So far I've never had them fail on the machines I've built. Heck I don't recall having an Intel CPU fail on me (or people I know) in the past decade or so.
At one place they had a crappy OEM 1-U server where the Intel CPUs were thermal throttling whenever they had stuff to do. But the server still kept running. I certainly don't blame Intel for the insufficient cooling in that case - it's most certainly the OEM's fault.
The fans might not be that quiet. But they're not that noisy either - in an "average" PC, they're certainly not the noisiest items (HDDs, case fans, GPU fans).
Uh, if AMD CPUs were really that great, Intel wouldn't be able to sell crippled CPUs like this. This is like Intel slowing down on the race track just to keep AMD in the rear view mirror. Maybe even doing some "donuts";).
AMD's performance price curve is the way it is because AMD has to cut their prices so much to sell their inferior chips.
My guess is Intel's CPUs have got to the stage where their crappiest batch of desktop chips is inherently as fast or even faster than AMDs fastest desktop stuff, so at the low end Intel have to cripple lots of CPUs. And some bright spark says hey we've got lots of spare resources (AMD isn't making them break a sweat) why don't we do _this_ instead.
If Intel thought AMD had a chance of pulling something much faster out of the bag, they wouldn't bother with this. But it sure looks like AMD has nothing - I don't see anything AMD has in the horizon that would overtake Intel's stuff.
So when AMD releases their next generation, even if AMD charges lower, Intel can cut the upgrade fee (if necessary). Then the customer's old slow Intel CPU, suddenly runs as fast as some of AMD's new desktop CPUs (or maybe faster than all of them). No need to ship anything except a license key/code.
They (which includes AMD) already do it. When they get many good batches but more people are only willing to pay for a cheaper and slower CPUs, they just cripple some of them.
They used to do this by zapping stuff (fuses etc). But this sort of crippling is usually permanent ( there are some exceptions I guess).
Now they are just making it nonpermanent, so you don't actually have to buy a new CPU (or a new laptop -it's not usually easy to physically upgrade the CPU on a laptop). IBM has been doing this stuff for years too - ship a server capable of doing a lot more, when the customer needs more performance/capacity, they call IBM, and IBM unlocks it. Sometimes even temporarily- the customer might only need the extra capacity temporarily and so only pay for it for that period.
What this shows is AMD isn't competitive enough. If AMD's CPUs were much faster, Intel wouldn't be able to do this - since they wouldn't be able to make enough money from selling slower CPUs.
And what galls me even more is when the "inventors" think they're so innovative and so they DESERVE to be rewarded with a monopoly.
The real innovative ones typically don't even realize their "little steps" are huge leaps for other people. Or they're just too busy trying to turn dreams into reality for any of that bullshit (having them think daily whether what they've done each day is patentable would just slow them down).
Yes doing away with the patent system would cause the "small time" inventors to lose out. But given the state of things now, they're already screwed right? They are unlikely to be able to produce their own inventions without infringing on dozens or more other patents.
Some plugins and addons use/leak lots of memory. I notice Firebug makes firefox bloat up a lot, so I generally don't have firebug or disable it.
Anyway I run chrome, firefox (and IE sometimes), and chrome uses a lot of memory too - the difference is chrome tends to have multiple processes, so you don't see one big fat process, but you get multiple processes which add up to about the same thing. But it's easier to free up memory with chrome - since if you close one chrome process, the memory is freed up by the OS, none of that "I hope Firefox starts freeing up some memory before I have to kill the entire browser".
Cool, the future is already here... Oh wait... is embedded perl6 out yet?;)
I hope google or some other geniuses find a way to make python/perl faster. I just find coding in these less painful (currently work involves VB.Net, vbscript, batch files, with some perl thrown in where I can).
Getting the economics to work out is largely a matter of tweaking the sensitivity of the trigger and designing the consumable properly.
That's the problem with patents nowadays. even if you don't make the stuff (or produce any stuff except bullshit patents), you can collect a tax on people who work out the actual tricky practical details.
The innovative part is actually creating something that fits in/on/around a phone that can tell the difference between it falling to the floor and it falling in your pocket towards the floor or you running with it in your hand/handbag and swinging it around, or you dropping the phone into a bag or onto a bed. Or does it such a way where false positives don't cause big/expensive problems.
