AMD Promises Open Source Graphics Drivers

MoxFulder writes "Henri Richard, AMD's VP of sales, has promised to deliver open-source drivers for ATI graphics cards (recently acquired by AMD) at the recent Red Hat Summit. A series of good news for proponents of open-source device drivers. In the last year, Intel, the leading provider of integrated graphics cards, has opened their drivers as well. But ATI and NVidia, the only two players in the market for high-performance discrete graphics cards, have so far released only closed-source drivers for their cards. This has created numerous compatibility, stability, and ethical problems for users of Linux and other open source OSes, and prompted projects like Nouveau to try and reverse-engineer NVidia drivers. Hopefully AMD's decision will put pressure on NVidia to release open-source drivers as well!"

Comments

I could not read the summary

[niceone]

I'm sorry, I could not read the summary. I have worked in R&D... I got as far as "VP of sales has promised" and had a panic attack.

Re:I could not read the summary

[xenocide2]

Actually, thats one of the things that a VP could probably promise without severe problems for engineering. I guess you'd probably have to filter out the curse words in comments though...

Nice

[suv4x4]

They're just trying to get them some press. Unfortunately Linux gamers are an edge case. People needing video card support on Linux above vanilla SVGA as a whole is an edge case.

Re:Nice

[Anonymous Coward]

No, some other users just want fully operational 2D graphics with dual head support. More especially for dual DVI cards where the external TMDS is not supported under X.

Re:Nice

[someone1234]

Well, if there is a good video card support on linux, linux gaming will just strengthen. It isn't a godgiven that you can play games only on Vista.

Re:Nice

[suv4x4]

Well, if there is a good video card support on linux, linux gaming will just strengthen. It isn't a godgiven that you can play games only on Vista.

The problem is, most Linux desktop users use it to develop or manage it as a server. They won't pay for a game. They will not pay for anything at all, most of the time. The only companies paying for Linux are enterprises who need support and closed software. The rest are there for the free ride.

Sophisticated open source games basically can't happen, since you need a lot more than good developers to make a good game. Check the open source games around. You'll notice lots of smart programming, poor graphics and poor or non-existing story line. Let alone high quality music and voice overs.

We've seen how much the open source community understands design and graphics with GIMP, and it's best we stop beating that dead horse.

Re:Nice

[dc29A]

They're just trying to get them some press. Unfortunately Linux gamers are an edge case. People needing video card support on Linux above vanilla SVGA as a whole is an edge case.

Having solid drivers isn't just "an edge case". Go install the default ATI or Nvidia driver on a recent linux distro then upgrade it to a non open source one from the company. It's like day and night. I noticed a huge difference between having a default driver vs company made one, silly things like dragging a console with transparent background is no longer a pain, it's smooth. The desktop feels fast and I don't even have any 3d desktop installed.

Then you got things like multiple monitor support. My Feisty install without closed source drivers just wouldn't work. It kept resetting the screen resolution after reboots, wouldn't recognize my second monitor, I couldn't even force it, it was a black screen. Once I installed the closed source driver, shazam! All my video worries are gone. Now I am happily using a 2560 x 1024 dual monitor setup with hardware acceleration.

Also you got 3d desktops like Beryl. With eye candy being a major selling point in some operating systems, 3d features will become important if desktop linux wants to get more popular. I hope all graphic card companies will develop good drivers for Linux.

Re:Nice

[gsasha]

Wrong.

3D should not be about gaming only. Right now there are 3D-based window managers, and it's not inconceivable to have more real 3D-based applications. The fact that some mainstream cards have problems with drivers does nothing to help these use cases.

Re:Nice

[mackyrae]

Linux gamers are an edge case because it's too hard to game when your graphics look like $*&#!

Re:Nice

[OmegaBlac]

The problem is, most Linux desktop users use it to develop or manage it as a server. They won't pay for a game. They will not pay for anything at all, most of the time.

I, a Linux user, am more then willing to pay for the same commercial games that are available for Windows. Matter fact I have payed already when I purchased UT, UT2k4, Quake 4, and Doom 3 which I have installed exclusively to play on Linux. I have no idea where you got the idea that most Linux users are unwilling to pay for software let alone games. Did you poll every single Linux user? Or did you form your ignorant opinion out your ass? I'm sure there is a large number of Windows users that don't pay for their software hence the existence of warez groups offering commercial software for free. And what about the large number of people unwilling to pay for Windows or games which drives many software companies to develop draconian drm & copyright protection measures?

Re:Nice

[evilviper]

Go install the default ATI or Nvidia driver on a recent linux distro then upgrade it to a non open source one from the company.

I do it all the time, and absolutely don't see any of the 2D performance differences you've listed.

Xv (video playback) is slower with nv than nvidia, but not significantly so. And there's no way to compare the same between ati and fglrx, since the later doesn't even have Xv support at all.

The open source drivers for ATI and NV of course don't have GL support, or at least, not very good support. Unless you're using GL, you shouldn't notice a significant difference.

Things like support for dual monitors and TV can be tricky, but usually doable with a proper modeline, as divined by the X11 gods...

Of course, I do agree with you that proper open source 3D/GL drivers are important. With ATI seriously depreciating Xv, it won't be surprising if they start to omit hardware 2D support and leave the software drivers to emulate/convert it. With HDTV gaining popularity, GL video output is simply much faster than Xv overlay. And, of course, it's hard to say what other clever things people might come up with when those features are easily and widely available.

Re:Nice

[lakeland]

Not any more. Compiz and Beryl are becoming the standard way of drawing onscreen in much the same way as aero and quartz. That means unless you have decent 3D you will stuff up desktop performance. Gamers might have much higher demands, but the days of 2D chips being adequate for desktop use are over.

Re:Nice

[Lumpy]

Nice to see you dont know squat about linux.

3d acceleration and the Video acceleration is used daily by EVERY linux user (short of text based server installs.

What you just said is as redicilous as saying "Vista users dont need anything but 2d Svga."

I run Wxvga all the time WITH 3d and guess what I dont play games in linux at work.

And I am not a "edge case" but a typical linux user.

Re:Nice

[Anonymous Coward]

The problem is, most Linux desktop users use it to develop or manage it as a server. They won't pay for a game. They will not pay for anything at all, most of the time. The only companies paying for Linux are enterprises who need support and closed software. The rest are there for the free ride.

FUD, FUD, FUD, FUD, FUD, FUD, wonderful FUUUUUUUD! Not even good, new FUD either. Just the same old shit.

Re:Nice

[Tack]

And there's no way to compare the same between ati and fglrx, since the later doesn't even have Xv support at all.

Are we sure about this?

[tack@caladan ~]$ /sbin/lsmod | grep fglrx
fglrx 523792 9
[tack@caladan ~]$ dmesg | grep fglrx | head -2
[fglrx] Maximum main memory to use for locked dma buffers: 927 MBytes.
[fglrx] module loaded - fglrx 8.32.5 [Dec 12 2006] on minor 0
[tack@caladan ~]$ cat /etc/X11/xorg.conf | grep fglrx
Driver "fglrx"
[tack@caladan ~]$ xvinfo | grep Adaptor
Adaptor #0: "ATI Radeon Video Overlay"
[tack@caladan tv]$ mplayer -v *avi 2>&1 | grep Xvideo
Xvideo image format: 0x32595559 (YUY2) packed
Xvideo image format: 0x59565955 (UYVY) packed
Xvideo image format: 0x32315659 (YV12) planar
Xvideo image format: 0x30323449 (I420) planar
using Xvideo port 115 for hw scaling

Re:Nice

[yahooadam]

that was one large problem i faced when trying to move to Linux

i have dual head, just getting 1 screen to work can be a pain, but 2 is a nightmare

there are even tools around trying to help you get dual head, though that only works if the drivers are working properly

Honestly, ATM i wouldn't care if ATI drivers were open source, but ffs make them work AMD - of course open source means people can mess around with them, and then we are much more likely to get working drivers for all distributions

Re:Nice

[notamisfit]

With the r500 series, ATI's already eliminated separate 2d acceleration. It's only a matter of time before nvidia does the same.

Re:Nice

[Osty]

I, a Linux user, am more then willing to pay for the same commercial games that are available for Windows. Matter fact I have payed already when I purchased UT, UT2k4, Quake 4, and Doom 3 which I have installed exclusively to play on Linux. I have no idea where you got the idea that most Linux users are unwilling to pay for software let alone games. Did you poll every single Linux user? Or did you form your ignorant opinion out your ass?

It's pretty simple, really:

• You != an average Linux user.
• Loki Software proved the lack of market for Linux games 5 years ago when they shut down in 2002.

Now granted, part of Loki's problem may have been that they were shipping ports of Windows games several months after the game originally shipped on Windows, but the fact that Linux gamers wouldn't wait a month or three to support their platform of choice was damning.

I'm sure there is a large number of Windows users that don't pay for their software hence the existence of warez groups offering commercial software for free. And what about the large number of people unwilling to pay for Windows or games which drives many software companies to develop draconian drm & copyright protection measures?

The market for Windows software is large enough, even with the pirates, to make it profitable to sell them software. The problem with Linux is that the number of "Free or bust" users is large in relation to users who would pay for software, thus making it a losing proposition to try to do anything but niche software on Linux (by that I mean it may be worth selling Oracle at $10,000 per CPU, but it's not worth selling TurboTax at $30, because you have to sell a lot fewer copies of Oracle to break even or make a profit).

Re:Nice

[TheRaven64]

The high-end is still reserved for gamers, researchers, and people doing visualisation. A modern (cheap) GPU, however, does a lot more than a framebuffer. The most obvious thing it does is compositing. Pretty much every application does some form of alpha blending (see those icons on your toolbar?), even if it's with a 1-bit alpha channel, and there's no reason this couldn't be done in hardware. At the windowing system level it's even more important. Draw every window to a texture and let the GPU handle the shadows (not just a gimmick; on OS X is't a huge visual clue as to the active window) and overlaying.

Pixel and vertex shaders are a whole new ball game. There's a lot of text on my screen. All of it drawn from truetype fonts. A truetype font is basically a series of bezier curves. Microsoft Research released a paper a few years back where each of these curves was approximated to a triangle. A vertex shader program then inspects each of the rendered triangles and corrects the error between the triangle and the bezier. This allows an entire font to be uploaded to the GPU and rendered at any resolution with very little CPU load or RAM usage (compare this with Apple's hack of just storing a table of glyphs in the video RAM, which doesn't scale very well).

Pixel shaders can be used for a lot of things. With pixel shaders you can perform a lot of convolutions in hardware, giving some nice effects. You can use a pyramid algorithm to perform a number of things, like bi-cubic filtering, blurring, etc in a fraction of a second.

Sure, you could do a lot of these on the CPU, but the GPU is going to do them a lot faster, and probably use less power (important for mobile users).

Even without needing the 3D support, it's useful to have all of the features working correctly. Power management is a big one, since the kernel needs to be able to save the state of the GPU somewhere before turning it off, and Linux uses a lot of hacks to try to avoid needing to do this.

Re:Nice

[jZnat]

Why should software development companies waste money developing games for Windows when they could get a far larger market share by making games for Wii, Xbox 360, PS3, DS, PSP, etc.? The PC gaming market is much smaller than the console games market, and Nintendo is helping widen the gap with the Wii and DS which appeal to non-gamers as well.

Re:Nice

[howlingmadhowie]

you're not only looking at linux gamers, you're looking at the thousands of linux-workstations used in the special-effects-industry. if they could come with really stable drivers as part of the kernel, that may help sway the balance of power away from the good nvidia cards currently used.
there are a number of reasons why this would be in ati's interest:

1. the cards cost up to 4000 euros a shot, so it's a lucrative market
2. they could advertise with "we make the cards they used for harry potter 14"

opening up the drivers ensures that every linux kernel within a short time will come with a version of these drivers and there will be a good chance that the cards will just work and work well.

Re:Nice

[Osty]

Why should software development companies waste money developing games for Windows when they could get a far larger market share by making games for Wii, Xbox 360, PS3, DS, PSP, etc.? The PC gaming market is much smaller than the console games market, and Nintendo is helping widen the gap with the Wii and DS which appeal to non-gamers as well.

Many developers do exactly that. Some traditionally PC-only developers have been looking towards the consoles to keep them afloat (id). Other publishers (EA, Ubi) go for the spam approach -- develop the game for as many platforms as possible, which right now means release on Xbox 360, PS3, PS2, Wii, PC, DS, and GBA. The PSP doesn't get a port for the same reason as Linux -- the user base is too small (the PSP has a much larger userbase than linux, and is probably larger than the Xbox 360 and PS3 userbase as well, but it has a reputation of being a poor seller with users more interested in hacking for homebrew than buying games; thus, it doesn't get in on the port frenzy and has to hope for a one-off tie-in at some point later).

That said, the PC does have some strengths. Specifically, the mouse and keyboard layout is great for RTS games (I'd also mention FPS, but I believe it's easier to do FPS on a console controller than it is to do RTS on a controller so FPS is not a clear-cut PC win). Also, breaking into the PC market is much easier, given that anybody can develop a game and publish it on the web. Microsoft is trying to capture that market for the Xbox 360 with XNA, but it's still going to be a little while before XNA has enough support to make it worthwhile (needs network support, proper packaging to share games, the ability to play other people's games without spending $100/year on the creator club subscription, etc). I wouldn't be surprised to see big names like EA or Ubi decide to quit porting to PC, but even that won't totally kill PC gaming. It'll probably turn the PC into the platform for the "hardcore", with only hardcore FPS, RTS, or simulation games being profitable on the PC.

Re:Nice

[MoxFulder]

Having solid drivers isn't just "an edge case". Go install the default ATI or Nvidia driver on a recent linux distro then upgrade it to a non open source one from the company. It's like day and night. I noticed a huge difference between having a default driver vs company made one, silly things like dragging a console with transparent background is no longer a pain, it's smooth. The desktop feels fast and I don't even have any 3d desktop installed.

Agreed... this is why I was excited about possibly having open-source drivers, and posted this article. My current box has onboard NVidia, and a low-end ATI discrete PCIe card... frankly, I can't wait for *one* of them to have open drivers. Although using the binary drivers improves 3D performance and a lot of strange display bugs, as you point out, it's a huge pain to keep them up-to-date with kernel upgrades since they can't be bundled with the main kernel. I don't like putting a big binary blob in my kernel, which by all reports is out-of-date with respect to a lot of other kernel subsystems, and may open up security holes.

I don't do 3D anything (word processing, programming, web browsing mainly), but baseline unaccelerated SVGA is definitely *not* acceptable: 2D graphics acceleration is necessary for a smooth and productive desktop experience. The open-source 2D acceleration is actually pretty good at this point, but of course it simply DOES NOT WORK with a lot of the latest ATI cards in particular.

The current pace of open-source driver development is positively glacial, largely because most of the people who have sufficient documentation to easily improve the drivers are under NDA. Read this incredibly frustrating blog entry from a developer who's under NDA with ATI... using only a few hundred lines of code, he has patched the open-source Radeon driver to support most of the newer ATI cards... but ATI has spun its wheels for months without allowing him to release the code.

Re:Nice

[Stewie241]

* You != an average Linux user.
        * Loki Software proved the lack of market for Linux games 5 years ago when they shut down in 2002.


I think we also have to take into account the fact that the Linux landscape has changed drastically in the last five years. How popular was the iPod in 2002?

Re:Nice

[Aladrin]

You've missed 1 key point: 5 years ago!

Linux is a MUCH better desktop OS than it was 5 years ago. Coincidentally, that was about the same time I tried to use Debian as a desktop. It stunk and I quickly dropped it. Then 2 years ago, I found reason to try it out again. Slackware was pretty good, but still iffy for a desktop.

Now I've got Kubuntu. It's amazing, and definitely a good desktop OS. The home PC I have ordered was chosen based on the idea that it would only run Linux, and Windows didn't matter. (This one is going to be my 'game' PC in the living room now.)

Loki was too early. If they tried the same thing now, they'd have a LOT better success.

