aRTee

Why? They (and most) just want to do a basic install of Linux and at least get it to boot to a standard simple working VGA Mode on a 586+ processor (as suppose to be supported). If anyone who just starts out with a newer Linux OS has to revert to a text mode Linux, then that is absolutely no enticement to get them to switch from Windows, or even DOS 3.x for that matter. And if they got to do tons of editing (especially in text mode), etc. (and many people are clueless to do that) then that is not a good point. Bad point! Ehm, my point was that without graphics mode it could run a graphical desktop via vnc that one could connect to, and in that vein, a system with no good graphics is not useless at all. For instance, if he wants to learn to use Linux with an old spare machine, setting it up in text mode and running a vncserver, then connecting with a vncviewer or java capable web browser from another machine (which may well be windows too), will give him a nice desktop inside a window of any resolution and colour depth, completely unrelated to the state of the support of his graphics card. An easy way to toy with Linux, it will appear as a desktop inside a webbrowser or vnc-client window. From what I understood, the topic starter has another machine, he just wants to put Linux on some obsolete machine. Why 'should Linux be better than that'? A 586 class pc costs nothing more than the time and effort to pick it up - they are being thrown away by the dozens. I have a P166mmx that I can't seem to get rid of - no fun to use, but does run mdk10.1 (slowly but surely - only 48MB of RAM). And it's actually the distro builders that make it that way, Linux 2.6 compiles fine for embedded processors to 64cpu multiprocessor machines. It's just some compiler flags. Up until shortly, SUSE was still compiled for 386 (not sure if it isn't anymore - I think it actually is, despite distrowatch telling otherwise - in any case, I know many packages are still compiled for SUSE for 386). And all the rest is beside the point - as I said, if you are trying to run Linux on a 386 and you're not doing it for embedded development (where the 386 is the target, of course), you're getting yourself into more problems than this forum can fix. Even if you could have mdk 10.1 for 386 - who would want to run anything on that? No 386 can fulfill the demands of the most average computer user - it will be dog slow on regular things that the common user expects, like flash, java etcetc. You have a point here. Though with proper video drivers, this should not be necessary. I'm taking a wild guess: on systems with supported graphics, vesafb should not be necessary, and on others, vesafb will be the default one that gets set during installation. Who goes from a supported graphics card to a non-supported one, where he will need the vesafb driver anyway? I think in some cases vesafb isn't bad, but really, all video drivers should be in the system, so that vesafb is not necessary. I prefer the Xorg developers to focus on that instead of putting effort into making life less miserable for those few who go from a supported graphics card to a supported one. BTW, if you change graphics adapter, you should actually get the configurator when booting when new hardware gets detected. Also a better solution than using vesafb. So on second though, I actually don't think you have such a good point here either. Leaves none of them standing as far as I can see.

See your own comments below (486 etc. - you're in more trouble than this board can fix?). If they did in fact have a 486 processor, then Mandrake 10.1 wouldn't install, period. It only comes with the 586 Kernel. Please check what I wrote and tell me the relevance of what you're telling here. I realise very well that the topic starter has no such machine. No, Linux has to be compiled for a certain system (i586, i386, x86-64 aka amd64, IA64, Power, etcetc). But drivers, once in, are generally not taken out. In this case, there seems to be a problem with a 586 class machine that has graphics going with the vesafb driver. That driver is fine and included in any decent/recent Linux + X.org/XFree distribution. Note also that the topic started had the graphical installation, so vesafb was indeed working fine. Of course they do. Just look at this board's posted problems, and many unsolved ones. That goes for all OSes, and will continue to be so. Nonsense. If you do nothing to a well running Linux system which is properly configured, it will keep on running fine. Pointing out that there are many here who don't have the pleasure to own such a system is irrelevant. The fact of the matter is that: 1) on Linux, one has to do something for things to change - and in most cases, it's the user who has done something (made some change, as root) without realising it 2) people here usually have problems with configuration and hardware - things that are not so well supported OSes should (and most do) keep running fine once properly set up. I admit that it takes the proper hardware too. But the fact that my server, which has a standard mdk 9.2 install, without anything special, just urpmi'd packages and the configuration of my choice (which indeed did involve some text file editing), has been running now for over 1 year, functioning as an internet/print/scan/music/firewall server. Other OSes are very similar in this respect. Give a user a Linux pc, but make sure they don't do anything as root. Whatever they do (within reasonable limits - not dropping the machine, trying to break the hardware by constantly (un)plugging, etc), the system will not break - they may break their own account, but some other user account will not be affected and the system will start normally. It's only on (badly configured, as they generally are..) windows machines that this doesn't work this way. Which is why I said: Linux is not like Windows. The 586 quote was an example, so there's no point in implying that I mentioned it in direct relation to the graphics issue of the topic starter. It's not a bad comment from my part, I just failed to make it clear enough that I intended it as a more general example, not specifically related to this graphics issue but to the trend of recommending older Linux versions. To state it again, more clearly: if you have a 386 you have to change distros because Mandrakelinux is not going to cut it for you. BTW reverting to an older version of Mandrakelinux will not help if you have a 386 class machine (including 486 - so 32 bit pre-pentium, to be more exact), since Mandrake has been for i586 since quite a while - according to distrowatch, since forever, but I'm not 100% sure that's correct. It might well be though... Also, really, what exactly is your point in telling me (moi) what this board is for? :P Things get better on newer equipment/hardware only. And that is mostly due to it being programmed/designed for newer equipment/hardware. And the hardware itself is so much faster and better. If Linux is to be a GUI OS then the simplest VGA mode needs to always be accessible upon every video failure. MS does it very well, Linux does not. I agree on the last point - Xorg doesn't revert to the fbdev if nothing better is available. However - see below... at the very end... I disagree with all other statements. See just below. Nonsense. Check the LKML for proof of comparisons. 2.6 kernel series is faster (and more responsive - so feels even more faster) on the very same hardware. Search the LKML for "pre emptive' 'I/O scheduler' and then some. Loads of interesting things. I don't have to bench it, others have done loads of benching already, which is why Linus accepted lots of these real innovations into the kernel. Google is your friend. On a side note, I have run 2.6 and 2.4 kernels with the very same setup (same distro, installed software and services running) and I can tell you that the benchmarks don't lie - 2.6 is faster on the very same hardware. In the same vein, KDE 3.0 is faster than KDE2.x, and KDE3.n+1 is faster than KDE3.n, again: on the very same hardware. People who use GNOME a lot have mentioned the same for GNOME 2.8 vs 2.6 vs 2.4 - I have too little experience to be able to tell. The same goes for OpenOffice.org. 2.0 will be even much faster, but boy, was 1.0 faster than pre 1 versions! Again, on the very same hardware. Contrary to that, MSWin has been slower on the same hardware going from win3.1 via 95, 98 to ME (not sure, but I think 98SE was actually faster than 98 ), and from NT via 2000 to XP (and now 2003 server).

