Samsung: We can get to 5 nm without "any major technical difficulties"

[Mondozei]

And FinFET should be able to get them to 7 nm without major issues, or so Samsung's Kinam Kim says in a speech to the currently-ongoing ISSCC.

If true, then it appears that the 14/16 nm node was the real bottleneck. Intel are sounding the exact same things as Samsung re: their 10 nm node.

Since EUV litography is still in the middle of nowhere in terms of new technology - yes ASML, I'm looking at you - this is basically what we need.

 

[Mondozei]

What's interesting to note, however, is that Kinam avoids talking about silicone specifically. It's unclear whether he includes material in the term "technical difficulties".

Contrast this with Intel, who said that they will need new material after 7nm.

 

[videogames101]

no major technical difficulties for 7nm or 5nm?

who is this guy kidding

 

[ShintaiDK]

Samsungs PR machine in full BS. I guess they really copied everything TSMC did, and now going for more.

 

[jdubs03]

By "7nm" they really mean 10nm. Which makes sense considering their "5nm" will really be like 7nm. They can't out-run physics better than Intel, so trying to make it seem like they can get finFETs down to a "lower" node than Intel isn't going to work. They're going to run into the same issue as Intel down at those feature sizes.

However, when we're talking 5nm, I think this guy is throwing lots of pie in the sky.

 

[witeken]

Sure, 7nm (which SS calls 5nm) won't be the end of Moore's Law: Bohr claims to have hindsight up to the ~2-3nm node, so that's what SS means by no major technical difficulties. It will be a challenge like all previous node, except the ever increasing cost (and thus potentially timing) of getting there will be cumulatively higher.

 

[MisterMac]

You've still got to admire Samsung.


Once that behemoth machine corporation decides to do something - nothing is gonna stop them from doing it.
And i'm quite sure their gonna do it at profit as well -even if profit is small.

They will win some customers from TSMC with the PR stunts - you just know it.
Someone is going to discard substance sooner or later - just because... samsung.

 

[jdubs03]

You've still got to admire Samsung.


Once that behemoth machine corporation decides to do something - nothing is gonna stop them from doing it.
And i'm quite sure their gonna do it at profit as well -even if profit is small.

They will win some customers from TSMC with the PR stunts - you just know it.
Someone is going to discard substance sooner or later - just because... samsung.

Samsung I think has the potential to be THE most influential technology company in the world, just by the nature of the way their business is structured = conglomerate-esque. It's just amazing to see how much variety they have in their product lines. With their fabrication facilities, they definitely have improved significantly and I'm sure plenty of companies are lining up to port their designs to their process. TSMC won't go away but their stature in the foundry world isn't nearly as strong as it was in recent times.

 

[scannall]

And FinFET should be able to get them to 7 nm without major issues, or so Samsung's Kinam Kim says in a speech to the currently-ongoing ISSCC.

If true, then it appears that the 14/16 nm node was the real bottleneck. Intel are sounding the exact same things as Samsung re: their 10 nm node.

Since EUV litography is still in the middle of nowhere in terms of new technology - yes ASML, I'm looking at you - this is basically what we need.

TSMC and ASML have made great strides as of late on EUV. Interesting article here. EUV Progress

 

[krumme]

You've still got to admire Samsung.


Once that behemoth machine corporation decides to do something - nothing is gonna stop them from doing it.
And i'm quite sure their gonna do it at profit as well -even if profit is small.

They will win some customers from TSMC with the PR stunts - you just know it.
Someone is going to discard substance sooner or later - just because... samsung.

They get no customers from PR. Look the ppt crap gf spewed years ago - it doesn help but decrease credibility. The customers are pro.

As for ss Investment in semiconductor i think its gigantic 2015 forward. They have to protect their mobile business. The profit from it is vital for what ss is today. Without it they are just a bunch of concrete. They protect mobile with their tech in oled and semiconductor. We have seen nothing yet. Iphone6 have shown they can not sleep and just harvest. 2014 have been an wakeup call. Thats why they make all that noise.

 

[Idontcare]

Samsung: We can get to 5 nm without "any major technical difficulties"

And FinFET should be able to get them to 7 nm without major issues, or so Samsung's Kinam Kim says in a speech to the currently-ongoing ISSCC.

