Become a fan of Slashdot on Facebook


Forgot your password?
Get HideMyAss! VPN, PC Mag's Top 10 VPNs of 2016 for 55% off for a Limited Time ×

Comment Re:What took them so long? (Score 1) 112

Sure, but unless you've developed a superconducting substrate, or come up with a reliable, efficient 3D cooling system, or are willing to run the 3D transistors only at very low speed/power, you're going to run into serious heat dissipation problems.

Back then I was proposing a diamond semiconductor - supported and powered by water-cooled silver busbars. Diamond is extremely conductive thermally. The bandgap is 5.5V, corresponding to the deep ultraviolet, so you can run it very hot without fouling the electrical properties (though you have to keep; it below 752 F or it will gradually degrade.) I'd want to put it in a bottle with an inert atmosphere so it wouldn't oxidize at high temperature, either.

The flip side of the big bandgap is that it consumes more energy - and generates more heat - when switching than current silicon designs which run at about a third that voltage.

These days I'd probably go for layers of graphine, which conducts heat even better than diamond.

With a rectangular solid you can get a LOT of transistors (and their interconnects) into a few cubic feet. The original proposal was for a six-foot cube - 216 cubic feet. Powering and cooling on two faces gives you 72 square feet of heat and power transfer serice, with 432 square feet on the other two faces for optical I/O fibers. Nowadays I'd take a page from Gene Amdahl and go a tad smaller: so, like the 1960s-era cabinets for IBM compter components, the block of logic and its supporting structures would fit into a standard elevator.

Comment What took them so long? (Score 1) 112

The report adds that processors could still continue to fulfill Moore's Law with increased vertical density.

What took them so long?

I've been pointing out that a three-dimensional arrangement off components could continue FAR longer than an essentially single-layer arrangements since at least the 1970s.

Comment Re:How Much? (Score 1) 71

I thought that much was obvious, but for those who have not been paying attention, we are close to using up our hydrocarbons.

Maybe four centuries for all sources of fossil carbon, hydrogenated or otherwise, depending on usage rate.

Remember that "reserves" means "the stuff we already found while exploring". Nobody with a financial clue spends today's private money exploring for stuff they won't be digging up and selling for decades. So you only have more than about 20 years of "reserves" when there have been giant finds, the known reserves are too expensive to exploit and there might be easier stuff out there, or too much of the known reserves are unexploitable due to things like government intervention. There's no doubt quite a lot more out there, though it's still finite.

Running out is not a disaster. We can easily make all the stuff that's made from oil and there are other energy sources - including more coming down the pipeline. We're only digging/pumping up most of our energy and much of our chemical feedstocks right now because it's CHEAPER than the alternatives.

But it's not cheaper by much. (Photovoltaic is now becoming competitive with grid power in many areas, even without government market distortions, and the tech just keeps improving.)

By the time the fossil fuels run out we'll have lots of alternatives, and they'll run out by gradually getting more expensive, so people will smoothly transition to alternatives (thanks to Adam Smith's "Invisible Hand"). The main problem (if the CO2->global warming conjecture is true and substantial) will be keeping the Earth from crashing into the next orbital-mechanics driven Ice Age (as humans MAY have been doing for about the last 10,000 years or so, as the orbital climate-forcing has been curving down steadily.)

Comment Caller id spoofing already broke that. (Score 1) 120

The real way to handle it is to create an open source shared black list, have people sign up for a service, and vote when they answer a call on whether or not it is a telemarketer or robo-call.

Caller ID spoofing already broke block lists. By the time a call gets to your local telco there is no way even for them to tell where it really came from. They regularly spoof their identity - often as others they're robo-calling, or even as the phone they are calling.

IMHO the only way available currently is to trace back a particular call, from telco to telco, to see where it DID come from - then go after the actual robocaller. (Good luck getting that implemented, though. Or getting it to work across all countries, rather than letting the spammers run from safe havens.)

Comment Re:I really don't understand this drone applicatio (Score 4, Insightful) 43

Why would you use a heavier-than-air craft to essentially hover? Wouldn't an aerostat accomplish the same goal at a much lower cost, and lower risk of bodily harm should it fall from the sky?

I don't know why they chose it. Here's my take:

An aerostat requires tethers, which are points of failure, and has enormous wind drag. Lose the tether(s) and you lose control. Then you have a large, failing, floating device at the mercy of the winds, dragging first broken tethers, then its own large structure, on an uncontrolled path along the ground, wreaking unknown havoc.

A powered heavier-than-air (but still ultralight) has little drag and can also be made to change locations easily. With good design, if it begins to accumulate failures that jeopardize its continued operational ability, it can be made to fly to a repair site and land - after its backup has arrived to take its place.

If you have catastrophic events - like huricaines, tornadoes, or forest firestoms - it can easily be moved away (to land for shelter or fly around or above the storm) and brought back when the environment is calmer. You don't even have to take it out of service. Just fly it above the tropopause. The stratosphere is probably a good place for it to operate anyhow: Negligible weather, no cloud shadows for solar-powered planes, and gives you a lot of coverage per drone. (Balloons can get there, too, easily. But 50,000 feet or so is a LOT of tether.)

Comment But will they play Badger Badger Badger? (Score 1) 156

Over the past few years, Firefox has implemented Web APIs to replace functionality that was formerly provided only by plugins.

