Intel Shifts to TSMC for Chip Production: A Looming Threat for AMD?

[fansink]

.

...Or A Looming Truth For Intel?


Intel’s first processor using rival TSMC’s technology, the Lunar Lake, has officially launched, intensifying the competition with AMD. According to a recent report by TechNews, third-party testing has confirmed Intel’s claims: Lunar Lake is indeed the most energy-efficient x86 processor to date, outperforming Qualcomm’s Snapdragon X and even rivaling Apple’s M3, reminiscent of Apple’s groundbreaking M1 launch.

TechNews attributes this success not only to Intel’s redesign of power supply, frequency regulation, and packaging but also to the advanced TSMC N3B process.

Recently, Intel announced that in order to reduce costs and better prepare for its in-house 18A process, it has decided to abandon the introduction of the 20A process. As a result, the Arrow Lake chip launching this month will also use TSMC’s process. TechNews raised the question in their article: “With Intel’s new platforms expected to rely on TSMC’s process at least until 2026, will AMD face significant challenges?”

Can AMD’s Zen 5 architecture turn the tide?

TechNews noted that AMD’s current advantage over Intel rests heavily on using TSMC’s process. However, AMD is not alone in benefiting from TSMC’s power efficiency. Across the board, chips produced with TSMC technology have demonstrated superior energy efficiency, delivering high performance without consuming excessive power. But, with the efficiency gains from advanced nodes like M4 or A18 nearing their limits, chipmakers will need to adopt more aggressive power and frequency strategies to push performance further.

Lunar Lake’s impressive energy efficiency highlights both TSMC’s process advantage and Intel’s enduring design prowess. This should serve as a warning for AMD, which plans a major push into the laptop market in 2025. With the launch of Strix Point and Hawk Point this year, AMD aims to release five new platforms next year, targeting the mid-to-high-end laptop market. However, reviews of Strix Point already show that, while performance has improved, energy efficiency remains stagnant—a problem that could persist with future Zen 5-based products.

This opens a window of opportunity for Arrow Lake, which is now powered by TSMC’s process. If Arrow Lake can offer higher peak performance than Raptor Lake Refresh or Meteor Lake while maintaining strong energy efficiency—and with better OEM partnerships—AMD’s hard-earned foothold in the mid-to-high-end market may once again be overshadowed by Intel.

Facing competition shifts due to process changes is nothing new for AMD. As mentioned in the TechNews report, when NVIDIA launched the RTX 30 series on Samsung’s 8LPU (8nm) process, early issues with leakage and high power consumption gave AMD’s RX 6000 series GPUs, known for their superior performance and energy efficiency, a competitive edge. The high-end 6800 and 6900 models were even able to compete with NVIDIA’s RTX 3080. However, once NVIDIA returned to TSMC for the RTX 40 series, AMD struggled to keep up and eventually abandoned its high-end GPU plans, shifting focus to niche markets.

TechNews concludes that while next year may see the lowest degree of processor process diversity—since almost everyone is using TSMC—it will also be a critical year to evaluate the true design strengths of each semiconductor company. With AMD’s Zen 5 already on the table, all eyes are now on Intel’s Arrow Lake to see what surprises it brings to the market with TSMC’s technology.

[News] Intel Shifts to TSMC for Chip Production: A Looming Threat for AMD? | TrendForce News

Intel’s first processor using rival TSMC’s technology, the Lunar Lake, has officially launched, intensifying the competition with AMD. According to a ...

www.trendforce.com

Warning... Do not use the term TrendFarce, it is not an appropriate term.

 

[Daniel Nenni]

I do not even know where to start here.....

 

[Artificer60]

Overlooking the hyperbole, I do think there is one valid point in this article. Intel's design groups have taken a lot of heat for their poor processor designs, but they were tied to a process technology that was continually late and delivered below target for years. It has to be hard to design a competitive chip when you don't know how your transistors are going to perform. To their credit TSMC removes that obstacle for their customers.

