Forgot your password?
typodupeerror
X GUI

XIG Releases Commercial OpenGL X-Server 91

Posted by CmdrTaco
from the acceleration-never-hurts dept.
Ansgar Philippsen writes "XIG has released a first version of their hardware accelerated OpenGL X-Server. They offer a gamers edition (full screen only) and a professional edition (all goodies included). They support a wide range of chipsets. I would be very curious to see some comments if anyone has tested this product, especially under non-gaming conditions, e.g. a scientific OpenGL app running in a window. Additionaly, a comparison to XFree's upcoming 4.0 version will be very interesting. Unfortunately, I could not find any details on their implementation. XFree and PrecisionInsight have released a great amount of detail on their DRI/GLX implementation, and it sure would be nice to see how XIG did it." Interesting points: This is of course not open source, and at $99 for the game version and $250 for the pro version, its not cheap, but it does support several cards that aren't supported under XF86 (like my Number 9 Rev 4 for example).
This discussion has been archived. No new comments can be posted.

XIG Releases Commercial OpenGL X-Server

Comments Filter:
  • It may support that number nine card that I've never heard of, but I didn't see any 3dfx, nvidia, or matrox cards on the supported list. It seems strange that XIG would claim a "gamers" edition without *any* of the most popular 3d gaming cards out there being supported.

    Jeremy
  • by Anonymous Coward
    Unfortunatly, this release comes at a great expense to the free software community.
    XiG, orignally founded by ex-xfree developers, has hired away four of Xfree's most proficint developers, including Dirk.
    This has been a great loss to the Xfree project, and has been the reasons for the continied delays in the XF864.0 betas.

    So, It looks like we'll all be using XiG sooner or later.. Which is probably for the best.
  • Hrm, still no Diamond Viper 770. Under some other operating system, the drivers are superb - unfortunately, under Linux, we dont have the direct memory access from nVIDIA yet. Basic support under X is fairly decent though. Looking into my crystal ball, I see that I wont need to keep a Windows machine around to satisfy my Quake habits. (1600x1200 60FPS just ROCKS :)

  • It looks very interesting, but it's a Bad Thing that it's closed source :(
    Let's hope the XFree people learn all kinds of interesting stuff from the non-free server, so we can all enjoy a good opensource licensed OpenGL Xserver :-)

    Emphyrio
  • This gets posted every time anything related to X come up. Where is your proof?
  • it will be easier for xig to learn from xfree than the reverse
  • I have to agree! All the major news this year seems to have been about nVidia (personal preference) or 3dfx. I wouldn't consider running anything else for gaming...

    I guess they figure that everybody with a "Normal" card will be using XF86 instead...

    Bummer.
  • This reminds me of the situation with the LISP-machines in the early 70'th, and later the proprietary Lucid Emacs (AKA XEmacs).
  • check out MetroX from MetroLink http://www.metrolink.com.
    Although closed-source and non-free (both beer and software) it's damn good.
  • The Rev. IV is already supported in Xfree. See Number Nine's support page [nine.com] and the the XFree86 3.3.5 release notes [xfree86.org].

    Too bad 9 is bankrupt. The i128 series was and is a great card.

    -jwb

  • Hi all, If you want to run a Rev 4 under XFree86 I think you can use the T2R (Ticket to Ride) drivers. I dont think you can run games with it though. I havent done this but while I was working for NINE I heard this was possible. Since the Rev 4 is based on the T2R chip there isnt a problem. However, there is no accelerated 3d. I have used there beta product and even got it running under Linux and have run Q3 for the Rev 4. Hey, its not 3dfx but that is more a function of the hardware than the software. I think Xi has a pretty good product personally. -Thanks and happy new year, -Graphicsboy (xNiner)
  • Well, perhaps this is not proof, but I noticed this when looking at the XFree86 site.... The last update to this site and the last beta-relase were in the beginning of September. However, the site claims to release a new snapshot every 6 weeks or so... Has nothing happened since the beginning of September? Kind regards, Mark Wormgoor
  • Now that $Linux$ is such a hot commodity, more and more commercial software companies will start releasing products for the market and hiring away open source developers.

