It's nice to talk about what happens if you're on a laptop, but the fact of the matter is that there are still a lot of power-hungry applications out there where response-time is key. Mainly these are scientific simulations, genome processing, image processing, 3-d render farms and similar
Except this is explicitly *not* what TFA is talking about. TFA states the target application is "hard-to-parallelize applications, such as word processors and Web browsers". Laptop power usage is a killer for the very application they suggest they're targetting.
Yeah. This isn't about performance critical card, as TFA makes it abundantly clear. Such code is usually using a custom allocator that's *much* faster than malloc, and is very often easily parallelizable, so will be using all of the cores on your machine most of the time anyway.
The article makes it very clear that the target application is the kind of code that is not trivial, and specifically mentions "word processors and Web browsers". So laptop power use clearly *is* a concern.
When used for locking it is called spinning and not busy-looping, and stop your silly doomsday speak and grow a brain. The linux kernel itself more often use spinning than locking, because it is much faster and uses less cpu-cycles. You have busy-looping thousands of time each second when the kernel synchronizes threads and hardware, this is a no-go in application design, but a really common and efficient trick in low-level libraries and routines, and it will save you cpu-cycles and energy compared to semaphores, not use more.
Only if you restrict its use to occasions where you know the lock will become available quickly. The Linux kernel uses spinlocks for its internal structures where it knows that no other CPU is going to lock them for more than a few thousand cycles at most. I also believe it (usually) disables interrupts while the lock is held, so it knows that nothing will interrupt that operation prior to its completion. This is a very different situation from an environment where there may easily be multiple seconds between allocation requests.
But how much of your time is spent allocating memory? If you spend 5% of your time in malloc(), doubling its speed saves you 2.5% of your execution time.
Average is about 15%. Many programs spend nearly 50% of time in memory allocation.
One thing I noticed about the Linux version of flash is how CPU hungry it is, simple sites like Youtube, or any site with flash video playback would consume an entire core on my laptop. I spent some time researching why and everything I've seen so far indicates that Adobe has just implemented it poorly on Linux.
Have you tried comparing it against the Firefox Flash plugin in Windows, or are you comparing against IE? I've found even on Windows, Flash simply performs much better on IE than it does on Firefox. I have a 2.33GHz Celeron D machine, which slows to a crawl on any site with a moderately complex flash app under Firefox, but works fine on IE. YouTube video stutters like a bastard on Firefox, plays flawlessly on IE. I sometimes wonder if MS paid them off to make the IE version better than the others...
Re:where you at and one more correction.
on
iPad Review
·
· Score: 1
You don't need 3G to have GPS. Heck you don't have to have any cell phone connection to have a GPS. You need a GPS to have GPS.
OTOH, if you have 3G you don't need GPS for many applications: UMTS, the protocol suite used by 3G devices, is capable of providing your location to 50m accuracy guaranteed, usually better. This is good enough for most "location aware" applications, if not actually for providing directions.
I'd say any book people are still reading in significant numbers over 50 years after it was originally published is a classic, so, yes, Charlotte's Web is a classic.
You seem to have omitted the chemical prerequisites (i.e. you watched it while not off your head with some kind of mind-altering substance).
and Shakespeare
Shakespeare boring? Really? I'll admit I haven't seen _all_ of his plays, but those I have seen have generally been quite well paced and definitely worth watching. Just make sure you pick a genre you like (tragedies, particularly, aren't to everyone's taste, and can be quite depressing).
One question that the article does not pose (and can't answer due to its nature) is which is cause and which is effect. Is the reason that smokers have a lower IQ that the people that start smoking have a lower IQ, or does smoking damage your ability to reason logically?
Actually it can answer it. The study looked at two groups: fresh recruits and vets. We can assume an age difference of at least a few years between them, and the recruits are likely to be young enough that they've only been smoking for a couple of years on average. Therefore if the smoking were causing damage, we'd expect the recruits to show a less pronounced effect than the vets. As the article mentions no difference between the two groups, we can assume no significant such difference exists, and therefore (at least) no evidence for the latter proposition, and potentially evidence against it.
[Alan Kay's] hypothetical DynaBook bears striking similarity to what Apple finally came up with.
Except that the Dynabook would have been end-user programmable, with Kay intending to invest substantial research into designing a programming system simple enough for almost anyone to use.
Ok, so this is what I got from reading that short: well, this doesn't really address any of the concerns people have mentioned, but it's super duper powerful.
Except, you know, the average netbook has a processor that's 50% faster, 150% more storage capacity, a screen about 10% larger, plus the option of using a keyboard if you'd rather not play with handwriting recognition. Oh, and most have cameras, and quite a few have longer battery life.
There's actual volume. Who the hell is *buying* it?
Technical traders. Looking at the historic price chart, there's long term support between the 10c and 20c/share levels. It was overvalued at 50c, and should have fallen anyway even without this bad news. If you really and truly believe in technical analysis, you'd be buying right now.
