Vanilla Sky did not have a complex plot. It doesn't take much to figure out that after he passed out on in the street he was either a) dreaming b) hallucinating or c) living in some virtual reality system. Without all the hints about Benny the dog it would have been lots harder to figure out which of those is what's really happening. But there was nothing to remember, nothing to get confused about. Nothing. It was as straight forward as you can get. There was a very clear (and obvious) line when things started becoming 'ideal' and his life was 'perfect'. And *then* at the end they don't even give the audience the benefit of having a memory longer than an hour and they show you all the things you were supposed to have missed. I mean, come on, it's a good movie... but they didn't make it so complicated that joe average can't sit down and watch it.
In Vanilla Sky there are 5 major characters --- Black Hawk Down has a good 20-30 that are meaningful and they are broken up into many separate groups for much of the film. Keeping the characters straight can be difficult since they're all wearing the same clothes, all have the same haircut, all are dirty and bloody. IMHO, Black Hawk Down has a much more complicated plot than Vanilla Sky.
On the contrary, I am seeing significant and real benefits on my 1 and 2 CPU machines, especially under high load, everything just runs smoother, every process gets the % of CPU that it deserves. It's not that scheduling takes up a high % of the cpu time, it's that processes aren't scheduled "perfectly" under the current model.
But you are absolutely correct in that the scheduler improvements will be more apparent and dramatic on 4 and 8-way machines because of the elimination of the global run queue. Each CPU gets its own run queue and processes will only bounce around when other cpu's are idle. We finally have a scheduler that will work on enterprise class machines.
I must say that after using it for a few days, I'm impressed. It totally changes the characteristics of multiprocess servers like Apache and PostgreSQL under high load. For example, I've run ApacheBench against a mod_perl script that queries a pgsql database, in the new scheduler I get a mean response time that is N*1.05*concurrency with a standard deviation of less than 1% of the mean. In the old scheduler I'd get a mean that is N*1.07*concurrency with a sd of up to 75% of the mean. So in other words you get essentially the same throughput with both schedulers (O(1) appears slightly faster in my limited testing). But what's more important is that in the O(1) scheduler everyone is treated equally - they all get served in 1.05*N*concurrency, no more, no less -- while with the old scheduler some requests get a response that's 1*N and others get a response all the way up to 4*N*concurrency.
IMHO, it's better to give everyone an equal level of service than to randomly favor one group of users over another.
Don't you just hate that? The company I worked for last did the same thing, very paranoid. Showed complete lack of faith in its employees. It's not as though Google can relay pr0n to you. But I suppose you could use it to translate all that subversive foreign language literature.
No, it's not. Well, maybe it is, but that has nothing to do with the scale. It's based on the freezing and boiling of water just like Celsius. There's 180 degrees between boiling and freezing, go figure, just like a 1/2 circle. And it's unfortunate that Fahrenheit didn't want to use negative numbers on really cold days or he probably would have started the scale at zero. Instead he used the coldest temperature he could measure by making a slurry of salt and snow/ice. He adjusted it later when the boiling point was discovered to be 212 (and not 210) at sea level, by moving water's freezing point up 2 degrees -- though I'm still unclear on how he justified doing that, maybe changed the composition of the liquid in the thermometer at the same time. But anyway... there's no voodoo involved:-)
Re:has the targeted demographic really changed?
on
Attack of the Clones
·
· Score: 1
Doh, what the hell was I thinking/saying. I was thinking of Leigh Brackett who co-wrote Empire.
It's not a paradox that we die because we're fending off cancer. Essentially we die because our bodies are fending off mutation for our entire lives and we eventually just can't take it any more. A creator has nothing to do with it, and the finish line is not imaginary --- there is no reason not to believe that we will eventually be able to cure and prevent all forms of cancer. It's a slow process, but science is making real progress. But, trying to live forever is futile, we're all going to die either by untreatable disease, accident or violence.
Re:has the targeted demographic really changed?
on
Attack of the Clones
·
· Score: 1
Why was Jedi aimed at kids? Because of the Ewoks? There is lots of sex and violence in Jedi. In some ways more than the other movies. I don't see how it didn't deserve its PG rating. Sure, it wasn't ominous and dark like Empire was, but that was only partly because Empire's director died, mostly it was because it'd the end of a trilogy and it had a happy ending. One thing you didn't really get a feeling for was that the Ewoks were being oppressed by the Empire, that would have made the story more interesting at least.