    - The software won't come with source

    - The software will be expensive

    - RMS and many others will BITCH BITCH BITCH

    Congratulations - you wanted Linux to be popular? Well know it is, and the rules are going to change. All this "free stuff" is slowly going to start going away ...
  • What about the next pre-release xfree org would release? It should be released before the end of this year... that's in 4 days.
    Anyone know if they're gonna make it?

    see xfree [xfree.org]
  • by bbk (33798) on Monday December 27, 1999 @05:25AM (#1442963) Homepage
    Currently, the utah GLX project (glx.on.openprojects.net) has GLX support from Matrox G200 and G400 cards, ATI RagePro (and derivatieves), and some (not in the main source t ree) support for S3 virge cards.

    The speed of the project is amazing - after gX00 support was implemented, they got the ATI driver running in under a month.

    It's completely open source, and the only thing thats stops cards from being supported is the availibility of documentation. I'd rather see card manufacturers support this effort (some have - Matrox has a press release out touting the success of GLX and of opening their card specs- http://matrox.com/mga/press_room/lat_press_rel/G40 0_Linux.htm )
  • The problem with the Revolution IV under XFree 3.3.5 is that it is not 3D accelerated. It is more or less a generic SVGA card. Only the 2D acceleration is there. OpenGL is not that and cannot be gained by Mesa for the i128. You can't run quake 3 on it at all. It is the one thing that is sad about the card. Ticket to Ride is the only thing that works with the SGI Flat Panel, so I'm stuck with it for the moment, unless someone else knows a card that supports an OpenLDI connections.
  • Short delay in releases

    [December 1999]

    The XFree86 Project has been working very hard to get the 4.0 release out the door. It is taking a little longer than expected so we will be releasing the next pre-4.0 snapshot (3.9.17) before the end of the year. We expect to release 4.0 about two months later in mid-Q1/2000.

    XFree86 3.3.6 will be released in parallel with 3.9.17 as well.
  • Last I heard Dirk was still working for SuSE [suse.com]. At least, that's what it says here [xfree86.org].

    Chris
  • Congratulations. You've written something with no coherent meaning whatsoever.
  • Also: MetroLink is a big supporter of the XFree project. If you get MetroX, let them know that you appreciate their support!
  • by Greyfox (87712) on Monday December 27, 1999 @05:44AM (#1442969) Homepage Journal
    XIG's been up to their proprietary antics for years. I've heard horrible things about their support and software in previous /. stories. The only way you can expect a closed source package to survive is if 1) you offer considerably higher quality than any of the available free packages and 2) if you fill a niche where no free packages exist.

    In my experience, any time you have comparable open source and closed source software, the open source code will be of higher quality. Also in my experience, it's rare that any niche where no free packages exist stays that way for long (Though there are exceptions.)

    My philosophy is simple. I buy hardware to which free drivers are available. I do not buy hardware for which I'd have to pay extra for drivers for my OS. If a hardware manufacturer wants my money, they damn well better make their specs available. Most of the top-of-the-line hardware manufacturers are starting to realize this, so I'm not hurting for choices (I hope to pick up a Matrox G400 next. :-)

  • This is *very* interesting. I've been following their 3D pages for a couple of months now -- once I bought v5 AcceleratedX and I could *swear* there was mention of nVidia, 3dFX and Matrox support being some of the first.

    They also have a new mention of a tradeup from v5 to 3D -- that wasn't there before.

    Of course, with a Viper V770 Ultra it doesn't do me much good right now...

    -chill
  • Congratulations - you wanted Linux to be popular? Well know it is, and the rules are going to change. All this "free stuff" is slowly going to start going away ...