As for the 'waiting' issue... Was that the only issue? Or did Loki fail to advertise that they were going to be releasing that game in a few months? Because if I didn't know about it, I'd just pick it up for Windows, assuming Linux would never get it. Maybe there were other issues as well, that don't come to mind immediately.

Loki didn't prove anything except that they didn't make it.

Re:Nice

[Breakfast Pants]

World of Warcraft is one of the most profitable games ever created. If they had taken your advice, would it have been?

Re:Nice

[Stewie241]

And perhaps we could also add a link to http://appdb.winehq.org/. Here we see, that there certainly is a good deal of effort put into running Windows games on Linux.

Re:Nice

[Anonymous Coward]

Loki Software proved the lack of market for Linux games 5 years ago when they shut down in 2002.

Actually, you are incorrect. Loki Software died mainly because of managerial incompetence and mismanagement. You can read about some of that here.

And, as others have pointed out, the Linux desktop has matured a lot in the last 5 years. Even if Loki died because of a lack of customers (which is not the case), the same would not necessarily happen today.

Re:Nice

[tot]

I do it all the time, and absolutely don't see any of the 2D performance differences you've listed.

Text anti-aliasing is one 2D area where nv is very much slower since it lacks render extensions. The speed in text rendering is clearly noticable.

Besides, Google Earth is nice with nvidia driver, for what it's worth.

Re:Nice

[cheater512]

Pull your head out of your ass please.

I do my fair share of gaming under Linux.

And I'm not here for the free ride. The free part just makes the decision to switch easier.

There are quite a few large open source games.
Ever played Bzflag? God that game is addictive.
There is also Nexuiz which I hear is pretty good.
Then you get the small really really fun games like Frozen Bubble.

Re:Nice

[cheater512]

5 years ago there wasnt a big demand for Linux gaming.
Linux has gone a long way since then.

Your average user is now getting interested and is downloading isos.
And we have very good (binary only) graphics support.
We just need the games now.

Re:Nice

[db32]

I think linux users are probably more likely to pay for a game than windows users as long as we go down that road. I mean really the linux user is 100% legal in his usage of free software and is likely to just go buy the game because he doesn't run in the warez scene to get his software. The windows user who is probably running a pirate version is probably out downloading illegal software as we speak. Ultimately any given linux user is more likely to be using 100% legal software. And that is my stereotypical rant for the day.

Re:Nice

[phantomlord]

It's pretty simple, really: * You != an average Linux user. * Loki Software proved the lack of market for Linux games 5 years ago when they shut down in 2002.

Chalk me up to another non-average Linux user. I've got a 3 foot wide bookshelf with boxes of commercial Linux games I've bought (most of Loki's offerings while they were still operational, stuff from LGP, NWN, iD's offerings, etc). I wonder how many of us it takes to make a market.

Loki had a lot more problems going on than the lack of a Linux market at the time. They tried to be too big, too quick and go for too many AAA titles at once. Between having to pay large upfront license costs to port games (often six figures or more) and royalties from every sale on top of that, they just didn't have a business plan that met their market. They would have been much better off as a porting house rather than a self publisher (much like Ryan Gordon/icculus does now).

On the other side of the scale, LGP is working on a lot of B grade games. Some of them are very good but they're very, very slow and methodical in their porting. I've beta tested games for them which took more than a year to release after I got the first beta. They need to get stuff out the door if they want to be serious. Throw in Tux Games charging $50 for the exact same box you can buy in the discount bin for $15 (ok, here's $18.82 at walmart and you might skew the numbers because people aren't buying from them so they "don't get counted" as a Linux sale. In fact, you can pick up NWN, Quake 4 and Doom 3 from Walmart for the price of one game from Tux Games with shipping.

IMO, a lot of the problem is simply the game industry not understanding the linux market properly. A market exists but you can't go at it Loki style or you're doomed to failure, not because of the market, but because the business plan doesn't add up. Software houses should look toward portability when they design a game and the cost of a single developer to handle the Linux port of it would be pretty cheap in the overall development of the game (I haven't exactly done a poll but I wouldn't be surprised if you could find Linux geek willing to work for less than the average game coder just for the privilege of being able to get paid to program a game for linux). Another part of the problem are the publishers who dictate to the game houses what they're going to release so even if they want to do a linux version, it may not be possible.

Re:Nice

[MobileTatsu-NJG]

"World of Warcraft is one of the most profitable games ever created. If they had taken your advice, would it have been?"

That's an extreme 10,000:1 example. To put this in perspective, it's sort of like saying: "Microsoft has NEVER made a good product." "What about that optical mouse you're using?"

Re:Nice

[@madeus]

I'm not sure why you seem to think that undermines the previous posters point.

As they said, the console market is far larger than the PC market, and that console titles outstrip sales of PC titles significantly (and that they sell for more). World of Warcraft and Linage don't change that. Selling a million plus copies of a game in the first year is expected of a half decent console game (games EA's Madden do that and more in their first couple of weeks - games on the PC don't sell that well with anything like that frequency (nor do they tend to retail for as much).

Added to that, is of course the risk factor.

You might imagine that releasing an MMO is a better way to make money than a console game, but that's not borne out. MMO are far risker projects, requiring many times the capital investment, much longer development cyles (years longer), a more complicated business structure and business plan and have massive monthly outgoings (rather than just paying a small amount for a tiny helpdesk team and letting the staffers go once the game is out). When the risk is higher, you of course keep less of the return (and again, there is a lower return when you have high on-going costs).

As EA has discovered, it's a lot more economical to stick to releasing incrimental upgrades of existing tiles (from Madden and Fifa to the Battlefield series) than to take risks with PC MMO's, which ultimately fail far more often, and bring in less profit as a percentage of both investment and revenue even when they are successful. EA have even shut down 'successful' PC MMO's down, because they were not successful enough by their standards - there just wasn't enough of a return on their investment, that is, they were better off spending that money in developing a new console title, because it was almost certain to give a better RoI.

The overall numbers of gamers for Lineage 2 and WoW is large, but it's probably not as profitable as many think - a third of WoW subscribers are in China, and they pay just a fraction of the amount US and EU players pay. Several million more people in the US, EU and AUS have played the likes of Halo 2 on the X-Box than have played WoW on a PC or Mac.

All the above is is why there are so few MMO's, compared to console titles, as a business MMO's are simply less profitable (because they bomb more often every n attempts, and when the do go wrong they do so more spectacularly, in that a bigger hole is left in the publisher and/or developers wallet).

Re:Nice

[Goodgerster]

Blender _does not run_ without the binary blobs on my system. Who gives a damn about games?

Re:Nice

[Anonymous Coward]

I am a Linux user to and i am also more than willing to pay for the same commercial games that are available for Windows. The Problem is, this means no extra Money for the Publisher. The most People that use Linux are able to dualboot between Windows and Linux. If a good Game only have a Windows-Version many Linux Users will buy it for Windows. This means that a Linux Version of a Game will bring very view additional Buyers. But this will hopefully change in the Future, if Windows get a got copy protection and more Linux Users have Linux-only Systems.

Re:Nice

[redcane]

Have you played Nexuiz? Sure it doesn't haven't have a story line (nor does a fps arena style game need one), but it has fun gameplay, great graphics, and good music and voice overs.

Re:Nice

[Dersaidin]

I'd be on linux instead of XP right now if it natively ran as many games as windows...

Re:Nice

[Anonymous Coward]

Vanilla SVGA is 800x600 at 4-bit colour.

I think every Linux user who operates more than just a headless server needs support for more than vanilla SVGA.

Re:Nice

[sponga]

I will go back to the PC Gaming magazine article I believe it was where John Carmack everybody's favorite guy here around here said he was seriously disappointed on the sales of Doom 3 Linux and that would most likely affect future choices on development for games on Linux. Your weak arguments on DRM had nothing to do with the poor sales but rather a community who was once obsessed with the server market and shows up late to the desktop market expecting everybody to throw their hands up in a 'Halalu'. Reminds me of when the soldiers in Iraq were supposedly supposed to be greeted with flowers and cheers of finally being freed; sometimes people prefer the oppressive regime(Iraq, MS) because it works if that analogy works at all. So go ahead and tell John Carmack that he is wrong although it is nice to take into account also that they had no major retail shelf space; but even than he said he was still disappointed. Seems most of them just pirated the Linux Doom3 looking at the torrent count.

Every time some Linux person mentions that there are plenty of games for Linux I always get a laugh because they mention the same four games as they always do; come back to me when I can play CS/CS:Source, BF1942, BF2, BF2142,Test Drive Unlimited, Halo 2, Planescape, Call of Duty, Stalker, Spiderman 3, Crysis, Half-life 2, Jade Empre, Neverwinter Nights, Enemy Territory Quake Wars and the list goes on forever as it has been when the PC boom happened. Whats wrong you cant take the criticisim; well expect that as more people take to the Linux desktop that is exactly what the community is going to get and I am afraid of them having that power just to throw up their hands and say 'were done with you people because you are critical and just do it on your own'. Money doesn't grow on trees and there are actual human being developing these games that have children/family to support.

Another thing is that a lot of the Linux community cannot take criticisim of their stuff and they cannot accept that the developers of OSS had no serious 'interest' in developing for desktop/gaming for a very long time or not enough of them. OpenGL and OpenAL is there for everyone and yet developers are choosing DirectX more and more every year leaving OpenAL/GL in the dust to people like John Carmack who are firm believers in the Open Source community.
OSS community seems to be obsessed with the Commodore style gaming and have constantly snugged their noses at 3d intensive games or the corporations that actually make the game.

Just kind of 'cold hard truth' that it is gonna take a hell of a lot of effort to get the gaming scene going especially since a lot of the legacy technology was developed in the 90's for PC gaming and there was little to no interest in the desktop or gaming scene during that time.

Re:Nice

[Gareth Williams]

The open source drivers for ATI and NV of course don't have GL support, or at least, not very good support. Unless you're using GL, you shouldn't notice a significant difference.

I recently bought an old IBM T30 laptop. It has a 16MB ATI Radeon Mobility 7500. Not exactly flash :)

I slapped Ubuntu 7.04 on it, and was very surprised to find that OpenGL was working in the default install! Turns out it's using the open source 'ati' driver. So I popped a copy of the Quake III Arena demo on there, and was delighted to find I get a liquid smooth framerate. I ran a quick timedemo - 800x600 / 24bit / bilinear filtering / 'demo001' demo - it came back at 47 FPS :) In-game my FPS counter seems to hang around 60 FPS most of the time.

Frankly, that's a lot better than I was expecting of an old 16MB ATI card. And on a "out of the box" Ubuntu install, with open source drivers. All the compiz effects run nice & smooth too. So yeah, I think we're gettin there. Oh, they're dead stable too - haven't even hosed my X server yet, nevermind my kernel. More than I can say for NVidia's garbage. Hats off to the 'ati' driver guys, you bloody legends.

Re:Nice

[dbIII]

Loki were on schedule and everything was going to plan - then the dot com crash happened and the investors pulled out early. Technically they did everything well. Financially nobody expected a game software company to break even right away and they had planned to build up for a few years before the initial investment was paid off.

As for a good desktop OS - I'm still using the same Englightenment theme I was using in 1999 and it is still more functional than XP. It's been good for a long time but some distros made bad choices. Gnome was a pain for many things five years ago, I think that explans the earlier posters bad experience with Debian and delight at the new Kubuntu (which is KDE but a recent gnome desktop would also be an improvement from earlier).

Re:Nice

[timelorde]

What about that optical mouse you're using?

[looks down...]

Logitech.

[looks over at the 'ol Windows98 box...]

Logitech.

[goes downstairs, looks at wife's laptop...]

Logitech.

'Nuff said.

Re:Nice

[VagaStorm]

I think you forget a few really important element in your equation. 1. 5 years ago, linux where not as mature on the desktop as it is to day. I could, in fact, be a pain in the arse to get working even if you knew what to do. 2. 5 years a go, ppl where not so tiers of windows as they are to day. I think the lack of games on linux, is the nr 1. reason ppl still keep windows on their computer. I believe that is you could by a version of wow that ran on all the major distributions out of the box(and spent a few dimes mareding it so ppl knew it existed), it would be enough to shift a significant amount of users from windows to linux.

Re:Nice

[iminplaya]

How popular was the iPod in 2002?

I had something similar in 1963. It was a media player that fit in my shirt pocket. Though it had no storage to speak of(none), it played streaming media quite well. And because it was not stereo, I only had to wear one earphone, leaving the other ear open to hear the horn of the car that was about to run me down. We all thought it was pretty nifty.

Re:Nice

[Anonymous Coward]

Video games have been pushing milestones for many years.
What a blanket statement. You have to be sarcastic when you
sit and watch the latest shader 2.0 game run at speeds comparible
to slide-show movies of the 1980 elementary school classroom.

Take a pentium 3 celeron 1000mhz with a radeon 9800 and run the same game on
the xbox with its celeron 700mhz and nvidia chipset. Apart from some minute tweaks
the xbox will blow the pc out of the water. wow, 64 megs of ram compared to 1,000 Megs.

Now when you figure out that games can be written to haul ass on older pc's you should get really angry.

But hey its ok. They just want you to spend you money. Buy the new Pentium, AMD. We should long for the days
of old when people sought performance and not just good enough. Console people know the hardware they build games for.
The lie is so do PC people. Those labels in games "made for nvidia" , ATI branding.. ?bought and paid for son!

off to play my wii

Re:Nice

[Anonymous Coward]

English motherfucker, do you speak it? Your post was almost impossible to read. Seriously.

Re:Nice

[Anonymous Coward]

Yes, an edge case. Just like those silly Hollywood 3D animation companies that use Linux for content creation. You know, like Weta Digital, Dreamworks SKG, Pixar, Rhythm & Hues, Motion Pixel, ILM, Sony Pictures Imageworks, Amalgamated Pixels, Computer Cafe, Flash Film Works, Hammerhead, and a host of others. Silly little films from Titanic to Lord of the Rings to Shrek to Cars... Do ya think they might need 3D graphics that are fast? Do ya? Think a little!

Re:Nice

[StormReaver]

Loki Software proved the lack of market for Linux games 5 years ago when they shut down in 2002. Not quite. Loki Software proved that a small startup gaming company for a small gaming market cannot survive when its president is draining the company coffers for personal extravagances. If you factor back in all of the large amounts of money that Scott was taking out of Loki's operations and throwing away for personal use, Loki was more than self-sustaining.

Re:Nice

[rbanffy]

I see a lot of reasons to have dual-head and proper 3D acceleration under Linux.

Besides that, the platform-specific part of any current game is rather small compared of the overall effort. Much more is devoted to creating 3D content, music and motion data and to promote the game itself.

If one can develop the Unix/Linux version in a way it's not very incompatible with MacOS (or vice-versa), you would have an instantly higher market to address. A 2% increase in sales for a successful title may very well cover the costs of doing so.

Just like living creatures, as soon as an opportunity arises, businesses will seize it. Game development for Linux will start right after there are decent 3D drivers.

I would not be surprised if the current lack of interest on part of Nvidia and ATI is because someone told someone else "It's a nice display card you have there. It would be a shame if it became incompatible with Windows. A bad reputation is a bad thing to have, but, at least, you could have all those Linux users..."

Re:Nice

[shaitand]

'Loki Software proved the lack of market for Linux games 5 years ago when they shut down in 2002.'

A desktop easy enough for your average user is a relatively new phenomenon. Loki proved there was no market 5 years ago, not that there is no market today. Linux users are no more or less likely to be willing to pay for content than anyone else. Personally I avoid commercial software like the plague but games are an exception and I have purchased several that I play on Linux. There are also no small number of users who dual boot just for games.

Linux adoption has grown by leaps and bounds since 2002. The problem with many of these companies that tested the Linux waters is that they tested them too soon.

'The market for Windows software is large enough, even with the pirates, to make it profitable to sell them software.'

True, except that you mis-phrased it. The market for windows software is large enough to be profitable BECAUSE of piracy, not in spite of it.

Re:Nice

[bfields]

...some other users just want fully operational 2D graphics with dual head support.

