It strikes me that any such list should contain surprises only to people who never go to watch movies with these stars in the theatres. After all, 35mm film is still considerably sharper than even 1080 material.
The really interesting thing is the difference between how stars look when aided by camera angles, lighting, makeup, post-process filters, compared to meeting them in person. There's a staggering number of famous "beauties" who look VERY average when not filtered through the full set of beautifying tools of the filmmaking/photographic process.
You can't make the upscaled/interpolated versions contain more _accurate_ data than whatever was in the original stream. You can, however, create new "guessed" data that makes the resulting image or audio look and sound _subjectively_ better. It gives a more pleasant viewing/listening experience, which is the only thing miltimj is claming, despite attempts by responders to put words and claims in his mouth. Nowhere does he say it's an equal thing to a genuine 720 or 1080 source stream.
And as for upsampling CD audio - most decent standalone CD players and even soundcards _do_ just that, and have done for years, exactly because the resulting audio _sounds_ better, even if it contains no more "information" than the 16bit/44.1khz data on the CD.
What part of "don't care about data loss" did you fail to understand? Why would I want to waste %25 of my volume's storage capacity to get better data security on something where I don't _care_ about data security? And no - raid-5 doesn't match raid-0 for speed even on reads, at least not in my linux software raid setup. No "probably" about it - I have a raid5 volume running on the same hd's, where I keep data I actually care about.
If you need speed, but don't care about data loss raid-0 is entirely appropriate. For example, I keep my games installed on a 4-stripe raid-0 volume, since it helps considerably with level load times, and should a drive fail, it's easy enough to reinstall the games.
Re:And yet Europe seems to be doing fine
on
Pornified
·
· Score: 2, Informative
Well, the title IS meant to be funny. Åmål is the name of the small town, and it's used in the context "Why do I have to live in fucking ÅMÅL!?" in the line of dialog the film takes its title from. That said, I think most Swedes who scoff at the re-titling of the movie for english-speaking countries simply don't understand just how offensive a word "fucking" really is to most native english speakers. Especially since, if you don't know that Åmål is a small town, the context in which it's used here is by no means obvious.
If someone made a movie here in sweden with the word "knulla" or "fitta" in the title and put up posters for it in the subway or type the title out in 30cm tall letters on the movie marquee, i believe it would probably cause a bit of fuss, even here. Although probably just a little.
99% of the time someone says "glucose" they still mean dextrose. L-Glucose likely won't be encountered outside of the lab (hint - all glucose found in nature is d-glucose), and if anyone meant l-glucose, they'd specifically say l-glucose. Very lame excuse to quit your diet. It's not that hard to keep track of carbohydrate, since all you need to be careful about in the real world is avoiding too much fructose fructose, and making sure your total carb amounts are reasonably matched to your level of physical activity.
Erhm.... glucose and dextrose _are_ the same thing. And the body does process the sugar from apples the same as sugar from coke and pepsi - both sources provide fructose, glucose (some in the bonded form of sucrose, but that gets metabolized as 50/50 dextrose/fructose eventually). It's just that it's a fuckload easier to get unhealthy _amounts_ of fructose from coke and pepsi than it is from eating apples. Not to mention that there aren't exactly a lot of healthy fiber, vitamins in coke, unlike in apples.
The problem with really cheap graphics card is the often monstrously awful quality of their analog output. Not a problem if you use DVI w. flatscreen, I guess, now that even the cheapest cards usually have a DVI connector.
I have very similar experiences to yours when it comes to moving office/email/web-focused users to Linux - the need for help practically disappears. Once set up, it Just Works(tm). About your parents and the long distance support - I suggest creating a live CD which is properly configured for their broadband connection, and has an SSH account set up for you. That way, almost no matter how fuxxored their machine gets, they'll still be able to boot it on the CD, and allow you to log in and see (and probably fix, unless it's a dead hd or similar) what the problem is. Works great, and I've fixed some really massacred systems with this kind of rescue disk.
