AMD Confirms 2nd Generation Ryzen Processors to Debut in Q1-2018

[btarunr]

At a press event, AMD confirmed that its 2nd generation Ryzen desktop processors will debut in Q1-2018 (before April). It also clarified that "2nd Generation" does not equal "Zen2" (a micro-architecture that succeeds "Zen"). 2nd Generation Ryzen processors are based on two silicons, the 12 nm "Pinnacle Ridge," which is a GPU-devoid silicon with up to eight CPU cores; and "Raven Ridge," which is an APU combining up to 4 CPU cores with an iGPU based on the "Vega" graphics architecture. The core CPU micro-architecture is still "Zen." The "Pinnacle Ridge" silicon takes advantage of the optical shrink to 12 nm to increase clock speeds, with minimal impact on power-draw.

AMD is also launching a new generation of chipset, under the AMD 400-series. There's not much known about these chipsets. Hopefully they feature PCIe gen 3.0 general purpose lanes. The second-generation Ryzen processors and APUs will carry the 2000-series model numbering, with clear differentiation between chips with iGPU and those without. Both product lines will work on socket AM4 motherboards, including existing ones based on AMD 300-series chipset (requiring a BIOS update). AMD is reserving "Zen2," the IPC-increasing successor of "Zen" for 2019. The "Mattise" silicon will drive the multi-core CPU product-line, while the "Picasso" silicon will drive the APU line. Both these chips will run on existing AM4 motherboards, as AMD plans to keep AM4 as its mainstream-desktop socket till 2020.



View at TechPowerUp Main Site

 

[IceScreamer]

Nice, really looking forward to the new stuff. Also it seems they will be naming the CPUs after painters.

 

[GenericAMDFan]

usually I don't care about typos but there were 2 in this text;

[...]GPU-devioid silicon with up to eight CPU cores[...]

[...]The second-generation Ryzen processors and APUs will carry they 2000-series model numbering[...]

 

[Prima.Vera]

The core CPU micro-architecture is still "Zen."

So basically just a shrink...pfff. Was expecting better tbh...

 

[medi01]

So, AMD's "tock".
I wonder what performance to expect from APUs with 28 CUs. (8 are between 940MX and MX150)

...[piss on AMD's cookies, since it doesn't roll out major architectural changes about one year after triumphant Ryzen release]....

Yay, bad, baaaaad AMD!!!

 

[TheoneandonlyMrK]

So basically just a shrink...pfff. Was expecting better tbh...

Id say better than expected(certainly then i expected , a shrink and frequency boost , wouldn't mind that at all but then 12nm is probably just their name for 14nm+++ because I don't recall this 12nm node being talked about before.

 

[Mirkoskji]

Id say better than expected(certainly then i expected , a shrink and frequency boost , wouldn't mind that at all but then 12nm is probably just their name for 14nm+++ because I don't recall this 12nm node being talked about before.

Nvidia volta will be manufactured in12nm. Current Titan V is 12nm

 

[Shamalamadingdong]

Nvidia volta will be manufactured in12nm. Current Titan V is 12nm

Not the same though.
Nvidia uses TSMC and AMD uses Globalfoundries. And their process nodes are very different which also becomes apparent when looking at how Nvidia's GPUs behave when manufactured on TSMC's 16nm process versus on Samsung's 14nm process which Globalfoundries' process is based on.

Id say better than expected(certainly then i expected , a shrink and frequency boost , wouldn't mind that at all but then 12nm is probably just their name for 14nm+++ because I don't recall this 12nm node being talked about before.

Globalfoundries announced it sometime ago. Like many other node announcements, this is a small improvement to 14nm. Hardly even a half-node. It offers higher frequencies and some reduction in area. It's better tuned for performance as 14nm was based on Samsung's 14nm LPP which is made for low power/mobile.

 

[notb]

So basically just a shrink...pfff. Was expecting better tbh...

I'm waiting for comments like "5% improvement with each generation! AMD is awesome!"

 

[Imsochobo]

Id say better than expected(certainly then i expected , a shrink and frequency boost , wouldn't mind that at all but then 12nm is probably just their name for 14nm+++ because I don't recall this 12nm node being talked about before.

12NM node is talked about.

So basically just a shrink...pfff. Was expecting better tbh...

their APU's contain many many improvements over regular ryzens which should translate to lower power consumption, higher single->dual-Tri->quad core boosts ( maybe 100 mhz more on existing node)
Then add the new process and probably other arch features they added with APU's that translate to maybe 1-2% ipc + couple hundred mhz clock = good.
The memory latencies seems to be slightly improved with their apu's but I wouldn't say it for certain as there is only one CCX in apu and two in ryzen.

I think it'll compete better with intel in games while providing their value and core count.

 

[EarthDog]

Yay, bad, baaaaad AMD!!!

in the words of Little Wayne... "Dont start no sh1t, there wont be no sh1t".

Anyway, it should be interesting to see how much the clocks increase amd how much further, if any, it overclocks. If i were to guess, id say 200-300 mhz clock increase, but overclocking remains the same, barely past boost.

 

[newtekie1]

Anyway, it should be interesting to see how much the clocks increase amd how much further, if any, it overclocks. If i were to guess, id say 200-300 mhz clock increase, but overclocking remains the same, barely past boost.

