Well, how much slower would they be in the lower atmosphere? As the other poster pointed out, at the epicentre Nagasaki they had 1005 km/h winds. This has 3600s/h * 5km/s = 18000 km/h. So if at the ground the winds are "only" about 6% as bad, it would be "only" like being at ground zero of a nuke. For hours. That seems pretty bad all right:P
If you think that's bad, I humbly submit the following personal anecdotes:
1. Oblivion. So there's this mess of cultists and the high priest is right in front of them preparing to sacrifice someone. Being the sneaky barsteward I am, I plug him right in the head with an enchanted bow. So not only he does a spectacular back-flip in front of everyone, but he bursts into a very bright and spectacular flame too.
So the cultists freak out and start running around, don't find me. One minute later, they calm down and one of them goes, "It must have been the wind."
I don't know what kind of weather they have down there.
2. NOLF 2. So they had actually gone through the trouble of scripting reactions when an NPC finds a body. They'd shake it, ask stuff like "are you alive, comrade??", flip out and search for the killer, etc. Must have been fun in the original version.
Except some retard decided to replace all corpses with backpacks in the German version. You can probably see where this is going.
Yep. Some soldier would find a backpack on a bed in the barracks, freak out, and go "are you alive, comrade??" and the whole circus. To a backpack. WTF.
First of all, we're talking a planet considerably heavier than Jupiter, so presumably a gas giant. Or anyway it will have quite the pressure.
Second they said it produced explosive winds, up to 5 km/s. (Or "fucking unbelievably fast" in imperial units;) Because the air heats from 500 degrees to 1200 degrees on the hot side within hours, and expands, rushing towards the colder side.
That's over 5 times the muzzle velocity of an M16, BTW.
So, yeah, the conditions on the surface might not be as bad as they seem... as long as you don't mind winds strong enough to blast you to bits, not to mention that they're 1200 degrees hot;)
Or that history repeats itself
on
Less Is Moore
·
· Score: 5, Insightful
Well, actually it's just proof that history repeats itself. Because this thing has happened before. More than once.
See, in the beginning, computers were big things served by holy priests in the inner sanctum, and a large company had maybe one or two. And they kept getting more and more powerful and sophisticated.
But then it branched. At some point someone figured that instead of making the next computer which can do a whole megaflop, they can do a minicomputer. And there turned out to be a market for that. There were plenty of people who preferred a _cheap_ small computer, than doubling the power of their old mainframe.
You know how Unix got started on a computer with 4k RAM, which actually was intended to be just a co-processor for a bigger computer? Yeah, that's that kind of thing at work. Soon everyone wanted such a cheap computer with a "toy" OS (compared to the sophisticated OSs on mainframes) instead of big and powerful iron. You could have several of those for the price of a big powerful computer.
Then the same thing happened with the micro. There were plenty of people (e.g., DEC) who laughed at the underpowered toy PCs, and assured everyone that they'll never replace the mini. Where is DEC now? Right. Turned out that a hell of a lot of people had more need of several cheap PCs ("cheap" back then meaning "only 3 to 5 thousands dollars") instead of an uber-expensive and much more powerful mini (costing tens to hundreds of thousands.)
Heck, in a sense even multitasking appeared as sorta vaguely the same phenomenon. Instead of more and more power dedicated to one task, people wanted just a "slice" of that computer for several tasks.
Heck, when IBM struck it big in the computer market, waay back in the 50's, how did they do it? By selling cheaper computers than Remington Rand. A lot of people had more use for a "cheap" and seriously underpowered wardrobe-sized computer than for a state of the art machine costing millions.
Heck, we've even seen this split before, as portable computers split into normal laptops and PDAs. At one point it became possible to make a smaller and seriously less powerful PDA, but which is just powerful enough to do certain jobs almost as well as a laptop does. And now it seems to me that the laptop line has split again, giving birth to the likes of the Eee.
So really it's nothing new. It's what happens when a kind of machine gets powerful enough to warrant a split between group A who needs the next generation that's 2x as powerful, and group B which says, "wtf, it's powerful enough for what I need. Can I get it at half price in the next generation?" Is it any surprise that it would happen again, this time to the PC? Thought so.
Of course the 68k architecture was easily 50 times better than x86, so that helped too! lol. I still think its a tragedy that x86 went anywhere, the whole design is $*}#...
Admittedly, my experience is quite the opposite: I worked mainly on the x86 (and 8080/Z80 before it), where having your counter in the CX register actually made a difference, and having your lookup table in BX made another difference, and so on. And adding some SIMD instructions to it only made it even more quirky.
Still, even on more orthogonal architecture you can usually squeeze _something_ out for your own peculiar case, that the authors of a generic compiler never thought about. A lot less, admittedly.
The other issue was of course that case where the compiler just missed something entirely, but those cases are almost always ones where you can tweak the source a bit and fix the problem or use a different flag.
Very much so, but my point was exactly that you need:
A) to at least have seen what the compiler produced, or failing that
B) at least some vague notion of the machine behind, and what the compiler does,
to spot the need for a source tweak.
The fact is, I work all the time with people who've never even seen assembly, and have really no flipping clue of the code behind it all. It shows. E.g., their assumptions of the costs of various operations is usually missing the mark by a mile. You see them assuming, for example, that speed or cost is measured in lines of code, and the fact that one is a call to bogo-sort and the other is "i++" goes right over their head.
Basically the fact that you've worked on those drivers, know what the machine can do, and actually compared C compiler output to your own assembly... that IMHO made you a better programmer, whether you're aware of it or not. Even if you write C for the rest of your days, you'll write better C code for it. IMHO.
As far as efficiency goes, in the old days it was true that a coder with an intimate knowledge of the architecture could usually hand code more efficient assembly. Modern C compilers however can do a LOT of optimization and generally the resulting code is faster than anything that could be coded by hand, or at least AS fast.
The funny thing, though, is that I've heard that since at least the 80's. (Never mind that it was provably wrong, too.) But there was always some crowd repeating the same mantra. There was always some nebulous X years ago when competence was worth it, and a now where the compiler does a better job than anyone imaginable.
I'd imagine that it might be as old as when v2.0 of the first compiler was available. See last year's FORTRAN was worse than hand-optimized assembly, but this new compiler does everything better than any human! Knowing the underlying architecture is soo 50's. Move on to the 1960's already, you assembly has-beens. Heh.
The problem with the above line of thinking is assuming that a human would do the same optimizations that a compiler does. And in that case, yes, might as well let the compiler do it. What a human does better though is changing the algorithm, if another one fits the architecture better. The main advantage is probably not even the actual coding of assembly itself, but seeing actually what's the resulting code, exactly which registers are used and what the timings are for doing things the X way vs the Y way. Seeing that the X way actually needs swapping values to memory, while the Y way fits them nicely in the available registers. _That_ is the advantage.
Additionally, almost no compiler yet is _that_ complete. Most have basically a table of usual patterns that come with the associated clever optimizations, but are sometimes thrown off the hook by as little as writing a comparison the other way around. Java's JIT is for example well known and documented for it. And yes, the first Hotspot versions were documented to be thrown off the loop by as little as changing an "if (a == b)" to "if (b == a)" in a popular benchmark at the time.
