By: Michael S (already5chosen.delete@this.yahoo.com), June 13, 2022 5:20 am
Room: Moderated Discussions
Rayla (rayla.delete@this.example.com) on June 13, 2022 5:56 am wrote:
> Michael S (already5chosen.delete@this.yahoo.com) on June 13, 2022 5:13 am wrote:
> > David Kanter (dkanter.delete@this.realworldtech.com) on June 12, 2022 7:05 pm wrote:
> > > Hi Everyone,
> > >
> > > Happy Sunday! I am thrilled to bring you my latest article, which focuses on the Intel
> > > 4 process that was presented at the VLSI Symposium in Hawaii. Here's a quick summary:
> > >
> > > The Intel 4 process achieves 20% better performance and scales logic density by 2X while reducing costs
> > > through extensive design co-optimization, adoption of new materials, and judicious use of EUV lithography.
> > > The first product, the Meteor Lake compute tile will ramp to high volume manufacturing in 2023.
> > >
> > > Read more here: Intel 4 Process Scales Logic with Design, Materials, and EUV
> > >
> > > And as always, feel free to discuss or drop questions in this thread!
> > >
> > > David
> > >
> >
> > Isn't it strange that Intel does not offer a foundry node based on their 14nm process?
> > I was under impression that 14nm has the lowest cost per transistor
> > of all processes Intel ever had and also the highest fabs capacity.
> >
> >
>
> They do. You can even order it on MOSIS. The article mentions it, too (though
> calls it Intel 16 for some reason, a term that Intel AFAIK doesn't use.)
Article says that Intel 16 is what they called previously 22FFL - not the same process that used for 5 or 6 generations of Skylake. It has similar fins shape to Skylake, but much lower density. According to my understanding, the density of 22FFL is lower than that of process used for Haswell. I am not sure how it compares to even older 22nm variant used for Ivy Bridge.
> Michael S (already5chosen.delete@this.yahoo.com) on June 13, 2022 5:13 am wrote:
> > David Kanter (dkanter.delete@this.realworldtech.com) on June 12, 2022 7:05 pm wrote:
> > > Hi Everyone,
> > >
> > > Happy Sunday! I am thrilled to bring you my latest article, which focuses on the Intel
> > > 4 process that was presented at the VLSI Symposium in Hawaii. Here's a quick summary:
> > >
> > > The Intel 4 process achieves 20% better performance and scales logic density by 2X while reducing costs
> > > through extensive design co-optimization, adoption of new materials, and judicious use of EUV lithography.
> > > The first product, the Meteor Lake compute tile will ramp to high volume manufacturing in 2023.
> > >
> > > Read more here: Intel 4 Process Scales Logic with Design, Materials, and EUV
> > >
> > > And as always, feel free to discuss or drop questions in this thread!
> > >
> > > David
> > >
> >
> > Isn't it strange that Intel does not offer a foundry node based on their 14nm process?
> > I was under impression that 14nm has the lowest cost per transistor
> > of all processes Intel ever had and also the highest fabs capacity.
> >
> >
>
> They do. You can even order it on MOSIS. The article mentions it, too (though
> calls it Intel 16 for some reason, a term that Intel AFAIK doesn't use.)
Article says that Intel 16 is what they called previously 22FFL - not the same process that used for 5 or 6 generations of Skylake. It has similar fins shape to Skylake, but much lower density. According to my understanding, the density of 22FFL is lower than that of process used for Haswell. I am not sure how it compares to even older 22nm variant used for Ivy Bridge.