News Intel Completes Development of 1.8nm and 2nm Production Nodes

[Admin]

Intel finalizes specifications of its Intel 18A and Intel 20A process technologies.

Intel Completes Development of 1.8nm and 2nm Production Nodes : Read more

 

[zecoeco]

I don't see it coming to consumer products before at least 2025.
Plus, delays are always possible...

 

[jkflipflop98]

Buy that INTC now while it's on sale.

 

[waltc3]

Wake me when they begin shipping CPUs in 1 to 2 nm.... Intel has been absolutely super during the last five years at announcing things for the future that for one reason or another never materialize.

 

[thisisaname]

Wake me when they begin shipping CPUs in 1 to 2 nm.... Intel has been absolutely super during the last five years at announcing things for the future that for one reason or another never materialize.

Aye research is the "easy" part putting it into mass production is a lot harder.

 

[Kona45primo]

They have had terrible yeilds on 14 & 10nm and missed the ARC deadline by a solid year... "The earliest Arc products will be released in "the first quarter of 2022" and will be based on a GPU codenamed "Alchemist," "

That being said if they can get this to actually turn into viable production before it's outdated that would be amazing.

 

[InvalidError]

"Power through the backside" sounds like something that could be a pain in the ass to implement. Sounds like it is going to require a crapton of vias, a suspicion further reinforced by "PowerVias" being a key point.

I suppose if you unflip flip chips so they lie fat-silicon-down using TSVs for power and substrate IO, that makes the active electronics ~0.2mm closer to 3D-stacked stuff on top and the IHS for heat dissipation.

 

[Giroro]

This is "Intel 20A", not "20 angstrom", "Angstrom class", nor 2nm. The branding is designed to trick people into thinking it's related to size, when really it no longer has connection to any kind of sizing, whatsoever.
It's not as bad as when comcast writes Mbps as MBPS or when a phone salesman straight tells you that Mbps stands for "Megabytes per second" which would be 8x faster than what they are actually selling ... but it's almost as bad. I'd still put it in the to 10 computer marketing lies. I'd say it's slightly better than Seagate confusing a judge into letting them redefine the measurement 1 GigaByte as 1 Billion Bytes, and slightly worse than Intel letting their partners pretend Optane SSDs counted as RAM. So, and about the same badness as Intel's P/E-core mismarketing nightmare.

That said I'm already getting lost on what TSMC process Intel 20A is supposed to compete against. IIRC It will go up against and probably slightly behind TSMC N4 "4nm".
If I remember right, Intel wishes they could be competitive with the following:

Intel	TSMC
10 nm = Intel 7	"7nm "N7
7nm = Intel 4	NFF7+ or N6
Intel 3	N5
Intel 20A	N5+ or N4
Intel 18A	N3

 

[JamesJones44]

This is "Intel 20A", not "20 angstrom", "Angstrom class", nor 2nm. The branding is designed to trick people into thinking it's related to size, when really it no longer has connection to any kind of sizing, whatsoever.
It's not as bad as when comcast writes Mbps as MBPS or when a phone salesman straight tells you that Mbps stands for "Megabytes per second" which would be 8x faster than what they are actually selling ... but it's almost as bad. I'd still put it in the to 10 computer marketing lies. I'd say it's slightly better than Seagate confusing a judge into letting them redefine the measurement 1 GigaByte as 1 Billion Bytes, and slightly worse than Intel letting their partners pretend Optane SSDs counted as RAM. So, and about the same badness as Intel's P/E-core mismarketing nightmare.

That said I'm already getting lost on what TSMC process Intel 20A is supposed to compete against. IIRC It will go up against and probably slightly behind TSMC N4 "4nm".
If I remember right, Intel wishes they could be competitive with the following:

Intel	TSMC
10 nm = Intel 7	"7nm "N7
7nm = Intel 4	NFF7+ or N6
Intel 3	N5
Intel 20A	N5+ or N4
Intel 18A	N3

I'm no fan of Intel, but TSMC does the same thing. N5, N5+ and N4 are actually the same node size. There isn't a great way to compare the two with current metrics TBH. Density would be a better way, though still not prefect, but I doubt we will get a better apples to apples way to compare nodes anytime soon.

 

[icmn223]

"Wang Rui, president and chairman of Intel China, said at an event that the company had finalized the development of its Intel 18A (18 angstroms-class) and Intel 20A (20 angstroms-class) fabrication processes."
Wait, does that mean they are developing/manufacturing this in China? If so, isn't that a national security threat? I thought we were trying to keep the latest and greatest semiconductor tech. out of their hands for precisely that reason.

 

[bit_user]

Nice work, @jkflipflop98 !