I just have to take this in steps, there are too many issues to respond to and this board doesn't allow enough nesting/quotes to do so properly... Absolutely not, it fails on the setting after that basic resolution setting that comes after the X, which is no longer the simple VGA install resolution. And if the new setting/resolution/driver cannot install, then it is useless as a basic installable simple GUI desktop OS. And if the test fails, you go back and drop back to the X, then that is a failsafe dropback not at the requested, supported (if correct) resolution. Wrong, the initial problem indicated that when testing the configuration in the summary the blue screen of life of X.org came up, with a mousepointer that moved: This means that the X window system is up. There is something wrong in getting the rainbow image to show up. X is not likely the issue, rather some script that has to fetch this image. In any case, X is up and running at the chosen setting in the config, which means that that setting is fine as far as X is concerned. That the mousepointer is moving also indicates that all is fine up to that point. So things go wrong beyond that point, not at that point. Bad move, as a wrong setting can actually damage a video card or display. And bypassing the test then also leads to not knowing for sure if the resolution or the actual hardware even supports it. Then upon bootup failure, who's to say it is a video problem or not afterwards. Or you can just blow up something. Well, I think I just explained above why this is not going to be the matter. Blowing up graphics cards doesn't happen, and blowing up monitors, which is most likely what you are thinking of, only happens when you set too high a frequency and your monitor has no protection against that. The viewsonic vx900 is a TFT monitor, and TFT monitors are protected against too high frequencies, as are probably most CRT monitors made in this millenium. As for blowing up hardware with wrong settings, if the hardware is according to IBM PC spec (see below for your shot at the title in terms of the LG-CDROM reference), the only hardware that you can blow up with the wrong settings are non-pnp ISA cards and/or ISA bus scans (not sure what that was, maybe 2 cards that both try to put data on the bus, thereby short circuiting their outputs) and driving the monitor frequencies out of spec. Both cases are dying out - ISA is not used anymore and monitors have proper protection circuits. Graphics cards don't break, X just doesn't start. How? By blowing the video card/monitor out as a possible very bad scenario. That testing is there to do, else why bother put it in period. You may want to take chances, but most people will not. And I don't think the Mandrake (and others) wants things like that to happen either (see issues about ruining CD-ROMS too). Blowing up of up to spec hardware is not going to happen. Beside that point, the guys X was up and running, so the card and the monitor had no issues with that X setting. Ruining cd-roms that are not according to spec is like ruining a leaky boat. It seemed to work fine, but it was an accident in progress. I agree that the LG cdrom troubles were bad, but 100% the fault of LG. And they did admit as much. And yes, it did inconvenience (drive to panic levels) many Linux novices and experts alike. I had such a drive (still have it, it's out of service but it does work), I revived it with some effort, but otherwise exactly as prescribed by the LG info as put forward at the time.

Short reply: if you get the blue screen with the X mouse pointer and it moves, your graphics settings are fine. No need to use something else for that reason. There can be some issue with that testing screen, which leads to you not getting the rainbow colours. How about just not testing the graphics setup? Just make sure you finish the installation, and sort things out later? If you can get the blue-screen-of-life, X works, but if there are other problems, you may just have to sort those out on the command line - we're here to help, no worries. Side comment: I don't get it, every other time I see a comment about something not working, people start advicing to go back to older Linux versions. Linux is not MSWin, things don't just go bad. Sure, Mdk is for 586 or newer, but if you're still in the realm of 486 or older, you're in more trouble than this board can fix.... Disclaimer: this is not an attack on anyone here, just noting a strange trend - and I'm not claiming that an older version will not work, just that as a general rule, on Linux things get better, not worse (except for old hardware that they just don't support anymore, ... Linux is still made to run on 386 machines,....), this includes that Linux has gotten faster over the years: the kernel, Xorg, windowmanagers, everything. Anyway, you could run vnc on it and connect to it with your other pc, assuming you have a network. I have an old PII 350 as a headless server, works great, uptime slowing moving towards 400 days...!!! No need to throw anything out if the graphics are not playing nice. This is Linux, lots of cool things to do, and a headless server is as good as any machine to play around - just make sure you can properly connect to it once it's running. SSH and on a windows machine Putty should do fine, followed up by VNC.

Run 10.1. Kernel 2.6 is really faster than 2.4. Use icewm or something similar for window management, rox as filebrowser - or use the cli. Note that the standard icewm terminal icon starts konsole - not helpful. Use rxvt instead. KDE might run, but won't leave any memory free for running any apps.. the command 'free' will tell you how you're doing. I installed Mdk10.1 OE on a P1 166mmx with 48MB of RAM. 5 minutes to boot... 2 minutes to start OOo writer. But it did run. With 96MB of RAM you can do lots more. Oh, and on that machine I had to do a non-graphical install, the graphical one would hang/crash, with a hint about the text installer and not enough RAM..

Hmmm, well, if you went here (I link to this): http://www.neowin.net/forum/index.php?showtopic=204593 you would have read this: So this is not so strange... it's just very much in development. If you want to get this fixed, I suggest that you open a separate topic about it... To be honest, it's cool but mostly because it shows what's coming next...