But will they play Badger Badger Badger?

Until that can be emulated on the "replacement functionality", removing Flash is a fundamental impact on the Internet Experience. ;-)

Comment Look huge to me. (Score 2) 40

Google has how many subscribers now? Somehow, these numbers look astonishingly small

Google has how many employees to process these requests now? Somehow, these numbers look annoyingly huge.

How much does this cost Google to process? How much more does this cost to resist if Google wants to try to protect its customers' data, how much more to research whether each particular customers deserves this effort?

Can Google bill the governments for this service? Does this qualify as a fifth-amendment "taking"? Can google sue for reimbursement of these costs?

How much does this cost Google in lost revenue from people who bail out, or don't join, rather than leave their sensitive data where it is subject to search without their knowledge, and potential disclosure?

How much does incurring these costs result in raised costs or reduced services for Google's customers? How many, and what, services might they have to terminate, or never deploy, or never even develop, because the money that might have provided them is instead eaten by servicing government information requests? How badly does this impact their business models, their stock price, their investors' returns?

Comment I'm a boomer, but... (Score 2) 606

Oh, and by "we", I mean "baby boomers". I'm gen X and wasn't old enough to vote when all this shit really started in the 80s.

I'm a boomer - but I voted against pretty much all of this stuff. And campaigned against it, too. Virtually nobody I ever voted for was elected.

As for the political institutions: The generations before ours held onto power until quite recently (and have bequeathed it to individuals who are their ideological colleagues among later generations). Their crooked lock on the voting process has kept them in power. Look at the ages of the congresscritters and presidents. Even Bill Clinton was a pre-boomer - conceived DURING WWII, and growing up in a cohort where children were scarce and pampered, rather than a flood to be "channelled" into government-approved career paths (by threat of the draft during the Vietnam adventure).

Don't fall for the "blame the boomers" line: It's another instance of the power elite playing divide-and-conquer, to cut you off from potential allies.

Comment Useless - they're probably already filtering. (Score 2) 37

... most of the shmucks that ask for numbers like this use robo callers.

And the schmucks in question are normally cluefull enough to program their robots to NOT call the "premium content" number ranges. (Which is also what anyone programming a service that includes a callback feature should also do.)

Not doing this for cellphone ranges or numbers on do-not-call list doesn't impact a phone-pimp's bottom line. Trying to scam a pay-to-talk line does. It might not cost enough to bankrupt them, if their scam is lucrative enough - but even for those it would be a drain on the swag.

Comment Drop in the bucket (Score 1) 415

Note that the big-sounding numbers are in DOSES.

Divide by 365 for days in a year. Be generous and then divide again only by two (rather than, say, six or eight as is typical for painkillers). You're already talking 730 doses for ONE drug for ONE chronic pain patient.

So numbers like 265, 541, and 562 fewer doses correspond to less than one patient per doctor. Even the 1,826 for painkillers is less than the 2190 annual doses of a 6-per-day prescription for one chronic pain patient.

Yes, with 854,698 active physicians in the U.S., it does add up. But generic painkillers, antidepressants, and the like are cheap. This is a drop in the bucket compared to the more than $400 billion US market for all prescription drugs - or likely even the amount the drug companies spend on Congress to lobby for the drug war.

For me the big take-aways from this article are:

  * The impact of Medical Marijuana on overall drug company revenue is miniscule. Unless a fad catches on among doctors and they start switching some classes of patients en masse to M.J., the drug companies are unlikely to see any substantial drop in revenue, and would be ahead to save the lobbying money.

  * They might be much FARTHER ahead to start selling, reasonably cheaply, purified, standard-dose, convenient oral tablets of the several active compounds. Especially if they can get the government to declare them "orphan drugs" or some new category, so they don't have to spend a bunch on research or accept large-scale liability for possible side-effects, and can let the non-drug-company-funded researchers in the medical community continue to identify the conditions (such as intractable seizures) that these compounds improve.

Comment Re:Free-speech-for-all spectrum? (Score 1) 64

When you develop a network technology that can handle numerous transmitters on the same frequency---some of which are far more powerful than the one you're listening to---go ahead and let the rest of us know.

It's a class of modulation schemes called "spread spectrum". Been around for decades. If it's done right you can't even tell there are other signals out there unless you know THEIR spreading codes because they look too much like thermal background noise.

Yes, they "make the grass taller" and the total bandwidth is still limited by Shannon's criteria (with the interfering signals acting as "noise" in the "signal-to-noise-ratio" term.) So being located too close to "someone shouting" still makes it hard to hear. (Like being next to a waterfall.)

On the other hand, licensing doesn't seem to do what it claims, either. I've got a vehicle with a Sirius satellite receiver, and some company has, over the last couple years, been installing what appear to be cellular mini-towers that create block-wide dead zones all over Silicon Valley. (Some of them make my 2G phone drop connections, too, and some other spots kill the phone without killing the satellite feed.) The zones are too large for the 4-second buffer to cover. My commute is about 14 miles, almost all on 880 and/or 680 in Fremont and Milpitas (if I don't shunpike) and I can't find a route with less than four dead spots for the phone and three for Sirus-XM.

Slashdot Top Deals

The reason computer chips are so small is computers don't eat much.