Catch up on stories from the past week (and beyond) at the Slashdot story archive


Forgot your password?

Comment Re:20 cores DOES matter (Score 1) 167

Since the inception of HT, is there a reason CPU design hasn't advanced to the point of executing 4 threads per core rather then the 2 it always has been?

Workload and system balance, mostly.

If you look back several years (2008? earlier?) you'll see some Sun Sparc designs, and some IBM POWER designs, that supported 4 or 8 threads per core. They worked well for very specific workloads and applications.

The Sun Sparc designs with 8 threads per core were mostly tailored for "simple" highly-scalable web servers, where a thread is blocking on I/O most of its time, and a web server could spawn many many threads to support many simultaneous connections. Worked very well for that purpose.

IBM did stuff like that with their POWER architecture for terminal servers and financial transaction processing, where, again, the thread spends most of its time blocking on I/O.

You don't get that so much for Intel x86/x64 systems, because, on the desktop side, frankly, most users don't use 4 cores well, and the few that do aren't doing I/O-blocking tasks, they are doing CPU-bound tasks, video encoding, stuff that hits the SIMD units hard. HT doesn't benefit nearly as well for CPU-bound tasks, and that market is small enough not to be worth the extra architecture/development time. For x64 servers, there is a bit more of a market there, but Intel would much rather serve that market with their high-end Xeon 4-socket systems. 10 cores per CPU, 4 CPUs, you get 40 cores and 80 threads. Oh, and you pay about $4,000 per CPU that way. That also gets you ridiculous amounts of RAM, and better networking support too. Usually you want both of those on your 80-thread server system, anyway.

So I suppose the answer is, basically, it has, but only where it's worthwhile.


Comment Re: Not a chance (Score 1) 631

The problem in the sig
          1^2=1; (-1)^2=1; 1^2=(-1)^2; 1=-1; 1=0.
is really at the step
          1^2=(-1)^2; 1=-1

Because that is wrong. The step pretends to be sqrt() on both sides, but that's not what it is, because sqrt() really has 2 answers. sqrt(1) is either +1 or -1, simultaneously. To get the step as shown, what is really happening is abs(sqrt()).

Which obviously yields invalid results. You can't do abs() on both sides of an equal sign and expect to keep a valid equation.

But I agree with you, it looks like it should end "2=0", if you allow the wrong sqrt() step.

Comment Re:By Country (Score 1) 199

But maybe the real root of the problem is, why does the US truly need to 'project force' in a unilateral sense?

Because the 1800's taught us that wars on our own soil are bad. See, for example, the British burning Washington D.C. during the War of 1812 (which actually lasted 1812-14). The reason it is called the WHITE House was a not-so-subtle "Screw You!" to the British when we rebuilt it after they burned it to the ground.

The 1900's taught us that wars "over there" are much better. The US was the major economic and miltary superpower from 1945 to 1990 because our infrastructure wasn't destroyed by a continent-wide war 1939-1945.

A slightly different question is "Do we need to meddle in other people's affairs?" but, as a species, humans aren't good at not poking the bee hive.

Comment Re:Sigh (Score 2) 341

If there is evidence that a legislator is guilty of insider trading, or any other crime, they should be tried by their peers in congress, not by the justice dept.

The problem here is that doesn't work in practice. For evidence supporting this statement, I give you every internal police investigation into officer wrong-doing ever. If you haven't found it yourself before, how about this article written by an Air Force colonel whose son was shot in the head by police while hand-cuffed in custody. The officers were cleared of wrong-doing by an internal investigation. I don't expect an unbiased viewpoint from this man, but the stats he found don't make "tried by their peers" sound like anything resembling a workable solution.

People and groups put in charge of writing laws, with a history of making laws that benefit themselves and hold themselves to a lower standard than the rest of the population, need more oversight, not less.

Try again.