That is why I think Pat Gelsinger was right to insist that Intel's number 1 priority had to be to regain process leadership (or at least parity) and to hit process improvement schedules. Whether or not it was actually 5 nodes or not isn't nearly as important as whether or not 18A is a competitive process at the leading edge. Closing that gap in 4 years is a huge step for a company that was stuck on the 10nm hurdle for the better part of a decade.

 

[Daniel Nenni]

Overlooking the hyperbole, I do think there is one valid point in this article. Intel's design groups have taken a lot of heat for their poor processor designs, but they were tied to a process technology that was continually late and delivered below target for years. It has to be hard to design a competitive chip when you don't know how your transistors are going to perform. To their credit TSMC removes that obstacle for their customers.

That is why I think Pat Gelsinger was right to insist that Intel's number 1 priority had to be to regain process leadership (or at least parity) and to hit process improvement schedules. Whether or not it was actually 5 nodes or not isn't nearly as important as whether or not 18A is a competitive process at the leading edge. Closing that gap in 4 years is a huge step for a company that was stuck on the 10nm hurdle for the better part of a decade.

I agree, 18A is a huge achievement for Intel, especially for the internal Intel design groups.

If you compare Intel to Samsung Intel is way ahead. The gap was closed and Samsung has been lapped by Intel.

So congratulations to Intel, absolutely.

 

[samwilde]

Recently, Intel announced that in order to reduce costs and better prepare for its in-house 18A process, it has decided to abandon the introduction of the 20A process. As a result, the Arrow Lake chip launching this month will also use TSMC’s process.

Last I heard some Arrow Lake chips use Intel others use TSMC process. Is this accurate that they moved everything to TSMC?

TechNews noted that AMD’s current advantage over Intel rests heavily on using TSMC’s process.

Not true. AMD Zen 5 also has the advantage that because it is all symmetrical cores, you can for example use AVX-512, while in Intel you cannot if you are using one of their desktop processors with P and E cores. For whatever unfathomable reason Intel continues to refuse to implement AVX-512 in their E cores which means they have to disable AVX-512 on the whole processor. Because otherwise what would happen if the OS moves a task between cores with different ISAs?

 

[Daniel Nenni]

Last I heard some Arrow Lake chips use Intel others use TSMC process. Is this accurate that they moved everything to TSMC?

It is not accurate.

 

[dkr1986]

Last I heard some Arrow Lake chips use Intel others use TSMC process. Is this accurate that they moved everything to TSMC?


Not true. AMD Zen 5 also has the advantage that because it is all symmetrical cores, you can for example use AVX-512, while in Intel you cannot if you are using one of their desktop processors with P and E cores. For whatever unfathomable reason Intel continues to refuse to implement AVX-512 in their E cores which means they have to disable AVX-512 on the whole processor. Because otherwise that would happen if the OS moves a task between cores with different ISAs?

My read on this is AVX-512 is not used often (or not useful) in Desktop/ Client segments. So, Intel's desktop/client CPUs don't have that anymore (also due to E cores). It's useful for server workloads and Intel's XEON lineup has AVX-512 implementation. But E core being not compatible with AVX-512 is supposed to be fixed with AVX10.1/AVX10.2 implementation coming soon with next gen.

Intel's New AVX10 Brings AVX-512 Capabilities to E-Cores

Future hybrid chips will have the goodness of AVX-512 features on both P- and E-cores.

www.tomshardware.com

 

[milesgehm]

Is this accurate that they moved everything to TSMC?

It is a farce not a fact

 

[MKWVentures]

Is this accurate that they moved everything to TSMC?

It is a farce not a fact

Its absolutely not accurate! Intel hasn't announced plans to move 2025 products to TSMC yet.

 

[siliconbruh999]

I have heard the reverse and according to TSMCs statement themselves this rumour doesn't make sense TSMC says N3P is PPA competitive with 18A both launch at same time frame according to public timeline so why would Intel outsource majority to TSMC for 2025 if TSMC is providing same PPA as Intel node according to TSMC this time it doesn't make sense

 

[johnson.z.wu]

I agree, 18A is a huge achievement for Intel, especially for the internal Intel design groups.