I hear that you should be able to get dual-head support with recent Intel graphics chips, free x.org drivers, and one of these.

Re:Nice

[shaitand]

'Loki Software proved the lack of market for Linux games 5 years ago when they shut down in 2002.'

A desktop easy enough for your average user is a relatively new phenomenon. Loki proved there was no market 5 years ago, not that there is no market today. Linux users are no more or less likely to be willing to pay for content than anyone else. Personally I avoid commercial software like the plague but games are an exception and I have purchased several that I play on Linux. There are also no small number of users who dual boot just for games.

Linux adoption has grown by leaps and bounds since 2002. The problem with many of these companies that tested the Linux waters is that they tested them too soon.

'The market for Windows software is large enough, even with the pirates, to make it profitable to sell them software.'

True, except that you mis-phrased it. The market for windows software is large enough to be profitable BECAUSE of piracy, not in spite of it.

'but it's not worth selling TurboTax at $30'

You are probably right. The market is too small to be the sole support for many applications. However, if an application is properly written to begin with the amount of additional effort required to support Linux, or even Macs is relatively small. The market doesn't have to be large enough to offset the entire development cost, it only has to be large enough to pay for the additional development. Even if a software development company broke even on the additional platforms it would be beneficial to have a presence.

Re:Nice

[jZnat]

No wireless. Less space than a nomad. Lame.

Re:Nice

[jZnat]

Several of those games you mentioned either work well in WINE or haven't even been released yet, so I don't see why you're complaining.

Re:Nice

[Stewie241]

Totally beside the point. I was only trying to emphasize how quickly markets can change, and so to illuminate the fact that 5 year old statistics on the demand for 3d drivers on Linux may not be an authoritative basis on which to make decisions today.

Re:Nice

[Anonymous Coward]

"redicilous" is not a word. Buy a dictionary or use a spellcheck program.

Re:Nice

[Cal Paterson]

I have no idea where you got the idea that most Linux users are unwilling to pay for software let alone games.

Perhaps from the way that Linux people generally ship software with no cost to the user. How many Linux users out there are currently paying for any part of the software on their machine?

Payment for software only exists in the higher end server sector, and even here, it's more like payment for service.

You've drawn a stupid conclusion. What percentage of Doom 3 sales happened because of it's x86 Linux support? Linux users are largely uninterested in any of the following;

1. Games
2. Paying
3. Paying for games

Re:Nice

[Anonymous Coward]

"What about that optical mouse you're using?"

I had one. It worked great. So great, I bought another one. Then I discovered they were on the same wireless frequency - so I'd move one mouse, and the cursor on my other machine would move and click stuff. No big deal, the first one died shortly thereafter anyway - replaced with a Logitech.

I've owned many Logitech wireless mice before, and none of them talk on the same channel - even two that were identical models. You'd think MS would have learned from Logitech's multiple channel technology, but no...

I can't tell you how much my house wreaked of "stupid Microsoft not-think" when I discovered that flaw. I won't say they've never made a "good" product, but I won't hesitate to assert that they've never made a "great" product. At the very least, their products are the very bare minimum expected quality for what you paid for it - in the worst case, it's like buying a Harbor Freight quality tool for a Snap On price (like the $4,000 DST patch for Win2k).

So yeah, that was probably a bad example; MS pretty uniformly turns out products of (at best) mediocre quality.

Re:Nice

[Lesrahpem]

Not when you consider that most 3D animation and design is done on *nix systems.

Re:Nice

[Anonymous Coward]

I too am a Linux gamer and I too have purchased my games. I think that you completely misunderstand the Linux crowd. I didn't switch to Linux because it was free. I switched in the days of Windows 98 because I found a system that didn't crash and that didn't have to be rebooted with every minimal change in the system. Now, Windows has gotten much better than what it used to be, but I still prefer Linux. Firstly, because it's in my comfort zone and secondly because I get better performance out of it.

Examples:

Games on Linux typically run 10-20 FPS faster than they do on the same machine using Windows (I base that on games that I've installed on my personal dual booting machines).

Speed: Try to unzip/decompress a 61.5 MB in Windows. I tried several machines including a brand new machine running Vista. Typical results for windows were 10-16 minutes. Linux.... 8 SECONDS. Not minutes, seconds.

3D: 3D desktop that makes Linux look like Windows 2026. Enough said.

So, you're wrong when you state that people use Linux because they're too cheap to buy software. Sure, free is better, but when there's no free alternative, or a better commercial product exists, I'm more than willing to use my money to make a purchase. So, let's not start overgeneralizing the Linux population now. Saying that everyone that uses Linux won't buy software is like claiming that every Windows user is a pirate. It's far from true.

Re:Nice

[Draek]

I will go back to the PC Gaming magazine article I believe it was where John Carmack everybody's favorite guy here around here said he was seriously disappointed on the sales of Doom 3 Linux and that would most likely affect future choices on development for games on Linux.

please do, because I've never heard of such thing as "Doom 3 Linux" being for sale, and ID Software's own Doom 3 Linux FAQ states that you need a retail copy of Doom 3 for Windows in order to install the Linux version, which is how I installed and played it btw, so either your memory is failing you, or you confused DooM3 for Descent 3 or something (the last game I recall had different retail versions for both Windows and Linux, as I found out the hard way), but Descent wasn't made by John Carmack so I'd bet on your memory failing.

ohh and btw, both Neverwinter Nights and Enemy Territory run flawlessly on Linux without Wine in case you're interested.

Re:Nice

[EvilSporkMan]

For the record, the Xbox's operating system isn't nearly as full-featured (in an operating system kernel sense) as that on your PC and it probably also doesn't support multiprogramming, which means things like "timer interrupts" and "context switches" aren't there to slow the game down, and it doesn't have to make pesky "system calls" either. Operating systems impose a lot of overhead. Google xbox "operating system" for more and see also http://www.openxdk.org/xbox.html

Re:Nice

[suv4x4]

I think we also have to take into account the fact that the Linux landscape has changed drastically in the last five years. How popular was the iPod in 2002?

Apparently it has, since a company saying they may release proper drivers for Linux is some kind of very hot news.

Oh yea, things have changed, baby.

Re:Nice

[suv4x4]

Actually, you are incorrect. Loki Software died mainly because of managerial incompetence and mismanagement.

Did part of that include marketing games for Linux?

You know, since I'm not guite the gamer, I can't bring game company examples, but I've lots of experience with development tools.

What happened to Kylix, which Borland developed for Linux (Delphi for Linux). They had almost no sales and abandoned it a couple of years later.

Macromedia was actively researching Linux version of their entire Studio MX, and as an initial version rolled out a modified version which uses Wine to provide a "test port" to research the market for Linux Studio MX. They abandoned efforts in about year, due to no interest at all.

And now we have Dell about to sell Ubuntu machines.

Thing is, Linux users are VERY vocal, VERY demanding, and promise VERY big things (offer us the world and we'll buy it!). But when it comes to action, most of those users don't do anything "hey I just thought about the other folks who might want that, I'm just fine as it is", or stick to "free" (or cracks on wine).

It's the same story every time. Apparently we're through it once more.

Re:Nice

[suv4x4]

Chalk me up to another non-average Linux user. I've got a 3 foot wide bookshelf with boxes of commercial Linux games I've bought (most of Loki's offerings while they were still operational, stuff from LGP, NWN, iD's offerings, etc). I wonder how many of us it takes to make a market.

Oh, well about a hundred of you will be just great!

Re:Nice

[Hal_Porter]

Vanilla SVGA is 800x600 at 4-bit colour.

I think every Linux user who operates more than just a headless server needs support for more than vanilla SVGA.

Actually, you could get higher resolution modes by setting them up with the VESA Bios at boot time or when you need to change mode. XP can do this with it's vanilla SVGA driver. The kernel supports calls to the video Bios, at least on x86.

But the thing that kills you is that then you can't use any hardware acceleration, and unaccelerated BitBlts are really slow. You can see this if you disable the custom driver for your graphics card and get the vanilla SVGA one as a default. But I know of at least one highly customised Windows 2000 system which has an extra 256 grayscale display driven by an FPGA. The driver uses hardware to do BitBlt (easy cases), line drawing (ditto) and has a hardware cursor. That's pretty much enough to get a decent 2D desktop performance.

Re:Nice

[Brandybuck]

I need decent dual head support for my laptop. Otherwise my external monitor is useless, and I can't use the laptop to make presentations. Standard VESA mode works fine for what I do, but I'm severly limited by the lack of multiple display support.

I'm also concerned that near future Qt/KDE and GTK/GNOME are going to be requiring full XRender and OpenGL support for full functionality.

RE:Nice

[Anonymous Coward]

"Thing is, Linux users are VERY vocal, VERY demanding, and promise VERY big things (offer us the world and we'll buy it!). But when it comes to action, most of those users don't do anything "hey I just thought about the other folks who might want that, I'm just fine as it is", or stick to "free" (or cracks on wine)."

Eerily the movie and music industries get the same song and dance. There must be a connection.

Re:Nice

[azenpunk]

don't forget that each separate console is a whole new platform, it's not PC games vs. console games. It's pc games, vs. wii games, vs. playstation 2, vs playstaion 3 (heh), vs. xbox 30, vs... you get the idea. i'd be willing to bet the pc game market stacks up fairly well agains any specific console, which is the fair comparison.

Re:Nice

[supergnom]

I use the closed ATI drivers for my laptop (using an X1600 card), and one of the real bonuses (except the 3D-beryl-accelereated wonderness), is the fact that I can clock it down when I want. So when unplugging the power chord, it is clocked from 470mhz to about 100. Still gives fluent video, but gives me about an hour extra battery time.

Re:Nice

[Max Littlemore]

People needing video card support on Linux above vanilla SVGA as a whole is an edge case.

Hello, 2000 called and said they'd like their "Linux is just for hard core geeks who do everything in an xterm" cliche back.

Re:Nice

[masikh]

Don't shout, it only makes you being frowned upon!

Re:Nice

[mrsteveman1]

Im not sure the wine people should be wasting time developing direct3d functionality, it has no use in Linux but for games, any other 3D functionality would use opengl, and lets face it games are not important.

When Wine is capable of running most of the normal windows apps, allowing for additional libraries needed etc, THEN you can spend your time screwing with games, but until then work on things that matter to the majority of Linux users, that is NOT games.

Re:Nice

[Anonymous Coward]

What happened to Kylix, which Borland developed for Linux (Delphi for Linux). They had almost no sales and abandoned it a couple of years later. What happened? It was crap. Crap won't sell very well on any platform.

Macromedia was actively researching Linux version of their entire Studio MX, and as an initial version rolled out a modified version which uses Wine to provide a "test port" to research the market for Linux Studio MX. They abandoned efforts in about year, due to no interest at all. Crap. Maybe they should've done the real port instead of libwine crap. Did I mention nobody is interested in crap?

You're right in that Linux users are demanding. They won't accept things that don't work. Is that supposed to be a bad thing?

Re:Nice

[FooBarWidget]

"What happened to Kylix, which Borland developed for Linux (Delphi for Linux). They had almost no sales and abandoned it a couple of years later."

You've got to be kidding. Kylix sucked. It was slow, and the binaries it generated somehow didn't work with the system's native Qt library (it gets into an infinite loop at startup). I was a huge Delphi fan but I looked at Kylix and never used it again. If Kylix was *good* then things would have been different.

Re:Nice

[NickFortune]

Unfortunately Linux gamers are an edge case. People needing video card support on Linux above vanilla SVGA as a whole is an edge case.

Then again, as I understand it, gamers aren't the target audience. The reason that NVidia wrote Linux drivers in the first place was that Hollywood CGI shops like Industrial Light and Magic had a big investment in *nix apps. NVidia more or less inherited that business from Silicon Graphics, but they weren't going to keep it long unless they came up with drivers that worked on Linux.

Now (again, as I understand it) NVidia gets enough of an income stream from that business to make it worthwhile keeping a Linux team on the payroll. ATI came late to the party, and their Linux driver were more an attempt to break into the market, but they never generated enough income to support a full time dev team like at NVidia. WHich is one reason why AIT Linux drivers were often late in being released and buggy in comparison to NVidia's offering.

That's also why it makes sense for ATI to open their driver source code. They can't maintain the code base themselves, at least not well enough to keep up with NVidia and kernel developments both. On the other hand, the penny seems to have finally dropped that maybe they don't have to.

At worst, this gives them something comparable to NVidia. The open factor is probably going to make ATI a lot more attractive to the Linux geeks at the various render farms. And, you know, they might even see some income from Linux desktops as well. Certainly, if they pull this off, I'll have bought my last NVidia card.

Re:Nice

[Jesus_666]

I have a Mighty Mouse, so no microsoft there either, but at one point I had a Microsoft AP/router that almost didn't suck. Okay, I bricked it when I tried to reflash the firmware with something better (which required me to build my own serialJTAG cable) and it only worked well when you used Microsoft's weird config tool (which didn't work that well either)... But in theory it was worth its money. When I bought it on eBay. Used.

Re:Nice

[Anonymous Coward]

> I do it all the time, and absolutely don't see any of the 2D performance differences you've listed.

The first thing I noticed when I tried using a recent nVidia card with Fedora Core 6's default nv driver was an extraneous "dot" just above the top of my mouse cursor. Next, I noticed that the vertical line text cursor (used in applications like Firefox of OOo) was also buggy. Unusable, in fact. As I typed, a "trail" of former cursors was left between every character I typed. If I needed to move back through the text to edit something, I had to wait a few seconds for the "real" cursor to start blinking before I knew how far the cursor had actually moved.

Switching to the closed-source nVidia driver fixed both issues.

I would heartily applaud open-sourcing the official video drivers, and not only for any 3D performance benefits. I'm sure that drawing a text cursor is a purely 2D operation, and nv even gets that wrong. It'd be nice if I didn't have to worry about nVidia deciding to stop supporting my card in the future (it's inevitable that it'll eventually happen), leaving me stuck with a binary-only driver that won't work with the new kernel that Fedora will inevitably force on me if I want to retain support for all the rest of the software on my system.

Re:Nice

[Jesus_666]

Let's take a look at Aspyr. They do pretty much exactly what Loki did, just for Mac OS. The Mac gaming market isn't that big, but Aspyr has been doing business just fine (and doing AAA titles no less). Granted, the Linux market is more diverse and more difficult to serve, but still there's no reason why it shouldn't work.

Maybe someone will get capital to start a second Loki.

Re:Nice

[FictionPimp]

If there is a linux version of the game, I buy it at release day. If there is only a windows verison, I wait until it hits bargin bin. This way cedega has a better chance of supporting it.

Re:Nice

[iago-vL]

I used to have a Microsoft keyboard/mouse at work. I hated both of them. When a co-worker went on vacation, I stole his Dell gear and replaced it with my Microsoft stuff. He hasn't come back yet, I wonder what he's going to say...

Re:Nice

[Chandon Seldon]

I do it all the time, and absolutely don't see any of the 2D performance differences you've listed.

This seems to strongly depend on exactly which card/driver combination you're using. Some of them work great with the Free drivers, and some of them work like crap. When they work fine, I mostly don't notice. The Radeon X1300 on one of my machines though - the Free drivers don't even do basic 2D acceleration, surfing the web is ass-slow because it can't redraw fast enough to scroll.

Re:Nice

[Chandon Seldon]

I've also purchased, with actual money, every one of the Linux game releases that I've been interested in over the past 8 years. That's only 14 major commercial game releases, but that's constrained by my selection rather than by my budget.

This year, I'm waiting impatiently for two major Linux game releases: ET: Quake Wars and UT3.

Re:Nice

[yourlord]

Well, the goal is to make an environment that will run windows software, and DirectX is a part of windows last I checked, so they should absolutely be working to address it.

On top of that, the ONLY thing that keeps the windows ball and chain attached to me is games. I do everything else under Linux, but to play most of my favorite games I have to reboot into that travesty of an OS called windows.

I would prefer native games, but I'm a realist. I've been waiting 10 years for that to materialize, and it hasn't and it's likely going to be 10 more at least, if ever, before linux gains any real traction. I bought descent 3 from loki back in the day.. I wish the option was still there.

