That's because of a rather famous incident with the *previous* generation of AMD chips.
Intel had recently introduced an overheat sensor into their CPUs. They still have them, I think. There's basically a thermal probe included in the CPU packaging, and if the temperature goes over a certain critical level, the CPU starts throttling itself down, until the temperature goes down to a safer level.
Tom's Hardware (probably being paid by Intel...) did a video experiment on this. They got an Intel (early P4, IIRC) and a then-current-gen Athlon, started them both playing Quake 3, then removed the HSF.
The Intel chip promptly throttled itself down to 400MHz or so, and kept running the game (rather slowly). The Athlon crashed, hit something like 200-300 degrees C, and burned a little hole in the motherboard.
After that little stunt, AMD started building overheat sensors into their CPUs quite fast.
I saw this in action on one of my own machines, a Shuttle SN62K, a couple of years back. That machine has a known issue with the motherboard fan headers dying after about a year of use. It's also a very quiet system. I was using a 2.4GHz Celeron in it at the time. The fan header died and the fan (only fan in the machine, if you know Shuttles) stopped working. The CPU throttled itself down to 800MHz and kept right on going, for two weeks, before I actually noticed.
Why are you talking in the past tense? GF/DM are both still going (and still wildly popular). V5 comes out next month, IIRC, in Japan.
Konami have taken some tentative steps towards the North American market lately. They ran a quite extensive location test of V4 in the U.S., slightly modified (same songs, but with the songs by American bands given the most prominence in the default list order). Of course, this is complicated by the fact that the American arcade market is basically dying in slow motion, in every market except ones with a significant Asian population (up here in Vancouver it's massive, because of the huge Chinese population, and we have large arcades which import most of the latest Japanese games).
Otherwise your post is basically accurate, except that there's *nothing* revolutionary about Guitar Hero. Nothing it does was a revolutionary change from GuitarFreaks. It was all just tweakage. That doesn't make its impact any less, but adding a couple of buttons, licensing songs and making them longer is hardly revolutionary.
GF/DM already effectively has 'downloadable' content, BTW. In Japan, the arcade machines are connected to the Internet. Every so often during the life cycle of one version of the game, new songs will be made available.
Harmonix didn't 'bring in to prime time', they brought it to a new market, that's all.
BTW, one thing I think you left out (and many people do) is the awesome presentation of Guitar Hero. GF/DM are just typical Konami rhythm games in presentation terms. GH had really amazing presentation; it was clearly designed by people who were totally involved in the scene they were trying to reproduce. That was a significant factor in its success, I think.
Oh, yeah, my credentials: I have five stars on all but I think four songs on Rock Band expert drums, and my DrumMania SP is around 945.:)
There's three editions of Mandriva: Free, One and Powerpack.
Free contains 100% free / open source software and doesn't cost anything. It's actually 'freer' than Ubuntu, as Ubuntu uses a fairly slack definition which allows them to include non-free firmware (they argue that Ubuntu contains no non-free software, but it _does_ contain non-free firmware; they draw a distinction between the two). Mandriva Free does not contain any non-free firmware.
One is a free-of-charge live CD, and it contains non-free hardware drivers and browser plugins - NVIDIA and ATI proprietary drivers, non-free firmware for wireless, stuff like that. All non-free software that can legally be redistributed to the general public is free of charge in Mandriva, it's available in One and in the public/non-free repository.
Powerpack has all that non-free stuff and also a few pieces of exclusive *commercial* software that we can't legally redistribute to the general public, only people who pay. Cedega, Fluendo codecs, stuff like that.
You don't need to buy a commercial edition of Mandriva to get full hardware support, as there are no hardware drivers that are not either free / open source software, or publicly re-distributable non-free. You only need to buy Powerpack if you want the commercial software it includes, or the printed manual and installation support, or just to support Mandriva.
"It really doesn't matter whether it is his decision to make, but what it does show is that when there is a decision to be made for the wider benefit of Gnome, if it is at odds with how Nokia views things, then we know which side the toast will fall on the floor."
It does not show that at all. The bug report is on Maemo, an open source project yes, but one that is controlled by Nokia, intended for Nokia devices, and not a part of GNOME in any way at all; it merely uses a lot of the same libraries. You cannot use a Maemo decision to draw any useful conclusions about GNOME.
