Jump to content

TSMC to have four EUV scanners in 2015, to use EUV tools for 10nm chips

Jerakl

Source:
http://www.kitguru.net/components/anton-shilov/tsmc-to-have-four-euv-scanners-in-2015-to-use-euv-tools-for-10nm-chips/
 

Taiwan Semiconductor Manufacturing Co. has ordered two additional extreme ultraviolet (EUV) lithography scanners from ASML Holding. The equipment will be delivered in 2015 and will complement initial two EUV scanners TSMC has today, which will be upgraded. Four scanners will let TSMC make chips using 10nm process technology using EUV tech in commercial quantities.

 
think this is the machine.
asml_twinscan.jpg
 


 

Next year ASML will deliver two NXE:3350B EUV systems to TSMC. In addition, TSMC will upgrade TSMC’s existing NXE:3300B systems with new light sources that will increase their production capacity to the levels of NXE:3350B, which will allow the world’s largest contract maker of semiconductors to start volume commercial production of 10nm chips using EUV lithography equipment sometimes in 2016 – 2017.

 
 Personal Thoughts:

10nm? V nice. I'm not going to pretend to know a lot about this topic. I don't, at all. I've done a bit of looking in the past, that's it.

Article wasn't too long, wasn't much I was able to glean much from it. Other then that in 2015 they will have 4 of these machines that will enable them to do 10nm.

 
 
Shouldn't be a repost AFAIK

Someone told Luke and Linus at CES 2017 to "Unban the legend known as Jerakl" and that's about all I've got going for me. (It didn't work)

 

Link to comment
Share on other sites

Link to post
Share on other sites

I call BS. That would imply TSMC is ahead of both Samsung and Intel. The entire world logically knows that isn't remotely possible.

Software Engineer for Suncorp (Australia), Computer Tech Enthusiast, Miami University Graduate, Nerd

Link to comment
Share on other sites

Link to post
Share on other sites

Lots of words I see here, few I comprehend.

All I can think is;

10nm. Gimme. *grabby hands*

CPU- 4690k @4.5ghz / 1.3v    Mobo- Asus Maximus VI Gene   RAM- 12GB GSkill Assorted 1600mhz   GPU- ASUS GTX 760 DCUII-OC 

Storage- 1TB 7200rpm WD Blue + Kingston SSDNow 240GB   PSU- Silverstone Strider ST75F-P

 

Link to comment
Share on other sites

Link to post
Share on other sites

I call BS. That would imply TSMC is ahead of both Samsung and Intel. The entire world logically knows that isn't remotely possible.

You probably know more then me. I can't understand half the words in the article. I tried.

 

Lots of words I see here, few I comprehend.

All I can think is;

10nm. Gimme. *grabby hands*

My thoughts were a little different but fundamentally the same.

 

Was mainly along the lines of;

ooh 20nm interesting.

 

Can't wait until that comes to market.

Someone told Luke and Linus at CES 2017 to "Unban the legend known as Jerakl" and that's about all I've got going for me. (It didn't work)

 

Link to comment
Share on other sites

Link to post
Share on other sites

For those of you curious, a render of what it looks like on the inside:

pennwell.web.650.390.jpg

▶ Learn from yesterday, live for today, hope for tomorrow. The important thing is not to stop questioning. - Einstein◀

Please remember to mark a thread as solved if your issue has been fixed, it helps other who may stumble across the thread at a later point in time.

Link to comment
Share on other sites

Link to post
Share on other sites

I call BS. That would imply TSMC is ahead of both Samsung and Intel. The entire world logically knows that isn't remotely possible.

 

Really? You think that 2016-2017 is BS for 10nm when some customers are getting back 16nm chips in March? I would assume that Samsung, Intel and TSMC are all on roughly equal footing with regards to process size - some are just less vocal about it than others

Link to comment
Share on other sites

Link to post
Share on other sites

I call BS. That would imply TSMC is ahead of both Samsung and Intel. The entire world logically knows that isn't remotely possible.

I call BS on your statement. Samsung and intel are on 16/14nm now. and when TSMC gets these machines in 2015, they wont be comercially operable until atleast (rough guess) 2016. And that roughly alligns it with intels 10nm roadmap. 

 

And so what if they are updating... yes they were behind for some time now, but they got the funds to update their equipement, and with that, they wont be behind anymore... thats how business works... youre (like always) looking at it from a too narrow point of view, time wise.

"Unofficially Official" Leading Scientific Research and Development Officer of the Official Star Citizen LTT Conglomerate | Reaper Squad, Idris Captain | 1x Aurora LN


Game developer, AI researcher, Developing the UOLTT mobile apps


G SIX [My Mac Pro G5 CaseMod Thread]

Link to comment
Share on other sites

Link to post
Share on other sites

Could somebody explain to me or show me where to read about how are they planning to deal with quantum tunneling? I mean at 20nm it wasn't that big of an issue but 10nm ? 

Link to comment
Share on other sites

Link to post
Share on other sites

Could somebody explain to me or show me where to read about how are they planning to deal with quantum tunneling? I mean at 20nm it wasn't that big of an issue but 10nm ? 

