No, not true - you can certainly use Optimus cards on Linux, you just have to choose between the integrated chipset or the dedicated chipset at boot time. What you don't get is the power savings from being able to dynamically switch between the low-power integrated Intel gfx and the high performance NVidia gfx. It's really not that big of a deal - the battery life on my thinkpad is just fine using the NVidia gfx 100% of the time.
In this case NVIDIA provides me with a driver that works. I'm happy. I don't care what Linus has to say. NVidia, in my book, is supporting Linux on the desktop and for that I am happy.
+1
NVIDIA has been providing stable, fast, feature complete drivers for years and have supported Linux and FreeBSD just as well as they support Windows.
ATI, on the other hand, released specs years ago and the open source drivers are still unstable, slow, and incredibly buggy. The Intel drivers seem a little more stable than ATI, but they're still ridiculously slow and not feature complete.
I develop 3d software on Linux (and OSX, Windows) for a living and I test NVIDIA, ATI, and Intel gfx hardware on a regular basis. The NVIDIA closed source driver is the only linux 3d driver that is acceptable for doing real work. Period. It'd be great if the open alternatives were decent, but they're not.
Too true - although I'm of the opinion that if SGI had ported their IRIX tools to Linux earlier and remained a *NIX company, they may have had a gentler fall to obsolescence and perhaps even survived as a relevant company. The abrupt switch to a commodity software platform and relying on hardware as a differentiator was horribly misguided. No one was going to pay $5,000 for an SGI machine when they could buy a $2500 Dell that ran the exact same software. By the time SGI tried to embrace Linux, it was far too late...
I think Nokia abandoning Qt is the real issue here. If they were going to maintain their own mobile development platform that ran on Windows Mobile, Symbian and MeeGo - that/could/ have been a winning strategy.
Qt has been modular since ver 4, so you don't have to include the GUI components if you don't want to. The API is clean, elegant and consistent, plus the documentation is great. I don't have anything bad to say about ACE or Boost - they're both high quality toolkits - but if I had to choose just one toolkit to use for the rest of my life, it'd be Qt, hands down.
While I completely agree with your premise - that usability is often the opposite of allowing configurable options for everything This simply isn't true - but the fact that many open source projects (particularly in the gnome camp) adhere to this is one of my biggest frustrations with desktop linux. Just type "about:config" into your firefox url bar and see how many configuration options there are... yet noone would argue that firefox suffers from usability problems. If you don't want to expose all the configuration options in the default gui - ok, fine... but having tweakable options so power users could have *exactly* the environment they wanted used to be what linux and open source was all about.
If someone wants simple "usability" with no configuration options, they can always buy a mac.
Too true - after a friend of mine bought a Blu-Ray player, I asked why he didn't just buy a PS3 (they cost the same). He returned the player and exchanged it for a PS3 (plus the DVD remote) and couldn't be happier - not only does it play Blu-Ray and DVD, he likes that it can do photo slideshows and play MP3s through a wireless connection to his PC. He doesn't own a single game.
IMO, Sony is making a big mistake by not marketing the PS3 as a home media hub.
But one note about Spielberg... he strongly resisted DVD when it came out in favor of Laserdisc - so he's not exactly above taking sides (although in that case it was based on picture quality...)
nVidia has dropped support for cards older than the GForce4. I have a GForce2 with 64MB and TV tuner that would benefit from this driver.
No, NVIDIA has not dropped support for older cards - they support a driver for legacy cards which continues to be updated for new X and kernel releases. Look for the "Legacy GPU" release: http://www.nvidia.com/object/unix.html
Cards older than GeForce4 were fixed function pipeline parts - programmable pipeline cards are a different beast, so a split in the drivers is reasonable.
Yes, some NASA money goes to big companies. But a lot more NASA money goes to universities and small bleeding-edge tech startups. NASA is a research institution on a tight budget - if something is or will be available in the free market, why pay research dollars for it? That's the whole idea behind NASA's Centennial Challenges: http://exploration.nasa.gov/centennialchallenge/cc _index.html
Just to put things in perpective: two months in Iraq costs about the same as funding NASA for a year.
I'm sorry, but there just isn't any two ways about it: Qt is an expensive toolkit compared to other commercial offerings.
True.
Qt was immature and incomplete compared to other commercial offerings
Incomplete, maybe. Immature, definitely not. Compared to other UNIX toolkits, it was and is still superb. Compared to other cross-platform toolkits, it is and never has been paralleled.