I haven't tried Ubuntu myself, but I'm hearing nothing but good things about it. However, you're wrong about there not being any easy ways to upgrade Fedora Core distros. I haven't reinstalled any of my personal machines for ages. My fileserver was originally installed with RedHat 7.1, and has been smoothly upgraded from that to the FC4 it's running now. And for the last few years, it hasn't even had a graphics card in it, so I've been doing the upgrades remotely, using apt for rpm. Not a single problem with it, so far.
My main box, due to it running a bunch more self-made rpms and tweaks, has turned up some minor strangeness with distro upgrades via apt, but nothing that was any actual trouble to resolve.
The officially supported upgrade path for Fedora/Redhat distros (Pop CD/DVD in, reboot, and choose upgrade) is "easy" too. It's just inconvenient compared to apt-get or yum "online" upgrades, since you lose the use of your computer during the upgrade, wheras with apt or yum, you can keep working as the upgrade process runs. Redhat really do need to get with the program and make apt/yum full distro upgrades officially supported, like they are in debian-based distros. Preferably, I'd like to see the Smart Package Manager replace apt/yum, once it's out of beta, because from trying it, it seriously kicks the butt of both apt and yum.
Occasionally, the reaction is the opposite. During the testing phase of a project I was working on as a consultant, the single closed-source product involved was the one causing the most trouble, and one problem was that it kept messing with a certain file in/etc, sometimes causing a situation where the system never fully booted. I had to create a small script to be run during init which looked for this condition and fixed it. I called this script $PRODUCTNAME_Unfuck. Every bootup, you'd end up seeing $PRODUCTNAME_Unfuck [OK] printed during the boot process. Eventually I managed to get $BIG_SOFTWARE_COMPANY to fix the bug in $PRODUCTNAME, and the patches got applied, and the insultingly named little script was no longer necessary. However, when I visited the customer, long after the project was finished and taken into production, I witnessed $PRODUCTNAME_Unfuck [OK] pop past as one of the systems booted. Turned out that due to still being not-exactly-happy with $COMPANY or $PRODUCTNAME (there had been plenty of other problems) my clients had left the script in (it was basically a NOOP at this point since the error was fixed) simply because they wanted to needle $COMPANY's consultants when they visited. I thought it was pretty funny.
This depends on what soundcard you have. Most should be supported, but there are a few exceptions. The nForce MCP onboard audio, for example, needs a binary driver from nVidia to get sp/dif working.
You don't need to reinstall; upgrading works, if you know what you're doing. My fileserver, currently running FC3, started out as a RH7.0 install, and has been upgraded from there to what it is now. The last bunch of upgrades have also been done remotely and "live", since the machine no longer has a graphics card in it. This is not something I recommend for newbies though; complete "distro-upgrades" with apt-rpm don't tend to be trouble-free. If you have the skills to work around the problems, it's quite doable, though.
I tend to upgrade all my machines that way; lets me continue working while the entire distro-version changes underneath.
Well, it depends. From a practical point view, you could say that you have "infinite bandwidth" the moment you have more bandwith in the interconnect than the connected equipment could fill even under peak theoretical throughput. For example, in a supercomputing cluster, if you interconnect is faster than the combined memory bandwidth of all connected nodes, you would pretty much have "infinite bandwidth". Otoh, 100Gb/sec doesn't sound like it comes even close to that, for a decent size cluster of modern nodes. I seem to remember the NUMA backplane on SGI's Origin 3000 servers having something like 760GB/sec... That, otoh, might almost fit the bill.
My little cluster of two dual Athlon's (one machine with 2x MP 1600+, the other 2x MP 2800+, each with 512MB RAM) compiles 2.6.4 in 1 min, 57 secs. Needless to say, I'm rather unimpressed with the benchmark of this 4-way Opteron. I seem to remember one of the big 12-way IBM Power4 servers doing a kernel compile in 6 secs... now THAT is impressive.:)
> I seriously doubt the validity of a test that > outright asks you "Would you torture people in > a concentration camp?"