To point 1 you make, the hardware point - with time, there will be enough users, and actually today there are devices of each type that work well with Linux. Scanners? Epson for instance. All in one? Epson, HP, etc. Syncing with a Palm, that sort of stuff is not hard anymore, maybe to hard for Joe B but not harder than on Windows. Apple users also have to pick their hardware carefully; somehow there it doesn't get mentioned as a negative point. I don't see that as a major blocking issue, just as a slowing down obstacle - people with incompatible hardware are just not going to get all new Linux compatible parts, even if they'd like to switch to Linux. More and more there is hardware with clear indications that it works with Linux. In some shops, the personnel will be able to inform you about Linux compatibility. This will get more and more common - but I agree it isn't yet today. Businesses are going to do what is cheapest for them. For some (Erny Ball) the switch to Linux has happened some time ago. Some will see the lower cost of maintenance, others will not. Business is about survival of the most efficient and cost effective - this doesn't mean the one that gets the job done for the lowest amount of cash, but the one who has the best ROI. In my work, Linux is scoring bigtime. Within 3 years, at least (almost) all servers will be Linux, and lots of desktops will be. I think that wine / crossover office may be ok for some businesses, but it's a stopgap measure. No superwine will ever exist unless MS makes one. They may well do this if they see Linux taking over too much marketshare - technically, they could pull it off, the Linux kernel is open source, and MS developers know exactly what MSWin should do. That would mean everyone can run on a Linux kernel but get the MS layer to run MS apps. This would mean MS would get to sell this layer, and on top of that MSOffice and such. All in all, I think that many businesses will manage to get away from MS, or at least many governments will manage, and they will never come back to Windows. At work, the average user will use whatever he is given. At home, he will use whatever he can get help with. That, and it must come preloaded. I agree with this only partially. People actually don't dislike change, if it's in the sense of a new cool thing. If it's in the sense of losing time and not getting things done as efficiently as before, it's bad. So again, if they know they can get help with this nice virus-resistive Linux machine, they (at least some) will give it a try. The part about the virus-resistivity is actually the difference they care about, to them this change is good and they damn well know it. On several occasions I've been asked advice about what to do, and for an older couple with no computer skills, I couldn't advice them to ditch Windows, even though the guy would have done it. I live too far away from them, and couldn't be around to help out. Buying an apple was also not on the program, they have quite a recent pc, as the salesman said: "fully internet ready". As if that's not the biggest lie... We need to build marketshare, preloaded systems, clear compatibility indications (even if only by the salesclerk from some list) and home user networking where they can ask each other help and exchange the latest distributions and tips. Every day more pc's get converted. Of all those who try Linux, many start dualbooting. Of all those who have been dualbooting for some time, some start booting only to Linux. Linux marketshare is growing, and considering where it's at, with all that's supposedly blocking, with things getting unblocked (ATI's new drivers, better and easier to install Nvidia drivers, Intel 'blessing' the name Centrino for use with Linux laptops, wireless working well, etcetc), Linux /OSS getting better - KDE, GNOME, OOo, FF, Xine, Mplayer, K3B, Xorg, etcetc, what's going to happen? Linux marketshare will suddenly stop growing? Does anyone have any indication that that might happen and what might be the cause of that? Because I don't see it. We are all discussing why Linux isn't growing faster than it is, when in fact it's actually growing at a very steady and nice rate. Just because it's so much easier to imagine (and dream?) that Linux suddenly has 15% marketshare and you won't have to search the right shop that has that one compatible piece of hardware (usb cam, scanner, all in one, whatever), that doesn't mean that Linux is not doing very well. The reality is that with a helping hand (mine in the example I'm about to give), Linux fulfills the needs and desires fine of Joe Blow (Jane Blow in this case: my wife,.. and no, her name is not really Jane :) ). Now if only I could be there helping all those other Janes without my wife knowing,.. ehh, strike that please.... :P A 1000 mile journey starts with a single step. Step by step we're in this Linux expansion. Linux doesn't have to take over. It's enough if it takes a more visible and notable place. This is 2005, and it will happen soon.

Yeah, but my comment was about Joe Blow. You seem to be under the impression that since you can keep your machine clean, and therefor MSWin is proven to be capable of being maintained well, my point doesn't hold. However, it does, as the average box is a malware infected machine, and if it's on a permanent internet connection, it's sending us spam as I write this. _You_ can keep your MSWin machine clean. I likely could too, but I have no inclination to learn how to do that. I don't care about games (the ones you can get for free with Mdk are more than enough for me), so I have had no issues taking distance from the platform known as Windows. In the same vein, some here are arguing from the point of Linux in the same way: I don't know much about Linux, yet I can get it running and have no big issues. The issues with MSWin in terms of malware are real. The issues in terms of Linux hardware support and limited software availability are real too. Face it, Joe Blow cannot keep his windows machine clean. He can manage, with help, to keep it running. And he can without too much problems, get his steering wheel to function. Joe Blow trying a Linux distribution without proprietary driver support cannot ask around to get his Nvidia graphics card running even if _we_ all know that that is really not so hard. The point is: people have come to accept how things are with MSWin. This is market inertia - much stronger than the corporate lock-in that's holding back corporate Linux desktop adoption. BTW did you know that early 2004 the statistics were that of all wireless LAN devices, 35% got returned to the shops... And those devices were returned because the consumer couldn't get them to function. Guess what: they functioned fine. (Note: 35% of all devices were returned to the shops because the buyer couldn't get them to work and the devices were found to work fine. This implies that more devices got returned, some were really defective.) Really, Joe Blow has no clue about computers, and until he can get help with Linux, his best bet is to stick with MSWin. Or, as has been mentioned, go for a MiniMe / MiniMac. But there again he will face the same issues: not so many users to share experiences/tips with, to ask for help and to get pirated software from. And things do work differently - why do you think did Apple have to spend so much on marketing with the switching campaign? 3) preloads solve reinstallation problems since all hardware is properly selected for Linux compatibility. If trouble arises, I agree things are bad, and point 5 does Linux in. 4) Not more than that people really should not be on MSWin95 anymore. BTW one point that people into Linux tend to miss: Linux today is miles better than Win95 and 98 at the time - save for incompatible hardware. One of the reasons that Win could break through is that MS had a good way to handle the competition into oblivion. Win had no strong competition, in the non-technical sense, so it could easily dominate. Now, Linux may be better, but it has to compete with the installed base, market inertia and lock-in. It takes more than a zero price tag to get in there. But things are looking good. Like I said, in one of the most inertia and lock-in hampered companies, I now have a Linux workstation/desktop. Contrary to my own work pc that I installed Linux on, this one is managed and supported by IT. Linux will get there, but it will not be by getting gamers to accept bad steering wheel action and Cedega messing about. It will be through those who have demands that Linux can meet and who decide to take the plunge. Plenty of those around. Things are moving, even if it's too slow to the taste of most. Slowly, but surely, is nicer than hitting a wall at too high a speed. Accelleration is coming (Brazil, Munich, Extremadura in Spain, Vienna, etc all help out in the credibility). 2005 is going to be a damn fine year for Linux, better than 2004 was - which was quite good too, actually. More preloaded machines, getting a few percent more users - compared to the whole home desktop market, it wouldn't surprise me if the total number of Linux desktop home users saw a 25% increase this year. I would estimate a 10% increase at least... More and more hardware will get open source drivers, and we will see the first Open Spec (relatively) fast 3D card in Q2. (There was another thread about that.) In short, things are looking good.