Qt managed to become a predominant toolkit only because it used KDE
You've got that backwards - KDE used Qt. Financially, KDE is useless to Trolltech - anything that KDE touches is GPL'ed, so commercial developers stay away from it.
these days, nobody "needs" Qt, since there are plenty of excellent alternatives that cost you nothing under less restrictive licenses
What other cross-platform toolkits can even come close to Qt? If you're only developing for a single platform, Qt may seem overpriced. But if you want a superb cross-platform toolkit, there really is no other option. Qt is straightforward and elegantly designed - a well designed toolkit leads to more productive development.
The entry price for Qt is hard to swallow, but if you factor in the enhanced productivity and *huge* amounts of time saved from dealing with cross-platform issues, it really is a bargain. Honest.
'feeble' is but one of the f-words that can be inserted. The FVWM FAQ does, in fact, state that the original author has "forgotten" what the original f-word was:
I work at a government research lab - home to many unix geeks and hackers. Three years ago, virtually everyone had a pc laptop w/ Linux installed. Today, out of 24 people in my group, only two are running Linux on their laptops - everyone else has a Mac. And if Apple made a laptop w/ more than one stupid mouse button built in, those last two Linux machines would disappear.
On the Mac, you can run proprietary software for all your multimedia, wireless connectivity, etc. and still have access to all the good Free stuff. It's the best of both worlds. If only they had more than one stupid mouse button.
There's something to be said about things just working.
And something to be said about letting others have the freedom to choose whether or not to use proprietary software.
And to counter your argument: what happens in two years when ATI and NVidia decide your card is too old to support, and yet it still performs very well but you NEED the features in the latest kernel and latest x.org?
Um, NVIDIA still supports the TNT on Linux, and that card was released in 1998.
Just because proprietary software vendors can be evil/irresponsible/negligent/whatever doesn't necessarily mean they will be. I think that, by now, NVIDIA has proven that they're a responsible player in the Linux arena.
I'm lucky (hah!) enogh to be using a driver from a vendor who shows a little more support for OSS, but while the software is quite stable, the actual hardware is crap (and utterly useless for games).
"Support" for the Unichrome drivers is 2D only. NVIDIA and ATI both have open source 2D drivers for X.
The article is about 3D drivers, which is a whole different ball of wax.
It's good to know that there is at least some effort for configurability. I can't seem to get the option to work, but that may be because it's not fully implemented yet.
I have a few observations: Cancel/Ok is bad, just like Ok/Cancel.
That was just an example. Of course using more descriptive words is better - the argument is about the placement of the destructive option. KDE does use more descriptive words.
Not to start a flame war, but Gnome was the one that went against the grain, so the burden should be upon them to be backwards compatible. Before they decided to change the placement of the destructive option, all of the existing major GUI guidelines for Linux (CDE/Motif, KDE/Qt, even Gnome/GTK) used the OK/Cancel ordering. I do, however, agree that KDE should have the option of using the Mac button ordering as well - especially because people use KDE apps on OS X.
Another thing that I don't quite agree with you is that people will use both GNOME and KDE apps.
That's just plain silly. Just because I happen to be running KDE, you expect me not to use Gimp or Firefox? That's retarded. I'll use the best tool for the job, as would any sane person. KDE and Gnome creating software with the exact same functionality is a complete waste of effort and counter to the whole concept of Open Source. Don't reimplement, reuse and contribute!
It just kills me that I use Firefox on Windows, reboot to Linux, and all the menus, dialogs, etc. look and behave differently. It's the same app, for Pete's sake!
Can we just have the "desktops" agree to disagree and have a configuration option for standardized dialogs and button order? It is absolutely retarded to have one app on your system have Ok/Cancel and the other app have Cancel/Ok.
Personally, I prefer the KDE style because I use Windows at work and dual boot at home. Ok/Cancel is what I'm used to, and it makes more sense to me. If Gnome users prefer the Mac way of doing things, hey - that's great. But no matter what *desktop* a Linux user is using, they are going to be using a mix KDE *apps* AND Gnome *apps*. Can we *please* just have a configuration option that switches button order, file browser dialog style, etc. based on what the *user* wants?
I've only talked to 2 people who tried BB's program. They both (former and current Netflix subscribers who tried it because of the price) said the selection was worse, the availability was worse, the turnaround was worse. Only the price was better.