The patent is probably so wide and generic that even if some genius figures out the details later (maybe a redeployable airbag that inflates not too quickly and can be re-deflated manually so as to not be an expensive "per pop" device that destroys stuff), they'd have to pay Bezos money even though Bezos might have zero idea of how to actually make something that solves all the problems I mentioned (not saying he has zero idea, just saying he doesn't have to have any idea about the details).
And that's why the patent system is broken and doesn't really encourage innovation.
I can write all the sci-fi bullshit I want, it doesn't mean that it will make the stuff appear, and if I can actually collect a tax on people who actually build stuff, does it really encourage innovation?
A good combination would be someone with experience and some bright 25 year-olds with the energy and willingness to learn from the experienced but slower person.
The experienced one can say better do it this way and not that way because of XYZ. and the bright 25 year olds can go, "Oh, good idea, didn't realize that" and do things the better way much faster than the experienced slower person could.
Thing is you'd need a project large enough to afford them all:).
The other problem is often you just need stuff out the door as soon as possible, even if it's crap. The "problem" is the experienced person might have forgotten how (or refuse) to write crap even though it's faster. Doing things right often takes longer. The experienced person when switching from one language/framework to another might go: how do we do logging and tracing _correctly_ in this environment, how about exception handling and "signals", how about data escaping for all the possible outputs, how about multilingual support, how do we do upgrades and patches? It takes time to learn how to do all that given a new framework and platform.
Time that the bosses often don't have. So they'll just go for "quick and cheap" and deal with the bugs/problems later after they get the $$$$$$ from the customers:).
Hey it worked for Facebook didn't it? Start quick and dirty with PHP and MySQL, once you get money, hire lots of very smart people to fix/workaround PHP and MySQL;). OK to be fair apparently Zuckerberg did do some work on memcached.
Saying they're not significantly different because they're all "turing complete" is silly when the speed of light is finite. Most of us live in the real world where it takes time to do stuff.
You can build a house in many different ways, some ways are much faster, easier and involve a lot less pain, while some ways are just pure theory only and not practical.
And some of us realize we're getting old and running out of time to do stuff.
You may be able to build a house brick by brick, log by log, but it's usually going to take a lot longer than if you used some prefab stuff. Learning enough about the details, problems and "gotchas" of the different prefabs (frameworks) takes a fair bit of time.
I just started on.Net stuff +SQL Server etc this year and I don't really like it. Worse is when you search on google, bing etc and the "answers" are from clueless idiots who don't know what they're talking about, and don't have the sense to shut up (that includes some of those MVPs on the "social.technet" forums).
Some people say it's all just syntactic sugar, but I find it frustrating when things that should take one or two lines of code have to take 10 or more lines (assuming you want to do it correctly without breaking just because someone sneezes on it). Why? Because I still have to frigging type those extra lines. Yes I can put them in a library, maybe even make my own framework but then someone else later would have to figure those out and documentation might have to be written for it. That means more extra lines to type, debug and support.
Which brings us back to the topic- we're all getting old and running out of time. Many people just don't realize it yet;).
That's why I think PCs are here to stay and it's the mobile/portable stuff that will cannibalize themselves, some to extinction.
Imagine in the future if brain-computer-interfaces worked well and were safe. You wouldn't need a touch screen or that many buttons. Lower power consumption if you don't need to light up a screen.
Why carry a phone with a touch screen when you could have seamless virtual telepathy and telekinesis? Carrying a phone would be as anachronistic as carrying a pocket watch.
In contrast a very powerful nonportable home computer/sever is likely to still be useful to many people. It might even be part of what makes the "virtual telekinesis" work in your home.
Well, in terms of dollars, the console definitely reigns supreme. The PC game software market is about $700 million a year. Console game sales are over $5 billion a year.
but you have to drive them in reverse for safety (unless you drive the small types)
All that talk about forklifts and safety reminds me of the "Staplerfahrer Klaus" video (WARNING: contains gore, violence and potentially disturbing scenes): http://www.youtube.com/watch?v=pVssz2VMJVM
Apparently it's not a real forklift safety video but people might be more careful around/with a forklift after watching it:).
Unnamed option 3 is that they would get sued to the stone age if a bomb really went off and people found out that they knew about it and did nothing. It's a "backside save manoeuvre" on their part.
At the check-in desk, your luggage is scanned immediately in a purpose-built area. Sela plays devil's advocate - what if you have escaped the attention of the first four layers of security, and now try to pass a bag with a bomb in it?