    *sigh*

    When are all you neophytes going to learn? Free, as RMS has explained a billion times, refers to the freedom to make changes to software as the user sees fit. It does not refer to the cost of sofware. You know, the old free speech vs free beer that RMS explains in the GPL. I suspect that you have never read the GPL or the voluminous debates regarding this distinction; or perhaps you have decided to ignore everything and disseminate your own FUD.

    If companies release commercial software and allow users to modify the software, I'm sure RMS would not have any problems.

    How you got moderated to insightful is beyond me. Ignorant would be more appropiate.

  • Open Source is a wonderful thing in that it gives people the ability to make their own changes -- make the program better themselves.

    They key is *better*.

    XiG competes with XFree not in cost, but in performance. I purchased XiG v5 even though XFree was up, running and working fine. Why? Because AcceleratedX smokes XFree in screen updates on my system (TNT2 card, dual P3-450).

    The biggest visual cue is logging out from KDE -- I can watch the screen paint in 1280x1024x32 with XFree when it does the full-screen mask just before logging out. Takes a full three seconds. XiG blinks and is done in less than one.

    Performance is what matters. If XiG can't get it's act together with nVidia, Matrox and 3DFx then they won't sell too many packages. If XFree handles those, it's going to be out in front.

    You don't need to boycott XiG if they can't get those chipsets and can't do it better than XFree. It won't be necessary.

    -chill
  • 3dfx, nVidia, and Matrox cards are already working in XF86, so there is no point in devoting the time, nor the energy into developing an accelerated 3d server. I don't know who will buy this. If you pay more for the software for a card...you got something wrong in your head!

    $100 for a voodoo3 2000 already running Quake3 quite nicely, or $100 for a *commercial* 3D server...do the math.

    bye,
    -jimbo
  • The link above must be to something else. If you got to This card link [xig.com] it has all the manufacturers and the cards supported along with whether or not they support gamma correction.
  • All this "free stuff" is slowly going to start going away

    Not quite. The "free stuff" will always be around.

    Yes, there will be more new non-free non-$0 stuff around, but it won't replace the free stuff.

    IMO it's a "if you don't like it, don't use it" thing - I personally think it's a good thing.
    While I prefer free software, I prefer having a non-free tool on a free OS over having to use a non-free OS to get a task done, or to play the latest great game.


    I think
    dd if=/dev/zero of=`fdisk -l /dev/hda |grep FAT |sed -e "s/ .*//"`
    is somewhat harder if we insist that everything has to be GPL.
  • I see your point on the "gamers" Issue, but they're talking about a full OpenGL implementation, which is *not* supported by 3dfx, at least last I checked it didn't. It uses Glide which partialy implements OpenGL standards.

    The matrox cards just recently got support for full OpenGL implemntation and well the nvida cards....that's their own fault for not supporting such a nice beast =).

  • The first link is to the cards that are supported in 3d-accelerated X, and the second link is the cards that are supported in Accelerated-X.

    Two different programs, and unfortunately the OpenGL version has some serious support shortfalls. (so far, anyhow)

  • umm xig isn't GPL either
  • XiG has hired away four of Xfree's most proficient developers including Dirk.

    I think something like that would be announced on the XFree86-devel list... (which hasn't happened.)
    This is just a rumor.

    This has been the reasons for the continued delays in the XF864.0 betas.

    The reason why XFree86 is progressing somewhat slower than most other open source projects is, IMO, the fact that its development is too much closed up.
    Subscription to the development list only after approval, no write access to the CVS trees, ...

    In the long run, XFree86 makes it very very difficult for new developers to get in; any developer leaving for whatever reason can't be replaced as quickly as it works in other projects.

    So, it looks like we'll all be using XiG sooner or later..

    Definitely not. If XFree86 gets stalled, there will be a more open fork, even if I have to start it myself.
  • The commercial closed-source software has and will always have strong competition from the Open Source community. Consider the advantages of XFree86 and other OSS efforts:
    • costs nothing
    • comes preinstalled with most distributions
    • Open Source hype
    If XiG and other companies want to compete on this playground, they'll have to offer real advantages, such as better hardware support, and even that doesn't appeal enough to some people to justify a price higher than that of a full-blown Linux distribution.