All true (although I've had good turnaround). Also, the website isn't as nice and the condition of the DVDs is generally worse. However, getting two free in-store rentals per month makes it worth it for me. For $9.99 a month I get unlimited one-at-a-time through the mail (two day total turn around time) plus two impulse in-store rentals every month.
For the budget-conscious and/or more casual movie renter, Blockbuster Online is pretty good.
iFlicks (read iTunes tv shows and movies) will make Netflix ob-so-lete.
Until High Definition becomes the norm. Lots more pixels, lots more download time.
I can't really speak to BB's online service; they might have similar selection and pricing, but they also have the same disadvantages.
Blockbuster's online service is actually fairly decent - the website/selection/availability isn't quite as good as Netflix, but you get 2 free in-store rentals per month included w/ your online subscription (even if you're at the $9.99 cheapo price point). So, you get all the advantages of the online model, plus the ability to go down to the store twice a month for impulse rentals. I like the flexibility.
Yes, I've had a lot of problems w/ my Radeon cards (original Radeon and a 9200 in my laptop) - both the Open Source and proprietary drivers have lots of problems (although I've had much more trouble w/ ATI's binary only drivers)
The thing that everyone seems to be missing is that the specs for r100 and r200 Radeons *are* available, and even after 5 years of open specs, the open source community still hasn't been able to write a high performance, high reliability driver for the early Radeons. Writing a good 3D driver is a *lot* of hard work. This guy and anyone who's going to help him would be much better off putting their efforts into bullet-proofing the Radeon driver - it'll be far cheaper and far more capable than a home-brew custom FPGA based card. At best, this card is going to have the performance of a Rage128 (which has open specs) or a TNT2, if he's lucky. Why would anyone pay $200 to get the performance of a $10 card? It's silly.
I completely agree. This is a very gray area and nobody should be jumping on Linspire for this without more information.
I wrote Chromium BSU and created all of the artwork for the game. On my site, and with the distribution, I have an explicit copyright notice. However, several distributions, including Mandrake, Redhat and SUSE have reproduced my artwork on their websites for self promotion. They never asked for permission, and it was never given implicitly or explicitly. Linux distributions use artwork without permission all the time. This really shouldn't be as big a deal as people are making it.
Slashdot Mirror
No, not true - you can certainly use Optimus cards on Linux, you just have to choose between the integrated chipset or the dedicated chipset at boot time. What you don't get is the power savings from being able to dynamically switch between the low-power integrated Intel gfx and the high performance NVidia gfx. It's really not that big of a deal - the battery life on my thinkpad is just fine using the NVidia gfx 100% of the time.
+1
In this case NVIDIA provides me with a driver that works. I'm happy. I don't care what Linus has to say. NVidia, in my book, is supporting Linux on the desktop and for that I am happy.
+1
NVIDIA has been providing stable, fast, feature complete drivers for years and have supported Linux and FreeBSD just as well as they support Windows.
ATI, on the other hand, released specs years ago and the open source drivers are still unstable, slow, and incredibly buggy. The Intel drivers seem a little more stable than ATI, but they're still ridiculously slow and not feature complete.
I develop 3d software on Linux (and OSX, Windows) for a living and I test NVIDIA, ATI, and Intel gfx hardware on a regular basis. The NVIDIA closed source driver is the only linux 3d driver that is acceptable for doing real work. Period. It'd be great if the open alternatives were decent, but they're not.
Too true - although I'm of the opinion that if SGI had ported their IRIX tools to Linux earlier and remained a *NIX company, they may have had a gentler fall to obsolescence and perhaps even survived as a relevant company. The abrupt switch to a commodity software platform and relying on hardware as a differentiator was horribly misguided. No one was going to pay $5,000 for an SGI machine when they could buy a $2500 Dell that ran the exact same software. By the time SGI tried to embrace Linux, it was far too late...
I think Nokia abandoning Qt is the real issue here. If they were going to maintain their own mobile development platform that ran on Windows Mobile, Symbian and MeeGo - that /could/ have been a winning strategy.
Qt has been modular since ver 4, so you don't have to include the GUI components if you don't want to. The API is clean, elegant and consistent, plus the documentation is great. I don't have anything bad to say about ACE or Boost - they're both high quality toolkits - but if I had to choose just one toolkit to use for the rest of my life, it'd be Qt, hands down.