Comment Re:Are they "small government" republicans ? he he (Score 1) 393

Extreme right: Cut spending to equal revenue.
Extreme left: Raise revenue to equal spending.
Center: Continue to give everyone whatever they want, and borrow money from China to pay for it.


I thought the Extreme Right plan was cut revenue to destroy government, the Extreme Left plan was raise revenue while raising spending more, and the Center just looks left and right with a bewildered expression while muttering "Don't you people have any brains???"

Most Extreme Right seem to want government services without paying for them, and/or want a system that is only good for the rich and lucky. I'm middle class with manageable debt, good job and health insurance, and no real medical problems. That counts as "lucky" in America.

Most Extreme Left seem to want the government to take most of the roles of family, and some of the roles of community (while usually not admitting that community used to equal church).

150 years ago it wasn't that unusual for 3 generations of a family to live in one house. That covered retirement for the old, child care for the young, and consolidated housing expenses that made life possible. We don't do that so much any more, and we are still trying to figure out how to make our new system work.

I admire the problem, but I don't have a good solution that scales to 300 million people.

Comment Re:If anyone actually cared... (Score 2) 710

I have a Maytag dryer that I bought in 2002. Bought the washing machine with it, like most people who buy appliances when setting up a new house.

That dryer developed a nasty high-pitched squeal - metal on metal rubbing that didn't used to do that. Spend some time looking around on google, and this model has a known problem with this. It's a front-load dryer, and the drum sits on two rollers. The right-side roller is directly under a vent and gets condensed water dripping on it. So the metal wheel inside the roller rusts, wears away, and the dryer starts squealing.

You can buy a replacement roller kit on Amazon for about $7. It takes 1-2 hours to take the dryer apart (door, front panel, drum access plates, drum) to get access to the drum roller on the back inside wall of the dryer, change the roller, and put it all back together again. There are videos on Youtube giving step-by-step directions for doing this.

I've done this twice now. The first time was a pain to do, the second was just annoying in a "What? Again?" kind of way. I assume I'll do it a third time in another year or two.

It's worth doing for $7 DIY.

No way I'd pay someone $100-$200 to fix it for me. The attitude is "The dryer's not worth that much!"

As you say, the labor costs kill the repair market.

But if you can use a screw driver and a pair of plyers, it's amazing how much stuff that is designed to fail with "planned obsolescence" you can fix and keep working.

I'm told you have the same issues with cars, but frankly there I'm willing to pay someone to do the maintenance for me. I think I'm mostly worried about an expensive "learning experience" that leaves me without a vehicle at an inconvenient time. But I've got friends that would never pay someone to do an oil change on their car. They can do it themselves easier and cheaper, plus they like working on cars.

It's personal comfort level as much as anything else.

Comment Re:If people would fight their tickets... (Score 1) 286

I've gotten several speeding tickets in the last 20 years. Only contested one of them, just paid the rest. I *was* speeding, after all.

The one I went to court for? 53 in a 45 mph zone. When I looked down at my speedometer, after seeing the police cruiser, my speedometer read 50. I probably coasted a few mph slower between seeing the cop car, cursing quietly, and looking down, so that's fine.

But the cop wrote up the ticket for an intersection that doesn't exist, a block over and up where the two roads don't actually cross. I'm anal enough, I wanted the ticket corrected before I paid it. So I go in to court, and the officer can't find the certification for his speedometer calibration within the previous 6-month period, required by Florida law at the time for moving-mode radar guns - he was driving past me in the opposite direction. So I sat quietly, didn't say a word, and the judge dismissed the ticket before they ever talked to me.

And I got rewarded for being picky about details. :) Doesn't hapopen very often, but I take my victories where I can.

Comment Re:sigh (Score 1) 627

And you "save" not tax with a 401(k), you just get to defer some. The Roth is the only one that lets you "avoid" tax.

You don't avoid taxes with either of them. Not exactly. You make a guess about your current tax rates versus your future tax rates, and act based on the guess.