Re:Nice

[Wolfger]

The problem is, most Linux desktop users use it to develop or manage it as a server. They won't pay for a game. They will not pay for anything at all, most of the time.

You, sir, are living in a cave. Not only to I pay for games, I also pay an additional $5/month for Cedega so that I can actually *play* those games. Linux gamers are more than willing to pony up the cash if game developers will pony up the code.

Re:Nice

[Pecisk]

And, in fact, Loki Software proved something totally different - that bad CEO can bring to halt almost every company, including Microsoft. Yes, it wasn't because of bad sales why Loki went down, but because of some jerk-thief in post of CEO who took all money for himself. Sadly, myths about Loki demise are spread again and again, even from Linux users who don't dare to check facts.

Also I completely agree with you - situation has changed anyway, Linux user marketshare has grown significantly. I think Loki would rule today with it's expertise in games' porting.

Re:Nice

[spiki]

I noticed a huge difference between having a default driver vs company made one, silly things like dragging a console with transparent background is no longer a pain, it's smooth. WTF? Draging console with transparent background IS "edge case". Normal scenario of using console is in full screen. If you need something cool and fancy try http://yakuake.uv.ro/

The desktop feels fast and I don't even have any 3d desktop installed. Well, that's why it's fast :)

Re:Nice

[cbreaker]

There's more you can do with 3D acceleration besides games, just FYI.

Re:Nice

[metamatic]

Go install the default ATI or Nvidia driver on a recent linux distro then upgrade it to a non open source one from the company. It's like day and night.

It certainly is. When I install the proprietary ATI drivers on my FireGL T2, textures no longer load properly in OpenGL applications and eventually the machine locks up.

Re:Nice

[triso]

English motherfucker, do you speak it? Your post was almost impossible to read. Seriously. Hey! You sound like you need a foot massage.

Re:Nice

[Anonymous Coward]

Sure but if i do not want onboard graphic card neither intel chipset your solution doesn't fit.
You win if you bet that i have AMD processors and that i dislike onboard GPU :)

Re:Nice

[evilviper]

I'd never be able to get anything accomplished if I had to write out every single exception to anything I say in passing...

Re:Nice

[Tack]

By "write out every single exception" I suppose you mean "verify that my claims are true"?

It's not like I was nitpicking. There is no exception here. You said that fglrx doesn't support Xv and that's simply not true.

Re:Nice

[Anonymous Coward]

rite on, openxdk rocks.

Re:Nice

[evilviper]

By "write out every single exception" I suppose you mean "verify that my claims are true"?

Quite the opposite. And may I remind you that those who live in glass houses should not throw stones.

On many cards, flgrx doesn't provide Xv at all. Where it does, it's almost always broken in very serious ways.

I'm not interested in spending a lot of time citing sources to prove something to an arrogant and ignorant fool on /. but if you'd like to educate yourself, a good place to start would be http://search.gmane.org/?query=fglrx&author=&group =gmane.comp.video.mplayer.user&sort=relevance&DEFA ULTOP=and&%5B=2&xP=fglrx&xFILTERS=Gcomp.video.mpla yer.user---A

Re:Nice

[Tack]

This arrogant and ignorant fool has spent a great deal of time hacking on various media related projects such as xine, mplayer, and freevo, and I follow the relevant issues on Linux carefully. Clearly this doesn't make me infallible, but it does make me a bit more qualified than your average person.

I believe the real reason for your hostile reaction is simply that I called you on your bullshit and you don't like it. I note that you've subtly adjusted your claim from "fglrx does not support Xv" to "fglrx does not provide Xv on some cards" or "Xv provided by fglrx is buggy." Neither is relevant. The context of the original claim is that one is unable to compare Xv performance between the OSS ati drivers and fglrx because fglrx doesn't support Xv. That's just demonstrably incorrect. Of course there are bugs, but the bugs I've encountered (in both nvidia and fglrx drivers) on the hardware I've used don't preclude a meaningful comparison with their OSS counterparts.

But hey, it's surely much easier to write me off as arrogant and ignorant and point me to some useless gmane search about fglrx bugs (shall we go through the same exercise with the nv, nvidia, ati, and radeon drivers?).

Re:Nice

[evilviper]

This arrogant and ignorant fool has spent a great deal of time hacking on various media related projects such as xine, mplayer, and freevo,

Yeah, I do remember you providing a few patches for some basic stuff. Nothing involving video output, of course, so I still seriously doubt your qualifications.

I note that you've subtly adjusted your claim from "fglrx does not support Xv" to "fglrx does not provide Xv on some cards" or "Xv provided by fglrx is buggy."

I haven't changed a thing. As I said, I simply didn't "write out every single exception".

And you're completely distorting what I actually said, even though you're putting your own words in quotes, as if I said it... Xv isn't supported on MOST cards. Where it is, it's almost always VERY buggy, to the point that it doesn't even provide a usable picture. There are some cards that it does work on, but those are rare exceptions. Of course, you're sure I'm wrong, after all, as you said "There is no exception here."

The context of the original claim is that one is unable to compare Xv performance between the OSS ati drivers and fglrx because fglrx doesn't support Xv.

The few cards that fglrx properly provides Xv is not a good statistical sampling, so it wouldn't be a fair comparison by any stretch of the imagination. Of course it would be irrelevant anyhow, as very few people can make use of fglrx's Xv to begin with.

But hey, it's surely much easier to write me off as arrogant and ignorant

I wasn't exactly just writing you off. It's quite clear you are both. I fully explained, the matter, and yet you continue to ignore and deny the facts.

and point me to some useless gmane search about fglrx bugs

If you checked more than one or two comments, you'd see they have nothing to do with "bugs", and everything to do with long-standing limitations of fglrx.

Re:Nice

[Tack]

I haven't changed a thing. As I said, I simply didn't "write out every single exception".

My objection was simply over this claim:

And there's no way to compare [performance] between ati and fglrx, since the later doesn't even have Xv support at all.

You weasel your way out of this by saying you didn't feel like enumerating the few exceptions. What you fail to realize is that you closed the door on this by saying "at all." You made an absolute claim and this is what prompted my reply, because I knew it to be wrong.

And for this I am an ignorant fool.

Re:Nice

[JonLatane]

Whoa! Did you just name something created by Microsoft worthy of a patent? Because even though using the video card to render vector fonts is obvious, that method sure seems novel to me.

I'm scared.

Re:Nice

[TheRaven64]

Microsoft does a lot of interesting and novel research. I do, however, always get the feeling that the people doing the research and the people building the products are not allowed to talk to each other. Pick any well-respected conference, and there will likely be a paper or two from MS Research containing something that makes you think 'hey, that would go really well in Windows/Word/etc.' Then come back in five years, and see if they've implemented it, and they almost certainly won't have done.

Re:Nice

[jZnat]

Do you not get the CmdrTaco (founder of /.) reference? That was his initial opinion on the iPod back when it was first announced. My, how things change. ...which is exactly what you were saying, and I agree. ;)

Re:Nice

[ClamIAm]

This argument is bullshit because it is unverifiable. It is unverifiable because many PC game sales are never figured into sales numbers. What I'm referring to here, of course, are things like shareware and other distribution methods that bypass Wal-Mart/Gamestop/etc. Indie game developers (think PopCap) are selling tons of games for tons of money, yet these numbers are not accounted for. So even your oversimplified "consoles are bigger" argument is BS.

Further, your comparison itself is flawed. You lump together all consoles into one category, which is very logically flawed. When a game is developed for a "console", this means a specific console. For a game developer to take advantage of your mythical "bigger market" that consoles offer, the game would have to be ported to all consoles, which would cost much more than developing for Windows.

I'll believe it when I see it

[The One KEA]

$SUBJECT. If AMD really means it, it bodes well for the future - I always hoped that their openness with the Linux community over the x86-64 porting effort wasn't a one-off.

The big question though is whether or not they will try for mainline inclusion, or if they will go with an out-of-tree effort.

Re:I'll believe it when I see it

[Anonymous Coward]

The big question though is whether or not they will try for mainline inclusion, or if they will go with an out-of-tree effort.

Why do you say that? As far as I'm concerned, that decision is of almost no consequence as long as there are open-source drivers that distros can package.

Ethics? Still, nice to hear.

[The Orange Mage]

This has created numerous [...] ethical problems for users of Linux and other open source OSes,...

Damnit, Jim, I'm a computer user, not a philosopher! But honestly, I think most of the people COMING to Linux in the Desktop world could care less about these "ethical" issues. Once again, it's just another thing that some of the Linux community puts above having things Just Work(tm). However, since some of these closed-source drivers aren't working for some, it's nice that AMD wants to open theirs so that eventually they can be modified until they work. A win for everybody, actually.

Re:Ethics? Still, nice to hear.

[Just Some Guy]

I think most of the people COMING to Linux in the Desktop world could care less about these "ethical" issues. Once again, it's just another thing that some of the Linux community puts above having things Just Work(tm).

Under Linux, most things Just Work(tm) because people with those ethical issues took the time to do something about it. You can't possibly claim that GNU or Linux exist in an amoral vacuum.

Re:Ethics? Still, nice to hear.

[TheRaven64]

Free Software has never been about ethics, it's always been about making stuff work. One of the pre-requisites to making stuff work, however, was the availability of the source code and the rights to fix or adapt it, and redistribute the changes (so not everyone needs to do the same fix themselves). BSD started with a bunch of guys trying to fix UNIX. GNU started with RMS trying to fix a printer driver. Linux started with a guy trying to fix Minix.

The last one is the most interesting, since fixing Minix ended up meaning completely re-writing it because (at the time) the license didn't allow redistribution of modified versions (only patch sets, and those were growing unwieldy).

To an outsider, it might seem that ethics or ideology were the motivating factors, but in reality it's just a desire for things to work. The problem with binary-only drivers is that they might kind-of work now, but at some point they might not and then there will be nothing we can do about it.

Re:Ethics? Still, nice to hear.

[Kjella]

Damnit, Jim, I'm a computer user, not a philosopher! But honestly, I think most of the people COMING to Linux in the Desktop world could care less about these "ethical" issues.

Ethical I doubt. But if you're moving to Linux, which is pretty much full of open source software and almost no commercial software, then you gotta have some belief that this model is working better and/or cheaper than the closed source model. And even if you're not the one doing the fixing, you have the belief that the availablity of source so someone else can fix it is important. I wouldn't touch a Linux kernel graphics driver with a ten foot pole, yet I certainly care about that. I'm fairly pragmatic about it, if there's no open source equivalent that fits my needs I'll go with the closed source. But I certainly care and have a preferance.

Re:Ethics? Still, nice to hear.

[LWATCDR]

Gee people are dieing from lack of food, medical care, shelter, and clean water. I am glad that this "ethical" dilemma if now over.
I am all for AMD/ATI open sourcing their drivers. If they do and they are stable I will even replace my Nvidia board with and ATI card. I need to one soon so I can run FSX. Yes I boot Windows and Linux on my PC.
Ethical?
Good freaking grief.

Re:Ethics? Still, nice to hear.

[Anonymous Coward]

I can't speak for all of "Free Software" (however you define that), but RMS is definitely not about "making stuff work"; read anything he's ever written and it's obvious that it's ONLY about ethics. For him, making stuff work is secondary.

Re:Ethics? Still, nice to hear.

[replicant108]

Free Software has never been about ethics

On the contrary, Free Software has always been about ethics.

From the GNU Manifesto:

"Why I Must Write GNU: I consider that the golden rule requires that if I like a program I must share it with other people who like it. Software sellers want to divide the users and conquer them, making each user agree not to share with others. I refuse to break solidarity with other users in this way.I consider that the golden rule requires that if I like a program I must share it with other people who like it."

http://www.gnu.org/gnu/manifesto.html

Re:Ethics? Still, nice to hear.

[Anonymous Coward]

"The problem with binary-only drivers is that they might kind-of work now, but at some point they might not and then there will be nothing we can do about it."

If anyone believes this statement is some sort of hypothetical scenario that is unlikely to occur, they're wrong. It happened to me a few weeks ago.

I recently installed a copy of European Air War (an old favourite) only to have it complain because recent nVidia drivers no longer support 8-bit textures. Bugger. So I installed a copy of Crimson Skies for some arcade tomfoolery. Well, that loaded, but I could no longer read most of the menu text, since they are rendered from textures that do not have base-2 dimensions which is no longer supported. Arrgh.

Won't somebody please think of the classics.

Re:Ethics? Still, nice to hear.

[grcumb]

Damnit, Jim, I'm a computer user, not a philosopher! But honestly, I think most of the people COMING to Linux in the Desktop world could care less about these "ethical" issues. Once again, it's just another thing that some of the Linux community puts above having things Just Work(tm).

Look, the majority of people visiting their doctor don't spend much time on the Hippocratic Oath, but if you don't think 'First, do no harm' doesn't have an impact on how medicine gets practiced, you've got another thing coming.

The Four Freedoms described by Richard Stallman aren't just nice to have, they define the nature of Free Software. They aren't just abstract constructs to be bandied about at the pub, these are the mechanisms by which software can be made accessible enough for it to exist in a vendor-neutral, independant manner.

The individual elements of the Hippocratic Oath are under constant scrutiny, and are constantly being challenged, and the process of doing so keeps medical practice healthy, by and large. It's not a debate that many patients enter into, but you can be darn sure it affects their lives directly. Precisely the same thing is true in terms of Free Software: Users may not care much about the arcana of the GPL and such, but adherence to its principles is integral to the entire FOSS system.

So please, stop putting quotation marks around the word ethical. You may not put much value in it, but those who think about these things understand that ethics are very practical considerations.

Re:Ethics? Still, nice to hear.

[16K Ram Pack]

That's just one viewpoint.

I worked for a long time on mainframes, and never heard of Richard Stallman or the GNU Manifesto during that time. Yet, I saw first hand how much easier it is to fix something if you've got the source code than not.

Richard may have an ethical or political view on this, but most businesses that use Free Software, do so because it's the practical solution. Because they think that the savings will outweigh the proprietary alternative.

That's why I like to use it (but I'm also nice enough to contribute my changes back!).

Re:Ethics? Still, nice to hear.

[stinerman]

No. Open Source is about making stuff work. Free Software is about ethics and freedom.

Re:Ethics? Still, nice to hear.

[Anonymous Coward]

For him, making stuff work is secondary.

The story is well known and a little research on your part would reveal it. From wikipedia:

In 1980, Stallman and some other hackers at the AI Lab were not given the software's source code for the Xerox 9700 laser printer (code-named Dover), the industry's first. Stallman had modified the software on an older printer (the XGP, Xerographic Printer), so it electronically messaged a user when the person's job was printed, and would message all logged-in users when a printer was jammed. Not being able to add this feature to the Dover printer was a major inconvenience, as the printer was on a different floor from most of the users. This one experience convinced Stallman of the ethical need to require free software.

Re:Ethics? Still, nice to hear.

[Anonymous Coward]

I know the story, but that was then and this is now. The point is that he values "Free" over "works". Obviously the software has to perform some function to be useful, but if it's not Free then it is not acceptable to him. So if working Free software is acceptbale but working non-Free software is unacceptable then one must conclude that "works" is secondary to "Free".

Re:Ethics? Still, nice to hear.

[adoarns]

It's just another thing that some of the Linux community puts above having things Just Work(tm)

I understand it differently. The Linux community works because of the GPL. You can't have Linux without the GPL. So this philosophy thing, this breaking of stuff that Just Works, is in the name of keeping together a much bigger and important thing.

Re:Ethics? Still, nice to hear.

[harves]

Can we perhaps agree that Free Software is all about "not screwing your users in the long-term"? You can claim its an ethical thing, or its a practical thing, but the result is the same. "Free software" says that you should not use your creation in a way that places your customers at your mercy.

Re:Ethics? Still, nice to hear.

[Brandybuck]

But honestly, I think most of the people COMING to Linux in the Desktop world could care less about these "ethical" issues.

