artificial network”, offers improved prediction instructions and paths through the processor pipeline. If that sounds familiar to you, and it should, you’ll recognize this as a branch predictor.
The goals put forward on this slide exactly describe the goals of prediction units on a processor, and have for as long as I have been reading architecture documents. “Building a model of decisions driven by software code execution” is an interesting way of describe using the history of previous code results to attempt to predict the next code result early. A prefetch can grab the next instruction ahead of time and, if it predicted correctly, is ready to execute immediately without time decoding.
The description of this as a “neural net” is just one of semantics. A quick look over at Wikipedia shows a variation of the predictor called a Neural Branch Predictor. The differentiator between this and a standard unit is that a neural unit can exploit long histories of previous data with only a linear growth in required resources. That comes at the cost of added latency though. When I have more time with AMD’s engineers we can get a better explanation of what is going on, but for now, just understand that Zen has an improved branch and path prediction unit to improve performance and power efficiency.
Smart Prefetch has a similar setup. A prefetch unit attempts to guess which data the application will need from cache before the request is made, improving performance by reducing wait times for memory access. The same argument can be made as to if this constitutes a neural net, or smart network or whatever – the important aspect is that AMD claims the improvements in this area of the architecture result in performance gains for IPC, helping to add up to that impressive 40% improvement over the last generation.
More Performance Previews
Back in August, AMD held a quick press conference in San Francisco to show off the first official performance metric for Zen. It compared an 8-core Zen processor against an 8-core Broadwell-E, both running at 3.0 GHz. The Zen CPU finished a run through Blender first, indicating that the IPC of Zen was higher or on-par with that of Broadwell-E.
For today’s Zen information blast AMD had another pair of data points that worth discussion. First, AMD pitted a Ryzen processor running at a fixed clock speed of 3.4 GHz, with no boost features enabled yet, against a stock Core i7-6900K, with boost enabled. For your reference, a 6900K is an 8C/16T part that runs at a base clock of 3.2 GHz and can clock as high as 3.7 GHz with Turbo Boost. When operating at full thread capacity, the 6900K is in the 3.4 GHz to 3.5 GHz range. The benchmark was Handbrake, a very common piece of software that most review sites use for testing and that many consumers use for media transcoding.
While not the most riveting video coverage, this will show the transcode running on both systems, fully utilizing 16 threads. The AMD processor finished a few seconds ahead of the Intel part, again indicating that it is leading or competitive against Intel’s current enthusiast class processors.
A power demonstration was run, though this time with Blender, showing total system power draw during a render.
While we don’t know the rest of the components on the system, and that this is total system, not CPU only power draw, so we must be careful about making definitive conclusions. The AMD Zen PC was running at ~188 watts, the Intel Core i7-6900K machine at ~191 watts. Even if you give the platforms +/- 15 watts for variances in motherboards and the like, that means AMD is within range of Broadwell-E.
More to come
Now we know the brand, Ryzen. Now we know the clock speed, a minimum of 3.4 GHz on a flagship part. We know something new about the architectural improvements for power, speed and efficiency. We do not know the release date (1H 2017) or pricing. And we don’t know the full story in terms of performance and where Ryzen will compete against an Intel lineup that will be shifting in the next 2-6 months substantially.
I’m more excited for AMD new processor than I was in August when they first show a performance demo that was very quickly followed by doubt and scrutiny on the workload. One more benchmark does not a CPU story make, but it’s clear that AMD remains committed to its stance on processor IPC and competitiveness in the market. Those of us that have been waiting for true competition to Intel’s Core i5/i7 processors from AMD, without the overarching caveats, will not have to hold out much longer. If you were ever doubting that 2017 was going to be a pivotal year for computing and PC gaming, this is yet another slap in the head with a large trout.
“Ryzen” sounds to much like
“Ryzen” sounds to much like “Ricin”, a very deadly and toxic seed that can kill humans. Wow AMD, thanks for a lame name.
It sounds more like an
It sounds more like an evolved Pikachu like name! So Zen becomes Ryzen with some moon stone Hocus-Pocus!
Exactly my thoughts.