There'll always be some piece of code which misses that completely.
Which brings us to another point: cheating. If you think that the compiler will perform as good on _all_ cases as it performs on the showcase benchmarks, I have some logging rights in Sahara to sell. The generated code for popular benchmarks, which performs as well as the best hand-optimized assembly... guess why? Because it _is_ hand-optimized assembly. The compiler is programmed to recognize whole chunks of that benchmark and essentially use the hand-optimized assembly that exists just for it.
Assembly has its own problems, don't get me wrong. It's _very_ expensive to produce, a nightmare to maintain, and isn't cross-platform either. I wouldn't advise it for whole projects or for most projects, for that reason. Bang per buck, the vast majority of stuff out there just doesn't justify using assembly. But the myth that an expert can't produce better assembly code, is still just that: a myth.
The real question is whether you want to pay a _lot_ more, for a little extra speed. But (unless you hired a monkey) that extra speed has always been there, and still is there.
A) power corrupt formerly honest and nice people, or
B) it's just natural selection at work, at the biggest turds float to the top?
It seems to me more like B, though I can't say I've done a real study or anything.
The thing is, if you have a dog-eat-dog set up, the ones who refuse to eat other dogs (e.g., because of having morals) never make it big in the first place. Either they don't get promoted, or they get their prices undercut by someone who saves by being a bigger fuck, and either go bankrupt or bought.
As an extreme example to illustrate a point, think, say, a third world country where it's not illegal to dump toxic stuff in rivers and safety laws are non-existent. So company A are the nice guys, they don't want to screw over their workers and community. They invest in filters, invest in safe equipment and training, doesn't bribe/deceive/lobby/make backroom deals, etc. So their products are more expensive. Their competitor, company B, are owned and led by a couple of greedy fucks, who just skip all that extra cost and do any tricks in the book to get a goverment subsidy or contract. If it's a big bribe or shady deal that gets that job done, so be it. So their products are cheaper. Do you have any doubts as to who's going to push the other off the market?
(It's not even as much a hypothetical example, because it used to happen in the first world too, in the not so distant past. E.g., back when the Titanic was built, the norm was IIRC to have one dead worker for every million dollars worth of ship built. The Titanic was remarkable in that they only had IIRC 3 dead workers in accidents during building. But anyway, roll that in your head, they actually made statistics and found it acceptable to kill people rather than spend money on safety. It's not a funny thought.)
It's easy to look afterwards at the big resulting conglomerate "B Industrial Corp" and think, "man, all that power corrupted them." But in fact they got to power by not being nice in the first place.
1. Yes, we have a bunch of _stories_. It's been a popular subject for novelists, playwrights, myths, etc, because we all _like_ to believe that everyone is the same, and the scary criminals out there can be turned good by just appealing to their humanity.
It's like why crucifixes or garlic are supposed to help against vampires, a bit of iron helped against the sidhe (elves), silver keeps werewolves at bay, and why bringing offerings to the tribal totem was supposed to help you get a good hunt and keep you from getting eaten by a wolf. It's a scary thought that it's all a big crapshot and something or someone could terminate your life, without any "saving throw", just because you were in the wrong place at the wrong time. So for all dangers -- real and imaginary alike -- we've invented some bullshit defense. Just carry a bit of garlic with you against vampires, a bit of iron against the sidhe, and remember to appeal to the killer's humanity. Heh. Snake oil at its finest.
We have no scientifically documented cases of anyone who stopped being a sociopaths. So if your being "scientific" is believing myths and legends, well, you might have a nasty surprise at some point.
What we _do_ have, though, is people realizing "man, if I don't stop, they'll hang me". Or "lock me up for life." That's why, say, three strike laws actually made a heck of a difference.
What we also _do_ have is people who, basically decide to take it all with them, or rather blow their ill gotten gains on something that will get them remembered. Sort of like Bender building a giant monument to himself in that episode. So sociopathic robber barons in their old age blow most of their wealth on some charity or such, to carry their name for another century. But, here's the fun part, while being still just as ruthless to their workers, business partners, etc. Downside, that tends to only work for guys who managed to rob a few billions, not for small fish like this one.
In effect, all that can happen is the sociopath putting on another _mask_, if you can give him enough of a reason to. Elliminating the disincentive not to, well, I fail to see how that'll help.
2. Well, let's return to this case, and how much humanity he's shown so far.
At some point when one of his script-kiddie apprentices got second thoughts, do we see this guy thinking, "hmm, maybe he's right"? No, we see him telling said apprentice to "stop being a bitch."
And let me remind you that he has actually _seen_ the effects of his actions, because he was consulting for security to the same people. He _saw_ the havoc and stress he created, first hand. Then stole their passwords and pwned their machines.
Do I believe that he'll get reformed by seeing it one more time? Not bloody likely, I would say.
Look, I'm not going to say that they're 100% irreparable. I'm saying that a lot of psychologists and psychiatrists tried to repair them, and nothing worked. I'm saying that
A) we _don't_ _know_ how to repair them, and
B) if there's a way, it's certainly not as simple as pouting and showing them the people harmed. That's been tried, it never worked. I fail to see how repeating that we should try that again is going to make any difference.
1. The people in prisons score on the average over 20 on the antisocial personality disorder scale, which is to say you have a spectrum ranging from borderline sociopathic to outright psychopaths. A normal person scores 2-3.
2. There is no known way to turn a sociopath into a normal person. Trying to psychanalyze them just teaches them to fake the answers that will hide their callousness better.
3. Showing one the damage he's done and the people whose life he's destroyed... does nothing whatsoever, since a sociopath doesn't give a fuck about other people in the first place. They live in a single-player world, with them as the player and the rest being about as important or empathy-worthy as the NPCs in <insert MMO or RPG>. You can lie to them, manipulate them, cause all the harm you can get away with, whatever advances your quest or keeps you entertained. It doesn't matter, they're just NPCs. That's the kind of world a sociopath lives in. It includes even their own children, not just strangers who downloaded a virus.
4. They have a tendency to not have a sense of personal responsibility. They'll just shift the blame to someone else (e.g., the victim for being too stupid to download a virus) or rationalize it in any other way.
So, seriously, if you know some way to "undo" sociopathy, by all means, we'd all be very interested to hear it. But otherwise let's bury the retarded idiocy already that prisons should be some touchy feely school in respecting other people's feelings. These guys just can't do that.
The only thing they do understand is, basically, "let's not do something that will get me locked up for good". Well, some of them. Turning it all in just a slap on the wrist and some pouty "you've been a meanie and upset people" lesson will just remove that deterrent too.
To sum up: Logic is about truth, not expertise, or anything else.
Actually, nope, logic is merely about whether the conclusion follows from the premises. It says nothing about truth of the conclusion. You can still reach a wrong conclusion from a false premise.
E.g., "the Earth is flat => from a tall enough tower you can see any point on Earth" is provably logical. Well, adding a couple of minor premises that have to do with geometry. But its main premise is false, so the conclusion is false too. But the inferrence is perfectly logical.
