AMD Details ZEN Microarchitecture IPC Gains

[Cvrk]

then there's the price.
i hope is not gonna be more expensive than i5 6600k . from what i read, 258$ is the expected price

 

[EarthDog]

$260 for a 8c/16t CPU that performs as good as Intel? No way in hell.

 

[JrockTech]

$260 for a 8c/16t CPU that performs as good as Intel? No way in hell.

Way. The value train is coming.

 

[Fx]

Eh. I don't think that 40% over excavator is being too optimistic.

I don't either, but I'd rather choose to be pessimistic and then be treated with 30-40% improvement. That would really make me feel good about dropping all of that money on a new system that I will be using for another 2-4 years.

most people here understand that whatever benchmarks come out of a company's marketing division are going to be pr crap.

Aye.

 

[EarthDog]

Way. The value train is coming.

Only time will tell. If AMD were halfway intelligent about it, they would raise prices while still undercutting intel comparables by a decent amount (10%-20% or so) This will increase both desktop market share and profits...

If I had to guess... the flagship zen processor with 8c/16t and IPC within 5% of intel, you are looking ~$399. That's undercutting intel 6900K by a ton and will be a HUGE win for AMD and consumers in general. Again, no way they release at $260 if they are that comparable to Intel.

Consider this: 6700K costs what, $350 (MSRP)? Who (in their right mind) wouldn't pay for strikingly similar performance but with 8 'real' cores for $400? If you can't spend that cash, get the Hex or Octo priced less...

... but $260? No friggin way.

 

[qubit]

massive 40 percent gain in IPC (translated roughly as per-core performance gains), over the current "Excavator" microarchitecture.

Sounds like this is improved enough to compete with Intel's current gen and Kaby Lake. Let's hope the performance of the overall product is good enough to give Intel a kicking.

I can't believe my almost 7 year old Sandy Bridge is still fast, fluid and more or less current tech. Its performance and features should have been completely eclipsed by modern CPUs from both companies by now. A competitive AMD will help to finally make it obsolete, giving me a compelling reason to finally upgrade.

 

[TheGuruStud]

Sounds like this is improved enough to compete with Intel's current gen and Kaby Lake. Let's hope the performance of the overall product is good enough to give Intel a kicking.

I can't believe my almost 7 year old Sandy Bridge is still fast, fluid and more or less current tech. Its performance and features should have been completely eclipsed by modern CPUs from both companies by now. A competitive AMD will help to finally make it obsolete, giving me a compelling reason to finally upgrade.

For realzies. Sandy ppl have been waiting far too long. AMD could definitely steal some enthusiast sales when they upgrade.

 

[BiggieShady]

How "lower power" translate into IPC gain?

That would means an efficiency or performance gains, which is reasonable since the slideshows are actually never indicate anything about these improvements results in IPC gain.

Or maybe lower power allows them to use more complex cores.

It does not result directly in IPC gain, but when you focus on low power design methodologies in core components you have a greater freedom in silicon layout and ultimately (you are right) it allows them to make more complex cores that require less voltage.
Also, when you are generous with the fast cache you are automagically increasing IPC and lowering power at the same time.



I wish they've said something about zen's IMC

 

[Totally]

Only time will tell. If AMD were halfway intelligent about it, they would raise prices while still undercutting intel comparables by a decent amount (10%-20% or so) This will increase both desktop market share and profits...

If I had to guess... the flagship zen processor with 8c/16t and IPC within 5% of intel, you are looking ~$399. That's undercutting intel 6900K by a ton and will be a HUGE win for AMD and consumers in general. Again, no way they release at $260 if they are that comparable to Intel.

Consider this: 6700K costs what, $350 (MSRP)? Who (in their right mind) wouldn't pay for strikingly similar performance but with 8 'real' cores for $400? If you can't spend that cash, get the Hex or Octo priced less...

... but $260? No friggin way.

They tried that I read an article not too long ago to sum it up they stated marketshare and it mindshare is what is important and it not simply a matter of making a product that equals or betters the competitor and undercutting them. Consumers tend to stick their perceptions and what they are used to .i.e. just because amd is now making a better product in their eyes Intel still makes a good product.

tl;dr AMD has the reputation of being a poor man's intel/value chip, they need to change that image before things like price wars have any benefical gains in the long run. People who say that they vote with their wallets are full of crap or are limited to the enthusiast population which are a small %.