So how much gain in performance (or apparent performance) should one expect after applying this combined patch? Are the performance gains only applicable under special circumstances? Are they focused more on desktop apps than server?
High Fidelity and Chasing Amy are not good enough films to be in anyone's top 5. I loved both of them, and I wouldn't even put them in my top 50 even though I probably gave them a 10 at IMDB. Try Memento, American Beauty, Usual Suspects, Fargo, and Taxi Driver. Those are movies that are (or will be) 'classics' in the true sense of the word and would have a much better shot at sneaking into most people's top 5. Yes, they do show a strong bias to what I like to see in movies. Age *does* have something to do with it, I'm a couple years older than you and I know why you chose those movies, they are about young people trying to find their way in life. They are not movies that would be selected by the vast majority of people or film critics to be in the top 5. I gave Fellowship of the Ring a 10 at IMDB, does that mean that it's one of my top 5? No, not yet anyway, maybe in a few years when the series is complete. But what it means to me is that it's better than 90% of the crap that the studios are feeding us. IMHO, you can only grade films on a curve to grade them fairly, they must be compared to other films - it's just the nature of the medium, nothing really stands on its own.
I don't buy it, you must be thinking of Pocket PC and not CE as a whole.
NT on an Alpha was just a waste of a good Alpha.
Kinda like NT on the PowerPC. Boxes that could have been perfectly good Macs, but running Windows instead. A friend of mine bought one a while back, it came with NT preinstalled on the harddrive. We wiped it almost immediately and installed LinuxPPC.
I know, I think it ran on MIPS as well?... Anyhow, I don't remember it ever running on x86...
Actually it runs on 12 processor architectures, including x86 and MIPS.
Pocket PC is kind of a separate beast from Windows CE. It's basically CE plus a bunch of extensions that make it fit the needs of PDA users better. It may very well only be available for StrongARM.
Actually CE was designed for a range of processors that meet a certain spec. StrongARM is just one, there's also the Hitachi SH3.
Re:bios flashing, not just from DOS anymore
on
Kernel 2.4.17 Out
·
· Score: 2, Informative
You should try out/dev/bios. It's a very cool little kernel module. I recently used it to flash the bios on one of my linux boxes that didn't have a floppy installed. Just cat the file out to/dev/bios and reboot. It's a beautiful thing, though obviously you probably wouldn't want to have it installed all the time:-)
Hmmm, works for me... Maybe you're using the Desktop Manager powertoy? That overrides your wallpaper setting - you have to set it in the Desktop Manager config. Just a wild guess as to what might be wrong. Otherwise, I guess I'd report it to M$.
Sure, you don't, I don't either. But talk to most real unix sysadmins and you'll discover that most of them spend 99% of their day logged in as root. It's been the same wat at every company I've worked, and these were large corporations. And I'm sure that many linux newbies users probably become frustrated by the lack of power granted to their lowly user accounts compared to what they can do in Windows. As a result they end up either giving up on linux or spend most of their time su'd to root.
Sure, but you could say about the hundreds of services that run above 1024 as well. Look at nearly every database server (mySQL, PostgreSQL, MSSQL, Oracle), proxy services, irc, those services don't run as root. A better answer is to lock down permissions on every port individually, set them to be owned by a specific user. So root doesn't have to be the one to start httpd, let 'www' do it. There has been a portfs patch for linux for years to allow filesystem like access to ports. Why it hasn't gotten into the real world is another thing that makes no sense to me. My whole point was that 1024 is a very arbitrary number. It's silly that we're still forced to deal with it.
True, but how often in the real world are the users logging in at the console of their own computer NOT root/administrator (or equivalent). The reasoning behind non-root users being unable bind to ports below 1024 (and a few other things) has always baffled me. The original intention as I understand it was to say that one could trust the service running on that port since it was started by the owner of that system. But this was back when the only unix systems out there were at universities and cost absurd sums of money. Now fast forward to today... unix can be run on commodity hardware and therefore nothing stops a blackhat from setting up his own server and running services on low ports or spoofing packets. So why does this restriction on ordinary users really increate the security of a system in the real world?
The solitaire bug seems to be fixed in XP. Woohoo - go Microsoft! Way to fix the important stuff...
Wow - my birthday too! (not the same year though)
on
Happy Birthday Perl!