The "likelihood" of correctness is not relevant in classifying fallacies. You either have proof or you don't.
In formal logic, yes. In informal logic it's a lot fuzzier. You apply lots of things daily which _probably_ are correct.
Your example is a textbook case of the fallacy, because that's exactly the kind of argument that people tend to use: "this guy is an expert, therefore he must be right". A poignant counterexample is when you have two experts who disagree. They can't both be right, yet according to your "logic" they have to be.
Actually, that's your own mis-reading. I've never said that either _has_ to be 100% right.
It is also possible that Obama's intuition about dark matter turns out to be correct in the end. Then what? Hence, the converse of the fallacy is that you are discounting non-experts just because they are non-experts. If a high-schooler tells you 1+1=2, is he wrong because he got a D in Algebra?
Here again you seem to mis-understand what logic _is_. An argument like "the earth is round, therefore 1+1=2" is still false, even if the "1+1=2" conclusion is correct.
And an appeal to authority of the kind "X is an authority, he says 1+1=2, therefore 1+1=2" can be dismissed even earlier if X isn't, in fact, an authority in maths. And someone with a D in maths certainly isn't.
Basically, as formal logic goes "A => B" doesn't say anything whatsoever about B, if A is false. B could be true, or could be false, but that inference certainly didn't prove either.
Any attempt to prove B starting from a false A is doomed to be illogical. So anything that starts with "X is an expert in maths, therefore..." is a priori false if X had a D in maths. The conclusion may be right or wrong, but the "proof" is just not valid.
That was formal logic.
In informal logic we deal more with probabilities, even if most often subconsciously and without putting an exact number with lots of decimals on it. But basically most of your decisions are based on a vote in that neural network you have in your head, and fuzzy ("gut feeling") estimates of outcomes. You don't deal with "B is true" or "B is false", but with a P(B), which is a fuzzy probability of B being true. Whenever you decide whether to buy brand X or brand Y, essentially you don't have formal binary-logic proofs, you run a vote through that neural network and whichever decision seems more _likely_ to be correct, wins.
And there "X says B" ends up estimated on how likely you can trust X. If X won the Nobel Prize on that topic, then (in the absence of other data) P(B) is very high. (But again, I'm not claiming 100%, i.e., he doesn't _have_ to be right.) If X is the high-schooler with a D in maths from your example, then (in the absence of other data) P(B) can be assumed to be 50%. Basically he could be wrong, he could be right, and you'll have to use other data (even if your own experiences or knowledge) to push that probability either way.
So to sum this long rant up:
1. in formal logic, any appeal to authority is a fallacy. If he isn't an authority at all in the first place, it just makes it even funnier, but just as much a fallacy anyway. Yes.
2. in informal logic, you essentialy evaluate how _likely_ something is. It's not a proof in the formal logic sense. But if two guys advise you on what
Actually, on one hand, indeed no authority is infallible or beyond questioning. That's one thing more people should remember.
On the other hand, I still think there is a very valid distinction to be made between (A) a real authority on the subject matter, and (B) an arbitrary authority on another domain, and who's probably talking out the arse.
Basically,
A) if someone whose Ph.D. is on ancient middle eastern history tells me "you should believe what I say because I am experienced and I know what I am talking about" about the Neo-Assyrian empire, chances are he _does_ know what he's talking about. He's of course not infallible and not beyond questioning, but chances are that he does have some years and many tomes that he bases his assertions on. That's a valid authority to trust, unless you have evidence to the contrary.
B) if the same claim is made by, say, someone whose whole claim to glory is having starred in some movie set in that age, then the situation is exactly reversed. I have not much reason to believe that he knows enough to make such a claim, unless he can present more credentials.
Same about any other domain. If a lawyer tells me that X is illegal, by default I trust that he's probably right. If Hans Reiser (an authority on filesystems at least) tells me about the law, I ask a lawyer too. If an eye doctor tells me I need glasses, I tend to assume that he's probably right. If an archaeologist tells me the same, no matter how famous or how much an authority on his own field, I go ask a doctor. Etc.
Lumping both extremes under "baffle them with bullshit" seems to me, sad to say, idiotic.
Actually, I think the key word is missing there. The real fallacy is "argument from false authority."
As a hypothetical example: If an recognized astrophysicist says that there's something fishy about the amount of existing dark matter, that's a real authority on the subject matter, and is certainly something to keep in mind. If Obama says it, he's just not qualified to make that kind of a judgment, and it's simply something to ignore. For all his authority in politics and law, he's as qualified to talk about astrophysics as the local barber.
In this case I don't think Linus is an authority on usability or anything even remotely relevant to KDE vs GNOME. It's his personal tastes vs yours, nothing more. Unless you happen to know that his tastes accidentally match yours to the letter, it's something to thoroughly ignore.
Of course, that won't stop people from being fashion victims and trying to imitate him anyway. That's why celebrity endorsements work. That's why you see video clips with Van Damme and whatnot saying that they play WoW, for example. Because a lot of John Does out there will try to be like monkeys imitating that celebrity. Or why you see Fatal1ty branded heatsinks, although I don't think he'd know enough physics to actually judge a design, nor the experience of having tested 100 heatsinks and picked the best. That's appeal to false authority.
I don't doubt that here too a lot of people switched to KDE just because Linus blasted GNOME, and will now hastily switch back to GNOME because Linus uses it now so it must be cool.
1. Well, as an aspie myself, I seriously don't understand his defense. Asperger's Syndrome is basically like being colour blind, except in this case we aren't wired to even notice (much less decode) body language. I can tell if somebody screams or laughs, but everything else doesn't even exist for me. And far as I can tell mom can't even tell if you screamed at her or not, and is constantly taking wrong guesses there.
So offending people face to face or commiting social faux pas is a lot easier, because where someone else would take a hint, you don't even have a hint. E.g., I've had stuff like being told "dude, why didn't you stop it with that terminally bored face in the meeting? Didn't you see the way the boss was looking at you?" And I was thinking he looks at me because he likes me or something.
It's also very easy to conclude stuff like "everyone else is stupid" when you lack the hints that she's just making conversation and trying to sound interesting (or so I'm told,) or he's lying to you and hoping _you_ are stupid enough to believe him. (I find that if you dig deep enough in why someone insists on something illogical, you'll actually find a hidden motive rather than complete idiocy.)
On the other hand, being an aspie is all about logic, so anyone who blames it for not understanding "break law => get punished" is talking out the arse.
And you _can_ learn to function pretty normally in society by using logic, an I mean in a lot more detail than "break law => get punished". I've read a lot about psychology and anthropology, for example, just to know what I'm supposed to do or not to do, in the absence of ad-hoc hints to change the course.
2. _However_, Asperger's Syndrome has a very high probability of co-morbidity with something else, like OCD, OCPD, ADHD and going all the way to sociopathy.
And it seems to me like the _real_ problem of both this guy and Hans (the other with the aspie defense) is actually sociopathy. Both seem to be self-centered arseholes, and both seem to think that the law is some kind of game, among other symptoms.