 

[cyneater]

Lets hope there are decent motherboards in decent supply to go with the chips.

was it the 939 or am2 that had a lack of motherboards when it came out... thus alot of people went to intel.

 

[EarthDog]

Lol, I don't ever recall a lack of motherboards, lol... I dont imagine this to be a problem.

 

[dyonoctis]

What I fear is that even if zen is good, that might not be enough to put AMD back on rails:

• Desktop pc's are not selling as much as they were ten years ago, Lot's of people are just fine with a laptop. (and 99% of the laptop I'm seeing are intel inside.)
• Intel got a better marketing image. Even in the pentium 4 era a lot of people who claimed to be aware of computer hadware, were saying that the pentium 4 was the best thing you could get. That's the power of marketing.
• Gamers, creative content creator, and other fields that need great cpu power may be the market were zen have the most money potential. However the creative/professional sector is dominated by Intel xeon. I have yet to see a Z workstation/dell precision/lenovo thinkcentre featuring an opteron). And even for the guy who want to make a professional system himself, it's so much easier to put together a xeon system. I haven't seen any post about guys bragging just how much a 400$ 12 cores opteron is a great value for multithreaded task.
• Apple may help AMD getting a better standing in the eyes of the basic buyers, but I really don't see them ditching intel cpu... I'm not even sure that they'll stay with Amd gpu, seeing how pascal is efficient.

 

[BiggieShady]

@dyonoctis What they are saying in the first slide is that the same zen core scales all the way down to 10-20 W for laptops ... it shouldn't be bad in an APU with a radeon gpu

 

[TheGuruStud]

@dyonoctis What they are saying in the first slide is that the same zen core scales all the way down to 10-20 W for laptops ... it shouldn't be bad in an APU with a radeon gpu

Which should mean powerful quadcore. None of this stupid dual core intel shit.

 

[medi01]

Remember Larrabee from intel ? We could have though that this thing was going to be awesome because intel cpu are good. Yet intel choose to give up on making this a gpu because it couldn't compete with what amd and nvidia were offering.

Larrabee was not a GPU.
It was an attempt to build something rather different, with x86 instruction set and idea that specialized hardware for z-buffering et all is not needed and it's better done in software.

It was basically a bunch of (simpler) x86 cores.

It would trounce usual GPUs at stuff like ray tracing and yadayada... but in the end Intel abandoned the idea as it didn't quite perform as expected.

 

[BiggieShady]

Which should mean powerful quadcore. None of this stupid dual core intel shit.

Don't know about that, quadcore cluster is made perfectly symmetrically - it could be cut vertically to make dualcores with four threads ... or laser cut off non working cores to improve yields

 

[TheGuruStud]

Don't know about that, quadcore cluster is made perfectly symmetrically - it could be cut vertically to make dualcores with four threads ... or laser cut off non working cores to improve yields

For cheapness, dual, but quad should become mainstream. Can't find a quad i7 unless you sell a kidney.

 

[BiggieShady]

For cheapness, dual, but quad should become mainstream. Can't find a quad i7 unless you sell a kidney.

Here's for disruptive technologies

 

[dyonoctis]

Larrabee was not a GPU.
It was an attempt to build something rather different, with x86 instruction set and idea that specialized hardware for z-buffering et all is not needed and it's better done in software.

It was basically a bunch of (simpler) x86 cores.

It would trounce usual GPUs at stuff like ray tracing and yadayada... but in the end Intel abandoned the idea as it didn't quite perform as expected.

Yes, it wasn't a gpu on a technologic level ,but they wanted to sell this as a 3D gpu for the basic consumer. In the end they had to give up on that, and made it a coprocessor, a "GPGPU" like product. lt doesn't change the fact that intel tried to get against Nvidia/Amd high performance chip, and failed because they have little experience in that domain. I mean, they could have just tried to make a super beefy Intel graphic HD, but instead they went with x86, and it took them a while to realise that it wasn't going to work that well in video games.

 

[medi01]

lt doesn't change the fact that intel tried to get against Nvidia/Amd high performance chip, and failed because they have little experience in that domain

Except they NEVER have tried that "in that domain".
Larrabee is not "in that domain" and there was no company on this planet with experience "in that domain" at all, Larrabee was that unique.