But not everyone is using Linux. Shocking, but true. As a FreeBSD user, Linux-only proprietary drivers are a major pain in the ass. I'm sure the OpenBSD, NetBSD, OpenSolaris, etc., feel the same way. A video card should NOT dictate what operating system I use. I'm buying a new computer in a couple of months, and if there isn't a decent open source video driver for newer ATI cards, I'm going to be considering Intel instead. This isn't about boycotting AMD, it's about avoiding hardware that doesn't work.

Re:Ethics? Still, nice to hear.

[Anonymous Coward]

Under Linux, most things Just Work(tm)

Have y'all ever used GNU/Linux on the desktop? I think Linux makes for an amazing server OS but makes me vomit on the desktop. Although, I do not consider this the fault of any of the fine people who contribute to the desktop cause.

Re:Ethics? Still, nice to hear.

[drinkypoo]

I worked for a long time on mainframes, and never heard of Richard Stallman or the GNU Manifesto during that time. Yet, I saw first hand how much easier it is to fix something if you've got the source code than not.

Congratulations: You have discovered the benefit of Open Source. But you haven't discovered the benefit of Free Software.

Re:Ethics? Still, nice to hear.

[SanityInAnarchy]

I think most of the people COMING to Linux in the Desktop world could care less about these "ethical" issues.....However, since some of these closed-source drivers aren't working for some, it's nice that AMD wants to open theirs so that eventually they can be modified until they work.

And that's the point, really. Or more importantly: Open drivers for a given card will work forever, on any platform, no matter what the user wants to do to their system. Closed drivers for a given card will work as long as it's profitable for the company to make them work, on the platforms that are profitable, in the use cases common enough to be profitable. Or, maybe in the case of Linux, "less unprofitable".

So, for example: Open ATI drivers mean I'll get 3D on my Powerbook (which I believe has an ATI card), whereas nvidia Linux drivers are x86 or x86_64 only. Open ATI drivers mean we'll get compositing that works (like the current Intel drivers) -- currently, the only way this works on ATI is with XGL, and I still get random lockups (10 seconds or so at a time) with the binary nVidia drivers. Open ATI drivers mean I don't have to worry about upgrading my kernel as often as I want, especially if the drivers actually become part of the kernel -- with nVidia, it will probably work, but you never know.

And the list goes on and on and on.

Most people could care less about the "ethical" side, if by that you mean RMS fanaticism where RMS himself refuses to own a car (since most contain proprietary software). I still tell most people to buy nVidia, because it Just Works, even if the drivers are currently less open than ATI. But given a year or two for these drivers to actually come out, stablize, and get the features we've been begging for, I imagine I'll be leading people back to ATI. (One guy will miss his nVidia SLI -- boo-frickin'-hoo.)

Re:Ethics? Still, nice to hear.

[CTachyon]

Have y'all ever used GNU/Linux on the desktop? I think Linux makes for an amazing server OS but makes me vomit on the desktop.

I do. I have for 7 years, in fact. I remember the excitement (in the positive sense) of downloading new point releases of KDE 2.x and looking for the new features. I remember every release being an improvement, often a huge one, over the previous one.

And I also remember first installing Debian and being awed by apt-get. I remember finding out GNOME was the default on Debian, then finding out that GNOME was actually pretty damn good itself and there wasn't really any need to apt-get install kde. Since then, GNOME itself has gotten even better.

As you can probably guess, I have no idea what you're talking about.

Re:Ethics? Still, nice to hear.

[ClamIAm]

No. Open Source is about appealing to amoral corporations. Free Software is about Freedom.

can they?

[mastershake_phd]

The agp specification is proprietary and you need to pay (heavily) for the spec. Releasing their driver source would be like giving away the agp spec. It might not be legal.

Re:can they?

[Excelcia]

I don't think their driver source has anything in it that discloses AGP specifications. They've been using Linux apgart code for a while, in a manner that may have already been violating the license.

Re:can they?

[Anonymous Coward]

The agp specification is proprietary and you need to pay (heavily) for the spec. Releasing their driver source would be like giving away the agp spec. It might not be legal.

Copyright regulates "works" (i.e. expressions of ideas) not ideas. If what you said were true, reading, for example, a C++ book would mean that you can't write any C++ code anymore.

Re:can they?

[Anonymous Coward]

Wrong. google for "agp spec".
The spec is free, and always has been. You're confusing them with the PCI specs, which PCISIG charges lots of money for. Intel however, never asked for a cent for the AGP specs.

Re:can they?

[pbhj]

>>> "Releasing their driver source would be like giving away the agp spec."

Well, no it wouldn't.

Assuming you're right and the spec is proprietary - then releasing source would probably depend on the wording of any agreement that AMD (or it's partners / subsiduaries) signed.

I can readily imagine that the AGP folks didn't anticipate open-sourcing of driver code in the agreement - in which case AMD should be fine. IANAL.

Seeing is believing

[Anonymous Coward]

Don't buy ATI until they have followed through with that promise. As far as I am concerned, they have until July, when their new low end card becomes available. If there are no Linux drivers for that card then, I will buy an NVidia based card.

Re:Seeing is believing

[AC-x]

Right, so instead of buying a card from the company that has just promised to release open source drivers you'll get one from a company that hasn't, yeah that makes perfect sense.

Re:Seeing is believing

[incer]

Actually, I believe it's about buying a card from the company that has drivers that work. I switched to nVidia for that reason, for example.

Re:Seeing is believing

[MooUK]

At the moment I'd buy a decent Intel graphics chip. They may be generally inferior, but I'm used to being nowhere near cutting-edge with hardware, and out of all the options under linux they are the best.

Re:Seeing is believing

[AaronW]

I have one of their cards now in a machine at work... the ATI drivers are quite buggy compared to nVidia. I would just go with nVidia until ATI has gotten their Linux act together.

I curse at my ATI card constantly due to the frequent bugs I hit (often every few minutes or even seconds the text in my editor gets corrupted, and my mouse cursor is also corrupted).

Plus, Google Earth doesn't seem to work with ATI.

Re:Seeing is believing

[Aladrin]

I think the coward's point was that nVidia's drivers are better right now. If ATI hasn't made good on their commitment to open source (and thus foster an environment for better drivers) then he's just going to keep on as if nothing has changed.

Why should he believe the promises of an PR person and let that influence his buying decision?

I just ordered parts to build a new PC and the GPU is an onboard Intel X3000. Why? Open source drivers. If ATI has open source drivers the next time I buy parts, I'll probably choose them for the GPU, even if nVidia also has open sourced, despite the fact that I've been an nVidia fan. Why? ATI has better hardware! It's just too bad their drivers have always sucked, on all platforms.

The AC stated it like an ultimatum, but I think he/she was just stating a fact: Their next purchase would depend on ATI's action or inaction.

Re:Seeing is believing

[babyrat]

Good thinking!

Punish that damn ATI for not having an open source driver. Punish them by buying hardware from another company that doesn't have open source drivers!!!

Re:Seeing is believing

[Anonymous Coward]

As you probably know, the other company's drivers at least work: Working OSS driver > working closed source driver > incomplete OSS driver > incomplete closed source driver > no driver. To some people an incomplete OSS driver is better than a complete and working closed source driver. I'm not one of those people.

This is great

[C_Kode]

I only buy Nvidia because it just runs better under Linux even though ATI is better on Windows. I happen to run both and I want the best of both worlds. My guess is this is partly because of the change of momentum towards Linux on the corporate desktop over the last year.

Some people will be sure to downplay this, but I think this is really the beginning. It will take time, but I expect that Linux desktop graphics will closely compete with the Windows desktop soon.

Nvidia, this is your wakeup call. Follow suit, or my next graphics card will ATI.

Re:This is great

[WaZiX]

If they do it, whatever Nvidia does, my next _WILL_ be an AMD/Ati. To me, they deserve at least my next sale just for having started the movement.

Great News (let's hope it will actually happen).

Re:This is great

[Anonymous Coward]

but I expect that Linux desktop graphics will closely compete with the Windows desktop soon.

What are you saying? That the graphics on Linux are going to get slower and more ugly?

Re:This is great

[cp.tar]

On the other hand, if DirectX 10 is as good as some people tell me, Open Source community will still have to play catch-up...

Personally, I can't wait for really good open source drivers. Somehow it gives me a feeling somebody, somewhere is actually working on them.

Re:This is great

[Anonymous Coward]

Well not really, OpenGL allready has the extensions needed to compete with D3D10 on a feature basis. (nvidia even had their own vendor specific ones released a few months before DX10 was avaliable), so technically there isn't any catching up to do as both Windows XP and GNU/Linux had the ability to use those shiny features before Vista had them.

Re:This is great

[CuteAlien]

Same here. I did buy nvidia because drivers had less problems on linux. When ATI opens the source and nvidia does not my next card will be ATI. Simple as that.

Hopefully it will help corporate adoption of Linux

[Com2Kid]

Some of the most popular corporate laptops (Dell D600 I am thinking of you) have perpetually had nasty graphics drivers that have resulted in much suffering on the part of users.

For the D600, I have a rather nice choice of either good performance and much graphical corruption (weird cursors and such) with the official ATI drivers, or horribly slow performance and no corruption with the existing open source drivers.

Oddly enough this only happens with some Distros, so I am sure there is a magic setting I could change some where, but honestly, I shouldn't have to! Especially just to get basic desktop functionality.

Don't Promise it. Do it.

[TechyImmigrant]

Really it isn't hard. Identify the code you own, replace the code you don't, put on a GPL header and release.

Promises are cheap.

Re:Don't Promise it. Do it.

[TheRaven64]

GPL? All of the existing DRI drivers are MIT licensed, except for the small amount of kernel glue, which is the same license as the kernel (GPL on Linux, BSD on *BSD).

Re:Don't Promise it. Do it.

[Brandybuck]

Ditto. AMD isn't making money off their drivers, they're making money off their hardware. So make the drivers free... as in unrestricted. Keep the licensing baggage out of it. Don't call it "free" and then encumber it with a bunch of strings, conditions, clauses, and hairy lawyers looking for someone to sue.

Re:Don't Promise it. Do it.

[julesh]

Really it isn't hard. Identify the code you own, replace the code you don't, put on a GPL header and release.

Of course, in a significant software project, step 1 might take a month of lawyer's time ($35,000), and step 2 might take a man-year (or maybe even more) of software engineer's time ($80,000). If you think financing a $115,000 operation "isn't hard", I'd like to have your bank account.

Re:Don't Promise it. Do it.

[frogstar_robot]

There is also:

1. Identify the code you own and release it. 2. Release any register specs not clearly indentified in #1. 3. Leave it to the community to replace the non-owned code.

Also, $115,000 is chump change to a company like AMD. If that was all it cost, then doing this is a no brainer. The potential Linux market for their cards is bigger than that. Intel is certainly poised to cost them much more than that in market share loss due to their more Linux friendly stance. In all fairness, I think producing even a redacted code dump will cost them more than that in lawyer and developer time but the final number would still be closer to 100k than a million.

I have a question...

[Anonymous Coward]

I have absolutely no idea about video drivers. but can somebody explain why good open source graphic drivers are not made. Is it a question of amount of research involved or not able to match the performance expected......

Re:I have a question...

[RobertLTux]

video performance is kind of like trying to drive in a city where maps (actual correct as of this moment ones) are top secret burn before reading
some stuff you can just guess (main library police office fire department city hall factories over that away homes over this away) but some things you need to get a chunk of map (like 2nd 4th 9th streets all have half streets next to them and 8th st is north only) then you get into wierd stuff like both an airport and a seaport with situations where you can get a ticket if you drive the wrong car to one of the ports (or even that you are driving a CAR AT ALL).

plus some of the OEMS do tricks like have a flash upgrade change bits of the map or change chips on newer versions of the same board.

in short it becomes a game where not only do you not have the names of the players but you are not sure exactly what game is being played

Re:I have a question...

[Anonymous Coward]

Drugs are bad, m'kay?

Re:I have a question...

[dotgain]

Whiskey. Tango. Foxtrot.

Actually, I hope they don't.

[Excelcia]

Hopefully AMD's decision will put pressure on NVidia to release open-source drivers as well! Actually, being an ATI fan, I hope NVidia doesn't release open source drivers. ATI will get a lot of goodwill with this, at the very least, and at most, if Dell's experience with Linux is good, perhaps they'll get more sales. Go ATI. :)

Re:Actually, I hope they don't.

[Anonymous Coward]

Actually, being an ATI fan,

You're a *fan*, of a *company*?

God, marketting just loves brainless people like yourself.

Way to go!

Re:Actually, I hope they don't.

[MoxFulder]

Actually, being an ATI fan, I hope NVidia doesn't release open source drivers. ATI will get a lot of goodwill with this, at the very least, and at most, if Dell's experience with Linux is good, perhaps they'll get more sales. Go ATI. :)

I'm sorry, but that's a really stupid attitude, since you're a *consumer* of their products. You benefit from your favorite company's *innovation*, not from their sales figures.

Hoping that your favorite company's competition continues to fail basically ensures that your favored products start to suck... without NVidia and Intel at nipping at its heels, I can assure you that ATI cards would stagnate.

For example, I am an AMD fan when it comes to processors. I like the value of their mid-range offerings, I like HyperTransport, I like their innovation in the 64-bit area, and I consider them more friendly to open source. But does that mean I want Intel to suck? Far from it!! I want Intel and AMD to fight each other tooth and nail. I rejoice at the low-power Core 2 Duo processors, knowing it will force AMD to come up with something better. I delight in the price wars that have forced AMD to discount its processors, allowing me to buy a dual-core 64-bit Socket AM2 processor for $60 from Newegg.

The way I see it, I want my favorite products to encounter incessant and BRUTAL competition... and to triumph through innovation.

Dupe?

[Anonymous Coward]

Either way,

http://linux.slashdot.org/linux/07/05/10/1424224.s html

an alternative reading...

[Anonymous Coward]

VP of sales: We have to delay the r600 again!!
CEO: WTF?!?!? Ok, ok. Let me think a moment... with all those problems, it seems that we are not developing _anything_... I got it!! You just have to go to one con, one full of hippys with long hair, and make a new press release stating something wonderful of our new cards... No! much better! something wonderful of all of our cards. That is. This will give us some good press and nobody will remember the delay
VP of sales: what kind of press release can afford such incredible thing??
CEO: I don't know... well, these hippys are always requesting new and shiny drivers that works with their toy operating system. Just promise that.
VP of sales: But... we promised that 18 months ago, and still...
CEO: And?
VP of sales: Well... but our customers are not stupid.
CEO: We'll discuss that later. Just give me a month without news of r600 and I'll remember you in the next stock options party.

They're probably just press whoring, but...

[Deagol]

I'll bite, though. I've been a Matrox and Nvidia user ever since I gave a hoot about the card driving the pixels in my machines and stopped buying Trident and S3 cards (so.... maybe 10 years now). For some reason, I've never given ATI a second look, aside from the fact that they seem to be the chipset of all rack-mount servers I've ever used.

I'm in limbo now, though. I'd love to upgrade my Geforce FX 5200/128M card, but I just won't until I can get dual-head and acclerated 3D under FreeBSD/amd64 with an *open source* driver (it matters to me -- I'm sick of binary blobs). Period. The first company who gives me hardware and open sources the drivers (or at least the damned specs) will get my business the minute a driver is available for my platform, and I'll become their biggest volunteer fanboi/astroturfer to reciprocate the deed (on- and offline).

I'll believe it when I see it, though. I ain't holding my breath.

Re:They're probably just press whoring, but...

[Anne Thwacks]

I'd love to upgrade my Geforce FX 5200/128M card,

Id love it if mine worked. I had FreeBSD working on it for about 3 years, and last month I did a cvsup and portupgrade of xorg. Since then, I have been stuck in comamnd line mode!

I will never buy anything from NVidia again unless they opensource their drivers, and I shall make damn sure that everyone who shares the data centre knows why.

ATI Rage drivers for spark64 would be nice too, even if only in text mode!

The fact is, this closed source stuff completely sucks. Pay for it? Its not worth a fig!