I don't think we let people free just because they're sociopaths. In fact, most of the population in prisons scores over 20, a normal person scores 2-3, and 30 is the limit for outright psychopathy. That lack of empathy for their fellow man and society is usually what gets half of them into prison. (And the other half into upper management.)
But at any rate, that's a completely different mental disorder. And blaming it on Asperger's Syndrome does a disservice to everyone.
I don't know about every version of Windows, but I do happen to know that NT needed much more memory than Netware for the same performance as a file server, back when everyone ditched Novel in favour of MS. NT's saving grace was that NT _and_ the extra memory cost much less than a Netware upgrade. But the performance difference was there, and only excessive caching fixed it.
Actually, you illustrate exactly my problem. Better than I could. Thanks.
Let's put it like this:
1. _All_ a fragmented swap file (or any other file) adds is longer seeks back and forth to write the fragments.
2. Any competently written OS has _no_ excuse to block while waiting for a seek. It should just schedule another thread.
3. Windows does stall for seeks, as you've noticed.
Don't get me wrong, your point about the advantages of keeping the seeks short with a contiguous swap partition, is very valid too. The difference between 1-2 track seek and seek-across-the-drive is one order of magnitude.
But still, that's an optimization. That difference should just translate into maybe 1% difference in performance, as everything else (and especially the GUI threads) should be scheduled in that time. But, as you've noticed, the effect is a lot more dramatic.
It's not just the windows explorer copy that sucks. My point stands about just about any other way to do that. Total Commander, Install Shield, you name it. A lot of the IO wait translates into basically (a percentage of the) time where the whole machine does nothing else.
For the actual transfer, yes, you have to poll and suck in the bytes one by one. But AFAIK you don't have to wait in a loop for the seek to finish. Unless we're talking about old MFM drives on the XT. As I was saying, the PC architecture does have interrupts for just about everything. So you just start another process and wait for the interrupt.
Heck, even without interrupts, typical total HDD latency is between 6ms (Velociraptor) and 20+ms. You can just poll once every millisecond (or how often your scheduler interrupt comes) and do something else until it finishes.
Copying a file between two partitions is a scenario that's 90% dominated by waiting for the head to be in the right position, and for the right sector to come under it. Make that maybe 99% if the partitions are on the same physical HDD. Any competently written system, even in PIO mode has 90% of the CPU time there free for other processes.
At least NT obviously didn't have. As I was saying, with the drivers configured wrong in PIO mode, it just froze solid for a second when minimizing a window. I can't imagine any scenario where that's normal.
And the fact that Windows simply gets stuffed there, tells me that something's rotten in the kingdom of Redmond. Maybe it's not Windows itself but the IDE drivers or whatever. But something is thoroughly suboptimal in its disk IO IMHO.
Moraelin brings up the problems with an inaccessible swap. Well, if that's occurring, then whoever set up the system should have spent more time planning his/her partition layout so that a swap partition is mandatory.
I mean, for example, the HDD overheats and doesn't do more than throw fits. Err, faults. It happened to me, circa 2001, after stuffing increasingly hot components in a box that was obviously designed for old cool stuff.
We're not talking something that happens regularly. And just an old-and-upgraded home box, not some enterprise system with redundant hard drives mirrored.
Actually, while I do somehow sped more time at home on my Windows gaming box than under Linux (so this isn't a blanket Windows bashing,) my superficial and uninformed impression was that, all else being equal, any Windows box I've seen seems harder hit by IO than any Linux/Unix box I've ever seen.
Yes, you can get a Linux box to crawl too, if the hard drive is stuffed and it can't swap for example. Or if the chipset isn't supported well by the drivers. (Rarer these days, but certainly possible.) Or whatever.
But Windows... seems a bit special. I mean try to copy a directory between two hard drives, or better yet from a DVD to HDD, and Windows seems to me basically stuffed. Even notepad can get about as responsive as a narcoleptic snail. And you can just about forget about, say, playing a game while that happens.
And that's before you even add such brakes as an anti-virus.
I've seen that behaviour in any Windows, from 3.0 to Vista, including a detour through NT 4.0. In fact in Vista let's just say there's a reason why so many people were pissed off at the indexer kicking in all the time.
My subjective impression is that I've yet to see Linux get anywhere near that unresponsive, in a similar scenario. Again, assuming that you don't have a nearly dead HDD and the chipset is supported in DMA mode.
But heck, even in PIO mode, I've used Linux in PIO mode and I've used, say, NT in PIO mode. (Thanks to a retarded IT department which installed the wrong IDE drivers.) Linux did obviously have poor file IO performance, but NT just freaking _froze_ for a second or two, for example, when minimizing or maximizing a window. (Presumably due to aggressive memory management which swapped more of a process out when minimized.)
Now admittedly I haven't actually programmed an OS at any point, so I'm probably talking out the arse, but I see no reason why that should happen at all. Any common source of IOWait has an interrupt. Even in PIO mode you don't have to poll until it's done. And DMA, now that was invented for the precise reason and purpose of transferring some data while the CPU services another process. It's why it's there. So there's no freaking reason for the whole OS to just twiddle its thumbs and wait. Even if one process is waiting for _paging_, you can still yield to another process while waiting for the HDD.
1. Yes, it is entirely within the society's rights to decide that it requires certain standards of proof that your game doesn't cause people to run amok.
But even there it would first need to figure out such an impartial double-blind test, that's possible to pass. For medicine we have it. I don't think any of the "OMG, games turn people psycho" crowd actually made even a valid correlation, much less a scientific test there.
2. In medicine's case the FDA first had to make its own claim, and prove it, namely that the existing unchecked medicine can kill people. The idiot who sold a solution of sulpha in ethylene glycol, essentially made that case and proof for them. Somebody had sold a medicine which didn't just sometimes not work, and didn't just sometimes kill its patients. It killed _every_ single person who took it.
You're free to try to disprove their claim too, but they have a bullet-proof documentation of everyone who bought that medicine and died, and what are the biological effects of ethylene glycol.
Before the state could make that claim and proof, there was no reason to let them pass new legislation. As long as the "but someone could sell medicine which kills people" proof hadn't been made, we all just defaulted to the existing idea, namely, "nah, mate, the free market will solve everything, because nobody would risk the massive PR loss of face of selling poison as medicine."
So even there there was an implicit burden of proof on the FDA and generally those wanting more legislation.
It seems to me like it's not unreasonable to expect the same kind of burden of proof on those wanting to regulate games. As long as they don't make a reasonable proof that games do kill (e.g., by turning some people into amok psychos), we can and should tell them to take a fucking hike. Just like everyone did with the FDA before that incident.
I see no fundamental problem there. Essentially you're legally required to make a claim that your medicine (A) works better than placebo, and (B) you know and disclose the risks and potential side effects. And you're required to prove it.
The requirement part comes from having such bad experiences as someone selling sulpha dilluted in ethylene glycol... which is a very deadly poison, and actually killed everyone who took that medicine. In excruciating pain, over a couple of weeks. So now if you don't or can't make and prove those claims, you're not allowed to market that medicine. Or not as medicine.