If you compare Intel to Samsung Intel is way ahead. The gap was closed and Samsung has been lapped by Intel.

So congratulations to Intel, absolutely.

It is too early to say congratulation to 18A, as it is not done deal so far. Intel is still behind TSMC.

 

[Xebec]

Intel 20A was "cancelled" which was to be the node they some Arrow Lake compute tiles used. Definitely no Intel fabbed arrow Lake with the official 20A announcement..

 

[MKWVentures]

I have heard the reverse and according to TSMCs statement themselves this rumour doesn't make sense TSMC says N3P is PPA competitive with 18A both launch at same time frame according to public timeline so why would Intel outsource majority to TSMC for 2025 if TSMC is providing same PPA as Intel node according to TSMC this time it doesn't make sense

Cost and flexible capacity. Intel is dealing with Non-technical challenges.

I still think it is interesting that people are declaring 18A a victory today. No internal product will be sold on it until 2H 2025. No external foundry product will be sold commercially until 2026. There are one or two roadmap changes coming by mid 2025

 

[MKWVentures]

Intel 20A was "cancelled" which was to be the node they some Arrow Lake compute tiles used. Definitely no Intel fabbed arrow Lake with the official 20A announcement..

Arrow Lake Refresh might have some Intel Chiplets.... but not on 20A

 

[fansink]

Seemingly, Pat had an "obligation" to place 2nm order(s) with TSMC at the appropriate time, not knowing if/when 20A and 18A were going to have sufficient yields when promised, at least as a hedge.

 

[Xebec]

Arrow Lake Refresh might have some Intel Chiplets.... but not on 20A

Curious about all of the claims that "Arrow Lake Refresh is Cancelled". Will Intel do the "2 years between uarch" deal like AMD (against OEMs wanting new stuff every year), or are they pulling in Panther Lake-S enough that we'll have 18A Compute on desktop next year.

(I really hope for 18A Compute. TSMC N3 doesn't seem to be pushing the needle much on desktop vs N5/Intel 7 honestly.. and I get that all modern desktop chips are basically chips designed for laptop/server efficiency, but..)

 

[Daniel Nenni]

Seemingly, Pat had an "obligation" to place 2nm order(s) with TSMC at the appropriate time, not knowing if/when 20A and 18A were going to have sufficient yields when promised, at least as a hedge.

Last I heard Intel did not have a TSMC N2 wafer agreement. All internal design was moved to Intel 18A. Thus the comment by Pat that he was all-in on 18A?

 

[hist78]

Last I heard Intel did not have a TSMC N2 wafer agreement. All internal design was moved to Intel 18A. Thus the comment by Pat that he was all-in on 18A?

If all of Intel's new leading-edge production will be sourced internally, it seems at odds with the supply chain resilience that CEO Pat Gelsinger promotes. Why does Intel need to take such an unnecessary and huge risk by betting solely on 18A? Will Intel's future really depend entirely on Intel 18A?

 

[siliconbruh999]

If all of Intel's new leading-edge production will be sourced internally, it seems at odds with the supply chain resilience that CEO Pat Gelsinger promotes. Why does Intel need to take such an unnecessary and huge risk by betting solely on 18A? Will Intel's future really depend entirely on Intel 18A?

Either he knows something we are missing to take such a huge risk seems like he was not kidding when he said he bet the company on 18A let's see what happens in like 8-9 months

 

[Daniel Nenni]

If all of Intel's new leading-edge production will be sourced internally, it seems at odds with the supply chain resilience that CEO Pat Gelsinger promotes. Why does Intel need to take such an unnecessary and huge risk by betting solely on 18A? Will Intel's future really depend entirely on Intel 18A?

Intel can still use TSMC N3 for support chiplets, right? CPU/GPU/Compute chiplets can be 18A and why can't Intel use Intel 3 for supporting chiplets? I don't see a huge risk here.

The whole supply chain resilience thing is another way of saying we are late to the party so we must use an outside source. Intel must make as many wafers internally as possible to get the manufacturing costs down.