·
· Score: 1
What an odd coincidence! No wonder I like Perl so much... it's almost like Larry Wall gave me a gift on my 12th birthday. Well... maybe not.. but funny anyway!:-)
When The Simpsons is coming down the stream from DTV it's already highly compressed VBR MPEG2, it's a cartoon - think huge expanses of solid color. DirecTV adjusts the vbr compression ratios on a per channel basis, and (rarely) on a per program basis, to acheive a set maximum bandwidth for each channel to use all of their available satellite bandwidth.
The compression ratios on the DirecTV stream are already much higher than you would hope for since they're squeezing in 100's of channels (ppv's, music, movie channels, local rebroadcasts, basic cable channels, etc...). Don't get me wrong, I love my Tivo and DirecTV. They're way better than cable. But what you gain in it being digital, you sometimes lose in hugely lossy compression. I can't wait until DirecTV starts broadcasting HDTV for everything.
Besides, to get back to your original point, the hardware to do realtime VBR MPEG2 encoding is still expensive. The standalone TiVo's out there today use CBR which is relatively cheap. It would just be silly to reencode VBR into CBR as you'll always get larger file sizes unless you go much higher in compression ratios.
Slashdot Mirror
Good luck getting one of these through airport security!
In Vanilla Sky there are 5 major characters --- Black Hawk Down has a good 20-30 that are meaningful and they are broken up into many separate groups for much of the film. Keeping the characters straight can be difficult since they're all wearing the same clothes, all have the same haircut, all are dirty and bloody. IMHO, Black Hawk Down has a much more complicated plot than Vanilla Sky.
Right you are :-) I thought those were still the updates to the old scheduler - but it does say 'scalable scheduler' under pre10. Yay!
But you are absolutely correct in that the scheduler improvements will be more apparent and dramatic on 4 and 8-way machines because of the elimination of the global run queue. Each CPU gets its own run queue and processes will only bounce around when other cpu's are idle. We finally have a scheduler that will work on enterprise class machines.
http://people.redhat.com/mingo/O(1)-scheduler/
I must say that after using it for a few days, I'm impressed. It totally changes the characteristics of multiprocess servers like Apache and PostgreSQL under high load. For example, I've run ApacheBench against a mod_perl script that queries a pgsql database, in the new scheduler I get a mean response time that is N*1.05*concurrency with a standard deviation of less than 1% of the mean. In the old scheduler I'd get a mean that is N*1.07*concurrency with a sd of up to 75% of the mean. So in other words you get essentially the same throughput with both schedulers (O(1) appears slightly faster in my limited testing). But what's more important is that in the O(1) scheduler everyone is treated equally - they all get served in 1.05*N*concurrency, no more, no less -- while with the old scheduler some requests get a response that's 1*N and others get a response all the way up to 4*N*concurrency.
IMHO, it's better to give everyone an equal level of service than to randomly favor one group of users over another.
Don't you just hate that? The company I worked for last did the same thing, very paranoid. Showed complete lack of faith in its employees. It's not as though Google can relay pr0n to you. But I suppose you could use it to translate all that subversive foreign language literature.
http://www.zeosync.com/flash/pressrelease.htm
No, it's not. Well, maybe it is, but that has nothing to do with the scale. It's based on the freezing and boiling of water just like Celsius. There's 180 degrees between boiling and freezing, go figure, just like a 1/2 circle. And it's unfortunate that Fahrenheit didn't want to use negative numbers on really cold days or he probably would have started the scale at zero. Instead he used the coldest temperature he could measure by making a slurry of salt and snow/ice. He adjusted it later when the boiling point was discovered to be 212 (and not 210) at sea level, by moving water's freezing point up 2 degrees -- though I'm still unclear on how he justified doing that, maybe changed the composition of the liquid in the thermometer at the same time. But anyway... there's no voodoo involved :-)
Doh, what the hell was I thinking/saying. I was thinking of Leigh Brackett who co-wrote Empire.
It's not a paradox that we die because we're fending off cancer. Essentially we die because our bodies are fending off mutation for our entire lives and we eventually just can't take it any more. A creator has nothing to do with it, and the finish line is not imaginary --- there is no reason not to believe that we will eventually be able to cure and prevent all forms of cancer. It's a slow process, but science is making real progress. But, trying to live forever is futile, we're all going to die either by untreatable disease, accident or violence.
Why was Jedi aimed at kids? Because of the Ewoks? There is lots of sex and violence in Jedi. In some ways more than the other movies. I don't see how it didn't deserve its PG rating. Sure, it wasn't ominous and dark like Empire was, but that was only partly because Empire's director died, mostly it was because it'd the end of a trilogy and it had a happy ending. One thing you didn't really get a feeling for was that the Ewoks were being oppressed by the Empire, that would have made the story more interesting at least.