Hopefully the motion for dismissal will be granted and the defendant will be awarded costs and fees
Seems likely. I suspect that even if the guy could prove he was being harmed, his case would fail. Here in the UK, there's a provision of the relevant law (tort of nuisance) that states that if an activity only causes damage because of an "unusual sensitivity" of the damaged party, then it is not the responsibility of the defendant. This seems like a textbook case for applying such a rule, and I'd guess (or at least hope) the same rule applies in the US.
Single-bit errors shouldn't send the car out of control... there should be some checksum that shouldn't add up. When a fault is detected, it should go to a backup program about safely shutting down the car.
The problem, reading between the lines of the article and trying to guess what the researcher was actually talking about, appears to be that major components are implemented with FPGAs. Flip a bit in an FPGA and you could easily end up switching off your fault detection circuit. You could easily change logic like A & B to A & 0, or similar, or change where in the circuit an input signal comes from.
The only way to deal with this, as far as I see, is to duplicate the functions to 2 different chips and have a hardwired circuit (*not* an FPGA, or indeed a microprocessor-based system) that compares the results and falls back to a safe mode in event of any difference between them.
I'm not sure how that is going to go over with the availability of guns in the UK, and how easy it is to drive in London. Lets face it, half the fun is running down everyone and shooting stuff.
Gun crime stats suggests criminals can get guns just as easily in the UK as they ever have been able to, despite a string of legislation over the last 13 years that's supposed to prevent it. And driving in London's no worse than in any other major city...
I see no details in the article. Looking at the developer's site, it seems their actions are:
- Shutting down the facebook profile associated with the script. This is poor behaviour, but entirely within their rights: it's their web site, if they don't want to support stuff like this it is their choice to do so. - Threaten to take legal action to seize control of a domain called "facebookplus.org", which the author claims is entirely unrelated to him.
So, what's the big fuss about? The former is annoying, but hardly "threatening to close him down"; the second appears to be a case of mistaken identity which will go away if he ignores it. Or is there some other threat I haven't seen?
Slashdot Mirror
given you could do most things a typical word processor can do on a 486, why in the world would you need to parallelize it?
I'd like to see an implementation of grammar-check-while-typing that works realistically quickly on a 486.
It's nice to talk about what happens if you're on a laptop, but the fact of the matter is that there are still a lot of power-hungry applications out there where response-time is key. Mainly these are scientific simulations, genome processing, image processing, 3-d render farms and similar
Except this is explicitly *not* what TFA is talking about. TFA states the target application is "hard-to-parallelize applications, such as word processors and Web browsers". Laptop power usage is a killer for the very application they suggest they're targetting.
Yeah. This isn't about performance critical card, as TFA makes it abundantly clear. Such code is usually using a custom allocator that's *much* faster than malloc, and is very often easily parallelizable, so will be using all of the cores on your machine most of the time anyway.
The article makes it very clear that the target application is the kind of code that is not trivial, and specifically mentions "word processors and Web browsers". So laptop power use clearly *is* a concern.
When used for locking it is called spinning and not busy-looping, and stop your silly doomsday speak and grow a brain. The linux kernel itself more often use spinning than locking, because it is much faster and uses less cpu-cycles. You have busy-looping thousands of time each second when the kernel synchronizes threads and hardware, this is a no-go in application design, but a really common and efficient trick in low-level libraries and routines, and it will save you cpu-cycles and energy compared to semaphores, not use more.
Only if you restrict its use to occasions where you know the lock will become available quickly. The Linux kernel uses spinlocks for its internal structures where it knows that no other CPU is going to lock them for more than a few thousand cycles at most. I also believe it (usually) disables interrupts while the lock is held, so it knows that nothing will interrupt that operation prior to its completion. This is a very different situation from an environment where there may easily be multiple seconds between allocation requests.
But how much of your time is spent allocating memory? If you spend 5% of your time in malloc(), doubling its speed saves you 2.5% of your execution time.
Average is about 15%. Many programs spend nearly 50% of time in memory allocation.
One thing I noticed about the Linux version of flash is how CPU hungry it is, simple sites like Youtube, or any site with flash video playback would consume an entire core on my laptop. I spent some time researching why and everything I've seen so far indicates that Adobe has just implemented it poorly on Linux.
Have you tried comparing it against the Firefox Flash plugin in Windows, or are you comparing against IE? I've found even on Windows, Flash simply performs much better on IE than it does on Firefox. I have a 2.33GHz Celeron D machine, which slows to a crawl on any site with a moderately complex flash app under Firefox, but works fine on IE. YouTube video stutters like a bastard on Firefox, plays flawlessly on IE. I sometimes wonder if MS paid them off to make the IE version better than the others...
You don't need 3G to have GPS. Heck you don't have to have any cell phone connection to have a GPS. You need a GPS to have GPS.
OTOH, if you have 3G you don't need GPS for many applications: UMTS, the protocol suite used by 3G devices, is capable of providing your location to 50m accuracy guaranteed, usually better. This is good enough for most "location aware" applications, if not actually for providing directions.