The HR test doesn't ask that. The questions are not "would you..." this or that - the phrases are devoid of such context, and the test simply asks you to enumerate phrases such as "torture a person", "slavery", "baby", etc on a scale of badness/godness, from most positive to most negative.
I've been running linux software raid5 on numerous SMP systems for over 4 years, and I've never seen any trouble with it, either under normal operations, or after HD or controller failures. Might be filesystem-specific, though - I've used ext2, ext3 & xfs on these machines, never jfs or reiserfs, so I can't vouch for their behavior on this setup. Likewise, all the SMP systems have been x86-architecture. No experience with PowerPC, Sparc or Alpha SMP systems running linux software raid.
I've personally done a 7km caving trip in one go. Took nearly 9 hours, and left me more exhausted than I've ever been in my entire life, before or after. I was in pretty lousy shape at the time, though. The first 7 hours were great. The last two were torture. Anyway, to get to the point; the standard through-trip for this particular cave (Doolin Cave, in Ireland) is 3.5 kilometers. 4 km is _not_ a particularly long caving trip, if you're talking mostly level progress in a non-submerged cave. Spelunking rules.:)
Of course. And the IO on writes is _incredible_. Takes a hell of a lot of time to read the data back, though, but anyone who has worked with backups to tape libraries should be used to that.
Also, lets not forget rpm -V and rpm -Va, which verifies the integrity of the files in installed packages. I recently managed to mangle my system badly, with filesystem corruption on the root filesystem leading to multiple damaged executables and libraries. I could still recover it without a reinstall, by checking rpm -Va, and looking for md5sum and size mismatches on stuff in binary and library paths, and simply doing apt-get reinstall package on the things with bad files. Quite lovely.
Slashdot Mirror
It strikes me that any such list should contain surprises only to people who never go to watch movies with these stars in the theatres. After all, 35mm film is still considerably sharper than even 1080 material. The really interesting thing is the difference between how stars look when aided by camera angles, lighting, makeup, post-process filters, compared to meeting them in person. There's a staggering number of famous "beauties" who look VERY average when not filtered through the full set of beautifying tools of the filmmaking/photographic process.
You can't make the upscaled/interpolated versions contain more _accurate_ data than whatever was in the original stream. You can, however, create new "guessed" data that makes the resulting image or audio look and sound _subjectively_ better. It gives a more pleasant viewing/listening experience, which is the only thing miltimj is claming, despite attempts by responders to put words and claims in his mouth. Nowhere does he say it's an equal thing to a genuine 720 or 1080 source stream.
And as for upsampling CD audio - most decent standalone CD players and even soundcards _do_ just that, and have done for years, exactly because the resulting audio _sounds_ better, even if it contains no more "information" than the 16bit/44.1khz data on the CD.
What part of "don't care about data loss" did you fail to understand?
Why would I want to waste %25 of my volume's storage capacity to get better data security on something where I don't _care_ about data security? And no - raid-5 doesn't match raid-0 for speed even on reads, at least not in my linux software raid setup. No "probably" about it - I have a raid5 volume running on the same hd's, where I keep data I actually care about.
If you need speed, but don't care about data loss raid-0 is entirely appropriate. For example, I keep my games installed on a 4-stripe raid-0 volume, since it helps considerably with level load times, and should a drive fail, it's easy enough to reinstall the games.
Well, the title IS meant to be funny. Åmål is the name of the small town, and it's used in the context "Why do I have to live in fucking ÅMÅL!?" in the line of dialog the film takes its title from.
That said, I think most Swedes who scoff at the re-titling of the movie for english-speaking countries simply don't understand just how offensive a word "fucking" really is to most native english speakers. Especially since, if you don't know that Åmål is a small town, the context in which it's used here is by no means obvious.
If someone made a movie here in sweden with the word "knulla" or "fitta" in the title and put up posters for it in the subway or type the title out in 30cm tall letters on the movie marquee, i believe it would probably cause a bit of fuss, even here. Although probably just a little.
99% of the time someone says "glucose" they still mean dextrose. L-Glucose likely won't be encountered outside of the lab (hint - all glucose found in nature is d-glucose), and if anyone meant l-glucose, they'd specifically say l-glucose.