"I once put this question to Jacques Duchesneau (the former head of the Canadian Air Transport Security Authority): say there is a bag with play-doh in it and two pens stuck in the play-doh. That is 'Bombs 101' to a screener. I asked Ducheneau, 'What would you do?' And he said, 'Evacuate the terminal.' And I said, 'Oh. My. God.'
A screener at Ben-Gurion has a pair of better options.
First, the screening area is surrounded by contoured, blast-proof glass that can contain the detonation of up to 100 kilos of plastic explosive. Only the few dozen people within the screening area need be removed, and only to a point a few metres away.
Second, all the screening areas contain 'bomb boxes'. If a screener spots a suspect bag, he/she is trained to pick it up and place it in the box, which is blast proof. A bomb squad arrives shortly and wheels the box away for further investigation.
"This is a very small simple example of how we can simply stop a problem that would cripple one of your airports," Sela said.
Slashdot Mirror
Intel makes really crappy heatsinks,
So far I've never had them fail on the machines I've built. Heck I don't recall having an Intel CPU fail on me (or people I know) in the past decade or so.
At one place they had a crappy OEM 1-U server where the Intel CPUs were thermal throttling whenever they had stuff to do. But the server still kept running. I certainly don't blame Intel for the insufficient cooling in that case - it's most certainly the OEM's fault.
The fans might not be that quiet. But they're not that noisy either - in an "average" PC, they're certainly not the noisiest items (HDDs, case fans, GPU fans).
Uh, if AMD CPUs were really that great, Intel wouldn't be able to sell crippled CPUs like this. This is like Intel slowing down on the race track just to keep AMD in the rear view mirror. Maybe even doing some "donuts" ;).
AMD's performance price curve is the way it is because AMD has to cut their prices so much to sell their inferior chips.
My guess is Intel's CPUs have got to the stage where their crappiest batch of desktop chips is inherently as fast or even faster than AMDs fastest desktop stuff, so at the low end Intel have to cripple lots of CPUs. And some bright spark says hey we've got lots of spare resources (AMD isn't making them break a sweat) why don't we do _this_ instead.
If Intel thought AMD had a chance of pulling something much faster out of the bag, they wouldn't bother with this. But it sure looks like AMD has nothing - I don't see anything AMD has in the horizon that would overtake Intel's stuff.
So when AMD releases their next generation, even if AMD charges lower, Intel can cut the upgrade fee (if necessary). Then the customer's old slow Intel CPU, suddenly runs as fast as some of AMD's new desktop CPUs (or maybe faster than all of them). No need to ship anything except a license key/code.
They (which includes AMD) already do it. When they get many good batches but more people are only willing to pay for a cheaper and slower CPUs, they just cripple some of them.
They used to do this by zapping stuff (fuses etc). But this sort of crippling is usually permanent ( there are some exceptions I guess).
Now they are just making it nonpermanent, so you don't actually have to buy a new CPU (or a new laptop -it's not usually easy to physically upgrade the CPU on a laptop). IBM has been doing this stuff for years too - ship a server capable of doing a lot more, when the customer needs more performance/capacity, they call IBM, and IBM unlocks it. Sometimes even temporarily- the customer might only need the extra capacity temporarily and so only pay for it for that period.
What this shows is AMD isn't competitive enough. If AMD's CPUs were much faster, Intel wouldn't be able to do this - since they wouldn't be able to make enough money from selling slower CPUs.
Yeah that's the problem with this sort of shit.
And what galls me even more is when the "inventors" think they're so innovative and so they DESERVE to be rewarded with a monopoly.
The real innovative ones typically don't even realize their "little steps" are huge leaps for other people. Or they're just too busy trying to turn dreams into reality for any of that bullshit (having them think daily whether what they've done each day is patentable would just slow them down).
Yes doing away with the patent system would cause the "small time" inventors to lose out. But given the state of things now, they're already screwed right? They are unlikely to be able to produce their own inventions without infringing on dozens or more other patents.
For the benefit of the small-time inventors I say replace the patent system with "Prizes for Innovation" instead: http://yro.slashdot.org/comments.pl?sid=1156061&cid=27145551
http://tech.slashdot.org/comments.pl?sid=1817120&cid=33865688
Reminds me of this guy and his water+naptha stuff
http://inventors.about.com/od/wstartinventions/a/water_fuel.htm
http://www.epa.gov/etv/pubs/600r980035.pdf
And add carousel time too... ;)
Who is Larry and why would you blame him for javascript memory problems?