Yep, AMD is already pretty much pushing these things to their max with the stock clock speeds.

 

[bug]

And here we are, reading news sourced from wtftech

 

[laszlo]

usually I don't care about typos but there were 2 in this text;

[...]GPU-devioid silicon with up to eight CPU cores[...]

[...]The second-generation Ryzen processors and APUs will carry they 2000-series model numbering[...]

you can send a PM anytime instead of pointing .... is more polite don't you think?

however seems you care a lot about typos made by others instead of yours, in the four posts you have....

 

[TheoneandonlyMrK]

Not the same though.
Nvidia uses TSMC and AMD uses Globalfoundries. And their process nodes are very different which also becomes apparent when looking at how Nvidia's GPUs behave when manufactured on TSMC's 16nm process versus on Samsung's 14nm process which Globalfoundries' process is based on.


Globalfoundries announced it sometime ago. Like many other node announcements, this is a small improvement to 14nm. Hardly even a half-node. It offers higher frequencies and some reduction in area. It's better tuned for performance as 14nm was based on Samsung's 14nm LPP which is made for low power/mobile.

Which is why I implied it was 14nm+++ they reduced only one gate dimension to 12nm at best and greatly improved some other geometries to gain an alleged 30-40% die area improvement but its not a true new node like 10nm is and 7nm will be as you know

Err as far as i remember the 14nm+++ got talked about then forgotten about as it got renamed 12nm ,or that's how i percieved it but who knows what dimensions it's really packing these days.

 

[bug]

Which is why I implied it was 14nm+++ they reduced only one gate dimension to 12nm at best and greatly improved some other geometries to gain an alleged 30-40% die area improvement but its not a true new node like 10nm is and 7nm will be as you know

What's in a name?
What matters here is the price. At current prices, Ryzen is a compelling alternative to Kaby/Coffee Lake. A newer process automatically means higher frequencies (it remains to be seen how much higher) that will hopefully bring the performance at the same level as Intel's. But if the yields are poor and prices go up, things become more complicated.

 

[mouacyk]

I'm waiting for comments like "5% improvement with each generation! AMD is awesome!"

Jumping 50% improvement between two generations to nearly catch up to your competitor takes some tinted glasses to appreciate. It remains to be seen how well improvements scale from here on.

 

[Kaotik]

Seriously, TPU news quality is going down in a spiral and no-one is doing anything about it?

Even if the image isn't photochopped, it's from before Raven Ridge launch (note the launched vs launches). There has been no recent AMD even where they would have confirmed or said anything about them so at best this should be really, really old news and it should be mentioned that if they actually said something about Pinnacle Ridge, it happened ages ago.
You should also make note that Raven Ridge is 14nm since you specify that Pinnacle is 12nm.
Oh and that other image isn't confirmed to be real either, it could be fake just aswell.

 

[notb]

Jumping 50% improvement between two generations to nearly catch up to your competitor takes some tinted glasses to appreciate. It remains to be seen how well improvements scale from here on.

As I've mentioned many times: Intel had a similar performance improvement in the same period. So it's really nothing special.

 

[efikkan]

As expected; a new stepping, a small clock bump, some tweaks and some bug fixes.

Nvidia volta will be manufactured in12nm. Current Titan V is 12nm

These are not new nodes, just refinement of existing ones.

 

[Slizzo]

So basically just a shrink...pfff. Was expecting better tbh...

I bet there's probably some microarchitecture clean up that will have been done since the launch of Ryzen. Opportunity is there, and it'd be a shame for them to miss being able to implement what they've learned since launch.

If we can see IPC go up 5%-10%, and then see clocks rise to around 4.5GHz this will be very competitive.

 

[OSdevr]

I suppose we'll get a second generation ThreadRipper as well.

 

[efikkan]

I bet there's probably some microarchitecture clean up that will have been done since the launch of Ryzen. Opportunity is there, and it'd be a shame for them to miss being able to implement what they've learned since launch.

If we can see IPC go up 5%-10%, and then see clocks rise to around 4.5GHz this will be very competitive.

You'll have to wait for "Zen2" for IPC changes. There is too little time to overhaul anything for this stepping.

 

[bug]

I bet there's probably some microarchitecture clean up that will have been done since the launch of Ryzen. Opportunity is there, and it'd be a shame for them to miss being able to implement what they've learned since launch.

If we can see IPC go up 5%-10%, and then see clocks rise to around 4.5GHz this will be very competitive.

If there were architectural changes, AMD would be touting them all over the place. The term "optical shrink" used in the article means no architectural changes. But you;re free to bet on whatever you want, of course.

 

[TheLaughingMan]

I bet there's probably some microarchitecture clean up that will have been done since the launch of Ryzen. Opportunity is there, and it'd be a shame for them to miss being able to implement what they've learned since launch.

If we can see IPC go up 5%-10%, and then see clocks rise to around 4.5GHz this will be very competitive.

It is already competitive now, but could be much better in gaming. In this case you are not going to get an IPC gain + 500 MHz on the clock. We will likely see clock improvement 200 to 400 MHz which will net a 5% to 10% gain in performance. So of course it is a architecture clean-up which is what you due with 2nd gen chips as it has always been with both AMD and Intel.

What they need to focus on is the memory compatibility, latency, and infinity fabric performance.