But that's largely a legal construct, and has nothing to do with how logic works or how burden of proof works. We as a society decided that you _must_ make and prove that kind of claims.
But burden of proof even there works the old fashioned way. _You_ must prove it. It doesn't work like "it's good medicine unless someone else can prove otherwise."
Slashdot Mirror
Well, how much slower would they be in the lower atmosphere? As the other poster pointed out, at the epicentre Nagasaki they had 1005 km/h winds. This has 3600s/h * 5km/s = 18000 km/h. So if at the ground the winds are "only" about 6% as bad, it would be "only" like being at ground zero of a nuke. For hours. That seems pretty bad all right :P
If you think that's bad, I humbly submit the following personal anecdotes:
1. Oblivion. So there's this mess of cultists and the high priest is right in front of them preparing to sacrifice someone. Being the sneaky barsteward I am, I plug him right in the head with an enchanted bow. So not only he does a spectacular back-flip in front of everyone, but he bursts into a very bright and spectacular flame too.
So the cultists freak out and start running around, don't find me. One minute later, they calm down and one of them goes, "It must have been the wind."
I don't know what kind of weather they have down there.
2. NOLF 2. So they had actually gone through the trouble of scripting reactions when an NPC finds a body. They'd shake it, ask stuff like "are you alive, comrade??", flip out and search for the killer, etc. Must have been fun in the original version.
Except some retard decided to replace all corpses with backpacks in the German version. You can probably see where this is going.
Yep. Some soldier would find a backpack on a bed in the barracks, freak out, and go "are you alive, comrade??" and the whole circus. To a backpack. WTF.
First of all, we're talking a planet considerably heavier than Jupiter, so presumably a gas giant. Or anyway it will have quite the pressure.
Second they said it produced explosive winds, up to 5 km/s. (Or "fucking unbelievably fast" in imperial units;) Because the air heats from 500 degrees to 1200 degrees on the hot side within hours, and expands, rushing towards the colder side.
That's over 5 times the muzzle velocity of an M16, BTW.
So, yeah, the conditions on the surface might not be as bad as they seem... as long as you don't mind winds strong enough to blast you to bits, not to mention that they're 1200 degrees hot ;)
Well, actually it's just proof that history repeats itself. Because this thing has happened before. More than once.
See, in the beginning, computers were big things served by holy priests in the inner sanctum, and a large company had maybe one or two. And they kept getting more and more powerful and sophisticated.
But then it branched. At some point someone figured that instead of making the next computer which can do a whole megaflop, they can do a minicomputer. And there turned out to be a market for that. There were plenty of people who preferred a _cheap_ small computer, than doubling the power of their old mainframe.
You know how Unix got started on a computer with 4k RAM, which actually was intended to be just a co-processor for a bigger computer? Yeah, that's that kind of thing at work. Soon everyone wanted such a cheap computer with a "toy" OS (compared to the sophisticated OSs on mainframes) instead of big and powerful iron. You could have several of those for the price of a big powerful computer.
Then the same thing happened with the micro. There were plenty of people (e.g., DEC) who laughed at the underpowered toy PCs, and assured everyone that they'll never replace the mini. Where is DEC now? Right. Turned out that a hell of a lot of people had more need of several cheap PCs ("cheap" back then meaning "only 3 to 5 thousands dollars") instead of an uber-expensive and much more powerful mini (costing tens to hundreds of thousands.)
Heck, in a sense even multitasking appeared as sorta vaguely the same phenomenon. Instead of more and more power dedicated to one task, people wanted just a "slice" of that computer for several tasks.
Heck, when IBM struck it big in the computer market, waay back in the 50's, how did they do it? By selling cheaper computers than Remington Rand. A lot of people had more use for a "cheap" and seriously underpowered wardrobe-sized computer than for a state of the art machine costing millions.
Heck, we've even seen this split before, as portable computers split into normal laptops and PDAs. At one point it became possible to make a smaller and seriously less powerful PDA, but which is just powerful enough to do certain jobs almost as well as a laptop does. And now it seems to me that the laptop line has split again, giving birth to the likes of the Eee.
So really it's nothing new. It's what happens when a kind of machine gets powerful enough to warrant a split between group A who needs the next generation that's 2x as powerful, and group B which says, "wtf, it's powerful enough for what I need. Can I get it at half price in the next generation?" Is it any surprise that it would happen again, this time to the PC? Thought so.
Admittedly, my experience is quite the opposite: I worked mainly on the x86 (and 8080/Z80 before it), where having your counter in the CX register actually made a difference, and having your lookup table in BX made another difference, and so on. And adding some SIMD instructions to it only made it even more quirky.
Still, even on more orthogonal architecture you can usually squeeze _something_ out for your own peculiar case, that the authors of a generic compiler never thought about. A lot less, admittedly.
Very much so, but my point was exactly that you need:
A) to at least have seen what the compiler produced, or failing that
B) at least some vague notion of the machine behind, and what the compiler does,
to spot the need for a source tweak.
The fact is, I work all the time with people who've never even seen assembly, and have really no flipping clue of the code behind it all. It shows. E.g., their assumptions of the costs of various operations is usually missing the mark by a mile. You see them assuming, for example, that speed or cost is measured in lines of code, and the fact that one is a call to bogo-sort and the other is "i++" goes right over their head.
Basically the fact that you've worked on those drivers, know what the machine can do, and actually compared C compiler output to your own assembly... that IMHO made you a better programmer, whether you're aware of it or not. Even if you write C for the rest of your days, you'll write better C code for it. IMHO.
The funny thing, though, is that I've heard that since at least the 80's. (Never mind that it was provably wrong, too.) But there was always some crowd repeating the same mantra. There was always some nebulous X years ago when competence was worth it, and a now where the compiler does a better job than anyone imaginable.
I'd imagine that it might be as old as when v2.0 of the first compiler was available. See last year's FORTRAN was worse than hand-optimized assembly, but this new compiler does everything better than any human! Knowing the underlying architecture is soo 50's. Move on to the 1960's already, you assembly has-beens. Heh.
The problem with the above line of thinking is assuming that a human would do the same optimizations that a compiler does. And in that case, yes, might as well let the compiler do it. What a human does better though is changing the algorithm, if another one fits the architecture better. The main advantage is probably not even the actual coding of assembly itself, but seeing actually what's the resulting code, exactly which registers are used and what the timings are for doing things the X way vs the Y way. Seeing that the X way actually needs swapping values to memory, while the Y way fits them nicely in the available registers. _That_ is the advantage.
Additionally, almost no compiler yet is _that_ complete. Most have basically a table of usual patterns that come with the associated clever optimizations, but are sometimes thrown off the hook by as little as writing a comparison the other way around. Java's JIT is for example well known and documented for it. And yes, the first Hotspot versions were documented to be thrown off the loop by as little as changing an "if (a == b)" to "if (b == a)" in a popular benchmark at the time.
There'll always be some piece of code which misses that completely.