    The most dangerous threat to OSS I can see is that some of these companies are hiring away good OSS developers to work on closed-source products.

  • Well XiG does both. As of now they are higher quality than XFree, and XFree does not yet have a finished 3D implementation.
  • umm, that wasn't the poin of the post. Try reading next time.
  • Bahhh..

    You're full of poop. The reason for the continued delay is that XFree4 has so many new features to debug:

    * SilkenMouse
    * Multihead __&__ Xinerama
    * DRI/DRM - 3D
    * XAA and DGA completly rewritten
    * New modular server.. everything is dynamic load
    * Integration of new X sample implementation code
    And alot more..

    Panaflex
    "Sure.. I believe in black helocopters as much as the next guy"
  • Well.. the 3D support is either GLX or DRI. Doesn't have much to do with the base XFree code right now!

    Pan
  • by Kid Zero (4866)
    It's that 64meg Ram free above what the OS takes that hurts. Granted, I've got it, but still, If I want to use up all my memory, I'll keep Win98.
  • I don't completely disagree with you, but I work for a company that uses XIG's Xserver (in a product) for the simple reason that setup for the customer is much much easier than some of the other alternatives like MetroX/Xfree86. You don't have to worry about chipsets/Ramdacs/etc. Just pick the name of the card you have, amount of memory and you're done. No hacking on the config files, no tweaking the monitor timings, etc. This probably makes it less flexible for hacker types since you can't tweak things to your hearts content but in a commercial environment 99.9% of customers don't care. They want the product installed and running and XIG buys you that.

    As for support I've only delt with them once and it was a very good experience. It did sound like their support dept was pretty lightly staffed so it may have been that I received better treatment than most since I was one of their commercial customers.

  • I agree - performance is a key issue. However, 2d/3d acceleration has been available for Nvidia based cards for quite some time now (since 6/99 I believe). The old super-slow KDE logout hasn't been around since. Even at 32 bit. True, there is no 3d acceleleration at 32bpp, but, it'll happen. Nvidia has since opened their drivers and specs, and I suspect that the next set of drivers to be released will offer much better performance.

    Don't give up just yet...
  • Don't you recognize the automatic complaint-generator when you see it?

    -----------

    "You can't shake the Devil's hand and say you're only kidding."

  • by pneurk (122261)
    I have used XiG's X-server in the past when I have a video board not supported by xfree86. The situation is reverse this time, very unusual... My Matrox G200 is supported by xfree86 and the glx project (once they make a stable release), but not by XiG. This is completely opposite from what I have seen in the past. Is the open source movement gaining enough attention to surpass commercial software in this respect?
  • I've got a TNT2 running OpenGL, and I must say that, although it's a hell of a lot better than software OpenGL, it's also a hell of a lot worse than Windows OpenGL. I don't know when XFree 4.0 is going to be out, and when Mesa is going to support their DRI, but until then, getting a commercial server is an option... simply because there is no Open Source alternative.

    -----------

    "You can't shake the Devil's hand and say you're only kidding."

  • Yes they are very closed. It's a shame. The DRI project is more open than than XFree, and it is integrating XFree, DRI, and Mesa. Check it out:

    http://dri.sourceforge.net

    Roger
  • If you look closely, most of the cards listed are "professional" OpenGL cards. None of those are the cheap cards most gamers overclock to get an extra thousandth of a frame per second. How many "gamers" have an Oxygen?
  • I've used Accel-X and Metro-X... I went back to using XFree. The only good thing about a comercial X server is that you have someone who's obligated to fix your bugs. They may be able to get chip specs faster as they can eat NDAs with no problems, but that doesn't amount to much -- if the OEM doesn't want to release the specs to programmers, then don't buy their "crap".

