Yes, but I think you're missing the point that the OP is really making: they are asking why improvements to processor speed are so danged incremental. Processors are maybe 200x times faster now than they were 25 years ago, but the point is that we got here, so it was physically possible. What stopped us from condensing the last 25 years of progress into 5 years? Or 1 year? Why is the progress of Moore's Law supposedly so inexorable? Does this indicate a "learned helplessness" of the industry, transitioning from the view that Moore's Law was an interesting phenomenon that arose from the industry without collusion, to the point where it now dictates what the product targets should be for this year and next year? Why is nobody trying to dramatically outstrip Moore's Law? Is it even possible to jump more than one process node ahead at a time, or increase IPC by an order of magnitude at a time rather than by a small percentage?
If Google can't afford to eat the cost of FTTP, they should charge consumers for the installation cost. I would pay for it. Not having that as an option sucks.
In 2016, the battery pack cost is still ~$227/kWh, meaning that a 60kWh Tesla battery pack is ~$13600. The target cost for parity with ICE vehicles is $100/kWh, which is likely to happen sometime between 2025 and 2030.
This presumably assumes prices continue to fall at the current rate. However, the Gigafactory is no longer the only game in town, there are at least two or three other players in the process of spinning up battery production facilities of similar scale. Massive increased demand from not just EVs, but also residential and commercial energy storage, and eventually short-range flying electric vehicles, will drive the cost down much faster than the existing rate.
Also: Material Design is an awful UI paradigm. All those gratuitous animations, increasing latency all over the UI -- it irks the heck of out of me. (I have as many animations as possible disabled in Android.)
This is not a study about honesty, it's a study about whether or not people are honest about their use of profanity. (The second sub-test, about pronoun usage, seems dubious at best.)
This too shall pass. The days are numbered for any old-timers that are embracing such an inefficient and clumsy medium for nostalgic purposes -- and for hipsters embracing old technology to be retro.
My last manager at Google was a gaslighter par excellence. I filed an HR complaint, but he was a director, so they sided with him. He threatened me with a performance improvement plan in order to hold my feet to the flames to force me out of the company (my performance was great). So I quit. Sometimes you literally have no power to fix these sorts of situations. Companies that tolerate (and even protect) these types of behaviors put themselves on a self-harming trajectory.
If you're using AdBlock Plus, every single page load will block until AdBlock Plus has finished running expensive regexps against the DOM. (That said, the browser really is bloated and slow, but invasive extensions like ad blockers will slow down the entire experience.) Just disabling AdBlock Plus won't necessarily speed up the browser though, because ad bloatware may slow down the browser even more than AdBlock Plus. Instead, try uBlock Origin -- it is much, much faster than AdBlock Plus, and supports all the AdBlock Plus blacklists. Your web experience will fly again.
Javascript is single-threaded by design, and hitting a script tag stops all partially-completed layout and rendering until the script has been parsed, compiled and run. Hopefully things will get faster once a big chunk of the web has been rewritten in a language that compiles to WebAssembly -- but this will take at least a decade or more.
Both Uber and Tesla have plowed headlong into creating autonomous driving technology, without engaging in rigorous formal system safety proof and testing methodologies -- and this is reckless and irresponsible. Waymo has driven over a billion miles of road in a simulator as a result of regression testing every major software update -- other companies are not apparently doing this. Waymo technology has something like eight nested failure modes of degraded performance if sensors or other hardware starts failing, and they use formal methods to provide soft proofs that these failure modes work in reliable ways. The other companies (all 50+ of them now) are simply saying, "let's just apply deep learning, it will give us super-human performance with very little effort." No.
... in other words, they didn't find cancer yet, in this study. But there is no promise that the same will be true for humans. Cancer cells *are* cells that have been reset to some early proliferative stage of development.
Google has commitments from all the major carriers to fix this exact issue in the next few months -- and there is major work going on in the underlying layers of Android to enable Google to push out updates without having to loop in carriers (presumably with the exception of updates to the baseband radio firmware, which has to be tested on the networks and approved by the carriers).
Also: 3D editing is hard on a 2D screen with primarily 2D input devices. It will probably always be hard until we get really good Brain-Computer Interfaces.
Couldn't they have used a nicer word than "smeared", like maybe "interpolated"? (See also other aversion-causing words: "moist", "crevice", "phlegm", "penetration", etc.)
Somebody please re-run these experiments specifically for marijuana. Somehow, pot-smokers have universally come to the conclusion that smoking burning cannabis leaves (without a filter) is not at all dangerous to them in any of the ways that smoking burning tobacco leaves (through a filter) is. I guarantee that the free radicals in cannabis smoke are every bit as dangerous as the free radicals in tobacco smoke.
Slashdot Mirror
Yes, but I think you're missing the point that the OP is really making: they are asking why improvements to processor speed are so danged incremental. Processors are maybe 200x times faster now than they were 25 years ago, but the point is that we got here, so it was physically possible. What stopped us from condensing the last 25 years of progress into 5 years? Or 1 year? Why is the progress of Moore's Law supposedly so inexorable? Does this indicate a "learned helplessness" of the industry, transitioning from the view that Moore's Law was an interesting phenomenon that arose from the industry without collusion, to the point where it now dictates what the product targets should be for this year and next year? Why is nobody trying to dramatically outstrip Moore's Law? Is it even possible to jump more than one process node ahead at a time, or increase IPC by an order of magnitude at a time rather than by a small percentage?