Some plugins and addons use/leak lots of memory. I notice Firebug makes firefox bloat up a lot, so I generally don't have firebug or disable it.
Anyway I run chrome, firefox (and IE sometimes), and chrome uses a lot of memory too - the difference is chrome tends to have multiple processes, so you don't see one big fat process, but you get multiple processes which add up to about the same thing. But it's easier to free up memory with chrome - since if you close one chrome process, the memory is freed up by the OS, none of that "I hope Firefox starts freeing up some memory before I have to kill the entire browser".
Sea Turtles and Monitor Lizards can even get hundreds of pounds in weight yet none are warm blooded.
Apparently leatherback turtles are warm blooded:
http://www.nature.com/nature/journal/v344/n6269/abs/344858a0.html
http://www.bbc.co.uk/springwatch/meettheanimals/leatherback.shtml
There are warmblooded fish too, e.g. bluefin tuna and some sharks:
http://en.wikipedia.org/wiki/Pacific_bluefin_tuna
http://www.nsf.gov/news/news_summ.jsp?cntn_id=104543
Cool, the future is already here... Oh wait... is embedded perl6 out yet? ;)
I hope google or some other geniuses find a way to make python/perl faster. I just find coding in these less painful (currently work involves VB.Net, vbscript, batch files, with some perl thrown in where I can).
I'll try powershell one of these days.
What's a pain processor?
Getting the economics to work out is largely a matter of tweaking the sensitivity of the trigger and designing the consumable properly.
That's the problem with patents nowadays. even if you don't make the stuff (or produce any stuff except bullshit patents), you can collect a tax on people who work out the actual tricky practical details.
The innovative part is actually creating something that fits in/on/around a phone that can tell the difference between it falling to the floor and it falling in your pocket towards the floor or you running with it in your hand/handbag and swinging it around, or you dropping the phone into a bag or onto a bed. Or does it such a way where false positives don't cause big/expensive problems.
The patent is probably so wide and generic that even if some genius figures out the details later (maybe a redeployable airbag that inflates not too quickly and can be re-deflated manually so as to not be an expensive "per pop" device that destroys stuff), they'd have to pay Bezos money even though Bezos might have zero idea of how to actually make something that solves all the problems I mentioned (not saying he has zero idea, just saying he doesn't have to have any idea about the details).
And that's why the patent system is broken and doesn't really encourage innovation.
I can write all the sci-fi bullshit I want, it doesn't mean that it will make the stuff appear, and if I can actually collect a tax on people who actually build stuff, does it really encourage innovation?
Give it another 5 years and you might have embedded Python/Javascript ;).
A good combination would be someone with experience and some bright 25 year-olds with the energy and willingness to learn from the experienced but slower person.
:).
:).
;). OK to be fair apparently Zuckerberg did do some work on memcached.
The experienced one can say better do it this way and not that way because of XYZ. and the bright 25 year olds can go, "Oh, good idea, didn't realize that" and do things the better way much faster than the experienced slower person could.
Thing is you'd need a project large enough to afford them all
The other problem is often you just need stuff out the door as soon as possible, even if it's crap. The "problem" is the experienced person might have forgotten how (or refuse) to write crap even though it's faster. Doing things right often takes longer. The experienced person when switching from one language/framework to another might go: how do we do logging and tracing _correctly_ in this environment, how about exception handling and "signals", how about data escaping for all the possible outputs, how about multilingual support, how do we do upgrades and patches? It takes time to learn how to do all that given a new framework and platform.
Time that the bosses often don't have. So they'll just go for "quick and cheap" and deal with the bugs/problems later after they get the $$$$$$ from the customers
Hey it worked for Facebook didn't it? Start quick and dirty with PHP and MySQL, once you get money, hire lots of very smart people to fix/workaround PHP and MySQL
Saying they're not significantly different because they're all "turing complete" is silly when the speed of light is finite. Most of us live in the real world where it takes time to do stuff.
You can build a house in many different ways, some ways are much faster, easier and involve a lot less pain, while some ways are just pure theory only and not practical.
And some of us realize we're getting old and running out of time to do stuff.
You miss the point.
.Net stuff +SQL Server etc this year and I don't really like it. Worse is when you search on google, bing etc and the "answers" are from clueless idiots who don't know what they're talking about, and don't have the sense to shut up (that includes some of those MVPs on the "social.technet" forums).
;).
You may be able to build a house brick by brick, log by log, but it's usually going to take a lot longer than if you used some prefab stuff. Learning enough about the details, problems and "gotchas" of the different prefabs (frameworks) takes a fair bit of time.
