September 27, 2015
End Of Moore's Law For CPU Speed Increases Coming To End
The rate of computer processor power growth has slowed from doubling once every two years to once every 2.5 to 3 yeas. The doublings might stop entirely by about 2025.
Will other forms of technological advance speed up to take the place of CPU doubling speed as a driver of economic growth? I'm thinking of genetic editing with CRISPR-Cas9 and other genetic manipulation techniques yet to be discovered. Biotech continues to have a far smaller impact than the computer industry on economic growth and living standards. But in theory biotech ought to be able to deliver far larger benefits - most notably rejuvenation therapies. The question is when will it happen?
Randall Parker, 2015 September 27 11:57 AM
While it's true that transistor density is going to stop doubling, and we always knew it would, (They can't get smaller than atoms, given current understanding of physics.) there's still plenty of room for actual processing power to keep going up for a while, as the transistors get faster, the logic designs get more efficient, and, especially, as the code gets better.
For a long time now programmers have been relying on Moore's law to allow them to write really bulky, inefficient code, and just brute force run it. Most of the increased processing power has been sucked up this way. The nice thing, if you want to look at it that way, is that this means there's a lot of room now to get more out of the same power hardware. To start shedding layers of abstraction, and write code that actually makes efficient use of those transistors.
As well, the cost of individual transistors can keep dropping for quite a while; A thousand dollar multi-core processor is, after all, just a fancy few milligrams of a very common element; There's no floor on the price of such CPUs short of the cost of sand.
So, expect the processing power, and especially the *effective* processing power, per dollar, to keep climbing for quite a while to come.
There's probably room for more efficient organization as well as more efficient code. The human brain operates at a "clock speed" of a few tens of Hz, but the organization of e.g. the visual cortex makes it so good at image processing that computers are just barely catching up.
As for the impact of bio-engineering, I expect to see a tipping point phenomena. Government will resist the onslaught of change, but bio tech products are teeny-tiny, and the tech is portable.
I bought a computer 3 years ago, and went looking online recently. The exact same computer (with a +1 to its model number) is selling for the exact same price. It's absurd.
Seems clear enough that Moore's law COULD continue if manufacturers wanted it to, but nobody cares about PCs anymore. Mobile processors are still competitive.
"Will other forms of technological advance speed up to take the place of CPU doubling speed as a driver of economic growth?"--Randall Parker
Will these other forms of technological advance require FDA* approval? Now you know why semiconductor progress has outrun pharmacological progress despite both fields having had major breakthroughs in their fundamental concepts and methods since the 1950s.
*Don't suppose that the Silver Bullet of eliminating the FDA will make the biotech breakthrough-to-market bottleneck go away. The FDA is a major impediment but not the only one.
Sartrus, you're either missing under the hood details or got ripped off three years ago. The Core i3/5/7 in newly released PC models is not the same processors found then new models three years ago. The value of what you get for a set amount of money has improved significantly. This may not be obvious if you don't do anything that makes the system work hard but it's there. The exact same machine you bought three years ago, if anyone still has in their product line, should sell substantially cheaper today. My last laptop purchase was around $500 and the closest equivalent to that model currently sold runs around $300.
This is why newer server designs have more and more processor cores and those cores can support more threads of processing; and why virtual machine systems are coming more and more into play to get the most use out of all those cores and threads of processing.
Gejala Storoke ringan dan cara menanganinya secara alami harus diperhatikan selalu secara serius , karena saat ini kasus penyakit stroke banyak terjadi dikalangan masyarakat dari mulai stroke ringan hingga stroke berat.
Meski yang dialami masih gejala stroke ringan tapi penanganannya harus secara tepat karena bisa berakibat fatal jika tidak diatasi dengan obat stroke yang ampuh. Sebelum kita membahas lebih lanjut mengenai pengobatan stroke, maka kita harus mempelajari dulu tentang apa itu penyakit stroke, penyebab penyakit stroke, gejala stroke,
serta bahaya apa yang bisa muncul? Setelah itu baru kita bisa memilih obat stroke yang tepat yang akan kita gunakan.