Very lame excuse to quit your diet. It's not that hard to keep track of carbohydrate, since all you need to be careful about in the real world is avoiding too much fructose fructose, and making sure your total carb amounts are reasonably matched to your level of physical activity.
Erhm.... glucose and dextrose _are_ the same thing. And the body does process the sugar from apples the same as sugar from coke and pepsi - both sources provide fructose, glucose (some in the bonded form of sucrose, but that gets metabolized as 50/50 dextrose/fructose eventually). It's just that it's a fuckload easier to get unhealthy _amounts_ of fructose from coke and pepsi than it is from eating apples. Not to mention that there aren't exactly a lot of healthy fiber, vitamins in coke, unlike in apples.
The problem with really cheap graphics card is the often monstrously awful quality of their analog output. Not a problem if you use DVI w. flatscreen, I guess, now that even the cheapest cards usually have a DVI connector.
Braindead was Jackson's _third_ movie.
Between Bad Taste and Braindead, there was the hilarious little muppets-for-adults movie "Meet the Feebles".
And personally, as fun as Bad Taste is, I don't think it was better than Braindead.
I have very similar experiences to yours when it comes to moving office/email/web-focused users to Linux - the need for help practically disappears. Once set up, it Just Works(tm).
About your parents and the long distance support - I suggest creating a live CD which is properly configured for their broadband connection, and has an SSH account set up for you.
That way, almost no matter how fuxxored their machine gets, they'll still be able to boot it on the CD, and allow you to log in and see (and probably fix, unless it's a dead hd or similar) what the problem is. Works great, and I've fixed some really massacred systems with this kind of rescue disk.
Deep Impact released energy equivalent to only 4.5 _tons_ of TNT, not 4.5 kilotons.
I've lost count of how many times I've seen this mistake made.
I haven't tried Ubuntu myself, but I'm hearing nothing but good things about it.
However, you're wrong about there not being any easy ways to upgrade Fedora Core distros. I haven't reinstalled any of my personal machines for ages. My fileserver was originally installed with RedHat 7.1, and has been smoothly upgraded from that to the FC4 it's running now. And for the last few years, it hasn't even had a graphics card in it, so I've been doing the upgrades remotely, using apt for rpm. Not a single problem with it, so far.
My main box, due to it running a bunch more self-made rpms and tweaks, has turned up some minor strangeness with distro upgrades via apt, but nothing that was any actual trouble to resolve.
The officially supported upgrade path for Fedora/Redhat distros (Pop CD/DVD in, reboot, and choose upgrade) is "easy" too. It's just inconvenient compared to apt-get or yum "online" upgrades, since you lose the use of your computer during the upgrade, wheras with apt or yum, you can keep working as the upgrade process runs. Redhat really do need to get with the program and make apt/yum full distro upgrades officially supported, like they are in debian-based distros. Preferably, I'd like to see the Smart Package Manager replace apt/yum, once it's out of beta, because from trying it, it seriously kicks the butt of both apt and yum.
Occasionally, the reaction is the opposite. /etc, sometimes causing a situation where the system never fully booted. I had to create a small script to be run during init which looked for this condition and fixed it. I called this script $PRODUCTNAME_Unfuck. Every bootup, you'd end up seeing $PRODUCTNAME_Unfuck [OK] printed during the boot process.
During the testing phase of a project I was working on as a consultant, the single closed-source product involved was the one causing the most trouble, and one problem was that it kept messing with a certain file in
Eventually I managed to get $BIG_SOFTWARE_COMPANY to fix the bug in $PRODUCTNAME, and the patches got applied, and the insultingly named little script was no longer necessary.
However, when I visited the customer, long after the project was finished and taken into production, I witnessed $PRODUCTNAME_Unfuck [OK] pop past as one of the systems booted. Turned out that due to still being not-exactly-happy with $COMPANY or $PRODUCTNAME (there had been plenty of other problems) my clients had left the script in (it was basically a NOOP at this point since the error was fixed) simply because they wanted to needle $COMPANY's consultants when they visited. I thought it was pretty funny.