I'd say any book people are still reading in significant numbers over 50 years after it was originally published is a classic, so, yes, Charlotte's Web is a classic.
the horribly boring 2001: Space Odessey
You seem to have omitted the chemical prerequisites (i.e. you watched it while not off your head with some kind of mind-altering substance).
and Shakespeare
Shakespeare boring? Really? I'll admit I haven't seen _all_ of his plays, but those I have seen have generally been quite well paced and definitely worth watching. Just make sure you pick a genre you like (tragedies, particularly, aren't to everyone's taste, and can be quite depressing).
So do I, but we should be cautious that this correlation might not actually be caused by anything in the statement itself.
One question that the article does not pose (and can't answer due to its nature) is which is cause and which is effect. Is the reason that smokers have a lower IQ that the people that start smoking have a lower IQ, or does smoking damage your ability to reason logically?
Actually it can answer it. The study looked at two groups: fresh recruits and vets. We can assume an age difference of at least a few years between them, and the recruits are likely to be young enough that they've only been smoking for a couple of years on average. Therefore if the smoking were causing damage, we'd expect the recruits to show a less pronounced effect than the vets. As the article mentions no difference between the two groups, we can assume no significant such difference exists, and therefore (at least) no evidence for the latter proposition, and potentially evidence against it.
[Alan Kay's] hypothetical DynaBook bears striking similarity to what Apple finally came up with.
Except that the Dynabook would have been end-user programmable, with Kay intending to invest substantial research into designing a programming system simple enough for almost anyone to use.
Thus passes the glory of the screwdriver.
I just had the strangest urge to do this:
Sic transit gloria impellator-cochleae
Ok, so this is what I got from reading that short: well, this doesn't really address any of the concerns people have mentioned, but it's super duper powerful.
Except, you know, the average netbook has a processor that's 50% faster, 150% more storage capacity, a screen about 10% larger, plus the option of using a keyboard if you'd rather not play with handwriting recognition. Oh, and most have cameras, and quite a few have longer battery life.
You're arguing over nothing. GP asserts Darl works for the antichrist. You assert he was working for Microsoft. You say tomato, I say tomato.
There's actual volume. Who the hell is *buying* it?
Technical traders. Looking at the historic price chart, there's long term support between the 10c and 20c/share levels. It was overvalued at 50c, and should have fallen anyway even without this bad news. If you really and truly believe in technical analysis, you'd be buying right now.
To avoid all EM, you'd have to find a deep cave with heavy shielding rock miles into the earth's crust.
Yeah, but you'd better watch out for the neutron decay from all that rock. Seriously, that stuff's dangerous. Gives off radiation.
Hopefully the motion for dismissal will be granted and the defendant will be awarded costs and fees
Seems likely. I suspect that even if the guy could prove he was being harmed, his case would fail. Here in the UK, there's a provision of the relevant law (tort of nuisance) that states that if an activity only causes damage because of an "unusual sensitivity" of the damaged party, then it is not the responsibility of the defendant. This seems like a textbook case for applying such a rule, and I'd guess (or at least hope) the same rule applies in the US.
Do you have a source for that?
Not quite identical to the parent's story, but this is a reasonably close match: http://mybroadband.co.za/news/Wireless/11099.html
Single-bit errors shouldn't send the car out of control... there should be some checksum that shouldn't add up. When a fault is detected, it should go to a backup program about safely shutting down the car.
The problem, reading between the lines of the article and trying to guess what the researcher was actually talking about, appears to be that major components are implemented with FPGAs. Flip a bit in an FPGA and you could easily end up switching off your fault detection circuit. You could easily change logic like A & B to A & 0, or similar, or change where in the circuit an input signal comes from.
The only way to deal with this, as far as I see, is to duplicate the functions to 2 different chips and have a hardwired circuit (*not* an FPGA, or indeed a microprocessor-based system) that compares the results and falls back to a safe mode in event of any difference between them.
Where have I seen their logo before? It looks _very_ familiar...
Nope, I think yours is - note how both 0.0.0.0 and 127.0.0.1 act the same.
No, they don't. Attempting to connect to 0.0.0.0 doesn't work because it's *defined* not to work in the socket API:
Jules@minerva ~
$ ssh 0.0.0.0
ssh: connect to host 0.0.0.0 port 22: Cannot assign requested address
Jules@minerva ~
$ ssh localhost
The authenticity of host 'localhost (127.0.0.1)' can't be established.
I think your network is misconfigured.
I'm not sure how that is going to go over with the availability of guns in the UK, and how easy it is to drive in London. Lets face it, half the fun is running down everyone and shooting stuff.
Gun crime stats suggests criminals can get guns just as easily in the UK as they ever have been able to, despite a string of legislation over the last 13 years that's supposed to prevent it. And driving in London's no worse than in any other major city...
I see no details in the article. Looking at the developer's site, it seems their actions are:
- Shutting down the facebook profile associated with the script. This is poor behaviour, but entirely within their rights: it's their web site, if they don't want to support stuff like this it is their choice to do so.
- Threaten to take legal action to seize control of a domain called "facebookplus.org", which the author claims is entirely unrelated to him.
So, what's the big fuss about? The former is annoying, but hardly "threatening to close him down"; the second appears to be a case of mistaken identity which will go away if he ignores it. Or is there some other threat I haven't seen?