For a traditional IRA, or a 401(k), you contribute pre-tax dollars (salary that doesn't count towards income tax). This is generally assumed to be contributed at a time in your life when you are in a higher income tax bracket, and you are contributing dollars that would be taxed at a higher rate, 28%, maybe, and instead will pay tax on it when you withdraw it. The assumption is that, in your retirement, your "income" is lower, therefore your tax rate is lower, maybe only 20%, so you avoid paying 8% income tax by paying the tax later rather than sooner.

A Roth IRA is for people in one of two situations. They have maxed out their traditional IRA contributions (IRS only allows about $15,000/year), and/or they don't have a job-associated 401(k) or 403(b). Basically, a 403(b) is a 401(k) but your employer is a charity. For the Roth IRA, you pay taxes on contributions now, and when you take the money out in retirement, it is tax free because taxes were already paid on contributions. If your tax rate now is lower than you think your tax rate will be in retirement, this is a good option for you.

It's all about minimizing taxes based on expectations of current and future marginal tax rates.

Comment Re:Peak During the Day? (Score 1) 504

My point is only that excess residential solar has little value, since it's generated when it's least needed. [Because if it was needed, houses wouldn't be generating extra.]

That's not usually how these kinds of rooftop solar systems are designed. Or, at least, its not the only way they can be designed.

The other way - that causes the grip operator the most trouble - is to spec the rooftop solar system so that it generates, over a year's time frame, the amount of kWh that the house uses. That makes you "net-0" for grid usage. That means that during the day, when you are getting power from the rooftop solar, you are almost *always* generating more than you use. And at night, when you are generating none, you get power from the grid. But over the course of a year, the power you draw from the grid and the power you supply to the grid approximately equal.

Your power bill should be zero, right? Well, not really. Because you're using the grid as a big redundant battery for overnight and cloudy days. And you should pay for that.

This is the situation that most clearly shows the need to separate the grid charges into "plant maintenance" and "electricity production". Plant maintenance covers the cost of the electrical lines, transformers, sub-stations, some reasonable percentage of the cost of the generating system (the coal plant or whatever), and labor for maintenance, plus reasonable overhead. Electricity production covers the cost of the fuel/coal/gas/whatever, the "rest" of the cost for the generating system, maintenance, and again reasonable overhead.

Honest grid operators (contradiction? I hope not) and honest rooftop solar advocates (err...) should both be pushing for this.

Sounds fair, huh? Right, it'll never work.

Comment Re:It followed a few of the plot lines, but ... (Score 1) 726

The crime rate has plummeted in recent decades, you know.

Not the only factor, I know, but...

There's a really interesting 10-15 year lag from the removal of leaded gas from American society, and the drop in the crime rate. It's almost like exposure to lead in early childhood causes developmental problems in the brain related to anger management and impulse control in adults. Maybe there's even some medical studies on the effects of lead in people...

Didn't the Romans have societal problems when they introduced lead-lined aqueducts?

Causation, correlation, and coincidence. Whatever, it's a fun little statistic (not to be confused with useful data).

Comment Re:Cheap Hydrogen (Score 1) 55

Yeah, this is basically how I interpretted it. Where the hydrogen comes from is outside scope of the sales pitch. I don't care about a portable battery replacement though.

I have a house with "common asphalt shingles" like most home owners in the US. When that house needs to be re-roofed, I'd like to get a set of solar panels, if I can convince myself at the time that it is cost-effective. That will probably be in 10-15 years, as the house was built in the mid-1990s. A large part of the cost of consumer rooftop solar panels is the installation, not the panels. Double the number of panels, installation cost doesn't change that much, use the extra energy to split water into hydrogen, store the hydrogen and use it at night to power the house when the sun doesn't shine. Keep the electric utility connection (and, reasonably, pay some kind of "connection fee" even if I don't use any electricity, probably even if I net provide power instead of consume it) and I have self-sufficient home electrical power for a one-time payment. I can probably tax deduct the interest if I pay for it with a home improvement loan, too.