4gate transistors maybe? to increase the energy needed to tunnel to something controllable. thats why intel went 3gate for 20nm

"Unofficially Official" Leading Scientific Research and Development Officer of the Official Star Citizen LTT Conglomerate | Reaper Squad, Idris Captain | 1x Aurora LN


Game developer, AI researcher, Developing the UOLTT mobile apps


G SIX [My Mac Pro G5 CaseMod Thread]

Link to comment
Share on other sites

Link to post
Share on other sites

Really? You think that 2016-2017 is BS for 10nm when some customers are getting back 16nm chips in March? I would assume that Samsung, Intel and TSMC are all on roughly equal footing with regards to process size - some are just less vocal about it than others

2015! 16 is when Intel is due to have it released.

Software Engineer for Suncorp (Australia), Computer Tech Enthusiast, Miami University Graduate, Nerd

Link to comment
Share on other sites

Link to post
Share on other sites

2015! 16 is when Intel is due to have it released.

 

Which is why the article says delivered in 2015 and then:

 

"start volume commercial production of 10nm chips using EUV lithography equipment sometimes in 2016 – 2017"

Link to comment
Share on other sites

Link to post
Share on other sites

Which is why the article says delivered in 2015 and then:

 

"start volume commercial production of 10nm chips using EUV lithography equipment sometimes in 2016 – 2017"

yeah, youve proven him wrong. now wait for the irrational argument to start, because intel is best and noone else has any chance to compete ;)

"Unofficially Official" Leading Scientific Research and Development Officer of the Official Star Citizen LTT Conglomerate | Reaper Squad, Idris Captain | 1x Aurora LN


Game developer, AI researcher, Developing the UOLTT mobile apps


G SIX [My Mac Pro G5 CaseMod Thread]

Link to comment
Share on other sites

Link to post
Share on other sites

Don't get your hopes up... Just cause you have the technology doesn't mean you can do it... 


Edit: I realized right after posting this I was a tad vague I meant don't get your hopes on products 10nm anytime soon as TSMC has had 20 and 16nm machinery for a while and Nvidia's 900 series still wasn't 16nm like some rumors were implying and I doubt the 1000 series will be either

5820k4Ghz/16GB(4x4)DDR4/MSI X99 SLI+/Corsair H105/R9 Fury X/Corsair RM1000i/128GB SM951/512GB 850Evo/1+2TB Seagate Barracudas

Link to comment
Share on other sites

Link to post
Share on other sites

I call BS. That would imply TSMC is ahead of both Samsung and Intel. The entire world logically knows that isn't remotely possible.

Agreed.

 

Sure, TSMC will be able to build 10nm, who can't? The real question is what is their yield? If they print 10nm wafers and half of them fail, no one would buy from them. Maybe Nvidia since it seems like they are in bed together.

 

Heck even Apple is running from TSMC and going back to Samsung. You know if Apple is crawling back to Samsung, there's something very wrong with TSMC's 16nm yields.

Link to comment
Share on other sites

Link to post
Share on other sites

Could somebody explain to me or show me where to read about how are they planning to deal with quantum tunneling? I mean at 20nm it wasn't that big of an issue but 10nm ? 

4gate transistors maybe? to increase the energy needed to tunnel to something controllable. thats why intel went 3gate for 20nm

 

:mellow:  wut

Someone told Luke and Linus at CES 2017 to "Unban the legend known as Jerakl" and that's about all I've got going for me. (It didn't work)

 

Link to comment
Share on other sites

Link to post
Share on other sites

:mellow:  wut

We can't infinitely decrease transistor's size without ignoring quantum mechanics . 

Basically weird shit happens when particles are in small groups - whole classic physics is turned upside down. And 10 nm is small, really small for silicone.

Like quantum tunneling where a particle tunnels through a barrier that it classically could not surmount.(Imagine that you throw ball at wall and it passes throught it )

It's not really a bad thing , for example Esaki diode uses this phenomena . Thing is - we still need many research to fully understand and take advantage of tunneling in conventional computing . Until it happens, we're going to try make it harder for particles to tunnel - like Luka said - by increasing energy required for tunneling.

 

Yeah, we should stop here for a moment and realise how great it is that basically everyone has an access to cutting edge chips for as little as <100$. It's a true marvel of technological advancement and production efficiency. 

Link to comment
Share on other sites

Link to post
Share on other sites

We can't infinitely decrease transistor's size without ignoring quantum mechanics . 

Basically weird shit happens when particles are in small groups - whole classic physics is turned upside down. And 10 nm is small, really small for silicone.

Like quantum tunneling where a particle tunnels through a barrier that it classically could not surmount.(Imagine that you throw ball at wall and it passes throught it )

It's not really a bad thing , for example Esaki diode uses this phenomena . Thing is - we still need many research to fully understand and take advantage of tunneling in conventional computing . Until it happens, we're going to try make it harder for particles to tunnel - like Luka said - by increasing energy required for tunneling.

 

Yeah, we should stop here for a moment and realise how great it is that basically everyone has an access to cutting edge chips for as little as <100$. It's a true marvel of technological advancement and production efficiency. 

I understood. And I wanted to become a chip engineer. Sounds like complicated crap.

Someone told Luke and Linus at CES 2017 to "Unban the legend known as Jerakl" and that's about all I've got going for me. (It didn't work)

 

Link to comment
Share on other sites

Link to post
Share on other sites

I call BS. That would imply TSMC is ahead of both Samsung and Intel. The entire world logically knows that isn't remotely possible.

TSMC is still a huge company, they manufacture for AMD and Nvidia aswell as apple and plenty of lesser known people, they probably aren't beyond intel, but samsung quite possibly

Link to comment
Share on other sites

Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now

×