• Competitor rules

    Please remember that any mention of competitors, hinting at competitors or offering to provide details of competitors will result in an account suspension. The full rules can be found under the 'Terms and Rules' link in the bottom right corner of your screen. Just don't mention competitors in any way, shape or form and you'll be OK.

TSMC to start 16nm volume production in 1Q15, says report

Gregster said:
**** apple and there hogging of the wafers :mad:
Click to expand...

Haha, TSMC might not have taken kindly to Nvidia's and AMD's words about 20nm being basically useless, seems since then Apple have taken priority :p

Looking forward to when 20nm / 16nm eventually arrive for GPU's. Imagine how powerful a fully enabled 16nm part could be ! :eek: + Stacked vram, future GPU's are gonna be epic.
 
Boomstick777 said:
Haha, TSMC might not have taken kindly to Nvidia's and AMD's words about 20nm being basically useless, seems since then Apple have taken priority :p

Looking forward to when 20nm / 16nm eventually arrive for GPU's. Imagine how powerful a fully enabled 16nm part could be ! :eek: + Stacked vram, future GPU's are gonna be epic.
Click to expand...

loads new Tech, Epically Expensive:D:D:D:D:D:D:D:D:D
 
TSMC 16nm wafers coming in Q1 2015

50,000 wafers a month in Q2


TSMC will start producing 16nm wafers in the first quarter of 2015. Sometime in the second quarter production should ramp up to 50,000 wafers a month.

The new FinFET process will be utilised for ARM SoCs. We still don’t have much information on 16nm GPUs. It is rumoured that Apple is one of TSMC’s major 16nm clients, possibly even the biggest client. However, there is also a possibility that Samsung will take over part of Apple’s business in late 2016, using the company’s 14nm node.

Other major clients for TSMC’s 16nm FinFET+ process include MediaTek and Qualcomm, along with smaller chip designers. At this point it appears that TSMC will transition to 10nm sometime in 2016, but there is no exact timeframe yet.

More here, bring a translator.
Click to expand...

http://www.fudzilla.com/home/item/35588-tsmc-16nm-wafers-coming-in-q1-2015


I don't know what this means to us waiting to get off 28nm though.
 
Well, with rumours pointing to a Q2 launch for 20nm products, it would be very unlikely that they would transition straight to 16nm, volume production starting in Q2 for 16nm would point to GPU's on that process not being available until Q4 at least at a rough guess

It could make 20nm one of the shortest lived GPU generations though
 
50,000 a month in Q2 seems very small and those will all be snapped up by Apple I'd imagine, could be Q4 before the smaller companies get any meaningful amounts. It's nice to speculate but it's getting a bit boring with all the rumours especially when you like buying new toys but don't want to be stung by a much better card due within a year and taking a few hundred quid loss selling and buying.
 
Because those pesky laws of physics tend to get in the way

It is currently cheaper to make incremental changes to current technology than it is to switch to a completely new one, but they are reaching the limits of what is possible on silicon
 
We need advancements away from traditional silicon as the traditional wafer hasn't got long left. molybdenum disulfide (copy and pasted :D) is looking like a good alternative and with Intel on board, it could well be the next step after silicon.

I thought graphene was the next 'big thing' but who knows.
 
Ayahuasca said:
Didn't they find a flaw in graphene that has basically destroyed all hopes of seeing it used any time soon.
Click to expand...

Not sure. I remember reading up on it a couple of years ago and everything was looking sweet for the future. I will do some reading up tonight.
 
Sure it was something to do with not being able to switch graphene transistors off. They may have came up with a solution by now.
 
Do you remember when you read it? Pretty sure I remember seeing a couple of articles about actual working graphene chips - although I think it was for wireless communications.

EDIT: Never mind, read more into it. You can dope it to give it a small band gap enough for wireless circuitry, but it's still not good enough for transistors.
 
Last edited:
Ayahuasca said:
50,000 a month in Q2 seems very small and those will all be snapped up by Apple I'd imagine, could be Q4 before the smaller companies get any meaningful amounts. It's nice to speculate but it's getting a bit boring with all the rumours especially when you like buying new toys but don't want to be stung by a much better card due within a year and taking a few hundred quid loss selling and buying.
Click to expand...

True Apple will get first dips but then again their the biggest player in the industry and I would take an educated guess at saying mobile chips that consume a 1-3 watts are less complicated to produce then the performance class products that AMD and Nvidia sell.
 
You must log in or register to reply here.
Back
Top Bottom