Exactly my thoughts.
Could have a cyborg raichu so
Could have a cyborg raichu so that he can do stuff like in metal gear rising.
I have a Zen on rye with the
I have a Zen on rye with the fried tofu, mung bean sprouts, with Wafu dressing(heavy on the wasabi)! Yum!
Edit: I
to: I’ll
Damn Gin and
Edit: I
to: I’ll
Damn Gin and grapefruit juice, rally can not taste the Gin in grapefruit juice at all! I’ll perish without a hair of the dog!
Probably best not to eat it,
Probably best not to eat it, then.
Ryezen
Ryezen
Ryzen is awesome name, shut
Ryzen is awesome name, shut your lame mouth
Don’t be jealous of my RIZEGA
Don’t be jealous of my RIZEGA build.
You mean your verizon build?
You mean your verizon build?
I didn’t make the ricin
I didn’t make the ricin connection immediately, but now I’m picturing Pinkman giving this proc to a child.
The branding here is, IMHO, a real bummer. It just reeks of overcompensation or just trying too hard in general. I know branding is, in the 21st century, of prime importance but if the best moniker you can come up with is some lame pun then AMD just should’ve added the extra pins to the socket to truly make it 1337 and drenched the lid in blinkenlights.
And maybe switched the ‘e’ to a ‘3.’
rYz3n – itz so 1337. Now with free neon sunglasses.
If the 3rd party reviews bear edible fruit, I’ll jump aboard, I just wish it didn’t, like someone poignantly suggested, sounded like a Pikachu.
Yeah Ricin was the first
Yeah Ricin was the first thing that came to mind as well when reading how it's pronounced heh. Zen was a much better brand name IMO than Ryzen though I suppose they want something they can use beyond the Zen 1.0 arch. Oh well, what matters is price / performance and if its there I will probably upgrade this ol' desktop finally!
And, you’re right. Price and
And, you’re right. Price and performance are all that matters in the end and, if AMD can tick those boxes, the product will succeed.
I’ve also heard, from a hypothetical, imaginary source, that the APU’s will launch as “Shyne.”
[insert crickets chirping]
In a desperate race to claim market share, uninspired parlor tricks and juvenile branding may erode some deserved legitimacy a quality product earns.
I’m not trying to slam AMD, but sometimes not obfuscating a potentially quality product with a snazzy box is the right move. The name Ryzen directly acknowledges AMDs secondary status and conjures a notion of trying to catch up, thus further solidifying a perception of being second class. This looks especially desperate when said product is decidedly not an aesthetic piece and its worth is derived solely on performance and value quotient. This feels, again just in my humble opinion, like a preemptive cheapening of their product and a real miss.
An abstract, arbitrary, and science-y sounding name allows full focus to be on how performant the chip is. No need to feed fodder to the pretentious, naysaying wolves, which I have apparently become.
They spent a lot of time
They spent a lot of time showing it off against the 6900K ($1100) and the 6700K ($299), which makes me think that they will price it somewhere in between, but closer to the 6700K.
Really where does the
Really where does the competition dig you folks up from! That’s some double plus good quack speak! And you bloviate in the very same marketing speak as AMD’s, Nvidia’s or any other marketing major/comm arts major would! Being someone of that marketing monkey mentality! Your paid duty is to spin negative without out doing any real technological comparison and contrast between AMD’s or even the AMD competition’s salient technology advantages or disadvantages! This pegs you as one such spin for hire entity! And there will be more of this leading up to Zen/Ryzen’s(marketing name) release. Those focusing on the relative coolness of any marketing monkeys’ terminology meant to impress the mono-brow knuckle daggers really have no place discussing any highly technological matter’s technological pros and cons. You are most definitely here to do someone’s bidding!
You caught me!
and all i got
You caught me!
and all i got my my shilling was a lousy will i am cd
You caught me!
and all i got
You caught me!
and all i got my my shilling was a lousy will i am cd
for me it sounds like Risen
for me it sounds like Risen aka rising from the ashes
Maybe they watched too much
Maybe they watched too much breaking bad! haha
Kind of glad that I resisted
Kind of glad that I resisted $140 off the 6700K at Microcenter. This should be interesting.