1) The MSWin we should compare to is winXP sp2, because that's what Joe Blow will be using. Don't start about W2K because that's not what comes on the pc Joe gets from the store. 2) Windows is as safe as you make it? No, Windows is as safe as the regular user makes it. So, how exactly do you prevent email viruses from running if you're using Outlook (which is what most people do) and the user has admin rights? Same for IE and activeX. And why the user (your wife, kids whatever) have admin rights is easy: there are plenty of progams (lots of games) that don't run without admin rights. You have little choice in those cases, and this is about reality. And don't tell me that you can handle that, and that it can be easily done, this is about Joe B again. Windows is as safe as the user can make it. Realistically, this is not very safe on average. The average MSWin pc has tens or hundreds of malware programs installed; the percentage of spam-zombies is a multitude of the Linux pc's out there. 3) Linux and hw is easy to solve: make it preloaded. That's the only solution, and that's the game MS is playing and has been playing all along. For that Linux needs marketing and PR. 4) upgrading Linux is not necessary 2x a year, but reinstalling Win often is _for Joe B_... Then again, if your hardware is willing and you happen to have the commercial editions, you can do a clean install and all you will have to add are the win32codecs and libdvdcss. Or, use urpmi to upgrade - I have done it from 10.0 to 10.1 and had no problems. 5) MSWin has the big advantage for Joe B that his family, friends, neighbours and colleagues all also use it. Plenty of people to ask for help. This is the one biggest missing thing for Linux. (Darkelve, this is what my project is about. With any luck we can solve this.) Forget 'WIn is so much easier'. Think as Joe B: who's going to help you, the computer unsavvy, with all this complicated computer stuff? With MSWin the answer is easy, with Linux, it's just not. For those without interest in computers, who don't want to learn Linux and figure out what hardware to buy, how to use the (different) software, who don't want to go to some webforum and as for help, or (better) use google to figure things out - they have _no_ choice - they can't use Linux. They don't want to be on their own. With MSWin they are not on their own. 6) one of the other things that come with that: plenty of people to get software from,... I see why Adobe doesn't make PhotoShop available for Linux - I doubt they would sell many copies, but it would certainly bring (more) piracy to Linux. 7) ATI has to make drivers, since their stuff just doesn't work as well as Nvidia's with Linux. The point is: those on Linux just say: if you ever want to use Linux, better buy Nvidia graphics. And those on Linux also happen to have, on average, better computer skills and knowledge -so they are the ones that are asked for advice. Here for once, there is an advantage in the Joe Blow type that was playing out badly for ATI and well for Linux: if you explain that _if_ they want to try Linux (in a not so distant future), they better get Nvidia. Joe Blow is not educated about computers, all he wants is to keep his options open. 8) If Joe can handle MSwin so well, why are there so many MSWin spamzombies filling my inbox and getting my cpu load to 100% through spamassassin? Those who can deal well with MSWindows can deal well with Linux, and if they're on MSWin it's because they want/need something that Linux doesn't offer (games or whatever). 9) Video drivers are the easiest thing if you have a commercial edition. They autodetect ATI and NVidia graphics and install the driver. Nothing difficult about it. If you get a preloaded system, the drivers will be there. I don't believe that you can get preloaded systems with no separate /home. All systems that I've seen advertised as preloaded had a separate home, and in some cases you could even order your partitioning the way you want it. And what's this stuff about dirs being in use? Why make a tarball, you have a cd burner, just burn the things you want to archive. Don't tell me burning a dir is too hard with K3B, I have friends who use MSWin mostly but go to Linux for K3B to burn their photos - Nero is not easy enough and not powerful enough (yeah, surprised me to, but this is what they tell me). However, most MSWin preloaded machines have only 1 big partition. If that should ever go haywire, they get to use some rescue cd (well, ask someone to help them with that, this is still Joe B). Or just lose all data when the obligatory format comes along. BTW Joe does in fact get his system reinstalled, but needs help from some friend to do so. And yes, on Linux the upgrade can go wrong, but 1) it is not as necessary 2) usually works better than a MSWin upgrade or SP install I agree, this should be improved. Quote: Linux is great, as long as they are happy running outdated apps and packages in 6 months. Well, as if Joe Blow cares about running outdated apps. There are more people on Win95/98/98SE/ME than there are on Linux. Probably 3 to 5 times more. Lots still using IE 5 or older, or even netscape 4.7 or some such dinosaur. Please don't mix up the habits of Joe Blow with those of people who want the latest (less than 6 months old) software. Well, that and _freedom_. I know, Joe doesn't see it, but it's like all healthy people don't really appreciate that until they lose their health. FX, when you talk about Linux upgrading issues and downloading, you're talking from a computer-savvy perspective - on my wife's pc, I last upgraded in September 2004 (10.1OE), before that it was March/April (10.0OE - because I wanted to see how well it would run and how easy it would install - easiest machine to install on that one can imagine). Before that it was 9.1 I think, somewhere 2002? Never did it bother her that she wasn't running the latest stuff. Joe B doesn't bother will all that. At all. The first part I agree with. It needs to be made easier to use. But not to be able to stack up to MSWin, it can do that nicely as it is, technically, it 'just' needs to be preloaded and people need to have friends/acquaintances that can help them out. Lastly, most computer users only care about what they can do with their pc, not how it works or how to maintain it. They don't even want to learn how to use it - they will only put in the effort if they see that the results of that can pay off, in terms of what they wanted to do. People don't want to learn how to use PhotoShop (gimp, gaim, msn) to learn just that, they want to use it and are prepared to go through the 'bad' time of learning since they realise what the payoff is. Learning to use an OS for most is not what they find useful. And they're probably right. So they will always need someone to help them out when they get stuck. They don't care if their pc is a spam zombie, until the moment that they can't upload some pics to the online digital print shop, because their bandwidth is taken up by the spam mails. They don't even care about spyware and other malware, until the moment that their pc slows to such a crawl that they can't use it like before. And that's the point where they get in touch with that handy acquaintance, who will know just the medicine. And since they don't want to change their ways, and before the mess they actually liked their system, the medicine is very rarely Linux.

