Pages:
Author

Topic: FYI Intel & Altera to Build Next-Generation FPGAs on Intel's 14 nm Tri-Gate Tech (Read 3639 times)

legendary
Activity: 980
Merit: 1040

tried to get a more statistics out of Altera....

Need more info on engineering specs.... ifs its really that big of a game changer..... why am I not requesting samples?

Do you realize all that altera announced is that they will use intel's upcoming 14nm process? We are several years away from seeing those chips, the fabs still have to built afaik, the node isnt ready, the design of the fpga most likely has not even have begun.  Even intel itself is not expected to be delivering 14nm chips until next year, I wouldnt expect altera to ship a 14nm product in volume until 2015 or 2016.
sr. member
Activity: 420
Merit: 250
The next step is making chips comprised of millions of silicon layers, as opposed to just 10 or so. With each "layer" being a few nanometers thick, that leaves us a relatively big margin to improve performance (30 years of Moore's Law).
I thought the reason they keep it "simple" at only a few layers was heat? You can't just stack layers a half an inch thick, and expect to cool the entire chip evenly.

Negative - you can go pretty deep. There are even some chip designs that interleave metallic layers to transmit heat out from deep chips.
hero member
Activity: 507
Merit: 500
"Products are targeted at ultra high-performance systems such as military, wireline communications and cloud networking.

Intel has announced that it has reached an agreement with PLD (Programmable logic device) manufacturer Altera to product FPGAs (Field-programmable gate array) on Intel's 14 nm tri-gate transistor technology. According to the announcement, these next generation products will enable breakthrough levels of performance and power efficiencies not otherwise possible and further the company's ability to deliver on the promise of silicon convergence by delivering a more flexible and economical alternative to traditional ASICs and ASSP."

FYI



Source

tried to get a more statistics out of Altera....

Need more info on engineering specs.... ifs its really that big of a game changer..... why am I not requesting samples?
hero member
Activity: 924
Merit: 506
The next step is making chips comprised of millions of silicon layers, as opposed to just 10 or so. With each "layer" being a few nanometers thick, that leaves us a relatively big margin to improve performance (30 years of Moore's Law).
No doubt this allows for an increase in density and more efficient 3-dimensional routing. But how much will this help power consumption?

It seems that 3D architecture would miss the original spirit of Moore's Law. Moore's Law pertained to how many transistors could be put onto an IC component. But the prediction was done at the time with the understanding that IC chips were single layers of integrated transistors (2D).

Going vertical with transistors in the architecture doesn't really increase the density of the transistors per each 2D layer. And doesn't even really increase the transistor density by being in 3D space. Well, no more than cutting an iron plate and stacking it increases the density of the iron atoms in the whole.

The only thing is that you do have more transistors on a single component, and only in that sense is it a semblance of Moore's Law. But I think it still misses the mark of what Moore was thinking. I could be wrong, but that's the way I understand it.



mrb
legendary
Activity: 1512
Merit: 1027
The next step is making chips comprised of millions of silicon layers, as opposed to just 10 or so. With each "layer" being a few nanometers thick, that leaves us a relatively big margin to improve performance (30 years of Moore's Law).
No doubt this allows for an increase in density and more efficient 3-dimensional routing. But how much will this help power consumption?

Reducing voltages? Using superconductivity? Using photons instead of electrons? If I knew, I would be founding the next "Intel" company! Smiley
member
Activity: 88
Merit: 10
We've already seen how everything will get faster-more core in the one chip.

The chips are so small and heat such a non issue that they just put more cores into the chip. More cores produce more heat which then brings you back tot he same cooling requirements of the larger process chips.
But yes also means coming back to the same power usage per chip, but that chip does more because there are more processing units.

So they might make these 14Nm FPGAs the same wafer size, but with more transistors allowing more engines. So they might be able to compete speed wise with the better ASICs miners, but because they needed more hashing units to do it will they still be in the same power window?

The FPGAs might be a technology step ahead because of their wider commercial applications, but who knows in 2 years time BFL might also get access to the 14Nm process for their ASICs to jump ahead of the market and crush the FPGAs again just like they are with current FPGAs

Maybe somebody will take a new FPGA, put it on a PCB, and program it. I just hope they plan to be out of business within a year or two such is the tech market these days.
sr. member
Activity: 420
Merit: 250
Explain like I'm 5 anyone? How does this help bitcoin mining? Do these serve for mining just as well as "asics"? How do they improve on the structure?