My 2500K is still going strong at 4.2Ghz! Hope AMD can surprise me and get me back in their camp.
My 2500k @ 4.3Ghz has been
My 2500k @ 4.3Ghz has been holding up well also.
Pfft, pussies. My GODLIKE i7
Pfft, pussies. My GODLIKE i7 2600K has been OverClocked to 5GHz on water and has been running in that state for at least 4 years now, in a 24/7/365 mode.
Bullshit, its been restarted
Bullshit, its been restarted and we all know it.
Are you seriously going to
Are you seriously going to argue semantics? PUH-LEASE.
Also, my system sits on an SSD and according to diagnostic tools such as CDI – it was reset no more than 59 times so far, IN ALMOST FIVE YEARS. That’s essentially equivalent as not turning off the station AT ALL.
Yeah, my 2600k has been on,
Yeah, my 2600k has been on, except when it’s been off, too!
Also running at 7.31242123 GHz, unless I have it downclocked and undervolted to .0000001 Hz at -31.7 volts. I’m trying to stay efficient, ya know?
Very puny. Go get yourself a
Very puny. Go get yourself a cookie.
What voltage out of
What voltage out of curiosity?
100% BSOD – 1.425 and
100% BSOD – 1.425 and lower
Glitchy/buggy, but no BSOD – 1.428~1.455
100% stable – 1.458
Not the best ever, but some guys out there have it as bad with this “lottery” as to needing to go all the way up to 1.5 (and some even higher) to get it boot at all at that frequency, not even mentioning getting it 100% stable.
Wait so 59 now = zero? That’s
Wait so 59 now = zero? That’s interesting math you have there.
365×5 = 1,825 days, / 59 reboots = Bro you are rebooting once a month.
Lrn2Math
One full reboot in my system
One full reboot in my system takes less than 7 seconds and that’s while NOT applying “fast boot” settings in my motherboard’s BIOS (with “fast boot” enabled it loads to desktop in roughly 5 seconds). This is essentially a nonexistent “problem” you’re trying to artificially impose there, on me. Now you will count how many seconds are there in a year, huh? Sure, be my guest.
8 core * 3.4 ghz = 4 core at
8 core * 3.4 ghz = 4 core at 6.8ghz
RyZen IPC is above 40%, so it seem AMD is going to cleanup the content creation market.
actually, as i have said in
actually, as i have said in my comment, actually, that might still not be the case
many more programs are optimised for the long and old intel core architecture, but not for this new AMD architecture
which means, unless you use blender, intel wins by a bit
but the price would be another story!!
well intel had to stop
well intel had to stop handicapping amd cpus in their c compilers, thats why intel has been faster in some programs that used said intel c compilers in the past
now amd is free of those chains so they can freely show what they are capable off
i also suspect amd has optimized their architecture to behave a little more like intels, this machine learning thing sounds like a adaptive way to mimic the ways intel arch is doing without getting intel lawyers hot in their pants
It’s the job of the CPU’s
It’s the job of the CPU’s maker to provide the optimization manuals for that CPU maker’s brand of custom microarchitecture. And that custom underlying x86 ISA running microarchitecture from AMD is different from Intel’s custom x86 ISA running microarchitecture at the base hardware level. And each CPU maker has to account for the underlying hardware differences with a proper set of optimization manuals! So with the CPU maker’s optimization manuals in hand the compiler writers(Software Engineers) can create the optimization parts of the compilers to produce the code that will run optimized on a maker’s exact brand of CPU hardware!
Now any x86 CPU can run any x86 ISA based compiled code in an unoptimized state but there can be some performance deficiencies for unoptimized code. AMD will have to create some compilers of its own, but most likely for AMD that will be for some open source based compilers using the current open source code base that has been in use for some time. AMD does not have to create compilers from scratch it really only has to create the optimization component of a compiler’s back end that will parse the instructions and reorder them in such a way as to be optimized for AMD’s brand of x86 ISA running microarchitecture. So a little automated reordering/tweaking of the compiler/assembler generated Op codes to make sure that the code does not cause cache thrashing or other ill effects on AMD’s brand or custom x86 ISA running microarchitecture. Many compilers on the market both proprietary and open source have flags that can be set for both Intel’s and AMD’s optimization specific methods, so that’s not going to be hard to do for Zen and its underlying hardware.