Darkelve, what's with the pyramid text? ;) One thing on tcpa and drm and such: people will not be confronted with machines in the shop that don't rip cds, check before playing back videos if the viewer has paid or whatever. They will introduce it more gradually. IBM (of all corps) has been selling tcpa capable laptops for what, 3 years now? They will introduce it gradually and slowly. Then all hardware will be tcpa capable. MS has big interests that viruses and trojans stick around for a bit. At some point, it will be adviced to enable tcpa, against all trojans, viruses and other malware. This will create positive press - no more updating of virusscanners, great protection for the user, hooray. At some point, those not using tcpa will be classified with the pirates and hackers/crackers (well, the word hackers gets abused for that anyway). Then some law will pass that says all hardware must be tcpa capable, or you, as a manufacturer, will not be allowed to sell in the US. That's right, they will force vendors to make ALL relevant hardware that they sell ANYWHERE in the world tcpa capable. Then, at some point, they will enforce the use of tcpa, without it you may not even be able to go on the internet. Or at least, that's the dark plan. In any case, they won't be so stupid to directly put tcpa machines against non tcpa machines in the shops: "look, this one has no issues with viruses ever, but if you want to play any media, you do need an internet connection so it can check if you have purchased the right license. That one needs virus scanners and firewalls and updates, but you can do as you please..." It will be like this: "Look, these are essentially the same, but this one has one extra nifty feature, that allows you to forget about virus scanner updates and firewalls. You may have to switch that feature on and off, since it might conflict with some programs, so during that time, just sever your internet connection, no problem. Hey, never to worry about viruses anymore, wouldn't that be nice!" So how to counter? Get Linux desktop user marketshare in the 10+ % worldwide, or at least in the rich countries. We are working towards that. Ps darkelve - about 2 more months or so... :)

Just a quick reply: again, those who don't find their software desires and needs fulfilled on Linux are not a target for migration. If people are interested, first check the needs, and if satisfied, check the desires. If not satisfied, they may want to try dual booting - some people find out there are other things than what they wanted to do, like substitute games or so. There are loads of people out there who don't care about games, video or audio editing etc. but who are fed up or downright incapable of keeping their machine clean and secure. If they are fed up, they will usually not be too upset during the initial Linux phases, if they are incapable, they will need help one way or another. If you can help them to get started, that's fine. Gui tools for text file config: - I don't think you should have to hand edit XF86Config / Xorg.conf - the distro should do this for you, or better: a preloaded box should have you not worry about any hardware config. - /etc/fstab : use harddrake and other gui tools to get your partitions and shares in there. I agree that there are things to be improved, but there are still lots of things that are good enough for a large part of the desktop users out there that are still on MSWin but who would find all their needs _and_ desires fulfilled with Linux. Maybe not a high number in percentage (10%? 20%?) but in absolute numbers and compared to the current number of Linux desktop users a very high number. Currently I guess there are 5% Linux desktop users, so about 20x as many MSWin users. Or less. Linux will come through, if only because there are too many big corporations with a vested interest in Linux - IBM, Novell, HP, etc, but also all Streamium devices of Philips run Linux. In the embedded market, Linux is really getting big - mobile phones etc. Linux is not there in terms of marketshare, but of projects in development Linux does have a large 'market'-share. On happy penguin games - I have not had big issues to get those games with urpmi normally - but then again, I haven't tried so many... And again, if you have trouble compiling, don't compare that to the MSWin easy of point and click - the Linux equivalent for that is urpmi, the MSWin equivalent of compiling is compiling, and I have yet to meet a regular MSWin user who could and would actually do that, even with instructions on what needs to be done. ./configure && make && make install is really a walk in the park compared to whatever is necessary on MSWin after a clean install... BTW, it's aRTee, not aTree - I feel like your monkey (from your icon) is hanging on to me now.... :P [edit] of course, Darkelve's post wasn't there just now - this one is a reply to ljones

We all are. Some of us just disagree where to start.... :) But sure, you're right in that if you voice your wishes, the chances they will come true are not slimmer than if you don't voice them. I don't believe in emailing software manufacturers so much, but sure, go ahead... As for emailing hardware manufacturers, I have repeatedly communicated with them about Linux compatible devices, either to thank them (epson in my case), got a very nice reply, or to inform them about linux compatibility (Hama USB 2 cardreader), which lead to a nice discussion with their rep, who was interested in hearing my point of view. Oh, just remember, I have also emailed canon when I bought my epson scanner to tell them I just went for the competitors product, motivated by clear Linux compatibility and support (epson even has it's own GUI for scanning under Linux). They never replied, but surely they got the message. I think that would help, then they see a lost sale. The problem with game houses is that Linux users have no habit of paying for software, or at least they are in the books like that. Maybe you can do the same: buy a Linux native game of a certain class, say DoomIII and then email the software houses that make competing games and tell them you (and your friends!) chose DoomIII for the game itself but also for the availability of the Linux version. Anyway, many users already have the windows version of a certain game, and if it would be released for Linux, they would likely just pirate it, saying: I already paid my license. No extra sales made to those people... About getting out there and demand - would it be an idea to have sort of presales for some game, taking only creditcard info, and then saying: "as soon as the game is there, we can sell this many". Then get them to put in the effort to port the game, or collaborate with LGP or so. If there's money to be had, nothing is impossible. Green paper is the bottomline. I wonder how much money people would actually be able to get together. Lastly, I just made this one up: Linux isn't about the desires of the users. It's about the needs. Think about it: it would be nice to have lots of things. But anything that users/developers really need will be there. This includes semi pro audio and video editing, not necessarily the really high end stuff (yet). Anything that would be nice to have is just lower on the list. Besides, there's just not enough money going around in Linux for the home desktop yet.... in some parts of the world people cannot even watch a dvd legally on Linux (save for using the lindows dvd plugin).

