Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

IBM and Samsung Unveil Semiconductor Breakthrough

newsroom.ibm.com

Post reply on HN
Print view View original

31–40 of 78 posts

Re: IBM and Samsung Unveil Semiconductor Breakthrough

#31

Post by lnsru »

dogber1 wrote: post #18

Better gate control results in lower leakage compared to fins, and there are probably also some better thermals associated with this approach due to the surrounding metals. However, the density is crap - stackable nanoribbons are likely the better path. The rest is just atrocious marketing smoke and mirrors from a zombie company that is essentially dead, but somehow manages to linger on as a shell of its former glory…

It’s so sad seeing my back then dream company IBM Research becoming a zombie. Probably that’s normal cycle of some companies.

Re: IBM and Samsung Unveil Semiconductor Breakthrough

#32

Post by verytrivial »

jliptzin wrote: post #2

“Energy intensive processes, such as cryptomining operations and data encryption, could require significantly less energy and have a smaller carbon footprint.” Wouldn’t the difficulty just go up such that you’d need to expand the number of computers calculating hashes until the electricity usage again becomes the limiting factor?

I would not march in the streets in protest if laws were passed curtailing proof-of-waste cryptocurrencies.

Re: IBM and Samsung Unveil Semiconductor Breakthrough

#33

Post by sbierwagen »

TheSpiceIsLife wrote: post #21
sbierwagen wrote: post #7

Earlier quoted context omitted.

https://www.ebay.com/b/Virtual-Currency-Miners/179171/bn_164... Once a year or so I check on the prices of used miners to see if it would be reasonable to use one as a space heater. Payback period has never been low enough for me to be comfortable with it. (If you buy $300 of equipment and mining becomes illegal tomorrow, then those miners are worthless.)

Not entirely worthless, given you’ve already sunk the cost they’d still be an effective space heater.

"Sunk cost" is for money you've already spent!

The sunk cost here is my existing forced-air electric heat, which is already installed in my house and already turns 1 watt of electricity into 1 watt of heat, or the other space heater I bought ten years ago and can easily put anywhere I need spot heating. If the goal is turn electricity into heat and nothing else, then I am already well supplied with ways to do that.

Re: IBM and Samsung Unveil Semiconductor Breakthrough

#34

Post by marcyb5st »

whatever1 wrote: post #22

how do we cool all of the layers though? with single layer it’s easy, we just need to cover the top surface. With a 3d structure how can we do it efficiently? micro fluids ?

Cooling heatpipes going through the chips? I was wondering the same for 3D stacked nand, but in that case the thermal output could be so low that you might not need to.

Re: IBM and Samsung Unveil Semiconductor Breakthrough

#35

Post by ephbit »

verytrivial wrote: post #32
jliptzin wrote: post #2

“Energy intensive processes, such as cryptomining operations and data encryption, could require significantly less energy and have a smaller carbon footprint.” Wouldn’t the difficulty just go up such that you’d need to expand the number of computers calculating hashes until the electricity usage again becomes the limiting factor?

I would not march in the streets in protest if laws were passed curtailing proof-of-waste cryptocurrencies.

Let's chat about "proof-of-waste" again when in a few years central banks will have created so much inflation and enabled so much unsound government budgeting (simulaneously making inequality explode) that "proof-of-waste" will have become one of (if not) the most important non-manipulable linchpins that will coerce central banks to return to a less inflationary path.

Just listen to what's happending in Turkey right now.

Re: IBM and Samsung Unveil Semiconductor Breakthrough

#36

Post by rbanffy »

sbierwagen wrote: post #33
TheSpiceIsLife wrote: post #21

Earlier quoted context omitted.

Not entirely worthless, given you’ve already sunk the cost they’d still be an effective space heater.

"Sunk cost" is for money you've already spent! The sunk cost here is my existing forced-air electric heat, which is already installed in my house and already turns 1 watt of electricity into 1 watt of heat, or the other space heater I bought ten years ago and can easily put anywhere I need spot heating. If the goal is turn electricity into heat and nothing else, then I am already well supplied with ways to do that.

Since there's very little mechanical work on a miner, it also turns one watt of electricity into one watt of heat. What it'll also do is generate a small amount of bitcoin to help you pay for the electricity.

It's sunk cost if you already have a miner and need a room heater.

Re: IBM and Samsung Unveil Semiconductor Breakthrough

#37

Post by rbanffy »

mhh__ wrote: post #24
Tempest1981 wrote: post #15

Does this shift the balance between TSMC and Samsung and Intel? Or is it copyable? How soon could it be in production?

IBM are slightly weird in that they do ground breaking research and pump out some genuinely impressive products but then they don't seem particularly bothered selling these products to anyone with anything less than a board of directors.

They don't need to make money selling chips if they can license their IP to people who run the foundries. IBM's prodigious patent portfolio is a huge revenue source for the company.

Re: IBM and Samsung Unveil Semiconductor Breakthrough

#38

Post by metanomicon »

jliptzin wrote: post #2

“Energy intensive processes, such as cryptomining operations and data encryption, could require significantly less energy and have a smaller carbon footprint.” Wouldn’t the difficulty just go up such that you’d need to expand the number of computers calculating hashes until the electricity usage again becomes the limiting factor?

Yes, it's essentially a factor of cost, not electrical consumption per se. So at best I don't see it helping to reduce the carbon foot print. In fact, there is a chance the problem would get worse.

Consider the Braess Paradox as an analogy. https://en.wikipedia.org/wiki/Braess%27s_paradox It's possible that increased efficiency (more capacity) would incentivize a greater number people to mine until usage was higher than before - similar to how expanding roadways creates more traffic. Then again, there might not be the time for the problem to get worse before people realized that money is the bottom line, not conservation. Why bother with greener chips.. when it's faster chips that matter. That's one of the big problems with proof-of-work consensus. The game theoretics work against our survival interests.

Re: IBM and Samsung Unveil Semiconductor Breakthrough

#39

Post by verytrivial »

ephbit wrote: post #35
verytrivial wrote: post #32

Earlier quoted context omitted.

I would not march in the streets in protest if laws were passed curtailing proof-of-waste cryptocurrencies.

Let's chat about "proof-of-waste" again when in a few years central banks will have created so much inflation and enabled so much unsound government budgeting (simulaneously making inequality explode) that "proof-of-waste" will have become one of (if not) the most important non-manipulable linchpins that will coerce central banks to return to a less inflationary path. Just listen to what's happending in Turkey right…

So, let's solve corruption with more corruption. No. Regrettably I've read the first half of The Sovereign Individual and it's all about shifting obscene wealth around, not improving the lot of the average human -- that fig leaf covers nothing.

Re: IBM and Samsung Unveil Semiconductor Breakthrough

#40

Post by jktogjfnn »

whatever1 wrote: post #22

how do we cool all of the layers though? with single layer it’s easy, we just need to cover the top surface. With a 3d structure how can we do it efficiently? micro fluids ?

Maybe alternate each computing layer with a cooling layer made from gold (with holes for wires obviously).
Post reply on HN