Re:They're probably just press whoring, but...

[Deagol]

I'm not sure if you've noticed the ports freeze holding steady these past 2 weeks or so, but from what I hear, it's a freeze to get X.org 7.2 into ports (we're at 6.9 now). Maybe it'll work then. :)

Day one Sale

[Anonymous Coward]

I havent bought an ATI card ever (unless you count the GC and Wii). I would imediate buy one if they had robest OS drivers. Currently I always buy Nvidia.

It is that simple really, not about gaming nessesarily, more about trust, I trust OS more.

Re:Day one Sale

[mlts]

Same here. If ATI acts on this promise and gives out open source drivers which allow use of all features at normal speed, I'll be making the switch back to ATI.

Otherwise, its nVidia for me, period.

Now, if both ATI and nVidia make truly open source drivers, that will make the choice difficult, though it would be a quite pleasant situation to be in.

Re:Day one Sale

[Randseed]

I've the same opinion. It isn't so much whether nVidia's drivers are open source or not. Honestly, this is the one case where I really don't care. What I do care about is whether the drivers (in whatever form they are) work and are kept updated. None of this crap about "Well, you upgraded to a development kernel to fix some other bug, now your graphics won't work because we release binary only shit."

nVidia has historically been better to me than ATI has. I have one desktop with an ATI X700, and one box with a nVidia 7600. The nVidia machine works a hell of a lot better in Linux and, frankly, in Windows too.

Intel driver Open Source?

[slashdot.org]

Last time I looked at the Intel driver source, there were a ton of calls into the video BIOS. Not something I would call an "Open Source" driver. This may have changed since then,- I really hope so.

Why is it important to have more source you might ask. Well, for one thing it would be really nice if we can get rid of the video BIOS altogether. A full source driver which shows how to switch video modes is a very good start to accomplish this (although not necessarily enough).

And then you might ask, why do we need to get rid of the video BIOS? Well, when evaluating graphics chips for an embedded systems, I found out that the video BIOS can spend an insanely long time initializing stuff and displaying stuff that we don't want/need (some like several seconds). In general, video BIOSs are over-engineered and do waaaay more than needed.

If you are aiming to build a near-instant-on system, and/or something that doesn't look like a PC, you want this sort of flexibility. If AMD steps up to the plate, that would be awesome.

Re:Intel driver Open Source?

[notamisfit]

Intel's open source drivers really aren't; it's completely undocumented vendor supplied code. ATI followed the same model before the Xbox 360 contract came up, and it's likely what they're going to return to.

Re:Intel driver Open Source?

[Josh Triplett]

Last time I looked at the Intel driver source, there were a ton of calls into the video BIOS. Not something I would call an "Open Source" driver. This may have changed since then,- I really hope so.

Why is it important to have more source you might ask. Well, for one thing it would be really nice if we can get rid of the video BIOS altogether. A full source driver which shows how to switch video modes is a very good start to accomplish this (although not necessarily enough).

Look into the new "modesetting" branch of the Intel driver, currently moving towards the default. It moves all the work of modesetting and other related hardware manipulation from the video BIOS into the driver, and avoids the video BIOS entirely. This does indeed give the benefits you describe in your post. Some of this modesetting code also moves toward sharing between drivers, to support modesetting for all Xorg video drivers. (Some of it consists of driver-independent code, such as dealing with funky monitors.)

Vague...

[evilviper]

Anybody got any more details? They talk about the lack of a timeline, but "graphics drivers" is also vague, and could mean 2D, or just another small subset of features.

I'm certainly not going to go out and start buying ATI cards until all the details are worked-out.

Ethics? Yes.

[pedestrian crossing]

I think most of the people COMING to Linux in the Desktop world could care less about these "ethical" issues.

True, right now they don't care. But that doesn't make it any less important to develop Free drivers.

Richard Stallman had his realization that Free software is necessary based on his experience with a printer driver.

Re:Ethics? Yes.

[Anonymous Coward]

It is apparent that whoever purchased the printer never asked if the feature was available with
whatever software the company decided to accompany the printer with. Granted the printer was
industry's first but little bits like that make space for other manufacturers to jump in as well
and cover the gap (just recall how OS/2 lost it because of the manuals). Stallman never understood
that and believe me, he will never will. And I am pretty sure that the manufacturer of the printer
never promised Stallman contractually or otherwise that he would getting be his hands on the
company's IP. He never understood the difference between open/standardised APIs and IP.

Re:Ethics? Yes.

[Anonymous Coward]

I think we all can appreciate the efforts RMS has made, but saying things like "the prospect of charging money for software was a crime against humanity" is hard for most reasonable people to come to terms with. On the other hand, this is the same community that complains about George Lucas "raping our childhood memories"...

Re:Ethics? Yes.

[Anonymous Coward]

Considering his weight, taking a few preemptive walks down to the printer might have done him some good.

Re:Ethics? Yes.

[Virgil Tibbs]

its interesting you don't seem to have the guts behind you're statement

thats actually not what he is saying - go read the GPL - *PLEASE*

its not distributing the source code with the binaries or on request which he complains about

under the GPL you CAN charge money for your software
read:
http://en.wikipedia.org/wiki/PHP-Nuke

Re:Ethics? Yes.

[Anonymous Coward]

Right. What RMS said is all in our imaginations.

Re:Ethics? Yes.

[808140]

If you'd actually read the article you linked to, you would realize that the quote in question (about charging money for software being a crime against humanity) was not in fact said by Richard Stallman, but only attributed to him by someone else. There's nothing new here: people are constantly trying to claim that the GPL is somehow anti-profit. But the GPL has never been about restricting anyone's ability to profit: in fact, consider the following statement, from the GPL itself (emphasis added):

For example, if you distribute copies of such a program, whether gratis or for a fee, you must give the recipients all the rights that you have. You must make sure that they, too, receive or can get the source code. And you must show them these terms so they know their rights.

People who claim that the GPL explicitly forbids making a profit from selling software or that Stallman himself is against the sale of software are either deeply confused or dishonest. In fact, in the early years of the FSF, the foundation was primarily supported by the sale of its own, GPL-licensed software. In those days, the internet was still slow, unstable, and inaccessible to commercial interests, many of whom nonetheless ran UNIX systems and desired GNU tools. These were sold, on floppy disks by snail mail, to people who were interested. Yes, for a fee.

"Yes," says the businessman, "I understand that -- but the GPL effectively drives the profit from selling software to zero, even if it doesn't explicitly prevent me from selling said software, because the person I sell it to can turn around and distribute it to his friends for free, a right the GPL guarantees!"

He's right, in a certain sense. However, anyone with a cursory understanding of economics and the free market knows that healthy markets also serve to drive profits to zero -- that's how you know the market is functioning properly. Margins will decrease with competition by definition, and prices will drop until all that is being covered by price is operational cost. This is considered good for the consumer; it means that businesses are forced to innovate and develop new markets, because when a sector becomes saturated, profits will by definition lower until the margins are so thin that companies previously providing the service will no longer be interested in providing them; this reduces competition, prices rise, margins widen, providing incentive to other companies to re-enter the field and make a profit -- and so the cycle begins anew. Clearly, in a properly functioning market with low barriers to entry, margins will oscillate very close to zero. This is how it is supposed to work. You don't have the right to make a profit, not in capitalism.

The case with the GPL is analogous -- the freedoms it guarantees the user reduce artificial barriers to entry and make competition very easy. For example, had I decided to sell GNU software in the 1980s -- ie, compete with the FSF, who was making a tidy profit this way -- I could have. I would have figured out how much it cost to distribute the software, and made my price lower than theirs by reducing my own profit margins, thus stealing their business. They could then have undercut me, or perhaps a third party could have undercut us both, and so on and so forth, until there was nothing left to undercut, and the price of selling the GPL'd software was exactly equal to the cost of distributing it. I probably wouldn't see this as ideal, and I'd probably stop selling GNU software. Competition would ease. Prices would rise. Business people call this an arbitrage opportunity: a product is overvalued, so you sell it for less to bring prices down, until the opportunity vanishes and there's no more profit.

The catch is, with the internet, distribution costs are nearly -- although not exactly -- zero.

Clearly, there's not much money to be made selling Free Software, although you'r

Re:Ethics? Yes.

[NekoXP]

For example, if you distribute copies of such a program, whether gratis or for a fee, you must give the recipients all the rights that you have. You must make sure that they, too, receive or can get the source code. And you must show them these terms so they know their rights.

People who claim that the GPL explicitly forbids making a profit from selling software or that Stallman himself is against the sale of software are either deeply confused or dishonest. In fact, in the early years of the FSF, the foundation was primarily supported by the sale of its own, GPL-licensed software. In those days, the internet was still slow, unstable, and inaccessible to commercial interests, many of whom nonetheless ran UNIX systems and desired GNU tools. These were sold, on floppy disks by snail mail, to people who were interested. Yes, for a fee.

There is a flaw in this argument.

The GPL doesn't truly forbid profit, that's right, since this would be a restriction of rights. It does only mention that the software may only be shipped at reasonable cost; be that postage, media cost, or the 6 months work on your part which requires a salary, it makes no real differentiation.

It however then goes on to make true profit totally impossible merely on the basis of distributing software. That fatal flaw being that you can sell your code for a profit ONCE, and then that customer can distribute your software for free, thereby completely and totally undercutting you, and depriving you of any further remuneration for your software.

However if a developer's goal is to simply write great software and get his money back for the 6 months of work he put into it, that's easy; sell it to a company who wants to open source it. Well, it's not easy, but it has been done. There's also the opportunity as the copyright holder to re-sell the software and re-license the software for profit, however the FSF has a bee in it's bonnet about transferring copyright to the FSF meaning only THEY can re-sell and re-license the software effectively (i.e. you give up all your rights under the GPL too, willingly, for the privilege of having your name stripped from the code, and an FSF banner on top)

The GPL, and the organisation that developed and standardised the GPL, basically makes very clear that you can try and make money from GPL software, but you'll be wasting your time.

Re:Ethics? Yes.

[808140]

The GPL, and the organisation that developed and standardised the GPL, basically makes very clear that you can try and make money from GPL software, but you'll be wasting your time.

I'm not sure where you get the idea that they make it very clear that you'll be wasting your time; I'm also not sure where you get the idea that you would be wasting your time. As I said, the FSF ran on the profits generated by selling GPL'd software in the early years -- while it has admittedly become more difficult to make money on distribution alone, the fact remains that when the GNU project was conceived (1984) and the GPL v2 first written (1991) the internet was far less accessible (and far slower) than it is today and there was no reason for the FSF to think of their distribution based business model as unworkable in the long term. The internet revolutionized distribution, and they (and many other organizations and companies) were forced to rethink their business model.

But nowhere was there any idea that you couldn't sell your software for a profit, even for a large profit, if you could find some way to make money off of it.

As for wasting ones time, RedHat remains (although I'm personally not fond of them) the premier provider of Linux-based OSs in the corporate sector, and they do not provide their (GPLd) software for free. You have to buy it. People do, despite the fact that it's expensive. There's a simple reason for that -- the GPL itself (like most free software licenses) explicitly states that the software comes with NO WARRANTY -- NOT EVEN THE IMPLIED WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE. This is a scary, scary thing for a company that can't afford downtime, which is why serious businesses don't run Gentoo, even if it might be a good distribution (I don't know, I've never used it). They pay through the nose for support from RedHat -- not even superbly great support, ironically. There's a great deal of money to be made here.

So here's the reality: with the GPL, you can't make a shitty piece of software and then sit on it until its copyright expires, making money on sales for the rest of eternity. That may be a shitty thing for a software programmer, because hey, everyone likes to "do a little bit of work and then retire on royalties." But the market is efficient because it forces people to work, not because it lets you have a free ride like some Soviet Party Cadre. There's no such thing as a free lunch. If you're going to make money, you're going to have to work hard, and that means supporting your software as well as just developing it. To make things even better (for me), you aren't even guaranteed the support contract on software you've written, because if your competitor manages to support your software better than you yourself do, I might just buy my support contract from him.

Low barriers to entry -- that's what capitalism is about. Conventional copyright is an artificial barrier designed to prevent the entry of competitors into your market -- it guarantees the author of a work a temporary monopoly. Even though everyone knew this went against the fundamental tenants of the capitalist system, it was argued (correctly, in my mind) that offering the author the ability to temporarily monopolize profits from his work would act as an incentive for authors to create works in the first place. But when copyright is extended indefinitely, the market suffers, and the consumer pays. This might make you a little bit richer personally, but you're not just an author, you're also a consumer of other people's intellectual property, and perpetual copyright drives the cost of doing business in general up. This acts as a "tax" on the average consumer: everything costs more.

Is making a little bit more money really a good thing, if everything costs more? I live in Silicon Valley. Salaries are high here, much higher than the rest of the country, but, frankly, who car

Re:Ethics? Yes.

[NekoXP]

The GPL pretty much does say you'll be wasting your time; summarised over a few paragraphs and legal points of course, but "you can charge for your software" and "your customers are free to put it on a website and give it away" is exactly wasting your time trying to make a profit on it.

RedHat do sell their software, and a vast majority of it is under the GPL; however the tools they provide to keep your system up to date, the service contract they enforce on the thing, are NOT GPL. It is not the GPL software they are selling, but a raft of 'mandatory' services. You can't put the RHEL CD online for free, because contained on that CD is in fact some proprietary software. Be that Java or a Flash plugin or RedHat Command Center, Directory Server, Certificate System, Cluster Suite even something bigger, this is stuff Red Hat have licensed and are making you pay for, or it's proprietary RedHat software and support wrapped around Open Source software you can just compile yourself - if you're willing to fight and play and perform validation, performance testing, stability testing on your own systems through insane text file configuration systems and not a point-and-click UI. This is stuff only RedHat (not RedHat's customers) are allowed to distribute, or it's not economical for anyone else to distribute.

In effect they do not make a profit on the GPL software. They probably do not make a great deal of money on selling individual (even at $80) copies of RHEL. They are a services company, and a value-added company. You still can't sell GPL software for a profit, and RedHat are not anything close to an example anymore of "selling GPL software and making a profit" because the vast majority of their profits come from selling other things.

I don't think there is any company in the world that truly makes a profit - or any individual developer for that matter who has attempted it and not been shot down by rabid FSF fanboys - by solely selling GPL software for a profit and curtailing the rights of customers to redistribute so-called Free Software independantly of the author and/or copyright holder.

Mainstream gaming

[CustomDesigned]

I agree that hard core Linux gamers are an edge case. However, most of us would like to be able to play Penguin Racer or Tux Kart occasionally. Useless eye candy like 3D window switching effects help relieve boredom as well. This doesn't require the latest hot graphics card with dedicated cooling towers. However, it would be nice to have stable drivers that track kernel evolution for entry level 3D cards - sufficient for simple games and effects. The present situation is that old low end Vanta Nvidia cards (suitable for Tux Kart) still require proprietary drivers - and Nvidia is losing the motivation to keep them updated (they did patch old drivers for the security hole mentioned on Slashdot a while back).

IMO, using binary blobs that run in the card, not in the kernel (i.e. downloadable firmware), are a reasonable way for vendors to hide trade secrets while keeping the card updateable and the kernel driver open source. As long as shared memory between the graphics card and main system is restricted to a window, bugs in the firmware shouldn't cause security holes in the kernel. In fact, one benefit of micro-kernel architecture is that isolated drivers that run in their own process and address space, can run in an intelligent I/O card instead.

The IBM Series/1 was built on the principle. All I/O was done by intelligent cards with a common API: submit Device Control Block with command, memory block, and parameters to start an operation. Receive vectored interrupt and find results in updated DCB and memory block. Interrupt included address of DCB, so interrupts were trivially "object oriented".

Re:Mainstream gaming

[OrangeTide]

It sure is nice when GLX works and you can do CAD, modeling, simulations and 3D programming(OpenGL) on a Linux box. So there are practical uses beyond gaming for those fancy 3D cards.