Basically it's possible that if this technology is mass produced and becomes cheap enough... quick enough... that we'll be able to see multiple fpga products that will give existing asics a run for the money.
 
It really depends on how cheaply they're available.
legendary
Activity: 952
Merit: 1000
The next step is making chips comprised of millions of silicon layers, as opposed to just 10 or so. With each "layer" being a few nanometers thick, that leaves us a relatively big margin to improve performance (30 years of Moore's Law).
I thought the reason they keep it "simple" at only a few layers was heat? You can't just stack layers a half an inch thick, and expect to cool the entire chip evenly.
legendary
Activity: 1904
Merit: 1007
So this means we will be stuck? No more evolution? I said i don't care about the money because i am only interested in the technical aspect. From what you are saying maybe Intel and the other fabs will just stop at 10 or 5 nm for example, but something else must come next. We must keep evolving so i just wanted to know what could be next.
legendary
Activity: 980
Merit: 1040
@Puppet I don't care about the money.

You may not, but the companies that need to pay for it, they do. Most fab operators are already struggling to keep pace because of the astronomical costs of todays <45nm technology. The number of leading edge fabs shrinks every year and its basically down to a 2 or 3 horse race now (Intel, GF and TSMC).  If it costs $50B to build a fab for some future tech that offers no more than 2x higher transistor density than existing fabs, at some point it just wont be worth it to anyone.
legendary
Activity: 3430
Merit: 3071
The next step is making chips comprised of millions of silicon layers, as opposed to just 10 or so. With each "layer" being a few nanometers thick, that leaves us a relatively big margin to improve performance (30 years of Moore's Law).
No doubt this allows for an increase in density and more efficient 3-dimensional routing. But how much will this help power consumption?

and more to the point (as I believe this is the current research engineering challenge with this tech), how do you efficiently cool such a 3D stack? Particularly at clock speeds that we're used to for current 2D chips. No answers to this so far, or at least the last I heard there weren't. Given the trend for closed environments to achieve new technological ends in electronics and computing, I wonder what on earth we may end up with if this is really the avenue that gets commercially exploited. Cube chips in an oil cooled bath? Maybe a processing farm company could make a secondary business in commercial food frying... Grin
legendary
Activity: 1904
Merit: 1007
@Puppet I don't care about the money. They are not being an important factor. If there are money to be made then there are money to invest.

@Carlton Banks I wasn't referring to bitcoin mining. I was curious for general technology.

Thank you all for answers.
donator
Activity: 1616
Merit: 1003
The next step is making chips comprised of millions of silicon layers, as opposed to just 10 or so. With each "layer" being a few nanometers thick, that leaves us a relatively big margin to improve performance (30 years of Moore's Law).
No doubt this allows for an increase in density and more efficient 3-dimensional routing. But how much will this help power consumption?
mrb
legendary
Activity: 1512
Merit: 1027
The next step is making chips comprised of millions of silicon layers, as opposed to just 10 or so. With each "layer" being a few nanometers thick, that leaves us a relatively big margin to improve performance (30 years of Moore's Law).
hero member
Activity: 658
Merit: 500
The current evolution is to go massively parallel.
The next step will be to go from binary logic to quantum logic.
legendary
Activity: 3430
Merit: 3071
So we are just a few years away from hitting max silicon technology? Is there anything else being developed or we just stop at 5nm?

Desktop CPUs are apparently moving to stacking multiple cores (16+ I expect) in between on die RAM layers, creating a kind of core/RAM multilayer sandwich. But mining will not benefit from that change, or at least not the RAM from such a sandwich. I think the mentioned substrate changes might be the only way to go past < 10nm, but that's not gonna be a serious issue for a decade or so. Right now, BFL are struggling to get their 65nm part into customer units inside of 12 months, so I don't think worries about hitting the 5nm ceiling are quite on the table (yet)
legendary
Activity: 980
Merit: 1040
carbon nanotubes have the potential to carry us a bit further, but anything smaller than a few nm may never happen; even if it ever becomes technically feasible, its doubtful if it will ever be economically feasible (Intels latest 14nm fab  will cost > $5 billion and costs are increasing exponentially with smaller nodes).
legendary
Activity: 1904
Merit: 1007
So we are just a few years away from hitting max silicon technology? Is there anything else being developed or we just stop at 5nm?
donator
Activity: 1616
Merit: 1003
Quick noob question: what happens when technology reaches  1nm or 0.5 nm? What is next?
The lattice spacing between atoms in silicon is 0.54 nm. It is safe to say that 0.5 nm technology is impossible (at least with silicon).
Pages:
Jump to: