[HN Gopher] Intel's Take on the Next Wave of Moore's Law
       ___________________________________________________________________
        
       Intel's Take on the Next Wave of Moore's Law
        
       Author : rbanffy
       Score  : 42 points
       Date   : 2022-12-06 14:45 UTC (8 hours ago)
        
 (HTM) web link (spectrum.ieee.org)
 (TXT) w3m dump (spectrum.ieee.org)
        
       | FfejL wrote:
       | Nice to see an article address what Moore actually said
       | (transistor count doubles every 24 months) and not what people
       | often thinks Moore said (performance doubles every 24 months).
        
         | Retric wrote:
         | It started out as every 12 months in the original 1965 article
         | which lasted for about a decade before being updated to 18
         | months, and relatively recently people started talking 24
         | months.
        
           | hanniabu wrote:
           | Yup, Moore's law should have always been a regression
           | trendline
        
       | Animats wrote:
       | All that chip stacking. How do you cool it? What's the yield?
       | This is packaging, not higher density transistors.
        
         | rbanffy wrote:
         | A lot of it is just two layers - the chip and the interconnect
         | below. If you have a chip that generates a lot of heat, you can
         | stack another one that generates less heat on top of the hot
         | one and use it as a heat spreader (a lot of mobile chips are
         | like that). Also memories have only a small part of them being
         | used at any given time, so they are a good fit for tall stacks
         | (current flash memory does that too)
        
           | zasdffaa wrote:
           | Not sure if relevant but dram refreshes continuously.
        
         | Arrath wrote:
         | >How do you cool it?
         | 
         | I've wondered whether little micro-filament heat pipes
         | sandwiched right into the chips might be possible, all plumbed
         | into a heat exchanger on the side of the die that can be
         | directly cooled, facilitating the extraction of heat from deep
         | within the stacked chips.
        
           | mook wrote:
           | > I've wondered whether little micro-filament heat pipes
           | sandwiched right into the chips might be possible
           | 
           | Maybe; there's https://www.tomshardware.com/news/tsmc-
           | exploring-on-chip-sem... from about a year and a half ago.
        
             | Arrath wrote:
             | Interesting, thank you!
             | 
             | Direct water cooling huh. With channels right in the
             | silicon I imagine water purity is just that much more
             | important.
             | 
             | God help you if the water gets gunked up with
             | disintegrating gasket material[0] and blocks some of those
             | channels off, or even worse, conductive stuff.
             | 
             | [0]https://www.youtube.com/watch?v=jHdEqWpexH0
        
           | Coffeewine wrote:
           | I don't know about production chips with such things
           | integrated, but it's certainly been tried. The below article
           | is from 2008.
           | 
           | https://www.eetimes.com/ibm-git-demo-3d-die-with-
           | microchanne...
        
         | netr0ute wrote:
         | > How do you cool it?
         | 
         | That's the million dollar question that we're trying to solve
         | right now.
        
         | bmicraft wrote:
         | Cooling is probably less of a problem than many people think.
         | Modern cpus are (or rather would be) very efficient, if you
         | don't push them to the very end of their frequency-power
         | bathtub curve, but the latest generations of desktop cpus are
         | basically stock overclocked. They can achieve 80% of their
         | performance at 40% the power. That basically means you could
         | double performance by 2.5x-ing silicon for the same power/heat.
        
       | pdimitar wrote:
       | This somehow sounds like specialized CPUs to me.
        
       | mikewarot wrote:
       | I'm amazed that we got as far as we have. It takes a well
       | composed assembly of hundreds of machines, many costing USD
       | 300,000,000 to operate as a "Fab".
       | 
       | The coming credit crunch as Boomers retire will make this type of
       | investment much more difficult in the future.
       | 
       | We're at the end of an era.
        
         | gyulai wrote:
         | Can you elaborate? Why does Boomers' retirement lead to a
         | credit crunch?
        
           | sbierwagen wrote:
           | Imagine a country of a hundred 18 year olds. They all get
           | jobs on the same day, and start putting savings into the
           | bank. The bank takes those deposits and loans them out to
           | various productive (profit producing) enterprises.
           | 
           | Time passes. The hundred all turn 65 on the same day, and
           | retire. The savings rate whipsaws, since nobody is depositing
           | money anymore, and is instead withdrawing it. The bank can no
           | longer make loans. Credit crunch.
        
