The new 12LP technology provides as much as a 15 percent improvement in circuit density and more than a 10 percent improvement in performance over 16/14nm FinFET solutions on the market today. This positions 12LP to be fully competitive with other 12nm FinFET foundry offerings. The technology leverages GF’s expertise at Fab 8 in Saratoga County, N.Y., where its 14nm FinFET platform has been in high-volume production since early 2016.
“We are pleased to extend our longstanding relationship with GLOBALFOUNDRIES as a lead customer for their new 12LP technology,” said Mark Papermaster, CTO and senior vice president of technology and engineering, AMD. “Our deep collaboration with GF has helped AMD bring a set of leadership high-performance products to market in 2017 using 14nm FinFET technology. We plan to introduce new client and graphics products based on GF’s 12nm process technology in 2018 as a part of our focus on accelerating our product and technology momentum.”
Is this the final name of GF 14+ that people talk about for months?
”
Zen and Vega 11 / Vega 10 refresh on 12LP is pretty much confirmed. >10% perf improvement should bode well for AMD's max frequency on Zen+ in 2018. I think 4.5 - 4.6 Ghz avg OC should be possible on Ryzen 2000 series.
They both are. Also, is it the process, the design, or the design on the process. You need to be careful on just saying the process is inferior outright, as other things can potentially be at play.So... will AMD still use TSMC's 16nm FF for Vega 20? How good is this 12nm process? It is clear that GF's 14nm is inferior to TSMC's current 16nm... I also thought TSMC was going to go to 12nm too.
Here you go.Thanks for posting.
Edit: the link doesn't work.
I don't know what generation is supposed to be vega 20 but I hardly doubt they will use TSMC 16nm for that. TSMC 16nm has some advantage over GF in performance but it is a bit less dense too. If GF has this node ready for early 2018 all high end products from AMD will be manufactured at 12nmLP.So... will AMD still use TSMC's 16nm FF for Vega 20? How good is this 12nm process? It is clear that GF's 14nm is inferior to TSMC's current 16nm... I also thought TSMC was going to go to 12nm too.
LP matches the nomenclature for 7nm. This could be the IBM process as a stepping stone to 7nm, with practice honing the difference in processes.I don't know what generation is supposed to be vega 20 but I hardly doubt they will use TSMC 16nm for that. TSMC 16nm has some advantage over GF in performance but it is a bit less dense too. If GF has this node ready for early 2018 all high end products from AMD will be manufactured at 12nmLP.
I would not press much attention to noneclatures/names of process nodes haha... They are very misleading.LP matches the nomenclature for 7nm. This could be the IBM process as a stepping stone to 7nm, with practice honing the difference in processes.
That they are. But they would likely have to extend the Samsung license or IBM license as 12nm previously has not been disclosed. Considering it is easier to practice with the smaller node on a larger process, the potential use of IBM process seems practical.I would not press much attention to noneclatures/names of process nodes haha... They are very misleading.
LP matches the nomenclature for 7nm. This could be the IBM process as a stepping stone to 7nm, with practice honing the difference in processes.
That they are. But they would likely have to extend the Samsung license or IBM license as 12nm previously has not been disclosed. Considering it is easier to practice with the smaller node on a larger process, the potential use of IBM process seems practical.
I think its highly likely that 12LP was developed by the IBM semiconductor team which was acquired by GF in 2015.
http://www-03.ibm.com/press/us/en/pressrelease/45110.wss
https://www.globalfoundries.com/new...-acquisition-of-ibm-microelectronics-business
https://arstechnica.com/gadgets/2015/07/ibm-unveils-industrys-first-7nm-chip-moving-beyond-silicon/
14LPP was the last officially licensed Samsung process by GF. I think 12LP and 7LP are developed by GF with the resources gained from the IBM acquisition.
12LP is a tweak of 14nm LPP.
Is there a video to watch about the presentation?
They have another custom 14nm HP ( high performance ) for IBM.
https://www.globalfoundries.com/new...custom-14nm-finfet-technology-for-ibm-systems
Is there a video to watch about the presentation?
They have another custom 14nm HP ( high performance ) for IBM.
https://www.globalfoundries.com/new...custom-14nm-finfet-technology-for-ibm-systems
This doesn't mean it isn't using a tweaked 7nm design. Just that IBM is offering a 14nm SOI finFET. There is nothing to say the 7nm is an SOI finFET. In fact, it only says it is a finFET. So nothing is ruled out, just shifts probability on whose process on 12nm is used. Just saying.That's really too bad for AMD, that 10% perf boost is exactly what the doctor ordered vis-a-vis Ryzen. That would really have kept up the Ryzen excitement momentum among DIYers.
I didn’t say anything about SOI. Since 12LP is based on 14LPP, it's clearly bulk Si.This doesn't mean it isn't using a tweaked 7nm design. Just that IBM is offering a 14nm SOI finFET. There is nothing to say the 7nm is an SOI finFET. In fact, it only says it is a finFET. So nothing is ruled out, just shifts probability on whose process on 12nm is used. Just saying.
That's the "money is no issue, give me every MHz you can" node. No doubt far too expensive to use for AMD products.
Risk production in 1H 2018, so realistically volume production in 2H 2018.