My concern about skipping 0.55 numerical aperture naturally goes to costs, but I'm sure the bean counters will have run the numbers and checked that it's both economically viable and competitive.

In the nearer term, my main concern is with voltage and frequency scaling on Intel 4, but I guess we'll find out before long.

 

[Jimbojan]

This will put AMD way behind, Intel 4 is already better than TSMC (5nm) so is AMD CPU in terms of power efficiency.

 

[bit_user]

I don't see it coming to consumer products before at least 2025.

FWIW, Arrow Lake is scheduled to launch in 2024, using the 20A node.

 

[bit_user]

I'm already getting lost on what TSMC process Intel 20A is supposed to compete against. IIRC It will go up against and probably slightly behind TSMC N4 "4nm".
If I remember right, Intel wishes they could be competitive with the following:

Intel	TSMC
10 nm = Intel 7	"7nm "N7
7nm = Intel 4	NFF7+ or N6
Intel 3	N5
Intel 20A	N5+ or N4
Intel 18A	N3

I think it's more like:

Intel (old)	Intel (new)	TSMC
14 nm
10 nm
10 nm (SF)
10 nm (ESF)	7	N7/N6
7 nm	4	N5
	3	N4
	20A	N3
	18A	N2

 

[isofilm]

Pat Gelsinger secretes dishonesty, like a slug secretes slime.

 

[bit_user]

Wait, does that mean they are developing/manufacturing this in China?

No. China can still import & sell most Intel products to most buyers, and I think there are still some Intel production facilities there.

As long as Intel wants to continue both selling into the Chinese domestic market and to China-based OEMs (e.g. Lenovo) for export to markets around the world, they need to communicate with the market there, just like they do elsewhere. That's why he's talking about Intel's progress and their roadmap.

 

[bit_user]

Pat Gelsinger secretes dishonesty, like a slug secretes slime.

Personal attacks aren't helpful or appropriate. Much better to cite specific examples, if you have them.

The CEO is the public face of a company, and they're always going to be putting a spin on things. I haven't found Pat to be worse than I'd expect, in terms of spin. I'm still hoping he'll be successful in reinvigorating Intel. I believe that's entirely why he went back. I'm sure he could've simply retired, if he didn't believe in the cause.

 

[-Fran-]

So, is this going to be 10nm+++++ or 7nm++++?

Regards

 

[lenslens007]

Looks like vaporware to me. By the time Intel brings this to manufacture (if ever) others will still be ahead.

 

[RichardtST]

Well, they better hurry up and get it right this time. AMD is gonna need that extra capacity for their 70003D chips.

 

[shady28]

This is "Intel 20A", not "20 angstrom", "Angstrom class", nor 2nm. The branding is designed to trick people into thinking it's related to size, when really it no longer has connection to any kind of sizing, whatsoever.
It's not as bad as when comcast writes Mbps as MBPS or when a phone salesman straight tells you that Mbps stands for "Megabytes per second" which would be 8x faster than what they are actually selling ... but it's almost as bad. I'd still put it in the to 10 computer marketing lies. I'd say it's slightly better than Seagate confusing a judge into letting them redefine the measurement 1 GigaByte as 1 Billion Bytes, and slightly worse than Intel letting their partners pretend Optane SSDs counted as RAM. So, and about the same badness as Intel's P/E-core mismarketing nightmare.

That said I'm already getting lost on what TSMC process Intel 20A is supposed to compete against. IIRC It will go up against and probably slightly behind TSMC N4 "4nm".
If I remember right, Intel wishes they could be competitive with the following:

Intel	TSMC
10 nm = Intel 7	"7nm "N7
7nm = Intel 4	NFF7+ or N6
Intel 3	N5
Intel 20A	N5+ or N4
Intel 18A	N3

Nope. Quit making up stuff. Intel 3/4 are close to TSMC N3.

Source : https://fuse.wikichip.org/news/6720/a-look-at-intel-4-process-technology/2/

 

[Stevemeister]

Mmm - having specifications and targets is one thing. Being able to hit them is something altogether more challenging.

 

[jkflipflop98]

Aye research is the "easy" part putting it into mass production is a lot harder.

Actually, you couldn't possibly be more wrong. R&D of a device is 100x harder than putting it into HVM. It takes ~6 years from the design board to a finished process. It takes about 6 months to ramp to HVM levels.

 

[bniknafs9]

didn't Biden say it was american innovation ? i think he lyes and it's from taiwan not America ... /

 

[Device Unknown]

In other news. China introduces its new 2nm manufacturing process. Strangly looks exactly like Intels. /sarcasm off. You know dang well anything in the hands of China gets copied. American companies are so incredibly STUPID it's astounding.