           | belval wrote:
           | Baby boomers life savings are still mostly in retirement
           | funds being used to finance various ventures to get a
           | reasonably good ROI. As the generation ages, the more money
           | will be taken out of those accounts with fewer people to put
           | their new earned money in the fund. Thus causing a credit
           | crunch.
           | 
           | Demographic problems are a bit of an everything problem
           | though.
           | 
           | EDIT: My answer above is somewhat wrong because of how much
           | immigration the US is getting, the pyramid actually looks
           | fine: https://en.wikipedia.org/wiki/Demographics_of_the_Unite
           | d_Sta...
           | 
           | Compare that to the Canadian pyramid:
           | https://www.populationpyramid.net/canada/2020/
        
             | staticman2 wrote:
             | Investing in typical stocks doesn't directly fund anything,
             | you buy the stock from some other private party. No funds
             | go to the company itself.
             | 
             | If you believe you can predict stock market returns from
             | demographics you should form a hedge fund.
        
       | wwwtyro wrote:
       | I guess I'm skeptical. This sounds more like an optimization of
       | current technology than something along the lines of vacuum tubes
       | -> transistors.
        
         | xadhominemx wrote:
         | Yes, that's how Moore's law has worked for 60 years:
         | incremental advancements. The past 15 years have actually
         | involved far more radical step changes than the 45 years prior.
        
           | ilaksh wrote:
           | This seems false to me. For example, electromechanical to
           | relays, to tubes,transistors, integrated circuits, highly
           | parallel ICs are all paradigm shifts rather than incremental,
           | and yielded multiple orders of magnitude rather than
           | incremental improvements in price/performance.
           | 
           | The next really significant boost is likely something like
           | optical transistors, memristors, analog computing, bio-
           | electrical computing, or anything that is highly divergent
           | from the current base concept.
           | 
           | I mean, if the organization of the current basic idea of
           | silicon and the software can align with a compute-in-memory
           | paradigm, I guess that would be more than incremental in a
           | way. But the lady seems to be talking about doing that in
           | still kind of a half-assed way as far as that part.
        
             | xadhominemx wrote:
             | Moore's Law (formulated 57 years ago) by definition applies
             | to dense ICs. It's true, there were a bunch of advancements
             | before the invention of ICs, but those were 60-80 years
             | ago.
        
         | ilaksh wrote:
         | Correct. These are relatively incremental improvements. And I
         | would argue that recently they are more relevant in GPUs anyway
         | due to the scalability.
         | 
         | To resume "Moore's Law" at the pace it once was, we need some
         | of the truly new paradigms to mature. Ideas like memristors or
         | optical transistors.
         | 
         | You could even imagine something like a ubiquitous transparent
         | P2P background computing grid as a big shift in software that
         | could matter.
         | 
         | But for hardware, the next paradigm that brings a few orders of
         | magnitude boost to the baseline is going to actually be a..
         | different paradigm.
        
           | yjftsjthsd-h wrote:
           | > You could even imagine something like a ubiquitous
           | transparent P2P background computing grid as a big shift in
           | software that could matter.
           | 
           | That's not really Moore's law, though; it's changing how you
           | use the transistors rather than how many you get.
        
         | smoldesu wrote:
         | That's honestly what we need right now. There is a plethora of
         | new optimizations available (big.LITTLE, chiplets, 3D stacking,
         | et. al) and neither Intel, AMD or Apple are using them all.
         | There are also ISA optimizations still on the table for both
         | x86 and aarch64, so the _real_ race over the next 5 years will
         | be seeing which company is able to compile these optimizations
         | in a meaningful way.
        
         | djexjms wrote:
         | Yeah. After reading the article, I have to agree. But I'm no
         | expert in this area. It seems like it could certainly help. If
         | I had to guess, things like ECRAM and reconfigurable analog
         | computing chips will have more of an effect than this (assuming
         | they can be made to work).
        
       ___________________________________________________________________
       (page generated 2022-12-06 23:01 UTC)