Disgustingly, this is also true of big pharmaceuticals.
It's a natural result of monopoly rights.
Why would a company need to market their products if they had a monopoly? Marketing budgets typically go up when competition increases.
If the last three presidents admitted to or were caught speeding, would that make it hard to justify a speed limit?
If someone wants simple "usability" with no configuration options, they can always buy a mac.
Too true - after a friend of mine bought a Blu-Ray player, I asked why he didn't just buy a PS3 (they cost the same). He returned the player and exchanged it for a PS3 (plus the DVD remote) and couldn't be happier - not only does it play Blu-Ray and DVD, he likes that it can do photo slideshows and play MP3s through a wireless connection to his PC. He doesn't own a single game.
IMO, Sony is making a big mistake by not marketing the PS3 as a home media hub.
But one note about Spielberg... he strongly resisted DVD when it came out in favor of Laserdisc - so he's not exactly above taking sides (although in that case it was based on picture quality...)
nVidia has dropped support for cards older than the GForce4. I have a GForce2 with 64MB and TV tuner that would benefit from this driver.
No, NVIDIA has not dropped support for older cards - they support a driver for legacy cards which continues to be updated for new X and kernel releases. Look for the "Legacy GPU" release:
http://www.nvidia.com/object/unix.html
Cards older than GeForce4 were fixed function pipeline parts - programmable pipeline cards are a different beast, so a split in the drivers is reasonable.
Yes, some NASA money goes to big companies. But a lot more NASA money goes to universities and small bleeding-edge tech startups. NASA is a research institution on a tight budget - if something is or will be available in the free market, why pay research dollars for it? That's the whole idea behind NASA's Centennial Challenges:c _index.html
http://exploration.nasa.gov/centennialchallenge/c
Just to put things in perpective: two months in Iraq costs about the same as funding NASA for a year.
That would be Intel.
I'm sorry, but there just isn't any two ways about it: Qt is an expensive toolkit compared to other commercial offerings.
True.
Qt was immature and incomplete compared to other commercial offerings
Incomplete, maybe. Immature, definitely not. Compared to other UNIX toolkits, it was and is still superb. Compared to other cross-platform toolkits, it is and never has been paralleled.
Qt managed to become a predominant toolkit only because it used KDE
You've got that backwards - KDE used Qt. Financially, KDE is useless to Trolltech - anything that KDE touches is GPL'ed, so commercial developers stay away from it.
these days, nobody "needs" Qt, since there are plenty of excellent alternatives that cost you nothing under less restrictive licenses
What other cross-platform toolkits can even come close to Qt? If you're only developing for a single platform, Qt may seem overpriced. But if you want a superb cross-platform toolkit, there really is no other option. Qt is straightforward and elegantly designed - a well designed toolkit leads to more productive development.
The entry price for Qt is hard to swallow, but if you factor in the enhanced productivity and *huge* amounts of time saved from dealing with cross-platform issues, it really is a bargain. Honest.
-Mark
'feeble' is but one of the f-words that can be inserted.
The FVWM FAQ does, in fact, state that the original author has "forgotten" what the original f-word was:
http://fvwm.org/documentation/faq/#1.1
I work at a government research lab - home to many unix geeks and hackers. Three years ago, virtually everyone had a pc laptop w/ Linux installed. Today, out of 24 people in my group, only two are running Linux on their laptops - everyone else has a Mac. And if Apple made a laptop w/ more than one stupid mouse button built in, those last two Linux machines would disappear.
On the Mac, you can run proprietary software for all your multimedia, wireless connectivity, etc. and still have access to all the good Free stuff. It's the best of both worlds. If only they had more than one stupid mouse button.
There's something to be said about things just working.
And something to be said about letting others have the freedom to choose whether or not to use proprietary software.
Use the "Legacy GPU" drivers. GeForce 256 is still supported.
http://www.nvidia.com/object/unix.html
And to counter your argument: what happens in two years when ATI and NVidia decide your card is too old to support, and yet it still performs very well but you NEED the features in the latest kernel and latest x.org?
Um, NVIDIA still supports the TNT on Linux, and that card was released in 1998.
Just because proprietary software vendors can be evil/irresponsible/negligent/whatever doesn't necessarily mean they will be. I think that, by now, NVIDIA has proven that they're a responsible player in the Linux arena.