So how much gain in performance (or apparent performance) should one expect after applying this combined patch? Are the performance gains only applicable under special circumstances? Are they focused more on desktop apps than server?
High Fidelity and Chasing Amy are not good enough films to be in anyone's top 5. I loved both of them, and I wouldn't even put them in my top 50 even though I probably gave them a 10 at IMDB. Try Memento, American Beauty, Usual Suspects, Fargo, and Taxi Driver. Those are movies that are (or will be) 'classics' in the true sense of the word and would have a much better shot at sneaking into most people's top 5. Yes, they do show a strong bias to what I like to see in movies. Age *does* have something to do with it, I'm a couple years older than you and I know why you chose those movies, they are about young people trying to find their way in life. They are not movies that would be selected by the vast majority of people or film critics to be in the top 5. I gave Fellowship of the Ring a 10 at IMDB, does that mean that it's one of my top 5? No, not yet anyway, maybe in a few years when the series is complete. But what it means to me is that it's better than 90% of the crap that the studios are feeding us. IMHO, you can only grade films on a curve to grade them fairly, they must be compared to other films - it's just the nature of the medium, nothing really stands on its own.
NT on an Alpha was just a waste of a good Alpha.
Kinda like NT on the PowerPC. Boxes that could have been perfectly good Macs, but running Windows instead. A friend of mine bought one a while back, it came with NT preinstalled on the harddrive. We wiped it almost immediately and installed LinuxPPC.
Actually it runs on 12 processor architectures, including x86 and MIPS.
Pocket PC is kind of a separate beast from Windows CE. It's basically CE plus a bunch of extensions that make it fit the needs of PDA users better. It may very well only be available for StrongARM.
Actually CE was designed for a range of processors that meet a certain spec. StrongARM is just one, there's also the Hitachi SH3.
You should try out /dev/bios. It's a very cool little kernel module. I recently used it to flash the bios on one of my linux boxes that didn't have a floppy installed. Just cat the file out to /dev/bios and reboot. It's a beautiful thing, though obviously you probably wouldn't want to have it installed all the time :-)
Hmmm, works for me... Maybe you're using the Desktop Manager powertoy? That overrides your wallpaper setting - you have to set it in the Desktop Manager config. Just a wild guess as to what might be wrong. Otherwise, I guess I'd report it to M$.
Tip: Type the comment into a text editor and then cut/paste into the browser.
Sure, you don't, I don't either. But talk to most real unix sysadmins and you'll discover that most of them spend 99% of their day logged in as root. It's been the same wat at every company I've worked, and these were large corporations. And I'm sure that many linux newbies users probably become frustrated by the lack of power granted to their lowly user accounts compared to what they can do in Windows. As a result they end up either giving up on linux or spend most of their time su'd to root.
Sure, but you could say about the hundreds of services that run above 1024 as well. Look at nearly every database server (mySQL, PostgreSQL, MSSQL, Oracle), proxy services, irc, those services don't run as root. A better answer is to lock down permissions on every port individually, set them to be owned by a specific user. So root doesn't have to be the one to start httpd, let 'www' do it. There has been a portfs patch for linux for years to allow filesystem like access to ports. Why it hasn't gotten into the real world is another thing that makes no sense to me. My whole point was that 1024 is a very arbitrary number. It's silly that we're still forced to deal with it.
True, but how often in the real world are the users logging in at the console of their own computer NOT root/administrator (or equivalent). The reasoning behind non-root users being unable bind to ports below 1024 (and a few other things) has always baffled me. The original intention as I understand it was to say that one could trust the service running on that port since it was started by the owner of that system. But this was back when the only unix systems out there were at universities and cost absurd sums of money. Now fast forward to today... unix can be run on commodity hardware and therefore nothing stops a blackhat from setting up his own server and running services on low ports or spoofing packets. So why does this restriction on ordinary users really increate the security of a system in the real world?
The solitaire bug seems to be fixed in XP. Woohoo - go Microsoft! Way to fix the important stuff...
What an odd coincidence! No wonder I like Perl so much... it's almost like Larry Wall gave me a gift on my 12th birthday. Well... maybe not.. but funny anyway! :-)
Besides, to get back to your original point, the hardware to do realtime VBR MPEG2 encoding is still expensive. The standalone TiVo's out there today use CBR which is relatively cheap. It would just be silly to reencode VBR into CBR as you'll always get larger file sizes unless you go much higher in compression ratios.