 

[dyonoctis]

Except they NEVER have tried that "in that domain".
Larrabee is not "in that domain" and there was no company on this planet with experience "in that domain" at all, Larrabee was that unique.

I'm not saying that larabee wasn't a great product. It's a wonderfull coprocessor/"GPGPU like, BUT for a mainstream gaming hardware wich was one of the key points they had when developing larabee, the choices that they made weren't great. Nvidia engineers weren't agrree at all with intel choices, they predicted that it wasn't going to work as a gpu :

"They've put out a certain amount of technical disclosure in the past five weeks," he noted, "but although they make Larrabee sound like it's a fundamentally better approach, it sn't. They don't tell you the assumptions they made. They talk about scaling, but they disregard memory bandwidth. They make it sound good, but we say, you neglected half a dozen things."

"Every GPU we make, we always consider this type of design, we do a reasoned analysis, and we always conclude no. That's why we haven't built that type of machine."

"Intel is not a stupid company," he conceded. "They've put a lot of people behind this, so clearly they believe it's viable. But the products on our roadmap are competitive to this thing as they've painted it. And the reality is going to fall short of the optimistic way they've painted it."

"There's an incredible amount about Larrabee that's undefined," explained Keane, commenting on the specifications so far released. "You can't just say 'it's x86 so it's going to solve the massively parallel computing problem.'"

"Look at the PC," he continued. "With an OS they don't control, and applications coming from everywhere... to say arbitrarily that everything's going to scale to 32 cores seems to me to be a bit of a stretch. "

source: http://www.alphr.com/news/home-and-leisure/220947/larrabee-like-a-gpu-from-2006

 

[medi01]

Nvidia engineers weren't agrree at all with intel choices

They were scared to shit (imagine, if ray tracing and other promises would work... GG GPU manufactureres) and pissed on competitors cookies, hardly hardly an argument.

A lot was uncertain, they tried and failed and I applaud Intel for trying.

But that was not the point. You were drawing "Intel can't do competitive high end GPU"'s from Larrabee fiasco. Which is a correct conclusion drawn from irrelevant fact.

Iris/HD GPUs show what they are capable of on GPU front (nothing impressive, although they keep improving)

 

[Slizzo]

Only reason why I want to upgrade from my Sandy Bridge system is that it's an early SB system. I have a first gen Z68 board with PCI-E 2.0 slots and what not. I want to make sure I have USB 3.1, PCI-E 3.0 and DDR4 memory.

If I can get a Zen processor that is 6c12t or 8c16t for around $350, I'm sold.

 

[dyonoctis]

They were scared to shit (imagine, if ray tracing and other promises would work... GG GPU manufactureres) and pissed on competitors cookies, hardly hardly an argument.

A lot was uncertain, they tried and failed and I applaud Intel for trying.

But that was not the point. You were drawing "Intel can't do competitive high end GPU"'s from Larrabee fiasco. Which is a correct conclusion drawn from irrelevant fact.

Iris/HD GPUs show what they are capable of on GPU front (nothing impressive, although they keep improving)

It wasn't just larabee, Intel also released the Intel i740. At launch it was actually a decent gpu, but intel didn't manage to follow trough. The reason why I'm saying this is because Intel showed twice their ambitions to get in the gaming gpu market, but never succeded. Yet they have money, and a great foundry (maybe the best). When larabee failed, they just went on with the Xeon phi, and seems to have given up on middle range/end high gaming.

Yes intel is making great progress with their igpu, but they are still stuck in the entry level market. Seeing at what the gaming gpu market looks like, it's not like they don't have a chance with a good product. When everyone was stuck on the 28 nm, intel had a great 22nm and 14nm rolling out. Unless intel don't care anymore about the gaming segment, the only reason I can come up with, is that intel staff can't find idea that would be competitive with Amd/Nvidia. If intel could actually make something competitive even with a RX 470 or a GTX 1060, I would be happy. More competition is always good.

 

[P4-630]

A bit off topic but I was just reading this:

"The Plus Side of AMD:
Some benefits that AMD processors come with are that they run cooler in your system and are then more quiet and use up less power."



http://hubpages.com/technology/best-amd-processor