Re:Mainstream gaming

[howlingmadhowie]

the current situation is pretty rediculous. the way i see it, a graphics card manufacturer makes a card where the capabilities of the chips on the board are unknown, the firmware on the board in a binary blob, the meaning of the various pins in the pci-express slot is unknown and a second binary blob installed in the kernel of the operating system is also unknowable. how did it ever get this ridiculous?

Re:Mainstream gaming

[Sloppy]

However, most of us would like to be able to play Penguin Racer or Tux Kart occasionally.
...
IMO, using binary blobs that run in the card, not in the kernel (i.e. downloadable firmware), are a reasonable way for vendors to hide trade secrets while keeping the card updateable and the kernel driver open source.

Reasonable, until...

Time to sign the pledge!

[the Hewster]

Looks like it's time to sign the pledge to support open, 3d graphics drivers http://www.pledgebank.com/open3d and put some pressure on graphics card manufacturers.

Re:Time to sign the pledge!

[Tovok7]

No! The time to sign petitions is (hopefully) over! No it is time to vote with our wallets! Let's all go out and buy one ATI graphic card the second the drivers are released as free software!
Then sales will explode and that sets a huge sign for other hardware manufacturers far beyond the graphic sector. Eventually, in a few years, a manufacturer who is not supplying free drivers will not be able to compete on the market. But unfortunately I don't expect the wave of freedom reach unto the embedded device like cellphones.

What about licensed functionality?

[dave420]

If there's licensed proprietary code in the drivers that ATI/nVidia doesn't own, it'd take a great deal of lawyers and possibly price hikes to make it possible. Just because something's closed source doesn't mean to say the only reason it's not open source is laziness or hatred of F/OSS...

Re:What about licensed functionality?

[notamisfit]

Which is why ATI drivers before r300 were open source?

you missed a step

[Falladir]

If you're a corporation with deep pockets, you also have to make sure you haven't been infringing on any software patents.

(This is what bothers me most about our current patent laws: there is no burden on the patent holder to apprise possible infringers of the situation.)

Nouveau-equivalent for ATi?

[mqudsi]

Does anyone know if there is a Nouveau-equivalent for ATi? The FGLRX package feels so wrong!

ATI cards are terrible

[Anonymous Coward]

I stopped using NVidia and currently spec mobos with the Intel X3000 because of the driver situation. Not sure what it'd take for me ever use an ATI card - a lot.

Don't support Windows but I have installed ATI 'drivers' on a Windows box for a relative. Their 'drivers' include a media player and rely on the .NET framework, their linux drivers display similar incompetence and even copyright-infringement (needless bundling of linux AGP code).

Yeah, it'll take a lot for me to ever switch a client to ATI cards.

Re:ATI cards are terrible

[Tatsh]

Don't support Windows but I have installed ATI 'drivers' on a Windows box for a relative. Their 'drivers' include a media player and rely on the .NET framework, their Linux drivers display similar incompetence and even copyright-infringement (needless bundling of Linux AGP code).

So true. Their Windows drivers are horrendous, and their web site used to be so confusing to navigate. Their drivers for Linux are okay, at least their is not crap included like a media player. Unfortunately I'm stuck with dealing with them right now (my main PC is a notebook). Every time I've used Nvidia after using ATI for the same thing, Nvidia was always more stable and the driver support is so much simpler. Now ATI is just playing catch-up with Nvidia, as it always has been. They are hoping to get ahead with this announcement.

Excellent

[OriginalArlen]

This is turning out to be a good year for good news! It was only when I started thinking of things this is better than, that I realised we've also had the apparent collapse of DRM for music (it's not over yet, admittedly, but if you were around between 1987-1990 you may remember that from Glasnost to the collapse of the USSR was also a slow motion thing...), the apparent flop of Vista, the imminent failure of the whole Palladium/TCM foundation of Vista's treacherous computing, Dell shipping Ubuntu pre-installed... whatever next? Darl McBride's corpse dragged through the streets by an angry mob of lawyers? Bush / Cheney resign? :)

Re:Excellent

[boolithium]

Lol, you should have got at least a 3 and a funny. Damned stingy karma giving mods.

Re:Excellent

[PRMan]

You forgot the HD-DVD / Blu-Ray crack.

Re:Excellent

[babbling]

I agree with you that the announcements have been good, but none of the things you mention have actually happened yet. I'm not doubting that they will happen, just pointing out that nothing has improved just yet.

Soon we will be able to buy unencrypted music on iTunes.
Soon we will be able to buy Dell machines that come with Ubuntu preinstalled.
Soon we will be able to buy ATI graphics cards and use them properly.

Both the iTunes and Dell things appear to be imminent, though.

Damn you AMD!

[Anonymous Coward]

It used to be so easy, I didn't even have to consider ATI's video cards. Now I have to start to keep up with two lines of confusing model names.

I think I'm going to cry

[boolithium]

In 1997 I had a 486 with a 2.5 GB hard drive, which why that might seem behind the curve, my previous machine was C64, so it was a big upgrade for me. My choices seemed to be limited to Windows 3.1, Beos, OS2, or this little project called linux (I guess freebsd too, but I actually use my computer :p ). So I got one of my fancy friends to use his isdn line to download me a Mandrake install cd.

A decade and like 30 distros later I'm still a loyal user. In the past couple years I have watched the great evils of the world morph into our new allies (ie.. IBM, Novell, Intel). But this all came with a underlying Faust kind of deal. Each kernel was tainted with philosophical contradiction. That video driver was nothing more than a cheap whore while the wife was out of town. A big ugly binary among pristine virgin innocent source included apps. All you who think linux can't play games might want to note opengl, sdl, and openal are finding there ways onto your nazi boxes more and more. And bling, ha, one word beryl, Mac aint got shit on us. We're the beast in this mother fucker. Oh and by the way with all that bling and a tweaked kernel, the newest gnome and plenty of other gentoo goodness I'm at about 196MBs of ram used. We'll see where games can be played.

By the way in case anyone from nvidia is reading this, please go fuck yourself you "nvrm: xid" bug ignoring sons of bitches!

Re:I think I'm going to cry

[Anonymous Coward]

"XID" messages are a symptom, not a cause. There is no "XID bug".

Re:I think I'm going to cry

[Jesus_666]

And bling, ha, one word beryl, Mac aint got shit on us. We're the beast in this mother fucker.

I really like these two sentences, especially "we're the beast in this mother fucker".

Curious wording aside, I agree that Linux gaming might have a big comeback - after all, thanks to Vista's new driver model Creative will have to pipe EAX through OpenAL, which should give OAL more popularity among Windows game devs, making the games a bit more portable.

Re:I think I'm going to cry

[Anonymous Coward]

> with all that bling and a tweaked kernel, the newest gnome and
> plenty of other gentoo goodness I'm at about 196MBs of ram used.

Must not be running Firefox.
(Or for that matter OOo, Nautilus, Xorg, or.. Gnome)

Dual head

[markov_chain]

Now there's something to do if you had a million dollars.

Re:Dual head

[cerberusss]

Now there's something to do if you had a million dollars.

This was a discussion last week at work. "What to do with a million or more".

One of the guys would like a gold plated, diamond encrusted laptop.
The second would like said laptop, with the innards upgraded every 6 months to the latest technologies.
The third would like said laptop plus upgrades, together with a cabal of female Linux kernel developers ready to code up a driver when owner buys some crazy piece of unsupported hardware.

*sigh*

Re:Dual head

[sqrt(2)]

You made my morning, sir. Thank you.

Words are cheap...

[Anonymous Coward]

A VP of sales knows it all too well. I believe it once I see it.

I've got an older nVidia card now and I'm placing my hopes on http://nouveau.freedesktop.org/wiki/

I'm GNU/Linux all the way and will only buy hardware that supports it.

Nice-Gameplay.

[Anonymous Coward]

"Also you got 3d desktops like Beryl. With eye candy being a major selling point in some operating systems, 3d features will become important if desktop linux wants to get more popular. I hope all graphic card companies will develop good drivers for Linux."

Interesting how "eye candy" is a selling point for an OS, but not a video game.

User safety and trust, right to tinker

[Anonymous Coward]

Linux distributions are developed and marketed as secure, or to be more precise, that no code is deliberately contained in them that acts against the user's best interest.
What code a closed-source driver contains is unknown. It might contain a backdoor granting specific people root access to your computer. And the most important lesson anyone can learn about computer security is that "might" translates as "will". As long as you don't know absolutely sure that something doesn't contain a backdoor, you have to assume it will. And with closed-source, you cannot ever be sure about that.
Then there is the thing about your legal right to modify the driver. This is a practical as well as ethical point. In the world before open source, or rather free software, the wheel was necessarily invented again and again and yet again, because even if a software company knew the source code of a competitor's application, it could not legally use it. If you (a hobbyist, or a company) simply wanted to add some functionality to a piece of software, you could not legally do so.
Code could be licensed, but there was not always an incentive to do so for the owner, and it's unaffordable for hobbyists. Enter free software. People can share and use each other's code without worrying. Apart from technical problems (of which there are plenty, I know), the wheel will never have to be reinvented again, and everyone can add whatever functionality they whish to such free software. This will in the long run speed up the technical pogress in IT far beyond what proprietary software could hope to achieve.

Slashdot greatest hits

[Sits]

ATI Committed To Fixing Its OSS Problems was posted only a few days ago (that one came from Chris Blizzard's blog) and the cautious tone is backed up by other Red Hat summit reports. However, since we're here why don't we pick out the highlights (along with overlooked gems) from last time?

• Someone contacts ATI's sales about this announcement and is told to hold tight for more news (this is consistent with what was posted to German news site heise.de link found via liquidat)
  
• Thank you for your current drivers ATI
  
• Slashdot shouldn't speculate on what ATI's roadmap
  
• We want open source ATI drivers, not Linux drivers
  
• ATI don't support vblank in their binary drivers (open source drivers do support vblank but it's hard to find the cards)..
  
• ATI? You mean AMD!
  
• Marketing... Ha!
  
• This is because of Dell starting a Linux push
  
• Doing a Windows graphics driver wrapper wouldn't be feasible...
  
• Reverse engineering modern graphics cards is hard... because the API driving them is complex.
  
• Intel have released open source graphics drivers BEFORE their next card was released!
  

Elsewhere on the web folks are wondering whether this means that the a new GPGPU will be accessible but the actual graphics driver itself will remain closed. AMD/ATI has also announced open source drivers before which translated into more stable and more frequently released Linux binary x86 drivers...

H.264/VC-1 acceleration

[swmike]

What some might have missed would be the opportunity to sell gfx cards due to all of a sudden being able to do H.264/VC-1 offload/acceleration in linux for HTPC usage.

The first manufacturer to offer this would get a lot of sales due to people using their cards for HTPC. I know I would, because currently I can't play 1080p VC-1 and barely h.264 properly, even on a Core2Duo.

Too little, too late.

[Ant P.]

ATI has opened the driver because they realise it's now next to worthless anyway: The open source people have created something that just works better in every way (except maybe speed), and they did it just stabbing blindly at registers. Their in-house programmers have reams of documentation to work with, so they have no excuse whatsoever for being behind. They should be fired, after the suits that perpetuated the whole "we can't open the code" farce for a decade.

nVidia doesn't get as much flak because they at least pretend to care. That, and they don't leave early adopters with bricked hardware (not even 2D drivers) for over half a year.

ATI dri effort

[Sits]

See ATI dri drivers. Support for the r300 and above came about through reverse engineering effort so in many ways the ATI reverse engineering effort is far further along (in so much as there are end user drivers that are even capable of the basic desktop effects)...

Re:ATI dri effort

[TheRaven64]

The ATi reverse engineering effort isn't quite as advanced as Nouveau, or more specifically, REnouveau. REnouveau is a fairly automated application that sends OpenGL drawing commands to the driver and spies on which control registers it writes to. It then builds an operations table that can be used for the driver. Doing the same thing for ATi cards would be possible, but I don't think anyone has done it yet. The existing drivers were reverse engineered the 'old fashioned' way.

Re:ATI dri effort

[Sits]

The open source ATI folks have a tool for the r300 chipsets called revenge and another tool imaginatively called radeondump which I gather do something similar. However tools like the mmiotrace are useful for both efforts so I think both projects are as "sophisticated" as each other.

Re:ATI dri effort

[MrHanky]

The open drivers for ATI cards up to whatever the R300 code supports is far ahead of the open nvidia drivers. I get >100 FPS in Q3A @ 1600x1200, 32 bit colours and so on with my Radeon 9800 Pro, and can run Beryl as well. The closed drivers are probably a great deal faster, but not so much that I accept the crashes it causes.

Ethics? Yes.-Going the distance.

[Anonymous Coward]

"Richard Stallman had his realization that Free software is necessary based on his experience with a printer driver."

So were can I download an open source printer?

Re:Ethics? Yes.-Going the distance.

[alx5000]

There ya go. No need to thank me.

Beryl is a joke

[Anonymous Coward]

http://themes.beryl-project.org/

Isn't it just so telling that _all_ of the most popular themes are ripoffs of the Windows GUI?

Re:Beryl is a joke

[redcane]

Maybe it's the other way around, or more for the novelty factor. I visited a friend yesterday, and his theme made it look like he had switched to Vista, but he's actually been running that theme for two years, and it's just an unforunate coincidence that vista looks like that.

NVidia can still do this for little risk

[Midnight Warrior]

NVidia can still come off okay too, if they start now. They don't have to risk their IP either. If they contribute money and/or access to older video cards, then they have fostered the development of Nouveau. If they give the team the resources to succeed without giving them access to the source code to the proprietary drivers, then they have fostered the reverse engineering without committing their own staff or risking lawsuit from revealing potentially infringing/stolen intellectual property.

At the very least, they should review the Nouveau contributors and declare them to be free of NVidia influence (a condition of fending off an IP lawsuit). This may not do anything to foster the development directly, but it certainly gives the open source advocates all kinds of warm fuzzies that the company is genuinely interested in making quality open source drivers.

I'd settle for quality working drivers

[Anonymous Coward]

given how much I've paid, I've had nothing but problems with ATIs drivers. in fact, given my experience, you can expect:

complex and error prone installs esp. with AIW cards
devices which can't find or loses its drivers
BSODs, crashes and lockups
no real support unless it's self serve and MS related

of all my hardware, video cards are the worst and cost the most

"Just work"

[matt me]

Non-free drivers and specifications will NEVER work. They will always fail users, because the developers cannot possibly predict how users in the future will use their products. Will the drivers with an 8 year old non-standard printer/modem/camera packaged for Windows 98 work on Vista? Probably not. Have the developers producted updated drivers? No, why should they? The line is discontinued. But the hardware is still in perfect condition. If the drivers were free, someone could easily update them, but a lack of specifications makes that impossible.

Non free drivers will ALWAYS end up screwing users, because it's impossible to produce something futureproof. In ten years, your 2nd generation iPod won't play modern codecs, only obsolete ones.

    We already see perfectly functional hardware abandoned because of inadequate software.

Think about the OLPC. Why do the drivers have to be free? Because if not, they are dependant on the developers (at their liberty, even if they are still in business) to produce software to work with newer, superior protocols and technology.

There exists a hi-tech car park, where cars are filed by a robot into pigeon holes. The company that built it went bust, never releasing the software. If the software goes wrong, that's hundreds of cars irretrievable without demolition. No company to take responsibility, or to fix it.

So, I guess next time I recommend PCs using ATI!

[OldHawk777]

I always try to be fair and make exceptional recommendations and deals for the folks at work.

A couple years ago I turned a CFO projected enterprise $8M deal into a $5.5M deal while getting hard-drives sizes doubled, RAM doubled, all CRTs swapped to same size LCDs ... upgrades, an additional 100 desktops (900 total), 35 HP Intel Servers, 20 SUN-Cisco nodes, and three Alcatel-Lucent Omnicore switches. Yep also the cable, patch-panels, wire-racks, transceivers, and all the other required hardware and software trinkets (let me think, was that Gates-Arrow, Ingram Micro, or CDW we made the deal with? Dang, I forget...). Total screw-ups were kept at less than 0.5% of cost which the vendor we went with resolved at no cost. Saved $2.5M ... had a ~$27K problem resolved for $0.