Now, is it really economically feasible to do that? The rooftop solar panels and DC-AC converters, yeah, they tend to have an okay ROI now, less than 15 years for a system that should last 25-30 years.

Add in a water electrolysis system, hydrogen storage, and a hydrogen fuel cell? Okay, that's harder to make the numbers work out right. I'm still hopeful for 10 years from now, though.

It's hard to convince myself this won't become standard in the southern US in 20 years, if the engineering can get worked out.

Comment Re:jerk (Score 5, Interesting) 1440

Come on now. If you see a traffic cop, he's not there to "protect and serve." They are the Badged Highwaymen, state-sanctioned assholes whose job it is to flip the lights on behind random people in the universal cop-sign for "stick em up and hand over your wallet, brownie."

Seriously? As an honest reply to this (okay, I admit, I just got trolled) traffic cops are there for several reasons.
A) Revenue collection. I'd be dishonest if I didn't admit that up front.

B) Keeping traffic close to speed limits. Yeah, the definition of "close" varies from cop to cop, and that makes it hard for a driver to drive with a lot of confidence of just how fast you can drive without getting a ticket. I hate that. I'd like an up front admission of "The speed limit is 70, but we won't ticket anyone doing under 82 unless they are otherwise driving unsafely". We'll never see that. Besides, "driving unsafely" is hard to define, but it's easy to give the guy changing lanes unsafely a speeding ticket, and it punishes unsafe behavior about as well (which means, not very) as a reckless driving ticket does, but it takes less to defend in court.

C) Being nearby when there is an accident. A nearby traffic cop is a first-responder for a traffic accident, and that job saves lives. They also do care-and-comfort during and after accidents. You look in any highway patrolman's trunk, and you'll find a teddy bear to be given to the little kid that survived a traffic accident (whose parent maybe didn't).

Most good traffic cops (and almost all Highway Patrol) regard speeding tickets as a way to get traffic to slow down so when there is an accident, there will be fewer deaths. In their job, it's always "when" and not "if" there is an accident. Energy is mass times velocity squared, remember.

Doing A lets the state pay for more cops to be around for C. Can't really tell you if I like that trade-off or not.

And yeah, none of this stops me from being pissed when I get a speeding ticket. Don't they have something better to do than bug me when I'm not hurting anyone?!?! ;)

Comment Re:Funding isn't automatic now (Score 2, Informative) 522

This is one of those "Lying with facts" things that needs more context to correctly understand.

The House of Representatives is currently controlled by a Republican majority, 232 (R) vs 200 (D). A simple majority is all that is required to pass any Bill in the House of Representatives, therefore, so long as the Republican caucus can keep its members in line, they can pass anything, no matter how much Democrats hate it, with no thought at all about compromise.

The Senate, on the other hand, has a Democrat majority of 53 Democrats, plus 2 independents that caucus with the Democrats. That's 55 Democrati-caucussed Senators. That's a "Democratic-controlled Senate", true. However.... Functionally, the Senate can't pass much of anything, especially a budget bill, without a 60-vote majority. Therefore, they require at least 5 Republican Senators to agree to a mutually-acceptable Bill. Quoting myself above... "so long as the Republican caucus can keep its members in line"...

For the Democratic-controlled Senate to pass a budget bill, the Republicans and Democrats have to find an acceptable compromise.
For the Republican-controlled House to pass a budget bill, the Republicans don't have to care about an acceptable compromise at all.

The House passes a Budget Bill. The Senate doesn't. Pretending those are equivalent situations is lying with facts.

The larger issue is that neither side seems willing to compromise much at all, so finding an acceptable compromise is much harder that you'd normally think it would be.

Slashdot Top Deals

"355/113 -- Not the famous irrational number PI, but an incredible simulation!"