TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

Moore’s Law Is About to Get Weird (2015)

101 点作者 sjayasinghe将近 9 年前

13 条评论

ansible将近 9 年前
Well, none of those listed technologies is practical.<p>Ternary logic is really cool, but it only gets you about a 10% efficiency improvement. I think it would be awesome to go in this direction, but retooling our entire tech stack for that isn&#x27;t going to happen. I mean, look at x86.<p>The only thing that will really make progress is molecular nanotechnology. I had always believed that reaching the end of silicon process technology would necessitate the research into MNT.<p>What&#x27;s funny is that we have arguably already reached the end on the silicon process technology, if Intel&#x27;s desktop product offerings are used as a guide. There are gamers who are building new rigs using chips that are 2 or 3 generations old, because there isn&#x27;t really anything better available. What&#x27;s really hilarious are the guys building systems with used Xeon CPUs from retired servers!<p>So I don&#x27;t know what&#x27;s going on any more. You would think that the profits and stock prices of all the silicon companies would collapse if they weren&#x27;t able to continue producing improvements on their existing products. Intel is spending massively on R&amp;D, but the gains are nowhere close to what we saw 15 years ago.<p>Note that keep referencing Intel, because they&#x27;ve been at the leading edge. AMD doesn&#x27;t have 14nm in production, for example.
评论 #12225883 未加载
评论 #12225566 未加载
评论 #12225335 未加载
评论 #12225257 未加载
评论 #12228269 未加载
评论 #12227563 未加载
评论 #12227668 未加载
评论 #12225859 未加载
steego将近 9 年前
For all the interesting things happening today, this is an incredibly dull article. I&#x27;m convinced I read this same article in the 90&#x27;s.<p>If we&#x27;re going to see any developments in alternative architectures and hardware, it&#x27;s going to happen in the cloud computing space. From my perspective, this is the only sector that is in the unique position to sell specialized computing services to customers on a mass scale and make great margins in the process.<p>There are many interesting and useful things you can do outside of Von Neumann architectures and Turing Complete systems that were never practical on PCs. Simple services like Reddis could probably be implemented with alternative memory architectures like content-addressable memory. Many useful algorithms in the data analytic and machine learning space can be compiled down the specialized hardware using using a combination of transistors and memristors.<p>For companies like Amazon, Microsoft + Google, there&#x27;s actually a financial incentive to harmonize hardware and software much in the same way Seymour Cray harmonized the software and hardware of Cray&#x27;s supercomputers. These cloud companies don&#x27;t want to be constrained to selling commodity virtual machines. They want to sell you the next generation PaaS solution like DynamoDB, SQL Azure or Firebase so they can lock you into their cloud platform.<p>If there&#x27;s a future where developers are writing code in special DSLs that&#x27;s compiled to gates on an FPGA, the cloud computing guys will be exploring it because they have a huge incentive to get their customer&#x27;s into the walled garden that yields better margins. If you can create one platform service that developers love and runs on cheaper specialized hardware, you&#x27;ll be able to destroy your competitors and signal an alternative hardware arms race.
评论 #12227532 未加载
评论 #12226439 未加载
评论 #12227817 未加载
评论 #12227373 未加载
erdevs将近 9 年前
From the article:<p>&gt; &quot;...Moore’s Law, which states that the number of transistors that can be squeezed onto a semiconductor chip of a given size doubles roughly every two years, has held true since the mid 1960s...&quot;<p>I thought it was the case that Moore&#x27;s &quot;Law&quot; hasn&#x27;t held up since ~2012.<p>I believe Intel and others have stated that as of 2012 or so, with 22nm processes, transistor density now doubles every ~2.5 years and is likely to slow further within 5 years without more fundamental breakthroughs.<p>The related trend that processor clock speeds doubled every 1.5-2 years-- which is commonly mistaken as Moore&#x27;s Law in lay press but is actually more related to Dennard Scaling, Koomey&#x27;s Law and historical statements by Intel-- had ceased to hold as of the late Aughts as well, I believe.<p>Memsistors have been a promise for some time. I wonder how we will make the leap from current dense-transistor IC tech and processes to anything new, given the huge infrastructure investments in fabs, quality control processes, etc. It seems it will be difficult to scale out a new technology, whatever it is, and get it price-competitive with existing transistor-dense IC technology. Perhaps I misunderstand here though, and would be interested in thoughts from those with greater expertise in the field.
评论 #12226240 未加载
rm999将近 9 年前