That said most x86 based software should ship with fat binaries that have 2 or more sets of code with one set optimized for Intel’s CPUs and one set optimized for AMD’s CPU’s so any installed software is installed customized to the detected hardware in use. There has been intelligent software installers/linkers/compilers in use for some time now that can detect any CPU’s exact brand, stepping number and enabled feature set via an OS query so even more customization can be had.
It’s your job to ask any reviewer about the version/maker’s brand of any compiler that was used to compile their benchmarking software, or if they received or are using any Intel compiled benchmarking software binary blobs when testing AMD’s hardware. They really should be compiling any test code optimized for the exact CPU hardware that the tests are being run on.
I would never trust any Intel branded compiler to compile any test code or benchmarking software for any AMD CPU tests, and would hope that the reviewer would use only a neutral compiler with the ability to produce optimized code for both Intel’s or AMD’s brand of CPU hardware. Ditto for any binary blobs compiled for AMD’s CPUs using a neutral compiler.
god damit! a few years ago, i
god damit! a few years ago, i was complaining that almost no benchmarks had blender included in it
and now, they’re using blender
now as to why blender? because it’s not optimised, the cycles render engine is not skewed towards intel or amd, and is computationally intensive
this shows the raw compute performance of the CPU, and is probably most representative to the computer server side.
however, you would probably not see this slight advantage in other programs, as they’re optimised for intel
but… as a blender cycles user, i am very sad that i just perchased an 5820k last month…..
No you are not a Blender user
No you are not a Blender user because Cycles uses the GPU for rendering! They are using Blender Render(Uses the CPU) and NOT CYCLES(uses the GPU) on this demo because CPUs are not very good at rendering so Blender Render will stress all cores/threads on a CPU! Blender render is a good benchmark to use because it pegs all of a CPU’s cores/threads at 100% for the entire render!
CPUs suck at rendering compared to GPUs so AMD needs to get that Cycles rendering working on Blender for its GCN GPUs. Full Cycles feature parity with Nvidia’s Cycles rendering is not there yet for AMD but they are working on it with the Blender foundation!
P.S. That render would have
P.S. That render would have been done at the click of the mouse if it had be done on a GPU! CPUs are the MOOKs of the rendering world!
Cycles/other GPU accelerated rendering will free rendering from the ratty grasp of the CPUs once and for all! Wait for the Vega LiquidVR™ demos with Ray Tracing interaction calculations accelerated on the GPU! Free from the tyranny of the CPU at last for all rendering workloads done on the GPU!
Edit: be done
to: been done
Edit: be done
to: been done
Cycles has an option for GPU
Cycles has an option for GPU accelerated rendering, which if you were doing a CPU comparison benchmark you would obviously turn off; the fact that the option exists and should be used if available does not make Cycles exclusively a GPU-accelerated renderer. With GPU acceleration turned off, Cycles will attempt to fully utilize the available hardware the same as the old Blender Internal renderer would.
In the context of comparing two general purpose processors to each other, it’s irrelevant that a specialized processor like a GPU can handle the workloads more efficiently. Apples-to-apples vs apples-to-oranges and whatnot.
I’m looking forward to GPUs
I’m looking forward to GPUs getting dedicated Ray Tracing hardware units and who would want a CPU for Cycles rendering! I have to use the CPU(Quad core Ivy bridge i7) on my laptop because the laptop’s AMD GPU is not GCN. So no Cycles rendering on the GPU for the Terascale based 7650M(rebrand). CPUs take forever to render with the Ray Tracing sample rates turned up and the AO and other feature passes enabled! And yes that Blender Render done on the CPU is good for a benchmark for any CPU! But it’s time to move away from CPUs for rendering and getting the Ray Tracing accelerated on the GPU with its thousands of FP Units is the better way forward.
god damit! a few years ago, i
god damit! a few years ago, i was complaining that almost no benchmarks had blender included in it
and now, they’re using blender
now as to why blender? because it’s not optimised, the cycles render engine is not skewed towards intel or amd, and is computationally intensive
this shows the raw compute performance of the CPU, and is probably most representative to the computer server side.
however, you would probably not see this slight advantage in other programs, as they’re optimised for intel
but… as a blender cycles user, i am very sad that i just perchased an 5820k last month…..