I've no idea. The point is it really doesn't matter much to this debate, as he has no control and virtually no influence over GNOME's policies on such issues.
Nokia obviously does not want to support Vorbis. That's not Quim's decision to make. He can't change reality on the bug report and say "sure, Nokia will support Vorbis tomorrow, everything will be fine and dandy", because it's clearly *not going to happen*. But Nokia's policy is not GNOME's, and what Nokia does really has no implications for what GNOME does.
I really don't understand what you expect Quim to do on this bug report, or why you think it implies anything in particular about *GNOME's* policies, rather than Nokia's.
As was extensively explained in various GNOME places recently, Miguel is not GNOME, and has borderline zero impact or influence on GNOME at present (hence the best 'looks-serious' tag the author could find for him was "co-founder"; Woz was the co-founder of Apple, does that mean he's running iPod codec policy?) . Quim Gil is rather more directly involved in GNOME right now, but he also works for Nokia. He also clearly does not set Nokia's corporate policy. Therefore what he's doing on that bug report is reporting a corporate policy that stinks. This is obviously an uncomfortable position for him, but has sod all to do with GNOME.
Yes, we raised $1.6m for the Linbox acquisition. It was spent on the Linbox acquisition. We also raised a couple of million in venture capital. That *is* our capital for the foreseeable future. All of it.
People sometimes really don't appreciate the difference in scale between a company like Microsoft and one like Mandriva. Microsoft makes $28m *profit* in a day. For Mandriva, $2m of *capital* is a huge investment.
We never really implied that it was illegal, but that it was unethical. That's why Francois asked Mr. Ballmer how well he sleeps at night. That's not a question you ask someone you think has done something *illegal*, but a question you ask someone you think has done something *wrong*.
Personally I'd say, if you have a choice, it's far better to do something ethically right but legally wrong than legally right but ethically wrong. And if you can do ethically right and legally right, the choice seems obvious. =)
"One wonders if Mandriva bribed the gov't in Nigeria first.."
Heh. Even if you didn't believe we wouldn't be inclined to do that, we don't have the *capacity*. Microsoft can get $400,000 out of Bill's petty cash jar. We couldn't get $400,000 from anywhere.
If this really came down to a battle to see who could provide the biggest...er...factory-to-dealer incentive, we'd be dead in the water.
(I work for Mandriva, in case you didn't figure that one out yet).
This comment especially for those clever asses who will say "because they have to!" before reading the article: the author is not contending that Red Hat could somehow prevent CentOS from being made. He is wondering why Red Hat doesn't provide a low-cost, no-support, barebones edition of Red Hat to try and take some of the CentOS user share. And he has a fairly good answer to the question, too.
...why I'd want to pretend GMail was a dedicated email application running on my desktop, when I could just...run a dedicated email application on my desktop?
"What you're not realizing, or refuse to accept, is that there are many different legitimate and illegitimate reasons people have for wanting to do something outside that box."
None of which you have provided or explained, and whose notional existence I thus feel entirely justified in ignoring entirely.
"You apparently don't care about any commercial or proprietary software"
Why do you say that? There's no reason commercial and proprietary software cannot be managed within the central repository / package management system paradigm. Mandriva Linux 2008 has two repositories for non-free packages, both of which are populated and used.
"There are reasons to want a way to work outside the package management system."
Again, please provide them.
"Ever tried to install two different versions of the same software package?"
Sure. For instance, on Mandriva Linux 2008, try:
urpmi gcc urpmi gcc4.3
and you've got gcc 4.2 and gcc 4.3.
or try:
urpmi task-kde urpmi task-kde4
and you've got KDE 3 and KDE 4. Easy, ain't it?
"Ever wanted to install a package for only one user?"
RPMs are built to be relocatable (if you've ever looked at a spec file, that's why they use %{_prefix}, %{_bindir}, %{_libdir} etc etc etc instead of just/usr ,/usr/bin ,/usr/lib). You could install most 2008 packages inside a user's home directory if you really wanted to.
"Ever wanted to install some software that isn't available in your package management system and never will be?"
All software should be managed through a package management system. I won't say all software could be, as that's a hostage to fortune, but I will say some ridiculously high percentage could.
"Good luck with that. But more importantly, the main problem here is that you're assuming that because you see no reason to do it, there is no reason. That kind of thinking is arrogant, and wrong. And one of the reasons that package management systems aren't perfect for every user and every situation."
