William Holt, General Manager of Intel's Technology and Manufacturing group, has now gone on the public record declaring their future chips will be getting slower in the name of energy efficiency as Moore's law comes to an end (archived). It has been clear for some time that the end of Moore's law would be coming sooner rather than later. Now Intel has come out in the public record for the first time to unveil their coping strategy of hopefully being able to leverage quantum phenomena in to make gains in efficiency at the cost of speed. This shift in computing hardware, perhaps the most consequential since the move from vacuum tubes to solid state transistors, eliminates the long held assumption that computers inevitably get faster over time.1 This statement by Intel is another piece of evidence demonstrating why XTCoin could have never been Bitcoin and why markets are the answer for scaling Bitcoin, not maximum blocksize increases.
This assumption has survived the stall in single threaded CPU performance which began in 2004. Before 2004 a 50% improvement per year was standard. After 2004 13-21% became the new reality (archived). In the near future not regressing in performance will be the hope. ↩
For the record, I wish to be known as the Moore law ender. Because it is I to publicly say the common wisdom re "Moore law" isn't worth twopence back in the days of the "expert consensus" re Gavincoin, early 2015.
"Here lies a law, borne by one, Moore, buried by one, Popescu."
I dunno, this "actual physical reality" guy makes a pretty convincing case for the title.
Sure he does, but…. well, we can't all be first.
Maybe we decide it with a coinflip. Just gotta wait for him to show up so he can pick heads or tails.
It is in fact very regular, and very stable, and the old one will fade/die. Yes, supposedly significant reserve capacity exists.
The silence on this incredible announcement is deafening. Everyone has taken speed increases for granted; now barring some breaktrhough, there are going to be years of flat-line on the performance graphs. Expect a large increase in the size of the overclocking equipment market.
On a related note, the $50,000,000 United States Dollar investment in Blockstream is another indicator that sabotaging Bitcoin simply will not be tolerated, even by people not in the WOT.
SegWIt is not sabotaging Bitcoin? TIL
So long as it stays a fruity opt-in thing like RBF it mostly just creates a hazard for people opting in to it.
Well, good thing we still have Nielsen's Law, right guys? Phew…
Right..!?
Guys??!
That doesn't apply in Viacomcastland, does it ?
lol, pulling numbers out of their asses to try and make it look regressive which, you know.., is almost linear if you look at it from the right angle.. (notwithstanding that the original intention of the broken-ass law was users' bandwidth – not downtown data-center approximation). Goal posts. They are there to be moved.
I'm not following. How does a future of slower chips imply that XTCoin could have never been Bitcoin?
Gavin was yakking about how Moore's law will allow to process gigabyte sized blocks in the future, therefore allowing for gigablocks in future is OK.
Ah, right. I see now. Thank you