Maybe there will be discounts
Maybe there will be discounts on 5960X or 6900K soon? 🙂
No you are not a Blender user
No you are not a Blender user because Cycles uses the GPU for rendering! They are using Blender Render(Uses the CPU) and NOT CYCLES(uses the GPU) on this demo because CPUs are not very good at rendering so Blender Render will stress all cores/threads on a CPU! Blender render is a good benchmark to use because it pegs all of a CPU’s cores/threads at 100% for the entire render!
CPUs suck at rendering compared to GPUs so AMD needs to get that Cycles rendering working on Blender for its GCN GPUs. Full Cycles feature parity with Nvidia’s Cycles rendering is not there yet for AMD but they are working on it with the Blender foundation!
Actually Blender review show
Actually Blender review show the opposite.
http://www.blendernation.com/2016/02/29/new-cycles-benchmark/
”
Most strikingly so-far is that the performance of CPUs is in a similar range as GPUs, especially when compared to costs of hardware. When shots get more complex, CPUs win the performance battle
”
Result.
GPU – better performance in some scene for preview
CPU – best performance for production quality
Check the result also of a 8core xeon E5 vs a GTX 1080
example, KORO scene: 53 min on GTX 1080, 15 on 3ghz 8core Xeon
A direct link to the data
A direct link to the data please, and not some blog with a bunch of posts. The are qusetions about the server rack of Xeons used as well as the total price of the Xeon SKUs! What is the costs of the Xeons relative to even the high cost of a single GTX 1080! For all we know there are probably a rack of Xeons at $$$$$ compared to a single 1080 GPU. And that’s Nvidia’s SKUs with less FP performance relative to AMD’s SKUs that have more FP performance. The only reason that CPUs are used/needed is for doing the the Ray Tracing interaction calculations, but that LiquidVR™ software from AMD does Ray Tracing interaction calculations on the GPU and not the CPU. There is no mention of that “8core xeon E5” in the link you refrenced(?).
Sure you can throw $20000 dollars worth of CPUs at the rendering problem but at what costs in power relative to what a single GPU can do, your data is not properly refrenced and I’m not searching for it, Direct link to the data you quote.
Did you even bother clicking
Did you even bother clicking the link ?
https://docs.google.com/spreadsheets/d/1rybGWiISHtgaUI-E_DIOM0wf6DW5UG1-p1ooizHimUI/edit?ts=56d095bd#gid=0
As you can see in the Koro scene, it takes a quad GTX 1080 to match render time of an 8 core 3ghz Xeon E5 in some scene. ($400 chip)
https://code.blender.org/2016/02/new-cycles-benchmark/
“When shots get more complex, CPUs win the performance battle. That confirms our own experience that fast GPU is great for previewing and lighting work, and fast CPU is great for the production rendering. ”
If you dont agree with the results and the conclusion, go download the benchmark and run them.
The first link you posted
The first link you posted links to different content(no Link provided), and is different from the link that you have now provided.
I’m not very trusting of any .zip content or Google docs spreadsheet content so you will have to post the text content along with the author of the content’s name and the title of said document! Straight up data without the exact PC hardware’s specifications used in the testing and author/title information is completely suspect. Why is this information not listed in hypertext or text form? Is it from a published article or just from blog posts. I’ll trust mostly properly published info with an Author and article title over any other information/data-sets. I lack the hardware for any Cycles/GPU testing as any Cycles rendering will revert to Blender Render on my systems, even the laptop with the AMD discrete mobile(Non GCN) GPU.
It’s simply not good to reference any unverified unpublished or non peer reviewed content that can not be verified!