No, I am saying that the vast majority of usage scenarios can and should be handled through package management, and I am annoyed when people suggest silly ideas that would massively undermine the quality of package management on Linux systems purely to solve problems that aren't really there.
the new mandriva.com is a permanent design, it's not temporary for 2008. it'll stay like that.
for the Club, yeah, it's a bit less obvious, but then we're not really going to be offering downloads through the Club in future anyway with the changes in that area. There's a Download link in the bar of links near the top of the Club page. That takes you to the download section where Club-specific downloads (i.e. Powerpack editions) are hosted.
Has nothing to do with urpmi, but with the packaging of truecrypt. You can file bugs for issues like this at http://bugzilla.mandriva.com/ , where they will be addressed.
autopackage is not the answer to all ills. for instance, the way autopackage is configured by upstream to operate on Mandriva will interfere with the operation of the regular Mandriva packages. it's also completely unsuitable for anything particularly complex or low-level, it really only works well for simple application-level stuff. it's not a bad idea, but it does not solve all problems, and the way it's implemented currently is not great and is definitely not supported by the Mandriva development community.
I post in various places in a manner appropriate to that place. that is, I post on slashdot in a manner broadly appropriate to the typical approach of slashdot users - blunt and direct. I find this is appreciated more than using an all-purpose smiling corporate drone approach in all environments.
"You don't suppose they're launching Mandriva 2008 now because they know the release of Ubuntu Gutsy Gibbon is imminent, do you?"
No, we're not. We release on a six month cycle, and always have, except for 2007 which came on a one year cycle as an experiment. Excepting that release, we've released around every April and every October since before Ubuntu existed.
Maybe Ubuntu initially chose to do all their releases slightly after ours in order to try and capitalize on the buzz created by each Mandriva release...:)
First, as has been explained several times in this thread, there is absolutely zero point in including Firefox 2.0.0.7, as the only change in 2.0.0.7 is a fix which is entirely irrelevant to Linux.
OpenOffice.org 2.3.0 was released on September 18th. That is not 'a while'. We were already in the Release Candidate stage at that point. Would you expect Microsoft to do a major version update of, say, Windows Media Player or Internet Explorer between Vista RC2 and Vista final? Of course not.
"The few pieces of software that have pre-compiled downloadable Linux versions still need at least three different types of packages just to cover "most" of the popular Linux distros"
This is because the idea of having pre-compiled downloadable Linux versions is, frankly, silly. The package management system works best when people understand it, and worst when they try to do end runs around it. This is not surprising. Software writers should write, and packagers (who work for individual distros) should package. That system works great. It's when people start messing with it that you get problems.
"And then we wonder why nobody bothers to develop for "Linux""
We do? Can't say I find myself kept up at night wondering about that. Maybe because lots of people *do* develop for Linux. It's simple - release source code.
"And there is no single clear-cut procedure for installing software completely outside of the native package management system in a way that neither will ever interfere with the other."
Why do you want one? What's wrong with the package management system?
"Good God, when I think of all the man-hours that are being wasted with all this idiotic redundancy, and all the time spent by users complaining in forums that their distro-of-choice doesn't have the latest version of package X yet because the package maintainer is on vacation, it makes my head hurt."
Then go think about something else and quit trolling Linux threads. Good lord, if anyone's wasting their time around here it's you.
No. His point was that the Linux compositing managers did not rip off Windows.
Amazingly enough, nobody ripped anybody else off. Compositing is a fairly obvious idea. As the technology to implement it became available, it was implemented, by all the major operating systems.
Actually, sound is becoming troublesome again with recent motherboards. A lot of new motherboards use slightly differing implementations of the HDA audio codec, and each different ones needs minor tweaks to the snd-hda-intel driver to make it work 'out of the box'. I think we're up to dozens or hundreds of these tweaks now. If you went out and bought a random sample of modern laptops, the onboard sound in a lot of them would not work with, say, kernel 2.6.21.
Slashdot Mirror
Intel had recently introduced an overheat sensor into their CPUs. They still have them, I think. There's basically a thermal probe included in the CPU packaging, and if the temperature goes over a certain critical level, the CPU starts throttling itself down, until the temperature goes down to a safer level.