Wow, blast from the past! Back in 2009 I made a post about this exact same dude from Samsung (Kinam Kim) making the same assertion

Samsung's Kim Claims No Limit to Scaling

Laura Peters, Editor-in-Chief -- Semiconductor International, 6/4/2009

The ever-optimistic Kinam Kim of Samsung Electronics noted that some people in industry say the limit of scaling is ~5 nm, to which he said, "I do not agree." In fact, he said he believes that there are various possible paths to overcome obstacles of silicon scaling to continue to grow the silicon industry far beyond the nanometer range. Kim, executive vice president and general manager of Samsung's Semiconductor R&D Center and Samsung Fellow, presented at IMEC's Technology Forum in Brussels, Belgium, this week.

http://maltiel-consulting.com/Samsung-5nm_No_Limit_to_Silicon_Scaling_maltiel_semiconductor.html

6 years later and he is still undeterred! Good on him, you need enthusiastic leaders in visible positions to keep up morale and momentum :thumbsup:

 

[Arachnotronic]

Samsung I think has the potential to be THE most influential technology company in the world, just by the nature of the way their business is structured = conglomerate-esque. It's just amazing to see how much variety they have in their product lines. With their fabrication facilities, they definitely have improved significantly and I'm sure plenty of companies are lining up to port their designs to their process. TSMC won't go away but their stature in the foundry world isn't nearly as strong as it was in recent times.

TSMC is still at the top of its game today and is projecting very solid growth this year.

Everyone is too quick to count them out, when really I think they've done a superb job running their business.

 

[jdubs03]

TSMC is still at the top of its game today and is projecting very solid growth this year.

Everyone is too quick to count them out, when really I think they've done a superb job running their business.

I agree they have done quite well even taking those A8/X orders away from Samsung, and their "16"FF/+ feature sizes compare favorably to Samsung, so I'm not counting them out, I just think that Samsung has been able to progress to a point where they are a little bit out in front, mainly in terms of timeframe. I have no doubt that implementations based on TSMC's new process are going to be impressive. I can't wait to see how Pascal performs especially in HPC solutions. We could be seeing 500 petaflop supercomputers in the not too distant future (lets say 2017).

 

[krumme]

Wow, blast from the past! Back in 2009 I made a post about this exact same dude from Samsung (Kinam Kim) making the same assertion



6 years later and he is still undeterred! Good on him, you need enthusiastic leaders in visible positions to keep up morale and momentum :thumbsup:

Haha. They are good to have but man they can get tired some.

 

[ThatBuzzkiller]

Big claim for a relatively new player when it comes to competing against other state of the art IDM players ...

 

[Idontcare]

Big claim for a relatively new player when it comes to competing against other state of the art IDM players ...

Samsung?

There was a time when they fabbed the DEC (Compaq) alpha chip as second source so Intel could convince the FTC to let them purchase a bunch of the IP.

That must have been nearly 25 years ago. They aren't a new player, they just don't have much market share in MPU logic specifically.

 

[therealnickdanger]

Nanometer is the new core count is the new GHz.

 

[tweakboy]

Who cares its years away.....

 

[senseamp]

No major technical difficulties for Samsung, he's just waiting for TSMC to figure those out

 

[Arachnotronic]

No major technical difficulties for Samsung, he's just waiting for TSMC to figure those out

 

[Fjodor2001]

Nanometer is the new core count is the new GHz.

Yeah, in the end what matters is actual CPU performance. And lately we've not seen much improvement on that front at least on the desktop, despite advances in process tech.

 

[ShintaiDK]

Its been utterly amazing in the mobile and server space.

 

[Fjodor2001]

Its been utterly amazing in the mobile and server space.

Yeah, ARM/Qualcomm/Apple/TSMC/Samsung have been very successful in the mobile arena, true. But as I said, on desktop process tech has not brought as much improvement lately.

 

[krumme]

Its innovation the apple way. Whats wrong with that?

 

[Phynaz]

Yeah, ARM/Qualcomm/Apple/TSMC/Samsung have been very successful in the mobile arena, true. But as I said, on desktop process tech has not brought as much improvement lately.

That because desktop is a dying market.