I linked to the data,
I linked to the data, benchmark file, report, commentary etc..
All was accessible from the first link.
In contrast you have provided no data yo prove your point.
(that GPU are better are rendering then CPU in Cycle)
So at this time all the FACTS contradict your un-supported claims.
Now, maybe something changed since march 2016. And you are welcome to point to more recent benchmarks/evaluation to show GPU finaly overtook CPU in Cycle.
But I wont do your homework for you.
And if you dont trust Blender developer community and claim they are liars, that is your personal problem. I cant help with that.
Contact : Ton Roosendaal, Chairman Blender Foundation
If you beleive the article/ benchmark report is made-up and false to have it removed from their website.
So I suggest you do your own testing and get it publish “properly” to prove your point. Waiting for a link…
So I think I heard Ryzen demo
So I think I heard Ryzen demo was running at 3.4 GHz with no turbo.. so turbo (when enabled) would be beyond 3.4 GHz..?
Yes, 3.4GHz base clock, turbo
Yes, 3.4GHz base clock, turbo clock to be specified later.
Its premature, but I’m sold
Its premature, but I’m sold on the 8 core ryzen.
AMD might also be able to OWN the render farm business with their 32core model. (Or cause Intel to drop prices by 50%)
8 core/16 threads Intel
8 core/16 threads Intel performance at 95W. Forget my avatar, I just don’t believe it. BS. :p
If you watch the video it was
If you watch the video it was up to 180w intel was at like 190w
Sounds like Kabylake is still
Sounds like Kabylake is still going to defeat Zen for games (mainly due to clock speed and “enough threads”).. But for everything else Zen could be a really good choice — either “fast enough” for light tasks (like any modern processor), or “much faster” for heavily threaded apps , than the 4-core i7 series…
This should make a good server processor..
AMD’s Zen/Ryzen is not
AMD’s Zen/Ryzen is not expected to best Intel’s latest in the single core IPC metric. AMD’s graphics will be better than Intel’s for high end gaming and even APU/SOC integrated graphics as AMD at 14nm will get more CUs/ACEs on its APU based Zen/Vega APU SKUs. There will also be AMD’s APUs built on an Interposer with HBM2 and a separate Zen Cores die and separate larger Vega die for the server/workstation/HPC markets! And most definitely there will be consumer APU on an interposer variants derived from any server/workstation/HPC APU on an Interposer SKUs!
When AMD’s graphics is added into any Zen/Vega AMD APU SKUs versus Intel SOC with Intel’s graphics IP, AMD will be able to best Intel in the overall APU CPU/GPU total performance value metric and the Price/Performance metric! Intel is not Known for using its best graphics IP on its lower cost SOC SKUs! Intel also has no discrete GPU IP to speak of!
AMD will be able to offer Zen/Vega package pricing deals. So AMD can price some very attractive CPU, GPU and motherboard(with AMD’s motherboard partners) package pricing offerings that both Intel and Nvidia can not match. Intel has no discrete GPUs to offer, and Nvidia has no x86 CPUs to offer. Also Nvidia has no motherboard partners/Motherboard chip-set IP for any motherboard/partner pricing deals.
AMD can package price it x86 CPUs, Polaris/Vega GPUs, and with its motherboard partners work up some great package deal pricing CPU, GPU, and motherboard maker deals! There are a few of AMD’s motherboard partners that are also AMD AIB/GPU partners so even better deals can be offered for a whole CPU/GPU/Motherboard deal from some of AMD’s partners and AMD!
If Zen can deliver this much
If Zen can deliver this much performance when it ships then any win Kabylake will win out over Zen on games will be so small as to be negligible. Especially at 1440p or higher resolutions where you’re primarily GPU limited even with the highest end cards today.
Even today there isn’t much difference in gaming between SandyBridge and SkyLake at 1440p or higher resolutions. If you want to limit things to only 1080p then the difference still isn’t all that big. That is part of the reason why so many people are still holding onto SandyBridge CPU’s in their rigs to this day. There just hasn’t been a reason to upgrade given the price Intel was offering for any of their new generation CPU’s since.
