Intel 18A Is Officially Ready for Customer Projects

[AleksandarK]

Intel has updated its 18A node website with the message, "Intel 18A is now ready for customer projects with the tape outs beginning in the first half of 2025: contact us for more information." The contact hyperlink includes an email where future customers can direct their questions to Intel. Designed as a turnaround node for Intel, 18A carries industry-leading features like SRAM density scaling comparable with TSMC's N2, 15% better performance per watt, and 30% better chip density vs. the Intel 3 process node used in Intel Xeon 6, as well as PowerVia backside-power delivery to increase transistor density.

Other features like RibbonFET are the first to replace FinFET transistors, making gate leakage a tighter control. Interestingly, Intel's first products to use the 18A node are client CPUs "Panther Lake" and "Clearwater Forest" Xeon CPUs for data centers. External Intel Foundry customers using the 18A node include Amazon's AWS, Microsoft for its internal silicon for Azure, and Broadcom exploring 18A-based designs. The process of gaining customers for advanced manufacturing is complex, as many existing Samsung/TSMC customers are not risking their capacity and contracts with established advanced silicon makers. However, if Intel's first few customers prove successful, many others could flock over to Intel's fabs as geopolitical tensions are questioning whether the current models of the semiconductor supply chain are feasible in the future. If US companies and startups decide to move with Intel for their chip manufacturing, Intel could experience a complete recovery.



View at TechPowerUp Main Site | Source

 

[wNotyarD]

So, when should we expect Intel's own 18A products again? (I mean, Intel is a customer of Intel Foundries, no?)

 

[Denver]

If they regain the ability to produce their own chips, it will be a significant step forward.
I can't believe I'm about to say this, but… go, Intel!

 

[Vayra86]

Oh look, the undead have risen once more

I wonder how long the necromancy holds this time

 

[ScaLibBDP]

If they regain the ability to produce their own chips, it will be a significant step forward.
I can't believe I'm about to say this, but… go, Intel!

I wish Intel VIPs are reading our posts and Thank You All of Us for using Intel CPUs at work and at home. I hope times of Intel 14++++...nm nodes are over.

Note: My primary development system is a Dell Precision Mobile Workstation with an Extreme Edition of Intel CPU.

 

[TumbleGeorge]

It could easily be Intel's last lithography node. I don't think there's any point in digging any deeper in "angstroms" It would be so expensive that it's unlikely to produce affordable products that would offer enough performance gains over time. Of course, like 14++++++, it could get some improvements. You can't just rely on reducing lithography, to achieve performance gains, you need new processor architectures, eliminating bottlenecks and long latencies, and more efficient computation.

 

[tfp]

I wish Intel VIPs are reading our posts and Thank You All of Us for using Intel CPUs at work and at home. I hope times of Intel 14++++...nm nodes are over.

Note: My primary development system is a Dell Precision Mobile Workstation with an Extreme Edition of Intel CPU.

Intel 14++++ nm? What year are you posting from?

 

[Daven]

Intel 14++++ nm? What year are you posting from?

I guess the same year as TumbleGeorge.

 

[kondamin]

It could easily be Intel's last lithography node. I don't think there's any point in digging any deeper in "angstroms" It would be so expensive that it's unlikely to produce affordable products that would offer enough performance gains over time. Of course, like 14++++++, it could get some improvements. You can't just rely on reducing lithography, to achieve performance gains, you need new processor architectures, eliminating bottlenecks and long latencies, and more efficient computation.

not unlikely we're about at the point that it's going to grind to a halt for everyone.
3nm hasn't been a cake walk for TSMC either never mind Samsung.

 

[TristanX]

Process (PDK) may be ready, but fabs and yield, not. As Intel said, they still are installing equipment in fabs, then must prepare all these stuff to be working as expected, then it will be ready.
This 18A is comparable to N3 but with higher clocks, but still N2 is significally better.

Intel won't be process leader anymore, current processes are too hard for them, giving their cash and IQ, and next processes will be even more demanding. Only large scale business can create enough profits to move forward, and only TSMC is on that route.

 

[Nhonho]

Intel won't be process leader anymore, current processes are too hard for them, giving their cash and IQ, and next processes will be even more demanding. Only large scale business can create enough profits to move forward, and only TSMC is on that route.

Intel should sell its factories to an American investment group and return to the government the money it received from the "CHIPS and Science Act" and become just a chip and software design company, like Nvidia.

 

[Caring1]

Intel should sell its factories to an American investment group and return to the government the money it received from the "CHIPS and Science Act" and become just a chip and software design company, like Nvidia.

They haven't received any money from that yet, it is conditional and pending

 

[ScaLibBDP]

Intel 14++++ nm? What year are you posting from?

2025. Any more questions?

 

[Vayra86]

not unlikely we're about at the point that it's going to grind to a halt for everyone.
3nm hasn't been a cake walk for TSMC either never mind Samsung.

We hear this every new node shrink since 22nm

 

[leonavis]

If they regain the ability to produce their own chips, it will be a significant step forward.
I can't believe I'm about to say this, but… go, Intel!

yeah, things got quite crazy ^^

 

[kondamin]

We hear this every new node shrink since 22nm

And every “node” shrink has become less significant since.

 

[Prima.Vera]

I wonder what % are successful yields. If they have better % than TSMC or Samsung, they might have a winner on their hands.

 

[tfp]

2025. Any more questions?

Intel isn't using 14++++ nm any more and from the looks of it Intel7 isn't much worse then TSMC 5NM they are using one the current chips when it comes to performance.

Intel is doing poorly but get better jokes or at least attempt to be accurate.

 

[kondamin]

I wonder what % are successful yields. If they have better % than TSMC or Samsung, they might have a winner on their hands.

Yield depends on size and since they are working with tiles those are small

 

[docnorth]

18A is slightly behind N2 in terms of absolute density, otoh it may have an edge in terms of power delivery (TSMC reportedly postponed the so-called backside power delivery for next year). In the past Intel pulled aces regarding architecture, but it remains to be seen if they are still capable to achieve something similar.

 

[Scrizz]

This 18A is comparable to N3 but with higher clocks, but still N2 is significally better.

sources?

 

[docnorth]

sources?

Beep, beeeep.

 

[Prima.Vera]

sources?

here:
Trust me bro.