Tom's Hardware (probably being paid by Intel...) did a video experiment on this. They got an Intel (early P4, IIRC) and a then-current-gen Athlon, started them both playing Quake 3, then removed the HSF.
The Intel chip promptly throttled itself down to 400MHz or so, and kept running the game (rather slowly). The Athlon crashed, hit something like 200-300 degrees C, and burned a little hole in the motherboard.
After that little stunt, AMD started building overheat sensors into their CPUs quite fast.
I saw this in action on one of my own machines, a Shuttle SN62K, a couple of years back. That machine has a known issue with the motherboard fan headers dying after about a year of use. It's also a very quiet system. I was using a 2.4GHz Celeron in it at the time. The fan header died and the fan (only fan in the machine, if you know Shuttles) stopped working. The CPU throttled itself down to 800MHz and kept right on going, for two weeks, before I actually noticed.
The mind just *boggles*.
Konami have taken some tentative steps towards the North American market lately. They ran a quite extensive location test of V4 in the U.S., slightly modified (same songs, but with the songs by American bands given the most prominence in the default list order). Of course, this is complicated by the fact that the American arcade market is basically dying in slow motion, in every market except ones with a significant Asian population (up here in Vancouver it's massive, because of the huge Chinese population, and we have large arcades which import most of the latest Japanese games).
Otherwise your post is basically accurate, except that there's *nothing* revolutionary about Guitar Hero. Nothing it does was a revolutionary change from GuitarFreaks. It was all just tweakage. That doesn't make its impact any less, but adding a couple of buttons, licensing songs and making them longer is hardly revolutionary.
GF/DM already effectively has 'downloadable' content, BTW. In Japan, the arcade machines are connected to the Internet. Every so often during the life cycle of one version of the game, new songs will be made available.
Harmonix didn't 'bring in to prime time', they brought it to a new market, that's all.
BTW, one thing I think you left out (and many people do) is the awesome presentation of Guitar Hero. GF/DM are just typical Konami rhythm games in presentation terms. GH had really amazing presentation; it was clearly designed by people who were totally involved in the scene they were trying to reproduce. That was a significant factor in its success, I think.
Oh, yeah, my credentials: I have five stars on all but I think four songs on Rock Band expert drums, and my DrumMania SP is around 945. :)
'funnier-than-i-thought-he'd-be department'? well, fun trivia - Scott Card wrote the original sword fighting insults for Monkey Island...
There's three editions of Mandriva: Free, One and Powerpack.
Free contains 100% free / open source software and doesn't cost anything. It's actually 'freer' than Ubuntu, as Ubuntu uses a fairly slack definition which allows them to include non-free firmware (they argue that Ubuntu contains no non-free software, but it _does_ contain non-free firmware; they draw a distinction between the two). Mandriva Free does not contain any non-free firmware.
One is a free-of-charge live CD, and it contains non-free hardware drivers and browser plugins - NVIDIA and ATI proprietary drivers, non-free firmware for wireless, stuff like that. All non-free software that can legally be redistributed to the general public is free of charge in Mandriva, it's available in One and in the public /non-free repository.
Powerpack has all that non-free stuff and also a few pieces of exclusive *commercial* software that we can't legally redistribute to the general public, only people who pay. Cedega, Fluendo codecs, stuff like that.
You don't need to buy a commercial edition of Mandriva to get full hardware support, as there are no hardware drivers that are not either free / open source software, or publicly re-distributable non-free. You only need to buy Powerpack if you want the commercial software it includes, or the printed manual and installation support, or just to support Mandriva.
A spiral bound notebook and a film camera.
"It really doesn't matter whether it is his decision to make, but what it does show is that when there is a decision to be made for the wider benefit of Gnome, if it is at odds with how Nokia views things, then we know which side the toast will fall on the floor."
It does not show that at all. The bug report is on Maemo, an open source project yes, but one that is controlled by Nokia, intended for Nokia devices, and not a part of GNOME in any way at all; it merely uses a lot of the same libraries. You cannot use a Maemo decision to draw any useful conclusions about GNOME.
I've no idea. The point is it really doesn't matter much to this debate, as he has no control and virtually no influence over GNOME's policies on such issues.
How can he 'separate the two'?