Hello to the new board members. ljones, good very first post! It's nice to see _on topic_ contributions here... :P Anyway, yes, you are right: for professionals, in terms of video and audio editing, there is not enough in Linux. Some would go as far as to say that even for photography there is not enough, since there is no PhotoShop. The point is, pro users know which program they need, and based on that they choose their platform. Some will get a Mac, others a Win pc. But this doesn't mean much. At work I just got a Linux workstation. It's not even a truely fast pc (P4 3GHz, 2GB RAM), but it's faster than any of the very expensive HP servers we have, as long as the job fits into the RAM memory. If it doesn't, there are Opteron Linux servers that also run circles around the HP machines. So here's the point: the software we use (very professional I might say, all license cost totalled and divided over all designers leads to around 50K US$ - you can see that the license cost of MSWin/Office or RH is nothing around here) does _not_ run on MSWin. So? So what? Exactly my point. And for these niche users/markets, it doesn't matter much that Linux doesn't serve them. Games have a much larger impact, everyone wants to know that they are not limiting themselves too much. With MSWin you just keep your options open. On the other hand, Linux is quite ready to use for most tasks, even (simple) audio editing and video editing (you mentioned kdenlive, whatabout kino?), and dvd authoring - today still a bit hard, but I've seen some nice results from a colleague of mine, quite a few things there that you can't (or at least, not easily) do with the software for Win that he compared with (which I think he got with the camera). On urpmi: you seem not to be using it as it's meant to be used. No offence, and I may be misunderstanding you, but here's what I can tell you. 1) normally you don't have to download an RPM. If anything is remotely interesting, it will be in 'contrib' or 'plf' - if it's very interesting, it will be in 'main'. Just a matter of configuring urpmi properly (see this board or my site for more info). 2) if you download an rpm that is properly made, you should be able to do urpmi [downloaded rpm] and urpmi should solve all dependencies for you. This works fine, even if the package is not built for Mandrakelinux (suse, fc packages). But frankly, see point 1. 3) if urpmi finds a circular dependency on 2 packages you download, you tell it to install both: urpmi [package 1 with deps on package 2] [package 2 with deps on package 1] and it's solved. No big deal. I must admit you're not the first with this kind of problem, and I have not seen this, so I just can't know for sure. Maybe you did find some game/program that's really great and that was not in the repositories but there was an rpm (for which distro?) available that gave you a headache even though you did everything correctly. But from some of the remarks you made about urpmi I conclude that in any case you don't know all the ways in which to use it (yet).... So your problems may fall into the 'user error' category. Games: you should not have to compile them yourself, and getting the graphics to work fine is a matter of hardware setup - with nvidia (the only graphics hardware that works nicely with linux), it's not hard to do, and if you buy any kind of linux, this config will be done at install time. Either you spend the money or you put in the effort. Anyway, any interesting open source game should be available as RPM via the urpmi repositories. Makes it easiest to install them. You say that for windows it's easier, people can just click and install them. But for games that are in development, they can't even compile those - MSWin doesn't have a development system integrated. For games that are not in development, there should be an rpm within the distribution, and if there isn't, there should be an rpm that is general enough to be used with urpmi as in 1) above. I cannot say that I have seen the issues you mention about 3d games. Config files and the like? The only thing I can imagine is /etc/X11/Xorg.conf (or its predecessor XF86Config-4 ) Maybe I'm not trying to run the same games? If you have some examples, I can have a look. In any case, your email did get me to think about some issues around Linux: some things are not as easy as can be imagined: how to handle proprietary drivers, install certain software, etcetc. With other systems they aren't either though, think of registry issues on MSWin. But those things will be improved on within Linux. I see the big issues in preloaded Linux machines and PR, mostly...

devries, thanks a million for cleaning this thread up. Everybody, can we keep this ON TOPIC please? Steve, I agree and disagree. The points you mention are good for those who _care_ about running Linux, and only Linux, and games. _I_ don't care much about games. I have had Railroad Tycoon Gold for Linux on my shelves for ages, still haven't ever even installed it. Did install and play Majesty, my wife still plays it now and then too. And it makes for a nice section in my reviews, showing off that there is no Linux forking. On the other hand, there are people who just want to play cool games, the hot ones right from the cd presses. Those people are not going to hold off. And asking them to hold off is a waste of breath/bandwidth, IMHO. But you're right, the Catch 22 I mentioned is not really that strongly a true Catch 22 - there is a way out, namely not buying MSWin games, but spending money on Linux games. If there is a considerable market, there will be more game houses porting/making games. Simple economics. BTW read also this: http://games.slashdot.org/games/05/02/04/1...tid=206&tid=106 and the linked article: http://www.cooltechzone.com/index.php?opti...id=984&Itemid=0 I don't believe that sending email and nagging about Linux ports will really help much - I would guess they treat that kind of email the same as spam. BTW another good way to make Linux games more popular could be if you play lan/inet games of a type that exists for Linux. There will be some games to pick from, and if one of those is available for Win and Lin, and others only Win, the one that is available for Lin too gets the vote. This implies that you are part of a group that actually has playing together at a higher priority than playing a particular game. Not that I believe that this works for most people though.... In any case: those who want to stop dual-booting and those who just have Linux, should be pointed out that there are some Linux games of high quality, both FLOSS and proprietary. If they care about high quality proprietary games, they should go out and buy Linux games. That is all there is to it. If someone wants Splinter Cell, I'm not going to start on Frozen-Bubble. Get an XBox or BTW I wonder how much of a problem many of those Linux?-not-in-a-million-years gamers would have with Linux if they couldn't copy/pirate MSWin games so easily...