I just started on
Some people say it's all just syntactic sugar, but I find it frustrating when things that should take one or two lines of code have to take 10 or more lines (assuming you want to do it correctly without breaking just because someone sneezes on it). Why? Because I still have to frigging type those extra lines. Yes I can put them in a library, maybe even make my own framework but then someone else later would have to figure those out and documentation might have to be written for it. That means more extra lines to type, debug and support.
Which brings us back to the topic- we're all getting old and running out of time. Many people just don't realize it yet
That's why I think PCs are here to stay and it's the mobile/portable stuff that will cannibalize themselves, some to extinction.
Imagine in the future if brain-computer-interfaces worked well and were safe. You wouldn't need a touch screen or that many buttons. Lower power consumption if you don't need to light up a screen.
Why carry a phone with a touch screen when you could have seamless virtual telepathy and telekinesis? Carrying a phone would be as anachronistic as carrying a pocket watch.
In contrast a very powerful nonportable home computer/sever is likely to still be useful to many people. It might even be part of what makes the "virtual telekinesis" work in your home.
Even if it didn't you are more likely to die of embarassment after being pwn'ed with a salmon than with a knife. ;)
But yeah being hit by a king/chinook salmon could kill you, especially if it's frozen.
Well, in terms of dollars, the console definitely reigns supreme. The PC game software market is about $700 million a year. Console game sales are over $5 billion a year.
Where'd you get that 700 million figure from? WoW's revenue alone was 1.2 billion for 2010. See page 55: http://investor.activision.com/secfiling.cfm?filingID=1047469-11-1413
11 million WoW subscribers * 12 months * monthly subscription fee adds up to a fair bit.
You could say it's rent and not sales, but whatever it is the PC games market is clearly more than $700 million a year.
There were 4.5 million units of Starcraft 2 sold in 2010:
http://investor.activision.com/common/download/download.cfm?companyid=ACTI&fileid=440263&filekey=2a37de98-400f-4916-9bb3-ae5ddf1b86b8&filename=ATVI%20C4Q10%20slides%20FINAL.pdf
Add Valve's Steam and non-Steam sales, then add everyone else.
but you have to drive them in reverse for safety (unless you drive the small types)
All that talk about forklifts and safety reminds me of the "Staplerfahrer Klaus" video (WARNING: contains gore, violence and potentially disturbing scenes):
http://www.youtube.com/watch?v=pVssz2VMJVM
Apparently it's not a real forklift safety video but people might be more careful around/with a forklift after watching it :).
At higher speeds the OS stuff starts going back to polling to be less CPU intensive :).
http://en.wikipedia.org/wiki/New_API
How much does it cost to actually get the $631,135.18?
Unnamed option 3 is that they would get sued to the stone age if a bomb really went off and people found out that they knew about it and did nothing. It's a "backside save manoeuvre" on their part.
They should actually get sued for shutting down the airport. Because if they do things right they shouldn't have to:
http://slashdot.org/comments.pl?sid=2370330&cid=37031702
http://slashdot.org/comments.pl?sid=2370330&cid=37031702
The problem is the airport got shutdown. It shouldn't have to be shutdown if things were done properly: http://www.thestar.com/news/world/article/744199---israelification-high-security-little-bother
At the check-in desk, your luggage is scanned immediately in a purpose-built area. Sela plays devil's advocate - what if you have escaped the attention of the first four layers of security, and now try to pass a bag with a bomb in it?
"I once put this question to Jacques Duchesneau (the former head of the Canadian Air Transport Security Authority): say there is a bag with play-doh in it and two pens stuck in the play-doh. That is 'Bombs 101' to a screener. I asked Ducheneau, 'What would you do?' And he said, 'Evacuate the terminal.' And I said, 'Oh. My. God.'
A screener at Ben-Gurion has a pair of better options.
First, the screening area is surrounded by contoured, blast-proof glass that can contain the detonation of up to 100 kilos of plastic explosive. Only the few dozen people within the screening area need be removed, and only to a point a few metres away.
Second, all the screening areas contain 'bomb boxes'. If a screener spots a suspect bag, he/she is trained to pick it up and place it in the box, which is blast proof. A bomb squad arrives shortly and wheels the box away for further investigation.
"This is a very small simple example of how we can simply stop a problem that would cripple one of your airports," Sela said.
Note date on article: 2009-12-30