Which brings us to another point: cheating. If you think that the compiler will perform as good on _all_ cases as it performs on the showcase benchmarks, I have some logging rights in Sahara to sell. The generated code for popular benchmarks, which performs as well as the best hand-optimized assembly... guess why? Because it _is_ hand-optimized assembly. The compiler is programmed to recognize whole chunks of that benchmark and essentially use the hand-optimized assembly that exists just for it.
Assembly has its own problems, don't get me wrong. It's _very_ expensive to produce, a nightmare to maintain, and isn't cross-platform either. I wouldn't advise it for whole projects or for most projects, for that reason. Bang per buck, the vast majority of stuff out there just doesn't justify using assembly. But the myth that an expert can't produce better assembly code, is still just that: a myth.
The real question is whether you want to pay a _lot_ more, for a little extra speed. But (unless you hired a monkey) that extra speed has always been there, and still is there.
Well, I'm guessing the latter. I mean: does
A) power corrupt formerly honest and nice people, or
B) it's just natural selection at work, at the biggest turds float to the top?
It seems to me more like B, though I can't say I've done a real study or anything.
The thing is, if you have a dog-eat-dog set up, the ones who refuse to eat other dogs (e.g., because of having morals) never make it big in the first place. Either they don't get promoted, or they get their prices undercut by someone who saves by being a bigger fuck, and either go bankrupt or bought.
As an extreme example to illustrate a point, think, say, a third world country where it's not illegal to dump toxic stuff in rivers and safety laws are non-existent. So company A are the nice guys, they don't want to screw over their workers and community. They invest in filters, invest in safe equipment and training, doesn't bribe/deceive/lobby/make backroom deals, etc. So their products are more expensive. Their competitor, company B, are owned and led by a couple of greedy fucks, who just skip all that extra cost and do any tricks in the book to get a goverment subsidy or contract. If it's a big bribe or shady deal that gets that job done, so be it. So their products are cheaper. Do you have any doubts as to who's going to push the other off the market?
(It's not even as much a hypothetical example, because it used to happen in the first world too, in the not so distant past. E.g., back when the Titanic was built, the norm was IIRC to have one dead worker for every million dollars worth of ship built. The Titanic was remarkable in that they only had IIRC 3 dead workers in accidents during building. But anyway, roll that in your head, they actually made statistics and found it acceptable to kill people rather than spend money on safety. It's not a funny thought.)
It's easy to look afterwards at the big resulting conglomerate "B Industrial Corp" and think, "man, all that power corrupted them." But in fact they got to power by not being nice in the first place.
1. Yes, we have a bunch of _stories_. It's been a popular subject for novelists, playwrights, myths, etc, because we all _like_ to believe that everyone is the same, and the scary criminals out there can be turned good by just appealing to their humanity.
It's like why crucifixes or garlic are supposed to help against vampires, a bit of iron helped against the sidhe (elves), silver keeps werewolves at bay, and why bringing offerings to the tribal totem was supposed to help you get a good hunt and keep you from getting eaten by a wolf. It's a scary thought that it's all a big crapshot and something or someone could terminate your life, without any "saving throw", just because you were in the wrong place at the wrong time. So for all dangers -- real and imaginary alike -- we've invented some bullshit defense. Just carry a bit of garlic with you against vampires, a bit of iron against the sidhe, and remember to appeal to the killer's humanity. Heh. Snake oil at its finest.
We have no scientifically documented cases of anyone who stopped being a sociopaths. So if your being "scientific" is believing myths and legends, well, you might have a nasty surprise at some point.
What we _do_ have, though, is people realizing "man, if I don't stop, they'll hang me". Or "lock me up for life." That's why, say, three strike laws actually made a heck of a difference.
What we also _do_ have is people who, basically decide to take it all with them, or rather blow their ill gotten gains on something that will get them remembered. Sort of like Bender building a giant monument to himself in that episode. So sociopathic robber barons in their old age blow most of their wealth on some charity or such, to carry their name for another century. But, here's the fun part, while being still just as ruthless to their workers, business partners, etc. Downside, that tends to only work for guys who managed to rob a few billions, not for small fish like this one.
In effect, all that can happen is the sociopath putting on another _mask_, if you can give him enough of a reason to. Elliminating the disincentive not to, well, I fail to see how that'll help.
2. Well, let's return to this case, and how much humanity he's shown so far.
At some point when one of his script-kiddie apprentices got second thoughts, do we see this guy thinking, "hmm, maybe he's right"? No, we see him telling said apprentice to "stop being a bitch."
And let me remind you that he has actually _seen_ the effects of his actions, because he was consulting for security to the same people. He _saw_ the havoc and stress he created, first hand. Then stole their passwords and pwned their machines.
Do I believe that he'll get reformed by seeing it one more time? Not bloody likely, I would say.
Look, I'm not going to say that they're 100% irreparable. I'm saying that a lot of psychologists and psychiatrists tried to repair them, and nothing worked. I'm saying that
A) we _don't_ _know_ how to repair them, and
B) if there's a way, it's certainly not as simple as pouting and showing them the people harmed. That's been tried, it never worked. I fail to see how repeating that we should try that again is going to make any difference.
Actually, here's a fun thought:
1. The people in prisons score on the average over 20 on the antisocial personality disorder scale, which is to say you have a spectrum ranging from borderline sociopathic to outright psychopaths. A normal person scores 2-3.
2. There is no known way to turn a sociopath into a normal person. Trying to psychanalyze them just teaches them to fake the answers that will hide their callousness better.
3. Showing one the damage he's done and the people whose life he's destroyed... does nothing whatsoever, since a sociopath doesn't give a fuck about other people in the first place. They live in a single-player world, with them as the player and the rest being about as important or empathy-worthy as the NPCs in <insert MMO or RPG>. You can lie to them, manipulate them, cause all the harm you can get away with, whatever advances your quest or keeps you entertained. It doesn't matter, they're just NPCs. That's the kind of world a sociopath lives in. It includes even their own children, not just strangers who downloaded a virus.
4. They have a tendency to not have a sense of personal responsibility. They'll just shift the blame to someone else (e.g., the victim for being too stupid to download a virus) or rationalize it in any other way.
So, seriously, if you know some way to "undo" sociopathy, by all means, we'd all be very interested to hear it. But otherwise let's bury the retarded idiocy already that prisons should be some touchy feely school in respecting other people's feelings. These guys just can't do that.
The only thing they do understand is, basically, "let's not do something that will get me locked up for good". Well, some of them. Turning it all in just a slap on the wrist and some pouty "you've been a meanie and upset people" lesson will just remove that deterrent too.
Actually, nope, logic is merely about whether the conclusion follows from the premises. It says nothing about truth of the conclusion. You can still reach a wrong conclusion from a false premise.
E.g., "the Earth is flat => from a tall enough tower you can see any point on Earth" is provably logical. Well, adding a couple of minor premises that have to do with geometry. But its main premise is false, so the conclusion is false too. But the inferrence is perfectly logical.
In formal logic, yes. In informal logic it's a lot fuzzier. You apply lots of things daily which _probably_ are correct.