Particularly if you overclocked that SandyBridge chip even modestly.
This stream was actually
This stream was actually quite underwhelming, in all honesty. They’ve kept on boasting “matches Intel’s 6900K THAT HAS A PRICE OF 1100 DOLLARS” all the time, but they didn’t reveal the price of competing RyZen solution, so this all was just a big waste of my time mainly only DUE TO THE MERE FACT THAT TechPowerUp OUTRIGHT LEAKED 90% OF RyZen info SEVERAL HOURS BEFORE THE STREAM! Making watching this almost completely POINTLESS! All I wanted is the price, so that I would know for sure if it’s worth it against 6900K price-wise. But I’ve got none, and thus I strongly believe this was just a waste of my time. And also, that “one last thing” of Lisa’s was an ABSO-EFFING-LUTE DISASTER, COMPLETE AND TOTAL FAIL. Battlefront Rogue One? F-U-K-K OFF.
People that post in all caps
People that post in all caps are idiots that don’t know shit about computers. Go back to playing DJ on your MacBook Pro with Touchbar.
You’re not trying hard
You’re not trying hard enough, kiddo.
“DUE TO THE MERE FACT THAT
“DUE TO THE MERE FACT THAT TechPowerUp OUTRIGHT LEAKED 90% OF RyZen info SEVERAL HOURS BEFORE THE STREAM!”
Maybe the NDA had expired and it’s not like the press has not been to any exclusive briefings days earlier! One would think that AMD would want some infromation out there the day of the event so viewers and websites would have a little background information for anyone to refer to until the event could be recorded and copies placed online [In NON FLASH Player encoded formats]. I’m not watching any live event that requires any Adobe Flash Player pawning vector installed on my system just to view such an event!
AMD has yet to discuss the
AMD has yet to discuss the interconnect used between between the Complex(4 full fat Zen cores per complex with only a shared L3) modular units. So when will we know what type of on die complex to complex fabric that AMD is using for Zen.
Well… I guess we will find
Well… I guess we will find out soon enough why they bought seamicro. 😉
Anyone from seamicro still
Anyone from seamicro still working at AMD?
And I thought seamicro was only doing software ?
(some pcie arbitration stuff)
BTW, 300 million seem like a lot to pay for seamicro…
Specially when AMD sold the profitable mobile “ADRENO” division, customers and contracts to qualcomm for 65 million.
AMD only shuttered Sea Micro
AMD only shuttered Sea Micro and AMD still retains the Sea Micro IP, including that SeaMicro “Freedom Fabric” other SeaMicro IP. AMD has had in its IP portfolio plenty of on die CPU, and on die GPU, interconnect IP even before it acquired SeaMicro and SeaMicro’s relevant IP!
It may be this Infinity
It may be this Infinity Fabric tech.https://youtu.be/pN8P6jHAqlU
You are correct though, I dont recall them giving out any details on the complex to complex connection so far..
My Google-Fu attempts has
My Google-Fu attempts has only lead me to this Anandtech article(Page 5) speculation about the CCX to CCX connection fabric!(1) so on page 5 the Anandtech author states:
“It is worth noting that a single CCX has 8 MB of cache, and as a result the 8-core Zen being displayed by AMD at the current events involves two CPU Complexes. This affords a total of 16 MB of L3 cache, albeit in two distinct parts. This means that the true LLC for the entire chip is actually DRAM, although AMD states that the two CCXes can communicate with each other through the custom fabric which connects both the complexes, the memory controller, the IO, the PCIe lanes etc.
One interesting story is going to be how AMD’s coherent fabric works. For those that follow mobile phone SoCs, we know fabrics and interconnects such as CCI-400 or the CCN family are optimized to take advantage of core clusters along with the rest of the chip. A number of people have speculated that the fabric used in AMD’s new design is based on HyperTransport, however AMD has confirmed that they are not using HyperTransport here for Zen. More information on the fabric may come out as we nearer the launch, although this remains one of the more mysterious elements to the design at this stage.”(1)
They are not using HyperTransport but that “Infinity Fabric” MAY just be what is used. So there is still some unknowns as far as the CCX to CCX connection fabric is concerned! AMD will probably save that information until the Zen/Ryzen official release date.