Maybe next time I will look at the MB-graphic or cPCI cards and decide ATI is easier to support over the lifecycle requirement. It will have to prove a better business decision, but I will look, and if all is about equal well ATI will win my recommendation.

I still have another year before I need to seriously start thinking again about big problems, but I won't forget to look at the ATI and NVidia lifecycle supportability issue. The recent distributed content management and storage network was set too a non-proprietary architecture for lifecycle compliance requirements ... scaleability and upgradeability as CFO/CIO infrastructure like to call it.

I am seldom questioned ... when I make a recommendation %~$, most of the technophobes in management remain silent and just hope I screwup.

Yeah, yeah. Let's see....

[jimicus]

1. Does ATi (and hence AMD) have the right to opensource all the code in their drivers? If they outsourced or otherwise bought code in - maybe not.
2. What level of functionality will these drivers provide? I would point out that Matrox open sourced their drivers - but then squirreled away a number of the features in a binary HAL.
3. Open source can mean a lot of things. If you're Microsoft, for instance, "open source" can mean "you can look but you can't touch".

It's great AMD are saying this - and it's a lot more than ATi would have done - but as has already been said elsewhere in the thread, I'll believe it when I see it.

Kitten killing

[Sits]

Ah one of the classic "why the drivers are closed arguments". Dave Airled basically summarised all the reasons for keeping the drivers closed in his Open Source Graphic Drivers - They Don't Kill Kittens talk at the 2006 Ottawa Linux Symposium (a longer more detailed version can be found on page 19 of conference proceedings and there's also an LWN discussion of the talk). The basic arguments were as follows:

• Microsoft - Conspiracy theorists find a way to blame Microsoft for every problem in Linux. This time they point out when Microsoft decided to use a vendor's chip in the XBox consoles or chipset vendors puts DirectX 8.0 support you don't get specs any more.
  
• ??? - Patents and fear of competitors or patent scumsucking companies bringing infringement. Vendors claim releasing chipset docs to the public may make it easier for these things to be found; however, most X.org developers have no problem signing suitable NDAs .
  
• Profit - Graphics card manufacturing is a very competitive industry, especially in the high-end gaming, 3-6 month development cycle, grind-out-
  as-many-different-cards-as-you-can. Quake 3 speeds are spotted in binary drivers any way and it doesn't explain fglrx which are some of the most unsuitable drivers for gaming on Linux.

Read the proceedings for detailed explanation of why no more kittens need to killed!

I don't think everyone understanding the argument

[jabjoe]

I don't think everyone understands the argument here. There is a problem with closed source drivers. It's not just ideology. Closed source drivers means you can end up with no drivers for a device for your version of the kernel. Even if drivers for some different kernel version exists. A good example of this is old devices. If the manufacturer still exists, they probably don't care to do the work to update drivers for a device they no longer sell. Maybe there should be a device/kernel interface that stays the same for all time, but I think as a rule, people want the best interface possible, with open source drivers so devices can be kept up. You then of course get the advantage of open source so you can fix/work-round bugs (or improve it!).

Rumours of a further announcement

[Sits]

Over in my slashdot post roundup there are mutterings in a post of further announcements. The same mutterings reappeared elsewhere too...

Bur for which cards?

[r_jensen11]

I couldn't find anything that said which cards they were targeting here. Is it anything =Radeon 7500? =9200? All Radeons? How about FireGL's? Granted, I'll be happy if they release code for anything, but nothing's actually going to help me out unless it's code for newer cards since the computer I'm using has PCIe and not AGP.

Linux gaming is irrelevant

[Anonymous Coward]

In fact Linux is the #1 operating system for people who do high end special effects 3D work. The closed source drivers that exist for Linux only exist at all because of the movie industry.

eye candy and scientific workstations

[nanosquid]

With Compiz, Beryl, and XGL, excellent 3D graphics support has become a mainstream issue. Furthermore, Linux is widely used in science and engineering, and those users use excellent 3D graphics as well.

Is this real?

[RCHS-Svein]

Strangely this statement is not listed on their press-releases page? //Svein

technically speaking

[nanosquid]

Technically, even VESA drivers are also open source drivers for ATI cards. And that's provided that the announcement even means that they are going to be releasing open source drivers at all.

So, the question is: what exatly are they going to be releasing?

In general, I just wouldn't pay much attention to these announcements either way: it's open source once it's actually been released, no sooner.

Xenon

[khyew_yi_dee]

Will this mean we can expect the opening of drivers for the Xbox360's GPU as well?

Not Really The Linux ATI driver is most hated.

[Anonymous Coward]

Reason why its most hated is for running windows games under linux.

Free with a capital F

[replicant108]

I worked for a long time on mainframes, and never heard of Richard Stallman or the GNU Manifesto during that time. Yet, I saw first hand how much easier it is to fix something if you've got the source code than not.

"Free Software [1] is the term coined by Richard Stallman in the 80s to denote programs whose sources are available to whoever receives a copy of the software and come with the freedom to run, copy, distribute, study, change and improve the software."

http://adsabs.harvard.edu/abs/2003dsa..confE..57C
http://en.wikipedia.org/wiki/Free_software

Excellent news!

[sdhoigt]

This is excellent news! I'll actually be in the market someday soon for a new graphics card to replace my current 64MB NVidia chipset. I'm tired of the flaky NVidia drivers that freeze up my system at random intervals. Bring it on AMD!

SD

If tey do this

[iminplaya]

Will it be possible to defeat any DRM they have built into their cards? I thought they employed Macrovision or something, but can't remember the details.

mod parent up

[Anonymous Coward]

No. Open Source is about making stuff work. Free Software is about ethics and freedom.

Well said.

Bravo!

[kramulous]

Bravo! I was extremely happy to read this. The advantages, for both consumer and manufacturer, will be plentiful.

I know this will never be read, but just want to go on the record.

Not nightmare -requires reading manual or new gear

[dbIII]

Dual or triple head can be a bit tricky with old hardware and is often worth getting cheap new hardware if you don't have a good driver. In the end I went mostly for Nvidia cards and some second hand Matrox cards with dual head on board to upgrade some old systems to dual head.

The ATI drivers are a pain on some cards with two outputs - I was getting clone mode working but not a large single desktop on one. Some ATI cards worked perfectly on two screens so long as you had a real DVI monitor (adaptor lost blue to VGA). Getting on board graphics and a PCI card going for dual head often works as does PCI+AGP, but I got intermittant problems mixing ATI and Nvidia cards.

You have to follow the instructions from the people that give you the drivers and edit the config file - in my opinion the gnome tool is worse than worthless and has caused many hassles when users have found it and attempted to change their resolution.

In the end the dual output Nvidia cards are cheap, work very well in linux and are not hard to set up. With PCI there isn't much choice and the ATI stuff works well if you pair it with on board graphics form intel, via etc and ignore the second output. Matrox stuff worked well paired with on board hardware or in 32 bit dual screen, but dual screen 16 and 8 bit forced me into clone mode (only a problem running old software that is too stupid to run in 32 bit colour).

When the ATI drivers improve to be as good as the Nvidia ones for things like dual head it will no longer be a hassle. That said with two new LCD monitors and a new dual output ATI card it would have been easy to get an old PCI slot machine to go dual head - however I was restricted to CRT screens with VGA input there so it was not a driver issue there.

Open Graphics

[Joce640k]

This is what the Open Graphics project was supposed to be doing:

http://wiki.duskglow.com/tiki-index.php?page=Open- Graphics ...but it looks like the idiots have gone in and spoiled it. It's now years overdue and way over-specced.

Dell

[Nimey]

Assuming this happens...

I wonder if Dell is a big driver behind this. Think about it: Dell's going to make Ubuntu an option in some of their computers. It will be cheaper for Dell to sell those models with only hardware that Linux supports well (lower support costs, fewer configurations to test), and it will also be cheaper to have the same hardware be on both Windows and Linux installs.

Result: DAMIT cleans up their Linux drivers and they sell lots more hardware to Dell.

Result: Win for us Linux weenies.

Re:Yeah, yeah. Let's see....

[charlieo88]

Bought is one thing, but outsource is work for hire. It belongs to the people that paid for it.

I'll make it easier still...

[Anonymous Coward]

Open source drivers are desirable, and I hope ATI are sincere in their desire to provide them. But, for now, I'd settle for closed source drivers from ATI which were full-featured & working properly.

MLC!

[laura_glow]

Yeah!

People keep on saying "Linux" but...

[Anonymous Coward]

...wouldn't this be useful as far as Windows goes as well? This could create better drivers for everyone, not just Linux users.

Re:People keep on saying "Linux" but...

[triso]

...wouldn't this be useful as far as Windows goes as well? This could create better drivers for everyone, not just Linux users. People don't expect open-source drivers for Windows. However, they do expect working drivers and Microsoft tries pretty hard to make sure they don't start generating BSODs. [Like they did for Vista.]

One sale made...

[J.R. Random]

... if AMD/ATI actually carries through. I plan to put together a high end box this summer, and was planning to get an Nvidia card as their closed source Linux drivers suck a little less than ATI's closed source drivers. But if AMD actually has open source drivers for their high end cards by then I'll buy AMD. I don't care if Nvidia's comparable card gets 15% higher FPS.

Nice-$$$

[Anonymous Coward]

There's also the piracy issue as well. Yes consoles can be hacked, but it's harder. Brick and morter stores are lowering the number of PC titles they carry, both used and new because people buy, copy, and return. Not to mention the online piracy. If there is any "hardcore" left? It will be an expensive one in every sense of the word.

"That said, the PC does have some strengths. Specifically, the mouse and keyboard layout is great for RTS games"

Either a USB or bluetooth mouse and keyboard would take that advantage away. The only two that PCs have is greater expandability and graphics resolution (but that's narrowing with HDTV).

Too little - way too late.

[binaryspiral]

I've noticed that Lenovo dropped ATI chips in favor for Nvidia... could it be a coincidence that they also support Linux on their laptops? No... they fielded many calls from pissed off linux users who rely on Lenovo to leverage their purchasing power to get some better video drivers.

Instead of waiting until ATI gets around to it - they just switched suppliers for some of their laptops to give their customers a choice.

Im staying with Nvidia

[Tama00]

AMD customer: "ati drivers for linux suck"
AMD: "I know we will open source our crappy incomplete ati drivers and let the community create the rest of the drivers! It will also promote the use of ATI cards in Linux and we wont have to spend a cent!"

im still sticking with nvidia their drivers may not be open but atlest they sponser Linux distrubutions and provide drivers that actually work!

Re:Im staying with Nvidia

[Oblong_Cheese]

AMD should write good, open-source drivers for Linux, but they don't. I don't care if they open-source and contribute nothing further; they contributed the initial code base. Let the community make the drivers, and then we can have driver updates whenever we want, pretty much. This ensures future driver releases will be open-source, stable, feature-rich and of good quality.

Re:Im staying with Nvidia

[Tama00]

People have already got opensource drivers for nvidia and ati. Make AMD incomplete drivers open source will only contribute so far.. Sure its a step ahead but its more of a babies step.

All i can see their efforts but to use is in tweaking the already exsisting open source code or using the already existing open source stuff to tweak their open source code lol

still incomplete.

Re:Im staying with Nvidia

[Oblong_Cheese]

There is a world of difference between the reverse-engineered, 2D-only open-source drivers currently available for nVidia and ATi, and the straight-from-the-horses-mouth, full 2D/3D open-source drivers that could be provided by nVidia and ATi.

OT: Linus did NOT rewrite Minix!

[sgtrock]

Linux started with a guy trying to fix Minix.

The last one is the most interesting, since fixing Minix ended up meaning completely re-writing it because (at the time) the license didn't allow redistribution of modified versions (only patch sets, and those were growing unwieldy).

Sigh. Can we please stop spreading this particular piece of misinformation? From no less an authority that the guy who wrote Minix, Andy Tannenbaum himself:

Finally he asked me if I thought Linus wrote Linux. I said that to the best of my knowledge, Linus wrote the whole kernel himself, but after it was released, other people began improving the kernel, which was very primitive initially, and adding new software to the system--essentially the same development model as MINIX. Then he began to focus on this, with questions like: "Didn't he steal pieces of MINIX without permission." I told him that MINIX had clearly had a huge influence on Linux in many ways, from the layout of the file system to the names in the source tree, but I didn't think Linus had used any of my code. Linus also used MINIX as his development platform initially, but there was nothing wrong with that. He asked if I objected to that and I said no, I didn't, people were free to use it as they wished for noncommercial purposes.

Evil blobs

[CustomDesigned]

Even without downloadable firmware, any card or motherboard device could have evil permanent firmware. Downloadable blobs don't add any additional security risk beyond what the public internet presents. As long as the card hardware enforces memory windows and PCI protocol, and the kernel driver is secure, an evil firmware blob is limited - just like worms on the internet. Granted, a blob that exploits a security hole in the kernel driver or PCI hardware in conjunction with some means of propagating over the internet would be one killer worm...and I doubt anyone has done penetration testing on either. Just like sendmail, it will take the first damaging worm before PCI (or FutureBus) security is taken seriously.

GPU's merging with x86

[ichigo 2.0]

Open sourcing the drivers actually make perfect sense, when you consider that AMD and Intel are working on merging the CPU and GPU. When GPU's can be accessed through the x86 ISA, there will be no need for drivers, so handing over the source to soon-to-be legacy video card technology is the first step on the road. Intel also sees this, which is why they're doing the same thing. It will be interesting to see what Nvidia does to keep themselves relevant, will we get a third CPU manufacturer or will they stick with the sinking ship of discrete GPU's.

The final step in the evolution of Fusion is where the CPU and GPU are truly integrated, and the GPU is accessed by user mode instructions just like the CPU. You can expect to talk to the GPU via extensions to the x86 ISA, and the GPU will have its own register file (much like FP and integer units each have their own register files). Elements of the architecture will be shared, especially things like the cache hierarchy, which will prove useful when running applications that require both CPU and GPU power.

Re:So, I guess next time I recommend PCs using ATI

[Chandon Seldon]

Be careful. If this wasn't ATI it might be OK to trust them, but ATI has had a "commitment to Open Source" since 1999, and as a result we have really, really bad proprietary drivers for Linux.

My response to this announcement is: I will try even harder to buy Intel graphics over ATI, until a source release happens. They will get *nothing* from me on an "announcement" this time - this isn't "trick me once" or "trick me twice", we're closer to "trick me 18 times" at this point...

Once they really release Open Source drivers, or (even better) full interface documentation for their cards, then I'll consider their products again. But, until they actually do the whole thing this just looks like more lies.

Re:OT: Linus did NOT rewrite Minix!

[Bent Mind]

I'm a bit confused as to your point. Linus did start with Minix. Your Andy Tannenbaum quote confirms that. Andy also states in that quote that "...MINIX had clearly had a huge influence on Linux in many ways, from the layout of the file system to the names in the source tree...". No, Linux didn't have any of the Minix code in it. However, this quote does not refute the OP point. The OP said "...since fixing Minix ended up meaning completely re-writing it because (at the time) the license didn't allow redistribution of modified versions (only patch sets, and those were growing unwieldy)." Taking the concepts from Minix and using new code to express them would support both of those quotes. The Andy Tannenbaum quote does not refute the OP's point.

Re:OT: Linus did NOT rewrite Minix!

[sgtrock]

Re-writing it assumes that he started with Minix code. Tannenbaum states very clearly that in his opinion, Linus did /not/ use any Minix code. IOW, Linus used just as much in the way of concepts from Minix as he would have from any Unix variant.

Not just an edge case..

[Clith]

Knowing that a certain brand of video card has open source drivers would certainly make me buy them, even if I don't use them for Linux right now. So this could have a moderate effect on purchasing decisions. It's a checkbox item, if you will.

Re:So, I guess next time I recommend PCs using ATI

[OldHawk777]

Thanks for the history lesson, I will remember to confirm promise and performance on delivering GPL+ open-drivers and full disclosure of products' interface documentation.