Actually, that's your own mis-reading. I've never said that either _has_ to be 100% right.
Here again you seem to mis-understand what logic _is_. An argument like "the earth is round, therefore 1+1=2" is still false, even if the "1+1=2" conclusion is correct.
And an appeal to authority of the kind "X is an authority, he says 1+1=2, therefore 1+1=2" can be dismissed even earlier if X isn't, in fact, an authority in maths. And someone with a D in maths certainly isn't.
Basically, as formal logic goes "A => B" doesn't say anything whatsoever about B, if A is false. B could be true, or could be false, but that inference certainly didn't prove either.
Any attempt to prove B starting from a false A is doomed to be illogical. So anything that starts with "X is an expert in maths, therefore..." is a priori false if X had a D in maths. The conclusion may be right or wrong, but the "proof" is just not valid.
That was formal logic.
In informal logic we deal more with probabilities, even if most often subconsciously and without putting an exact number with lots of decimals on it. But basically most of your decisions are based on a vote in that neural network you have in your head, and fuzzy ("gut feeling") estimates of outcomes. You don't deal with "B is true" or "B is false", but with a P(B), which is a fuzzy probability of B being true. Whenever you decide whether to buy brand X or brand Y, essentially you don't have formal binary-logic proofs, you run a vote through that neural network and whichever decision seems more _likely_ to be correct, wins.
And there "X says B" ends up estimated on how likely you can trust X. If X won the Nobel Prize on that topic, then (in the absence of other data) P(B) is very high. (But again, I'm not claiming 100%, i.e., he doesn't _have_ to be right.) If X is the high-schooler with a D in maths from your example, then (in the absence of other data) P(B) can be assumed to be 50%. Basically he could be wrong, he could be right, and you'll have to use other data (even if your own experiences or knowledge) to push that probability either way.
So to sum this long rant up:
1. in formal logic, any appeal to authority is a fallacy. If he isn't an authority at all in the first place, it just makes it even funnier, but just as much a fallacy anyway. Yes.
2. in informal logic, you essentialy evaluate how _likely_ something is. It's not a proof in the formal logic sense. But if two guys advise you on what
Actually, on one hand, indeed no authority is infallible or beyond questioning. That's one thing more people should remember.
On the other hand, I still think there is a very valid distinction to be made between (A) a real authority on the subject matter, and (B) an arbitrary authority on another domain, and who's probably talking out the arse.
Basically,
A) if someone whose Ph.D. is on ancient middle eastern history tells me "you should believe what I say because I am experienced and I know what I am talking about" about the Neo-Assyrian empire, chances are he _does_ know what he's talking about. He's of course not infallible and not beyond questioning, but chances are that he does have some years and many tomes that he bases his assertions on. That's a valid authority to trust, unless you have evidence to the contrary.
B) if the same claim is made by, say, someone whose whole claim to glory is having starred in some movie set in that age, then the situation is exactly reversed. I have not much reason to believe that he knows enough to make such a claim, unless he can present more credentials.
Same about any other domain. If a lawyer tells me that X is illegal, by default I trust that he's probably right. If Hans Reiser (an authority on filesystems at least) tells me about the law, I ask a lawyer too. If an eye doctor tells me I need glasses, I tend to assume that he's probably right. If an archaeologist tells me the same, no matter how famous or how much an authority on his own field, I go ask a doctor. Etc.
Lumping both extremes under "baffle them with bullshit" seems to me, sad to say, idiotic.
Actually, I think the key word is missing there. The real fallacy is "argument from false authority."
As a hypothetical example: If an recognized astrophysicist says that there's something fishy about the amount of existing dark matter, that's a real authority on the subject matter, and is certainly something to keep in mind. If Obama says it, he's just not qualified to make that kind of a judgment, and it's simply something to ignore. For all his authority in politics and law, he's as qualified to talk about astrophysics as the local barber.
In this case I don't think Linus is an authority on usability or anything even remotely relevant to KDE vs GNOME. It's his personal tastes vs yours, nothing more. Unless you happen to know that his tastes accidentally match yours to the letter, it's something to thoroughly ignore.
Of course, that won't stop people from being fashion victims and trying to imitate him anyway. That's why celebrity endorsements work. That's why you see video clips with Van Damme and whatnot saying that they play WoW, for example. Because a lot of John Does out there will try to be like monkeys imitating that celebrity. Or why you see Fatal1ty branded heatsinks, although I don't think he'd know enough physics to actually judge a design, nor the experience of having tested 100 heatsinks and picked the best. That's appeal to false authority.
I don't doubt that here too a lot of people switched to KDE just because Linus blasted GNOME, and will now hastily switch back to GNOME because Linus uses it now so it must be cool.
Sorry to break it to you, but very few people do that great in a prison. It's supposed to be a punishment and deterrent, ya know?
So, yes, I probably wouldn't do well at all in a prison. Guess what? So I don't break the law.
It's unfortunate if that's his problem, but I don't think any condition should be a blanket ticket to break any laws without punishment.
Or in other words, thank you very much, but I'll reserve my compassion for victims, not for some guy who figured out he can ignore the law.
1. Well, as an aspie myself, I seriously don't understand his defense. Asperger's Syndrome is basically like being colour blind, except in this case we aren't wired to even notice (much less decode) body language. I can tell if somebody screams or laughs, but everything else doesn't even exist for me. And far as I can tell mom can't even tell if you screamed at her or not, and is constantly taking wrong guesses there.
So offending people face to face or commiting social faux pas is a lot easier, because where someone else would take a hint, you don't even have a hint. E.g., I've had stuff like being told "dude, why didn't you stop it with that terminally bored face in the meeting? Didn't you see the way the boss was looking at you?" And I was thinking he looks at me because he likes me or something.
It's also very easy to conclude stuff like "everyone else is stupid" when you lack the hints that she's just making conversation and trying to sound interesting (or so I'm told,) or he's lying to you and hoping _you_ are stupid enough to believe him. (I find that if you dig deep enough in why someone insists on something illogical, you'll actually find a hidden motive rather than complete idiocy.)
On the other hand, being an aspie is all about logic, so anyone who blames it for not understanding "break law => get punished" is talking out the arse.
And you _can_ learn to function pretty normally in society by using logic, an I mean in a lot more detail than "break law => get punished". I've read a lot about psychology and anthropology, for example, just to know what I'm supposed to do or not to do, in the absence of ad-hoc hints to change the course.
2. _However_, Asperger's Syndrome has a very high probability of co-morbidity with something else, like OCD, OCPD, ADHD and going all the way to sociopathy.
And it seems to me like the _real_ problem of both this guy and Hans (the other with the aspie defense) is actually sociopathy. Both seem to be self-centered arseholes, and both seem to think that the law is some kind of game, among other symptoms.
I don't think we let people free just because they're sociopaths. In fact, most of the population in prisons scores over 20, a normal person scores 2-3, and 30 is the limit for outright psychopathy. That lack of empathy for their fellow man and society is usually what gets half of them into prison. (And the other half into upper management.)