Nokia obviously does not want to support Vorbis. That's not Quim's decision to make. He can't change reality on the bug report and say "sure, Nokia will support Vorbis tomorrow, everything will be fine and dandy", because it's clearly *not going to happen*. But Nokia's policy is not GNOME's, and what Nokia does really has no implications for what GNOME does.
I really don't understand what you expect Quim to do on this bug report, or why you think it implies anything in particular about *GNOME's* policies, rather than Nokia's.
As was extensively explained in various GNOME places recently, Miguel is not GNOME, and has borderline zero impact or influence on GNOME at present (hence the best 'looks-serious' tag the author could find for him was "co-founder"; Woz was the co-founder of Apple, does that mean he's running iPod codec policy?) . Quim Gil is rather more directly involved in GNOME right now, but he also works for Nokia. He also clearly does not set Nokia's corporate policy. Therefore what he's doing on that bug report is reporting a corporate policy that stinks. This is obviously an uncomfortable position for him, but has sod all to do with GNOME.
Yes, we raised $1.6m for the Linbox acquisition. It was spent on the Linbox acquisition. We also raised a couple of million in venture capital. That *is* our capital for the foreseeable future. All of it.
People sometimes really don't appreciate the difference in scale between a company like Microsoft and one like Mandriva. Microsoft makes $28m *profit* in a day. For Mandriva, $2m of *capital* is a huge investment.
We never really implied that it was illegal, but that it was unethical. That's why Francois asked Mr. Ballmer how well he sleeps at night. That's not a question you ask someone you think has done something *illegal*, but a question you ask someone you think has done something *wrong*.
Personally I'd say, if you have a choice, it's far better to do something ethically right but legally wrong than legally right but ethically wrong. And if you can do ethically right and legally right, the choice seems obvious. =)
"One wonders if Mandriva bribed the gov't in Nigeria first.."
Heh. Even if you didn't believe we wouldn't be inclined to do that, we don't have the *capacity*. Microsoft can get $400,000 out of Bill's petty cash jar. We couldn't get $400,000 from anywhere.
If this really came down to a battle to see who could provide the biggest...er...factory-to-dealer incentive, we'd be dead in the water.
(I work for Mandriva, in case you didn't figure that one out yet).
This comment especially for those clever asses who will say "because they have to!" before reading the article: the author is not contending that Red Hat could somehow prevent CentOS from being made. He is wondering why Red Hat doesn't provide a low-cost, no-support, barebones edition of Red Hat to try and take some of the CentOS user share. And he has a fairly good answer to the question, too.
the stock Classmate configuration does not even *have* a hard disk, it has a (small) Flash storage unit.
...why I'd want to pretend GMail was a dedicated email application running on my desktop, when I could just...run a dedicated email application on my desktop?
"What you're not realizing, or refuse to accept, is that there are many different legitimate and illegitimate reasons people have for wanting to do something outside that box."
/usr , /usr/bin , /usr/lib). You could install most 2008 packages inside a user's home directory if you really wanted to.
None of which you have provided or explained, and whose notional existence I thus feel entirely justified in ignoring entirely.
"You apparently don't care about any commercial or proprietary software"
Why do you say that? There's no reason commercial and proprietary software cannot be managed within the central repository / package management system paradigm. Mandriva Linux 2008 has two repositories for non-free packages, both of which are populated and used.
"There are reasons to want a way to work outside the package management system."
Again, please provide them.
"Ever tried to install two different versions of the same software package?"
Sure. For instance, on Mandriva Linux 2008, try:
urpmi gcc
urpmi gcc4.3
and you've got gcc 4.2 and gcc 4.3.
or try:
urpmi task-kde
urpmi task-kde4
and you've got KDE 3 and KDE 4. Easy, ain't it?
"Ever wanted to install a package for only one user?"
RPMs are built to be relocatable (if you've ever looked at a spec file, that's why they use %{_prefix}, %{_bindir}, %{_libdir} etc etc etc instead of just
"Ever wanted to install some software that isn't available in your package management system and never will be?"
All software should be managed through a package management system. I won't say all software could be, as that's a hostage to fortune, but I will say some ridiculously high percentage could.
"Good luck with that. But more importantly, the main problem here is that you're assuming that because you see no reason to do it, there is no reason. That kind of thinking is arrogant, and wrong. And one of the reasons that package management systems aren't perfect for every user and every situation."