If Google can't afford to eat the cost of FTTP, they should charge consumers for the installation cost. I would pay for it. Not having that as an option sucks.
In 2016, the battery pack cost is still ~$227/kWh, meaning that a 60kWh Tesla battery pack is ~$13600. The target cost for parity with ICE vehicles is $100/kWh, which is likely to happen sometime between 2025 and 2030.
This presumably assumes prices continue to fall at the current rate. However, the Gigafactory is no longer the only game in town, there are at least two or three other players in the process of spinning up battery production facilities of similar scale. Massive increased demand from not just EVs, but also residential and commercial energy storage, and eventually short-range flying electric vehicles, will drive the cost down much faster than the existing rate.
Also: Material Design is an awful UI paradigm. All those gratuitous animations, increasing latency all over the UI -- it irks the heck of out of me. (I have as many animations as possible disabled in Android.)
Or just switch to Google Fi. You get 3G roaming data in something like 120+ countries at the same exact price as domestic 4G data in the US ($10/GB).
This is not a study about honesty, it's a study about whether or not people are honest about their use of profanity. (The second sub-test, about pronoun usage, seems dubious at best.)
Who the freak needs to give ANYONE permission to travel to the moon? I hate bureaucracy.
This too shall pass. The days are numbered for any old-timers that are embracing such an inefficient and clumsy medium for nostalgic purposes -- and for hipsters embracing old technology to be retro.
My last manager at Google was a gaslighter par excellence. I filed an HR complaint, but he was a director, so they sided with him. He threatened me with a performance improvement plan in order to hold my feet to the flames to force me out of the company (my performance was great). So I quit. Sometimes you literally have no power to fix these sorts of situations. Companies that tolerate (and even protect) these types of behaviors put themselves on a self-harming trajectory.
Does it run on GNU Hurd?
If you're using AdBlock Plus, every single page load will block until AdBlock Plus has finished running expensive regexps against the DOM. (That said, the browser really is bloated and slow, but invasive extensions like ad blockers will slow down the entire experience.) Just disabling AdBlock Plus won't necessarily speed up the browser though, because ad bloatware may slow down the browser even more than AdBlock Plus. Instead, try uBlock Origin -- it is much, much faster than AdBlock Plus, and supports all the AdBlock Plus blacklists. Your web experience will fly again.
Javascript is single-threaded by design, and hitting a script tag stops all partially-completed layout and rendering until the script has been parsed, compiled and run. Hopefully things will get faster once a big chunk of the web has been rewritten in a language that compiles to WebAssembly -- but this will take at least a decade or more.
Chris Chavez of Phandroid is going to make a killing off of this -- the Amazon link posted in the Fortune article has his referral code in it.
Oh yeah, I call Morgan Freeman all the time, too.
Both Uber and Tesla have plowed headlong into creating autonomous driving technology, without engaging in rigorous formal system safety proof and testing methodologies -- and this is reckless and irresponsible. Waymo has driven over a billion miles of road in a simulator as a result of regression testing every major software update -- other companies are not apparently doing this. Waymo technology has something like eight nested failure modes of degraded performance if sensors or other hardware starts failing, and they use formal methods to provide soft proofs that these failure modes work in reliable ways. The other companies (all 50+ of them now) are simply saying, "let's just apply deep learning, it will give us super-human performance with very little effort." No.
It's this sort of behavior that has been responsible for many forks in the history of software engineering.
... in other words, they didn't find cancer yet, in this study. But there is no promise that the same will be true for humans. Cancer cells *are* cells that have been reset to some early proliferative stage of development.
Hint: cancer.
"It's offensive, the way people ask about it," one fellow tells the reporter
Sounds perfect for entitled millenials. I wonder when the house of cards will come crashing down in their lives.
I have never paid Apple a single dollar.
Google has commitments from all the major carriers to fix this exact issue in the next few months -- and there is major work going on in the underlying layers of Android to enable Google to push out updates without having to loop in carriers (presumably with the exception of updates to the baseband radio firmware, which has to be tested on the networks and approved by the carriers).
Also: 3D editing is hard on a 2D screen with primarily 2D input devices. It will probably always be hard until we get really good Brain-Computer Interfaces.
Couldn't they have used a nicer word than "smeared", like maybe "interpolated"? (See also other aversion-causing words: "moist", "crevice", "phlegm", "penetration", etc.)
Somebody please re-run these experiments specifically for marijuana. Somehow, pot-smokers have universally come to the conclusion that smoking burning cannabis leaves (without a filter) is not at all dangerous to them in any of the ways that smoking burning tobacco leaves (through a filter) is. I guarantee that the free radicals in cannabis smoke are every bit as dangerous as the free radicals in tobacco smoke.
Just wondering.