I'm lucky (hah!) enogh to be using a driver from a vendor who shows a little more support for OSS, but while the software is quite stable, the actual hardware is crap (and utterly useless for games).
"Support" for the Unichrome drivers is 2D only. NVIDIA and ATI both have open source 2D drivers for X.
The article is about 3D drivers, which is a whole different ball of wax.
(see the gtk-alternative-button option)
It's good to know that there is at least some effort for configurability. I can't seem to get the option to work, but that may be because it's not fully implemented yet.
I have a few observations: Cancel/Ok is bad, just like Ok/Cancel.
That was just an example. Of course using more descriptive words is better - the argument is about the placement of the destructive option. KDE does use more descriptive words.
Not to start a flame war, but Gnome was the one that went against the grain, so the burden should be upon them to be backwards compatible. Before they decided to change the placement of the destructive option, all of the existing major GUI guidelines for Linux (CDE/Motif, KDE/Qt, even Gnome/GTK) used the OK/Cancel ordering. I do, however, agree that KDE should have the option of using the Mac button ordering as well - especially because people use KDE apps on OS X.
Another thing that I don't quite agree with you is that people will use both GNOME and KDE apps.
That's just plain silly. Just because I happen to be running KDE, you expect me not to use Gimp or Firefox? That's retarded. I'll use the best tool for the job, as would any sane person. KDE and Gnome creating software with the exact same functionality is a complete waste of effort and counter to the whole concept of Open Source. Don't reimplement, reuse and contribute!
It just kills me that I use Firefox on Windows, reboot to Linux, and all the menus, dialogs, etc. look and behave differently. It's the same app, for Pete's sake!
Can we just have the "desktops" agree to disagree and have a configuration option for standardized dialogs and button order? It is absolutely retarded to have one app on your system have Ok/Cancel and the other app have Cancel/Ok.
Personally, I prefer the KDE style because I use Windows at work and dual boot at home. Ok/Cancel is what I'm used to, and it makes more sense to me. If Gnome users prefer the Mac way of doing things, hey - that's great. But no matter what *desktop* a Linux user is using, they are going to be using a mix KDE *apps* AND Gnome *apps*. Can we *please* just have a configuration option that switches button order, file browser dialog style, etc. based on what the *user* wants?
Thanks
All true (although I've had good turnaround). Also, the website isn't as nice and the condition of the DVDs is generally worse. However, getting two free in-store rentals per month makes it worth it for me. For $9.99 a month I get unlimited one-at-a-time through the mail (two day total turn around time) plus two impulse in-store rentals every month.
For the budget-conscious and/or more casual movie renter, Blockbuster Online is pretty good.
iFlicks (read iTunes tv shows and movies) will make Netflix ob-so-lete. Until High Definition becomes the norm. Lots more pixels, lots more download time.
Blockbuster's online service is actually fairly decent - the website/selection/availability isn't quite as good as Netflix, but you get 2 free in-store rentals per month included w/ your online subscription (even if you're at the $9.99 cheapo price point). So, you get all the advantages of the online model, plus the ability to go down to the store twice a month for impulse rentals. I like the flexibility.
Yes, I've had a lot of problems w/ my Radeon cards (original Radeon and a 9200 in my laptop) - both the Open Source and proprietary drivers have lots of problems (although I've had much more trouble w/ ATI's binary only drivers)
The thing that everyone seems to be missing is that the specs for r100 and r200 Radeons *are* available, and even after 5 years of open specs, the open source community still hasn't been able to write a high performance, high reliability driver for the early Radeons. Writing a good 3D driver is a *lot* of hard work. This guy and anyone who's going to help him would be much better off putting their efforts into bullet-proofing the Radeon driver - it'll be far cheaper and far more capable than a home-brew custom FPGA based card. At best, this card is going to have the performance of a Rage128 (which has open specs) or a TNT2, if he's lucky. Why would anyone pay $200 to get the performance of a $10 card? It's silly.
-M
I completely agree. This is a very gray area and nobody should be jumping on Linspire for this without more information.
I wrote Chromium BSU and created all of the artwork for the game. On my site, and with the distribution, I have an explicit copyright notice. However, several distributions, including Mandrake, Redhat and SUSE have reproduced my artwork on their websites for self promotion. They never asked for permission, and it was never given implicitly or explicitly. Linux distributions use artwork without permission all the time. This really shouldn't be as big a deal as people are making it.