(1)
“AMD Zen Microarchiture Part 2: Extracting Instruction-Level Parallelism” (see page 5)
http://www.anandtech.com/show/10591/amd-zen-microarchiture-part-2-extracting-instructionlevel-parallelism
Likely the SeaMicro Freedom
Likely the SeaMicro Freedom Fabric they picked up – https://www.pcper.com/news/General-Tech/AMD-making-SeaMicro-walk-plank
So here is a snippit of a 3
So here is a snippit of a 3 page PDF from 2012, the 3 page PDF is very interesting read:
“The result is the Freedom™ Supercomputer Fabric, a three-dimensional torus, with both path redundancy and diversity. The fabric is based on tunneling packets using FLITs and route them through a wormhole fabric.”(1)
[Note: AMD is/was contractually bound to support any of the current (at that time) customers after AMD’s winding down of AMD’s SeaMicro business opertions. So some customers are still being supported]
(1)
“SeaMicro Technology Overview
Anil Rao, January 2012
http://www.seamicro.com”
http://seamicro.com/sites/default/files/SM_TO01_64_v2.5.pdf
[Also Note: Wormhole switching as the nearist wikipedia definition of what may be that “wormhole fabric” term listed in the PDF!(1)] (2)
(2)
“Wormhole switching”
https://en.wikipedia.org/wiki/Wormhole_switching
P.S. Jeremy, definitely read
P.S. Jeremy, definitely read that PDF from SeaMicro, there is a lot of innovative technology/IP listed in addition to the Freedom™ Supercomputer Fabric IP. AMD may have shuttered SeaMicro but that IP may still prove to be worth more than what AMD paid for SeaMicro!
Where can i download the
Where can i download the blender file to test.. they mentioned that the files should be available, unless i missunerstood?
http://www.amd.com/en-us/inno
http://www.amd.com/en-us/innovations/new-horizon
nice press review, good
nice press review, good demos, expect to see more after the holiday. at least i can say this chip will not be the boondoggle the FX turned into.
price point- i expect it to be a bit richer than the current FX chips – but it is better than they are. but i also do not expect them to hit the high intel prices.
i am really interested in the MOBO prices- i can usually scrape enough together for a chip= but when the MOBO is almost the same cost or 50 % of the cost- well that is a dealbreaker.
good job so far AMD
Fantastic presentation!
Fantastic presentation! 🙂
Can’t wait to work with a Ryzen+Vega rig!
I JUST KNEW someone was going
I JUST KNEW someone was going to post the ricin connection!!! I have to agree that AMD should have thought of the similar sounding names.
Competetive in Handbrake?
Competetive in Handbrake? x.264 (I assume that’s the encoder they used) has a lot of tuning for Intel processors. Being able to compete evenly with Intel using code optimized for Intel really speak a lot to how fast these must be–or how flexable. Both end up with higher performance for most apps.
I am looking forward to more data on this!
When do you think we will see
When do you think we will see motherboards on sale ?
Time to build a ricin’ PC to
Time to build a ricin’ PC to match the Daewoo I drive.
I am not kidding.
If Intel does react what
If Intel does react what stops them from using their market power to just smaller or no margin to win? This is if the AMD CPU is superior.
Seems like the same people only buy AMD if its a doorbuster deal but always buy Intel at 3x the price for a small bump real world performance will never pay a premium for a superior AMD. That is if you believe forum posters.
“The name [Ryzen] is solid –
“The name [Ryzen] is solid – not earth shattering though I foresee a long life of mispronunciation ahead of it.”
Kaby Lake is in the same boat with a long life of mispronunciation. On various podcasts I’ve heard it pronounced 3 different ways: “kaw-bee”, “kay-bee” and “cab-bee”.
Kaby Lake is an internal code
Kaby Lake is an internal code name though, like Zen. It is not a marketing name. Intel’s marketing name for their processors is just “core”, which they shouldn’t have been able to trademark really. At least Ryzen is a real name and not just a common word.