Intel's Future Chips: News, Rumours & Reviews

Page 3 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Status
Not open for further replies.
S|A has been saying that Apple is very capable of buying all chips production of TSMC, use whatever they can, and mete out the rest to whoever they want (QC, NV,AMD), at whatever the price.
Say one thing for Apple, say they are thorough.
 
But the move itself to 450mm has cost fukcing huge amounts, in direct investments, takeovers, and partnerships. What would the semiconductor industry do without Intel leading innovation and forcing others too ?

AFAIK, Intel will start producing 450mm this year, prolly in late Q3.So would those chips be production ready, or more of a "production test chips" ?
 


Don't think the average punter will get their hands on a CPU made on a 450mm wafer till 2016, it will probably be on the 10nm process.


EDIT : Actually, thinking about this a bit more, Intel will probably do a run of 14nm 450mm wafers first, as by then, 14nm will be a known quantity and thus well suited to trying out on this new wafer size. It would also make for some very cheap and plentiful Atom SoC's to target various sectors of the Smartphone market. This might mean they would come out in 2015.
 


Yep the Samsung/Apple battle is causing major collateral damage. That's why neither AMD/NV are talking about 20nm anything yet. Even for their 2014 parts its 28nm.

Meanwhile Intel will be at 14nm.
 
Intel is going BGA with Broadwell for majority of the product lines.
Only 1 DT line will have an LGA package, prolly the high end 'K' parts. So technically, they "are committed to the LGA form factor".
 


450mm is essential in order to move below 10nm. Without such reduction in fab equipment and per-die cost, it would be impossible to sustain a shrink every 2 years (and still make money). This is what Intel is banking on. The move to 450mm wafers is going to cost upwards of 6-10B (on top of the 11B/yr Intel spends on CapEX) over the next 5 years or so. Very few other foundries will be able to make this move. TSMC, Samsung, and probably IBM. We're going to see the number of leading node foundries drop off to maybe 3-4 (including Intel) after the 450mm shift.

But the move itself to 450mm has cost fukcing huge amounts, in direct investments, takeovers, and partnerships. What would the semiconductor industry do without Intel leading innovation and forcing others too ?

AFAIK, Intel will start producing 450mm this year, prolly in late Q3.So would those chips be production ready, or more of a "production test chips" ?

..

Don't think the average punter will get their hands on a CPU made on a 450mm wafer till 2016, it will probably be on the 10nm process.


EDIT : Actually, thinking about this a bit more, Intel will probably do a run of 14nm 450mm wafers first, as by then, 14nm will be a known quantity and thus well suited to trying out on this new wafer size. It would also make for some very cheap and plentiful Atom SoC's to target various sectors of the Smartphone market. This might mean they would come out in 2015.

We won't see 450mm on a high volume node until 2018.
 
With the initial BGA only rumors people feared that Intel would take over the motherboard market and put the Asus/Gigabytes/etc out of business. That's clearly not the case.
 
intel mobos were focused on stability rather than performance or features. the ones i know of lasted pretty long too. problem is - the prices are too high and asus, gb, asrock all offer better.. of everything.
 
Status
Not open for further replies.