Samsung Claims Higher 3 nm Yields than TSMC

[btarunr]

Competition between Samsung and TSMC in the 4 nm and 3 nm foundry process markets is about to heat up, with the Korean foundry claiming yields competitive to those of TSMC, according to a report in the Kukmin Ilbo, a Korean daily newspaper. 4 nm is the final silicon fabrication process to use the FinFET technology that powered nodes ranging between 16 nm to 4 nm. Samsung Foundry is claiming 4 nm wafer yields of 75%, against the 80% yields figure put out by TSMC. 4 nm powers several current-generation mobile SoCs, PC processors, and more importantly, the GPUs driving the AI gold-rush.

Things get very interesting with 3 nm, the node that debuts GAA-FET (gates all around FET) technology. Here, Samsung claims to offer higher yields than TSMC, with its 3 nm GAA node clocking 60% yields, against 55% put out by TSMC. Samsung was recently bitten by a scandal where its engineers allegedly falsified yields figures to customers to score orders, which had a cascading effect on the volumes and competitiveness of their customers. We're inclined to think that Samsung has taken lessons and is more careful with the yields figures being reported in the press. Meanwhile, Intel Foundry Services competes with the Intel 3 node, which is physically 7 nm FinFET, but with electrical characteristics comparable to those of 3 nm.



View at TechPowerUp Main Site | Source

 

[AnotherReader]

Intel 4, and by extension Intel 3, is much closer to TSMC's N3 than N7. As far as Samsung is concerned, given their history, it would be wise to be skeptical of their claims.

 

[tabascosauz]

Intel 4, and by extension Intel 3, is much closer to TSMC's N3 than N7. As far as Samsung is concerned, given their history, it would be wise to be skeptical of their claims.

So from the descriptions of their 3LPE, Samsung wants to tell the world that they are not only ahead on technology (GAAFET while TSMC has 1 gen left to go and Intel won't use RibbonFET before its Angstrom era), and also ahead on yields? hmmmmmmmm

We're inclined to think that Samsung has taken lessons and is more careful with the yields figures being reported in the press.

I don't want to write them off just yet, but given their continuous years of failures in their sub-10nm portfolio and also being caught red-handed regarding yields, Samsung isn't the first to come to mind when it comes to "taking lessons".

 

[Tomorrow]

Intel 4, and by extension Intel 3, is much closer to TSMC's N3 than N7. As far as Samsung is concerned, given their history, it would be wise to be skeptical of their claims.

That may be the case on paper but N3 is in production. Intel 4 is not. As is often the case with Intel these days. Ultimately the biggest win is mass production not great numbers in labs. TSMC is taking it step by step and not trying to make huge leaps that then inevitably stall and delay production.

Also common logic dictates that process that is more aggressive/dense should have lower yields even if it is in production.

 

[Denver]

This is the kind of news that the GPU market needs.

 

[Chrispy_]

Like others, I'm sceptical given Samsung's history of wild claims failing to match reality, but tentatively - Yayyy?

More competition is good for us, TSMC being the only high-end node in town really hurts our pricing as end-users.

 

[freeagent]

Samsung:

 

[Denver]

Samsung:

So the guy with the broken leg is intel?

 

[AnotherReader]

That may be the case on paper but N3 is in production. Intel 4 is not. As is often the case with Intel these days. Ultimately the biggest win is mass production not great numbers in labs.

You're right. TSMC is far ahead in the metric that matters most: first date of mass production. One reason is that they have limited risk by forgoing big jumps like Intel's 14 nm to 10 nm and opted to introduce intermediate processes such as 10 nm. I was just peeved by the source's inaccuracy in calling Intel 4 a 7 nm process.