• Notice regarding image hosts and attachments. See here.

  • 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.

Fermi Delayed Till 2015

Caporegime
Joined
8 Jul 2003
Posts
30,080
Location
In a house
Fermi delayed until 2015

IC Insights sees more delays for two emerging and key IC manufacturing technologies: 450mm and extreme ultraviolet (EUV) lithography.
450mm production fabs could be pushed out to 2015 or 2016 or two years later than some had hoped, according to analysts. And EUV may miss the insertion into the 16nm node, thereby pushing the technology out to the 13nm logic node in 2015, analysts said.

And another technology, 3D devices based on thru-silicon vias (TSVs), is still in the embryonic stages and "overhyped," said Trevor "Yancey, an analyst with IC Insights. There is a lot of hype about TSVs for 3D devices, but there are still issues about the substrates, testing and costs, he said.

Some think TSVs will take longer to ramp for mainstream applications. In other words, three enablers that can help pave the way for Moore's Law 450mm, EUV and TSVs are still more talk than reality.

It is no surprise that the 450mm transition could be delayed. "The recession put 450mm on the backburner," said Bill McClean, president of IC Insights, at an event here.

As reported, Intel, TSMC and Samsung are separately pushing for the advent of 450mm "prototype" fabs by 2012. The companies are looking for 450mm "demonstration" tools at the 32nm node and "pilot tools" at 22nm. Some believe 450mm fabs will never happen, saying the R&D costs are too expensive. No one is sure who will pay for the tools or the R&D.

http://vr-zone.com/forums/547191/fermi-delayed-until-2015.html

Yep, had me as well. :p
 
I expected the link to be SemiAccurate.com but tbh in hindsight there wasn't enough NVidia bashing for it to be Charlie.
 
Last edited:
Yeah, I think that's actually referring to the diameter of the wafer - i.e. the standard at the moment I think is 300mm wafers.

Yup, and as for the title, its called getting peoples attention and making them read a story they otherwise would never bother reading with a very slight dig at the size of Fermi and cost of producing so few per wafer. A wafer that would at a guess offer way more than a 50% area increase would offer Nvidia a way to make Fermi profitable.

In that sense I agree, I reckon it will be another 3-4 years before its possible to produce Fermi at a decent profit :p
 
Yup, and as for the title, its called getting peoples attention and making them read a story they otherwise would never bother reading with a very slight dig at the size of Fermi and cost of producing so few per wafer. A wafer that would at a guess offer way more than a 50% area increase would offer Nvidia a way to make Fermi profitable.

In that sense I agree, I reckon it will be another 3-4 years before its possible to produce Fermi at a decent profit :p

I first thought this was a misprint but it seem to suggest that a die shrink wont be enough to make the GF100 profitable :eek:

If this is even close to true, Nvidia should sack every one on the design team on the grounds of gross retardation.
 
i wouldnt be suprised if this was true, were are the fermi cards, come on someone, get some solid proof up they even exist
 
Back
Top Bottom