Smart package manager really rocks, and needs more publicity. It blows apt and yum utterly out of the water.
'nuff said.
This depends on what soundcard you have. Most should be supported, but there are a few exceptions. The nForce MCP onboard audio, for example, needs a binary driver from nVidia to get sp/dif working.
Check out the Alsa soundcard matrix at:
http://www.alsa-project.org/alsa-doc/
That should let you know what the status of support is for your soundcard.
Hubble was launched in 1990. The servicing mission wasn't until 1993. In between, they did indeed rely on math to get useable images out of hubble.
You don't need to reinstall; upgrading works, if you know what you're doing. My fileserver, currently running FC3, started out as a RH7.0 install, and has been upgraded from there to what it is now. The last bunch of upgrades have also been done remotely and "live", since the machine no longer has a graphics card in it. This is not something I recommend for newbies though; complete "distro-upgrades" with apt-rpm don't tend to be trouble-free. If you have the skills to work around the problems, it's quite doable, though. I tend to upgrade all my machines that way; lets me continue working while the entire distro-version changes underneath.
Well, it depends. From a practical point view, you could say that you have "infinite bandwidth" the moment you have more bandwith in the interconnect than the connected equipment could fill even under peak theoretical throughput. For example, in a supercomputing cluster, if you interconnect is faster than the combined memory bandwidth of all connected nodes, you would pretty much have "infinite bandwidth".
Otoh, 100Gb/sec doesn't sound like it comes even close to that, for a decent size cluster of modern nodes. I seem to remember the NUMA backplane on SGI's Origin 3000 servers having something like 760GB/sec... That, otoh, might almost fit the bill.
My little cluster of two dual Athlon's (one machine with 2x MP 1600+, the other 2x MP 2800+, each with 512MB RAM) compiles 2.6.4 in 1 min, 57 secs. Needless to say, I'm rather unimpressed with the benchmark of this 4-way Opteron. I seem to remember one of the big 12-way IBM Power4 servers doing a kernel compile in 6 secs... now THAT is impressive. :)
> I seriously doubt the validity of a test that
> outright asks you "Would you torture people in
> a concentration camp?"
The HR test doesn't ask that. The questions are not "would you..." this or that - the phrases are devoid of such context, and the test simply asks you to enumerate phrases such as "torture a person", "slavery", "baby", etc on a scale of badness/godness, from most positive to most negative.
There's an online Hartman Value Profile test/calculator at:
http://www.qis.net/~jschmitz/hvp/test1.html
I've been running linux software raid5 on numerous SMP systems for over 4 years, and I've never seen any trouble with it, either under normal operations, or after HD or controller failures. Might be filesystem-specific, though - I've used ext2, ext3 & xfs on these machines, never jfs or reiserfs, so I can't vouch for their behavior on this setup. Likewise, all the SMP systems have been x86-architecture. No experience with PowerPC, Sparc or Alpha SMP systems running linux software raid.
I've personally done a 7km caving trip in one go. Took nearly 9 hours, and left me more exhausted than I've ever been in my entire life, before or after. I was in pretty lousy shape at the time, though. The first 7 hours were great. The last two were torture. :)
Anyway, to get to the point; the standard through-trip for this particular cave (Doolin Cave, in Ireland) is 3.5 kilometers. 4 km is _not_ a particularly long caving trip, if you're talking mostly level progress in a non-submerged cave.
Spelunking rules.
Of course. And the IO on writes is _incredible_. Takes a hell of a lot of time to read the data back, though, but anyone who has worked with backups to tape libraries should be used to that.
Also, lets not forget rpm -V and rpm -Va, which verifies the integrity of the files in installed packages.
I recently managed to mangle my system badly, with filesystem corruption on the root filesystem leading to multiple damaged executables and libraries. I could still recover it without a reinstall, by checking rpm -Va, and looking for md5sum and size mismatches on stuff in binary and library paths, and simply doing apt-get reinstall package on the things with bad files. Quite lovely.