This is one of the most insightful things about Moore&#x27;s Law I&#x27;ve read: <a href="https:&#x2F;&#x2F;www.quora.com&#x2F;How-long-will-Moores-Law-continue-to-apply&#x2F;answers&#x2F;605270" rel="nofollow">https:&#x2F;&#x2F;www.quora.com&#x2F;How-long-will-Moores-Law-continue-to-a...</a><p>Moore&#x27;s Law is a limit that ultimately arises from economics. The costs to keep Moore&#x27;s Law going are exponential (just like the gain), and fewer and fewer companies are willing to pay the increasingly high price.<p>I was always primed to believe the exponential gains were essentially &quot;free&quot; and came with a fixed amount of investment into the industry. Understanding that this isn&#x27;t true explains why Moore&#x27;s Law will end unless demand is exponentially increasing.
protomok将近 9 年前
The chemical, wetware, fluid and ternary computing concepts mentioned in the article I guess make for a good headline but seem highly unlikely.<p>In the short term I really think we are entering the era of accelerators. Accelerators like Micron&#x27;s Automata Processor (1) for graph analysis, accelerators for compute intensive applications like Convolutional Neural Nets, continued innovation in DSPs, GPUs, etc.<p>Accelerators that can either outperform traditional processors or use significantly less power I think represent one of the major next steps after Moore&#x27;s law. What comes next may well be a slime mold computer but I think at this point is pure speculation.<p>(1) - <a href="http:&#x2F;&#x2F;www.micronautomata.com" rel="nofollow">http:&#x2F;&#x2F;www.micronautomata.com</a> - actually I heard about this on HN!
评论 #12227556 未加载
GrumpyYoungMan将近 9 年前
One interesting avenue of research is embedded reconfigurable computing technologies like eMIPS [0]. Unlike other reconfigurable computing approaches that simply place a FPGA side-by-side with a CPU in the same package to act as a co-processor [1], eMIPS integrates reconfigurable logic directly into the processing pipeline, allowing the creation of custom processor instructions on the fly on a per app basis to accelerate apps. Pretty cool stuff.<p>[0] <a href="https:&#x2F;&#x2F;www.microsoft.com&#x2F;en-us&#x2F;research&#x2F;project&#x2F;emips&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.microsoft.com&#x2F;en-us&#x2F;research&#x2F;project&#x2F;emips&#x2F;</a><p>[1] <a href="http:&#x2F;&#x2F;www.theregister.co.uk&#x2F;2016&#x2F;03&#x2F;14&#x2F;intel_xeon_fpga&#x2F;" rel="nofollow">http:&#x2F;&#x2F;www.theregister.co.uk&#x2F;2016&#x2F;03&#x2F;14&#x2F;intel_xeon_fpga&#x2F;</a>
api将近 9 年前
One thought that isn&#x27;t on here:<p>Once the process is &quot;mature&quot; and isn&#x27;t changing every few years, the error rate could be squeezed way down. That would make it practical and economical to make larger chips. There are cooling and power issues too but I&#x27;m sure those are solvable.<p>Make way for the 64-bit 128-core 1024gb RAM 12cm^2 SoC?<p>Call it More&#x27;s Law? Or a BFC? (for Big F&#x27;ing Chip?)
评论 #12225266 未加载
评论 #12225268 未加载
carlob将近 9 年前
Adamatzky seems to be churning out cookie cutter articles on slime molds and highways<p><a href="https:&#x2F;&#x2F;scholar.google.com&#x2F;scholar?q=Adamatzky+highways&amp;btnG=&amp;hl=en&amp;as_sdt=0%2C39" rel="nofollow">https:&#x2F;&#x2F;scholar.google.com&#x2F;scholar?q=Adamatzky+highways&amp;btnG...</a>
amelius将近 9 年前
How about the breakthrough in silicene production? [1] Could we use this to start stacking transistors in the vertical direction?<p>[1] <a href="http:&#x2F;&#x2F;spectrum.ieee.org&#x2F;nanoclast&#x2F;semiconductors&#x2F;materials&#x2F;breakthrough-in-silicene-production-promises-a-future-of-silicenebased-electronics" rel="nofollow">http:&#x2F;&#x2F;spectrum.ieee.org&#x2F;nanoclast&#x2F;semiconductors&#x2F;materials&#x2F;...</a>
评论 #12224899 未加载
tkinom将近 9 年前
Moore&#x27;s Law seems to reach its limit only for CPU clock. But it still moving very fast for GPU, FPGA, DDR, SSD, etc, right?<p>Or even in term of Network connection speed - 100mbps, 1G, 10G, 25G, 40G (at lease in data center where it matters a lot.)<p>The simple usb connection speed also continue to increase at a fast rate.
worik将近 9 年前
&quot;Nautilus uses cookies to manage your digital subscription and show you your reading progress. It&#x27;s just not the same without them&quot;<p>Well, no thanks. No, it is not the same.
james-watson将近 9 年前
As usual, memristor based computing architecture is left out.<p>Intel&#x27;s 3DXPoint has already brought memristor memory units to market. The beauty of memristor architectures is that they can be used for logic and storage, so you can create some radical new architectures with them.<p>I&#x27;m too lazy to link the talk by Stan Williams of HP, but if you google it it&#x27;s the ~45min one.<p>This is the next architecture because it&#x27;s already here. All that other stuff is vaporware.
评论 #12234697 未加载
评论 #12227588 未加载
TheArcane将近 9 年前
What about Quantum computing?
评论 #12225484 未加载
评论 #12225352 未加载
评论 #12225353 未加载
评论 #12225185 未加载