reb2, that is the group of MSWin users that should not waste their time trying out Linux, and you should not waste your and their time trying to get them to try it. Not that you are doing that, I'm talking in general. The point of Linux and games is simple: as long as people who might buy a Linux version of a game will still buy the windows version (and play it on MSWin, dualbooting, or on wine/winex/cedega), they still count as MSWin game buyers. No matter if they would have bought the game in the Linux version, had it been available, they still buy the MSWin version. So, for software houses, those people count right up with the MSWin users. They are not a lost sale if there is no Linux version. Now, those who _only_ buy native Linux games are the ones that game houses should port for. Now you see the Catch22. Those who buy games regularly buy them for MSWin, have MSWin games, play them in whatever way they can (win or lin+wine or whatever). Those who are really on Linux and only buy native games are rare - not true gamers. Most who are on Linux are perfectly happy with the games they can run, maybe mostly open source games, with some closed source/proprietary games. Those who can live without games must first come to Linux - gradually, there will be some that get curious about some game, and hopefully there will be more and more people that buy only native Linux games. Then there will be a potential, it will be up to the game houses to make money off of that. It's sad but true - the LInux users are just not real gamers, and the real gamers game with MSwin versions of the game. So the point is: we must first help lots and lots of others to move to Linux (not dualboot, really use just Linux). At the same time, they must be made aware that if they want to get some game, it would be best to get a native Linux game. The bad thing is, it seems it will take lots and lots more Linux users, since those on Linux are the ones not gaming (like those on Mac, maybe) so Linux may need to have 15% desktop/home user marketshare. The good thing is: there easily are that many people availble and ready (with some help) to use Linux. It's the help part and the PR part that are missing in this equation. BTW I have been looking at Linux games, open source. True, they are not in the same class as doom3 or neverwinternights or splinter cell - but that doesn't mean they are no fun. Check out: glchess glgo gl-117 globulation nethack_falconseye lincity atomix egoboo xarchon xlogical slune gcompris childsplay atanks boson bzflag battle for wesnoth open transport tycoon (well, needs some data from the cdrom) oh, and of course the best flight sim: flightgear lots more - see also http://www.happypenguin.org/ I haven't played all of the above, but those were easy to urpmi into my system. True, some are really not goodlooking, but they can still be fun to play - anyone remember archon of Commodore 64? Xarchon, there you go! I did check and like: slune (ok, political slant to it that some may dislike), gl-117, flightgear (although I never really played it, it's too realistic, but as far as flightsims go, this one is excellent with maps to cover the whole world), frozen-bubble, glgo, glchess, xlogical (also in the form of pathological - which needs pygame installed even though urpmi will not do that for you), battle of wesnoth, ..... Then there are some proprietary games available: http://www.tuxgames.com/ Linux has lots of nice games, lots of free games that are fun to play, some for just a few minutes, some for hours, and then some nice proprietary games. For the diehard gamer, this doesn't suffice, they want to play some online games with their buddies, but the incidental gamer can really find nice things, including educational stuff for the kids, fun 2 player games on 1 machine, network games, etcetc. However, there will not likely ever be a top notch native Linux game that is not available on Windows - anything good from Linux gets ported due to it being GPL/FLOSS. Making a closed source Linux game costs too much effort... The other day, after having had a look at glchess (with gnuchess as the engine), I saw a 3D chess game in the shop for 'only' 25US$ (29 CHF), and I just though: people just don't know better.... :unsure: Ah well... Ps great HHGTTG quote in your sig...

pbpersson, are you gunning for a high post count?? :P Anyway - keeping win2k or xp around if you develop software for it is the first thing to do. In that case you're not a regular home user, so why Linux at all (if not for the heck of it)..? Crossover is only somewhat the answer - it is not perfect, it costs, it keeps people locked in into MSO... Lots of computers come with OOo preinstalled now. Time to ditch .doc, .xls, .ppt, .pps, etc and move to .swx and Co. The last official figures I've seen are that OOo has 15% of the market. If anything, people should create proper MSO filters to be able to read (and if possible, write) OOo documents. On the whole Esperanto thing: your dad may have been saying that 40 years ago, it was nothing but a nice dream, nice but unrealistic. Even today, many villages have their own dialects here in Europe - Switzerland is particularly bad, in the German speaking part every valley speaks in different ways. I grew up in Nijmegen (the Netherlands), and some people there still speak the local dialect (which I can't even understand) - it's had ages to die out, but still refuses. Linux on the other hand, is picking up steam. The examples I gave are just that, examples. But in any case, today you can buy preloaded systems from plenty of places, Carrefour sells preloaded Mandrake in France, Switzerland, Luxemburg, Belgium. You can get preloaded laptops, etc. That was not possible 2 years ago (apart from some small scale trials). Lots of mission critical software now runs on Linux, for instance the programs we have at my place of work to design and test microchips with. The company I work for is a large, very slow, entity. If we're doing it, all others have done it for some years now. BTW Nvidia designs their chips on Linux machines, as does AMD. Not sure about Intel, could be the same, but could very well be something else. Linux is going places, not because some here (including me) are saying so, but because we see things happening. I don't talk about Linux in non-computing/technical environments, but now it's happened a few times already that I was at some party or so and people mentioned they were using it or trying to - in that latter case my wife is quick to push me forward to offer to help them... ah well. About viruses getting targetted to Linux - yes and no. The regular email viruses that are so very successful at the moment are not possible on Linux. Read the web, use google, I don't want to repeat that here, it's been said everywhere. So that leaves the trojans and such, programs that use loopholes in programs and services. As Linux has no messy stuff incorporated into the kernel (contrary to MSWin), just not installing unnecessary services already helps lots. Then you only have to make sure your firewall is up (or actually, if you have no services running, it may not even make a big difference) and your kernel is patched. The hole that msblaster walked through was a service in MSWin that users have no control over. No such holes in Linux, there's no need for that kind of stuff - services that run and that a software firewall cannot close off. Next to that, linux isn't linux. Every system has a different version of the kernel, different patchset, different services/daemons, etc. This makes the landscape much more heterogeneous, which makes for good health and defence against viral diseases. Even with a 100% Linux landscape. In the end, yes, there may be more virus attacks, but there's no logical reasoning that leads to the belief that there will be as many successful attacks on Linux as there currently are on MSWin. Oh, another thing: apart from the email viruses, I mentioned trojans and such - online infecting things, which also includes worms of which MSWin users have seen their fair share. Well, the thing is, the reign where worms are successful is the reign of always-on machines, those that are accessible 24/7. And guess what - many of those are on Linux. Still I don't see the news saying that half the websites were down. So apparently, Linux in terms of worms is holding its ground. And in terms of email viruses, there's not such a big risk. The n00bs that don't know that making an attachment executable can be dangerous, will also not be able to do so - creating an effective email virus on Linux will really take very special social engineering skills. Next to that, there's not so much spyware and malware. So the only big risk I see is if ftp mirrors get hacked, rpms swapped for trojan infected rpms, and people installing stuff with urpmi and then ignoring the 'incorrect signature' warnings. For trojans to be successful in that way on a large scale, lots of things have to go wrong at the same time. So I take issue with the statement that 'if Linux becomes more successful, there will also be more viruses, since then it will be a target' - the implication is that it will be in a similar state that MSWin is in today, and there are no grounds for believing that, considering the points I mentioned before: on the server side it already is a target (has been for a long time), and on the home user side, it will be a target for sure, but it's just a harder nut to crack. Note: Linux/apache/etc is not a failsafe option - lots of website defacements in that region. Lesson to learn: there are holes, and some people can get in. The point is, that it takes skills and lots of effort, instead of some silly script - and then they can still usually only get into the non-patched systems, from 2 years ago of people who think Linux doesn't need updating....

The playground is friendly enough, it's just a matter of getting them started. Start with those who are ready, and get those who are not yet ready warm for the idea of opening up. Get them started on FF, OOo. That should help against some viruses. Linux hasn't been hackers only since 2 years or more. My wife uses it without problems, she just doesn't like OOo since she was really very skilled with MSO. But that's about the only thing. Linux is there, most could use it. If only they knew. Ix has it damn straight - it's all about PR.