But at any rate, that's a completely different mental disorder. And blaming it on Asperger's Syndrome does a disservice to everyone.
I don't know about every version of Windows, but I do happen to know that NT needed much more memory than Netware for the same performance as a file server, back when everyone ditched Novel in favour of MS. NT's saving grace was that NT _and_ the extra memory cost much less than a Netware upgrade. But the performance difference was there, and only excessive caching fixed it.
Actually, you illustrate exactly my problem. Better than I could. Thanks.
Let's put it like this:
1. _All_ a fragmented swap file (or any other file) adds is longer seeks back and forth to write the fragments.
2. Any competently written OS has _no_ excuse to block while waiting for a seek. It should just schedule another thread.
3. Windows does stall for seeks, as you've noticed.
Don't get me wrong, your point about the advantages of keeping the seeks short with a contiguous swap partition, is very valid too. The difference between 1-2 track seek and seek-across-the-drive is one order of magnitude.
But still, that's an optimization. That difference should just translate into maybe 1% difference in performance, as everything else (and especially the GUI threads) should be scheduled in that time. But, as you've noticed, the effect is a lot more dramatic.
It's not just the windows explorer copy that sucks. My point stands about just about any other way to do that. Total Commander, Install Shield, you name it. A lot of the IO wait translates into basically (a percentage of the) time where the whole machine does nothing else.
For the actual transfer, yes, you have to poll and suck in the bytes one by one. But AFAIK you don't have to wait in a loop for the seek to finish. Unless we're talking about old MFM drives on the XT. As I was saying, the PC architecture does have interrupts for just about everything. So you just start another process and wait for the interrupt.
Heck, even without interrupts, typical total HDD latency is between 6ms (Velociraptor) and 20+ms. You can just poll once every millisecond (or how often your scheduler interrupt comes) and do something else until it finishes.
Copying a file between two partitions is a scenario that's 90% dominated by waiting for the head to be in the right position, and for the right sector to come under it. Make that maybe 99% if the partitions are on the same physical HDD. Any competently written system, even in PIO mode has 90% of the CPU time there free for other processes.
At least NT obviously didn't have. As I was saying, with the drivers configured wrong in PIO mode, it just froze solid for a second when minimizing a window. I can't imagine any scenario where that's normal.
And the fact that Windows simply gets stuffed there, tells me that something's rotten in the kingdom of Redmond. Maybe it's not Windows itself but the IDE drivers or whatever. But something is thoroughly suboptimal in its disk IO IMHO.
The hard drive overheated and died, that's how :P
I mean, for example, the HDD overheats and doesn't do more than throw fits. Err, faults. It happened to me, circa 2001, after stuffing increasingly hot components in a box that was obviously designed for old cool stuff.
We're not talking something that happens regularly. And just an old-and-upgraded home box, not some enterprise system with redundant hard drives mirrored.
Actually, while I do somehow sped more time at home on my Windows gaming box than under Linux (so this isn't a blanket Windows bashing,) my superficial and uninformed impression was that, all else being equal, any Windows box I've seen seems harder hit by IO than any Linux/Unix box I've ever seen.
Yes, you can get a Linux box to crawl too, if the hard drive is stuffed and it can't swap for example. Or if the chipset isn't supported well by the drivers. (Rarer these days, but certainly possible.) Or whatever.
But Windows... seems a bit special. I mean try to copy a directory between two hard drives, or better yet from a DVD to HDD, and Windows seems to me basically stuffed. Even notepad can get about as responsive as a narcoleptic snail. And you can just about forget about, say, playing a game while that happens.
And that's before you even add such brakes as an anti-virus.
I've seen that behaviour in any Windows, from 3.0 to Vista, including a detour through NT 4.0. In fact in Vista let's just say there's a reason why so many people were pissed off at the indexer kicking in all the time.
My subjective impression is that I've yet to see Linux get anywhere near that unresponsive, in a similar scenario. Again, assuming that you don't have a nearly dead HDD and the chipset is supported in DMA mode.
But heck, even in PIO mode, I've used Linux in PIO mode and I've used, say, NT in PIO mode. (Thanks to a retarded IT department which installed the wrong IDE drivers.) Linux did obviously have poor file IO performance, but NT just freaking _froze_ for a second or two, for example, when minimizing or maximizing a window. (Presumably due to aggressive memory management which swapped more of a process out when minimized.)
Now admittedly I haven't actually programmed an OS at any point, so I'm probably talking out the arse, but I see no reason why that should happen at all. Any common source of IOWait has an interrupt. Even in PIO mode you don't have to poll until it's done. And DMA, now that was invented for the precise reason and purpose of transferring some data while the CPU services another process. It's why it's there. So there's no freaking reason for the whole OS to just twiddle its thumbs and wait. Even if one process is waiting for _paging_, you can still yield to another process while waiting for the HDD.
Or from the FDA itself: http://www.fda.gov/oc/history/elixir.html
1. Yes, it is entirely within the society's rights to decide that it requires certain standards of proof that your game doesn't cause people to run amok.
But even there it would first need to figure out such an impartial double-blind test, that's possible to pass. For medicine we have it. I don't think any of the "OMG, games turn people psycho" crowd actually made even a valid correlation, much less a scientific test there.
2. In medicine's case the FDA first had to make its own claim, and prove it, namely that the existing unchecked medicine can kill people. The idiot who sold a solution of sulpha in ethylene glycol, essentially made that case and proof for them. Somebody had sold a medicine which didn't just sometimes not work, and didn't just sometimes kill its patients. It killed _every_ single person who took it.
You're free to try to disprove their claim too, but they have a bullet-proof documentation of everyone who bought that medicine and died, and what are the biological effects of ethylene glycol.
Before the state could make that claim and proof, there was no reason to let them pass new legislation. As long as the "but someone could sell medicine which kills people" proof hadn't been made, we all just defaulted to the existing idea, namely, "nah, mate, the free market will solve everything, because nobody would risk the massive PR loss of face of selling poison as medicine."
So even there there was an implicit burden of proof on the FDA and generally those wanting more legislation.
It seems to me like it's not unreasonable to expect the same kind of burden of proof on those wanting to regulate games. As long as they don't make a reasonable proof that games do kill (e.g., by turning some people into amok psychos), we can and should tell them to take a fucking hike. Just like everyone did with the FDA before that incident.
I see no fundamental problem there. Essentially you're legally required to make a claim that your medicine (A) works better than placebo, and (B) you know and disclose the risks and potential side effects. And you're required to prove it.
The requirement part comes from having such bad experiences as someone selling sulpha dilluted in ethylene glycol... which is a very deadly poison, and actually killed everyone who took that medicine. In excruciating pain, over a couple of weeks. So now if you don't or can't make and prove those claims, you're not allowed to market that medicine. Or not as medicine.
But that's largely a legal construct, and has nothing to do with how logic works or how burden of proof works. We as a society decided that you _must_ make and prove that kind of claims.
But burden of proof even there works the old fashioned way. _You_ must prove it. It doesn't work like "it's good medicine unless someone else can prove otherwise."