No, I am saying that the vast majority of usage scenarios can and should be handled through package management, and I am annoyed when people suggest silly ideas that would massively undermine the quality of package management on Linux systems purely to solve problems that aren't really there.
the new mandriva.com is a permanent design, it's not temporary for 2008. it'll stay like that. for the Club, yeah, it's a bit less obvious, but then we're not really going to be offering downloads through the Club in future anyway with the changes in that area. There's a Download link in the bar of links near the top of the Club page. That takes you to the download section where Club-specific downloads (i.e. Powerpack editions) are hosted.
Has nothing to do with urpmi, but with the packaging of truecrypt. You can file bugs for issues like this at http://bugzilla.mandriva.com/ , where they will be addressed.
autopackage is not the answer to all ills. for instance, the way autopackage is configured by upstream to operate on Mandriva will interfere with the operation of the regular Mandriva packages. it's also completely unsuitable for anything particularly complex or low-level, it really only works well for simple application-level stuff. it's not a bad idea, but it does not solve all problems, and the way it's implemented currently is not great and is definitely not supported by the Mandriva development community.
I post in various places in a manner appropriate to that place. that is, I post on slashdot in a manner broadly appropriate to the typical approach of slashdot users - blunt and direct. I find this is appreciated more than using an all-purpose smiling corporate drone approach in all environments.
I'm as English as Neil Hancock, I'm afraid.
:)
Truecrypt woes? Do elaborate.
"Unreal 2004 installer"
there's your problem. as I wrote in my original message.
Step 1:
:)
go to www.mandriva.com
Step 2:
click on the giant green button that says Download
Advice on how we could make it easier is very welcome.
"You don't suppose they're launching Mandriva 2008 now because they know the release of Ubuntu Gutsy Gibbon is imminent, do you?"
No, we're not. We release on a six month cycle, and always have, except for 2007 which came on a one year cycle as an experiment. Excepting that release, we've released around every April and every October since before Ubuntu existed.
Maybe Ubuntu initially chose to do all their releases slightly after ours in order to try and capitalize on the buzz created by each Mandriva release...:)
Sheesh, take a pill.
First, as has been explained several times in this thread, there is absolutely zero point in including Firefox 2.0.0.7, as the only change in 2.0.0.7 is a fix which is entirely irrelevant to Linux.
OpenOffice.org 2.3.0 was released on September 18th. That is not 'a while'. We were already in the Release Candidate stage at that point. Would you expect Microsoft to do a major version update of, say, Windows Media Player or Internet Explorer between Vista RC2 and Vista final? Of course not.
"The few pieces of software that have pre-compiled downloadable Linux versions still need at least three different types of packages just to cover "most" of the popular Linux distros"
This is because the idea of having pre-compiled downloadable Linux versions is, frankly, silly. The package management system works best when people understand it, and worst when they try to do end runs around it. This is not surprising. Software writers should write, and packagers (who work for individual distros) should package. That system works great. It's when people start messing with it that you get problems.
"And then we wonder why nobody bothers to develop for "Linux""
We do? Can't say I find myself kept up at night wondering about that. Maybe because lots of people *do* develop for Linux. It's simple - release source code.
"And there is no single clear-cut procedure for installing software completely outside of the native package management system in a way that neither will ever interfere with the other."
Why do you want one? What's wrong with the package management system?
"Good God, when I think of all the man-hours that are being wasted with all this idiotic redundancy, and all the time spent by users complaining in forums that their distro-of-choice doesn't have the latest version of package X yet because the package maintainer is on vacation, it makes my head hurt."
Then go think about something else and quit trolling Linux threads. Good lord, if anyone's wasting their time around here it's you.
No. His point was that the Linux compositing managers did not rip off Windows.
Amazingly enough, nobody ripped anybody else off. Compositing is a fairly obvious idea. As the technology to implement it became available, it was implemented, by all the major operating systems.
FILM AT 11!
Actually, sound is becoming troublesome again with recent motherboards. A lot of new motherboards use slightly differing implementations of the HDA audio codec, and each different ones needs minor tweaks to the snd-hda-intel driver to make it work 'out of the box'. I think we're up to dozens or hundreds of these tweaks now. If you went out and bought a random sample of modern laptops, the onboard sound in a lot of them would not work with, say, kernel 2.6.21.