You're contradicting yourself I think.... This I disagree with. This is right - so really, the problem is lock-in. Mentally, but still. The good old problem of inertia. It doesn't matter if Linux apps are more modern or open. That they are not the same is the problem - for the Windows users of today. Tomorrow things are different. Look at it this way: compared to win95, Linux really has an incredible amount going for it. But still, there are likely more win95 users out there. Even if you compare Linux to winxp (as you should, since that's what people get when they buy a pc), the balance isn't in favour of win in many respects. But this doesn't mean much. The truth is, people just don't care (enough). And they are locked-in - mostly by 'habit', not so much due to file formats. Nothing but time and keeping on going with spreading Linux is going to change it. But there is time, and motivation and efforts to spread Linux. All is well that ends well. Ps pbpersson, after figuring out the problem wasn't linux, are you keeping the server on win2k or going back to linux? Just curious.

Thanks for the feedback - I see what you mean, but since the article is so old I'm not going to do anything about it... And being impartial is what I was trying for, good that you saw that too.

Aussiejohn, I copy that... interesting, I hadn't heard that story, which is odd since I am Dutch and as such used to get anything Philips spoonfed - not just through the media, but also since in the Netherlands, everyone has some family or acquaintance working for Philips.... maybe I just didn't pay attention - mind you, I was under 12 when the videostandards war was decided.... Sellis, the thing is, in China and other such countries, the licence cost isn't an issue. Just as the extended editions of LoTR cost 1$ per dvd (totalling what, 12 dollars for the full set?), MSWin is just as cheap as any linux distro: the price of the media. Of course, for governments things are (somewhat) different. In any case, the point is that soon Linux cannot be overlooked, and hardware compatibility issues will be less of a problem. Meanwhile, Linux just keeps spreading, all around the globe. And Ix has one of the biggest points, there is no good PR. Where do you get the budget if you give away your stuff? It will come.

;) Yeah, but the problem was your setup, not Linux.... I agree that in some cases Linux is not tuned enough for newbies, but then again, how often do you have this kind of issue? Meaning: that it would work erratically, that you need more info which Linux doesn't give you and Windows does, and where Windows error messages actually make sense and help you fix the problem? If you really really think about it, Windows isn't really ready for the home users desktop either. It just managed to be there. Mac OS X would be the most ready. And surprise, it's not there. Maybe minimac will change things there, I don't know. Preinstalled preconfigured Linux machines is what it takes, and get this: even without that, Linux got to where it is today. Here's a quote: The avalanche has already started. It is too late for the pebble to vote.

Short remarks: Philips didn't invent the VHS cassette, they had their own competing system called Video2000 (VCC as abbreviation). Note that if you don't like Sony, you shouldn't like anything Disney - they tried to stop Betamax completely - didn't like the idea that people could watch videos over and over whilst just paying once. Google for it, lots to read. Anyway, I won't go into all the points of view as to why Linux is not progessing faster to become the main system on a large percentage of the worlds desktops. Instead, I'll just focus on what's going on with Linux. First, it was for and by hackers. Computer nerds. Then came more users, who have a high interest in computers - they got a system where they can be in control. Lots of those don't develop. Then the businesses got interested. IBM, HP, etc. for support, others as users. Some as both developers and users. Intel started running a whole waferfab on Linux, I think in 2002. I'm in the semiconductor industry, let me tell you, whatever you run your waferfab on is the most mission critical thing you have. Here the price of Linux is irrellevant, it's all about control. In the meantime, lots of internet stuff runs on Linux. Control, sourcecode, stability, security, you name it, all of it counts. Being able to support yourself counts too. Side note: Linux runs on so many embedded things, you can't even count them anymore. But this is not about embedded, so I'll leave that.... At work, the recommended design systems we have are RH Linux on Opteron dual cpu systems. This week I should get a Linux workstation too. On the home desktop, Linux has moved from 1% in 2000 to about 5% in 2004, against inertia and lock-in. If you have something new, there is no big push to hold you back. If you are moving 'against the stream' so to say, lots of inertia, people with interests contrary to yours, lock-in, etcetc are all blocking you. Now, Extremadura in Spain, Brazil, Vienna, etcetc are all looking at and implementing Linux. Take into account China and other far east regions, and the picture becomes clear: Linux won't be going away anytime soon. IDC and Gartner have independently calculated that the market share of desktop Linux machines will be 7 or 8% in 2008. This is just preinstalled machines. You can dislike these companies or not, believe their studies or not; I think this is a relatively cautious estimate. And who is going to make this all happen? You and me. And all those others. Mark my words.

I will buy 2: one for my pc, one for the pc of my wife. And, maybe, just to support this, one for my _headless_ server. You see, the point is, if this becomes a success, hardware manufacturers see that it's good to make friends with the community. If it doesn't work out, proprietary will be the way to go. Hopefully we can avoid that. Think back at Loki, if you were around at the time - even today, people take Loki as an example to 'prove' that there's no bread in Linux gaming, porting games, etcetc. For those who don't know,: Loki was working on porting popular games to Linux, and after some time, they went out of business. That was around 2001 (or even 2000), when there were about 1% of all desktop pc users on Linux. Today, it's closer to 5%, but still the argument stands. Let's not let this happen here. About the performance, this card/design will have 2 pipelines at 200MHz, NVidia and ATI's (sub-)top offerings have 8 pipes and run at 400MHz and up. Do the math, this card will not be capable of competing with todays high end cards, but it should compare ok or even favourably with GF2mx cards (see these specs: http://www.maximum3d.com/reviews/abitgf2mx.htm). On the other hand, I for one long for rock solid drivers for my graphics cards, and still want to be able to do some sort of 3d (so 'nv' doesn't cut it) - the community will surely make that happen. Oh yeah, and it was mentioned that you should be able to play Doom III on this card.. (vga res and other 'uncool' settings for sure, but still...)

No problem. On a side note, between LiteOn and NEC, I'd advice a NEC burner since there you can update the drives firmware from within Linux and start using it right after changing the firmware, no reboot at all! Plus, the quality of the drives is good, and they're cheap.

I have the same card, and it can work fine. Do your files actually have any surround content? Did you try playing a dvd with Xine, with xine set to 5.1 (or, in my case 4.0) speaker setup? Kmix should then let you control the front and rear channels without any problem.