    All of the comercial servers appeared to be bloated and overly complicated. That's not to say XFree isn't either, but "everyone" knows how to deal with XFree; and XFree offers alot more control over the display setup (manufacturer specs are often lies.)
  • They want to make money off of all of the linux gamers out there that may be frustrated that they can't seem to get quake to work in X windows with their less then Voodoo 3 or GeForce 256. If their product was open source, then anybody could download the source code and save themselves 99 bucks.


    -----

    I see this from another perspective. I have a TNT1 that works great for Q1 and Q2 in Windows. Unfortuantely, the nvidia accelerated support sucks, so instead of opting for a TNT2 ultra or a Geforce I went with a Voodoo3 for Q3A. Why? As I see it, 3dfx has supported linux more in the past, and as far as nvidia is concerned, too little, too late, they should have released their drivers a long time ago. I'll spend my money on the companies that support linux, nvidia be damned!! Besides ... what's $100 to play Q3A in linux .. :)
  • Matrox isn't exactly brimming over with published specs. They have yet to tell anyone (aside from XiG) about the MGA-TVO (aka. maven) chip used for TV output and the secondary RAMDAC on the dual head G400's. They also have not released the programming specs for the WARP enigne(s). They DID provide WARP microcode for the GLX project, but the GLX team doesn't know what they are pushing into the WARP engine.

    Note: I don't hate Matrox for holding back the WARP docs (that's Gx00 internal microcode,) but sitting on the TVO docs is just stupid and counter productive -- we will figure it out.

    (I'm fairly certain they are sitting on them just like they did the G100 specs. Read the last developer newsletter for the word on the G100 specs.)
  • Don't hail Matrox too highly... there's still a large amount of documentation they are sitting on. They just released the G100 specs a few weeks ago -- the reason for not posting it with the others? none. (I don't guess there was much developer demand.)

    The entire community is greatful for the microcode Matrox provided, but we'd be much happier with the WARP instruction set (so would their competitors :-)) Matrox is making this the biggest PR move of the century.
  • by Anonymous Coward
    What? XiG wasn't founded by any ex-xfree86
    developers. And I'm not aware of them hiring
    any XFree86 developers EVER. Dirk Hohndel
    works for SuSE who pays him to work on XFree86
    full time. The "Missing details" post is complete
    nonsense.

    Mark Vojkovich (mvojkovi@XFree86.org)
  • Purchasing SGI OGL certification wont help you, its my undertstanding (I'm not a OGL person) that the certification is limited to a particular setup (OS/hardware/driver). ie you certify a particular implementation, not the client side API (Mesa).

    We live in a free country (free as in speech, not beer). It up to the individual to make the make the decision about where he wants to spend his money (for me, my time is way more important than money).
  • Na, that's a common misconception. Linux can map memory space which doesn't take up your physical memory. It may look like you're losing 64M, but that is just a mapped memory block reserved for the X-server. It doesn't use your physical RAM.
  • Your an idiot.
  • Hey Smart Guy!

    It should be "You're an idiot".
  • Still no descent RivaTNT suport in linux, ahhhhhh......., My Q3 framerate is just horrible.
    -entropy
  • It's nice that somebody finally made OpenGL work right with X, but they could charge just a little less for it. For that cost, I think I'd be more inclined to use WinNT/2k for professional uses. Sadly, win has more apps, and is about the same cost, if not cheaper. If they cut the cost down a little (maybe 25% or so), I'd definately purchase a copy, but for 300 bucks, it isnt worth it. Heck, last time I checked there were about 3 opensource (free) projects working on porting OpenGL to X. I think I'll just wait, maybe even contribute, for those projects.
    =======
    There was never a genius without a tincture of madness.
  • Actually, Glide exposes the 3dfx chip rendering functions and register set as an API. Mesa can use Glide to speed up rendering, but Mesa supplies the OpenGL compatible functions under Linux. Whether a card or chipset manufacturer has a windows based OpenGL driver or not means nothing in the Linux world. What matters is how much chipset information is available for the graphics chips. Is it enough to accelerate the 2d functions only or only enough to get the 3d engine rendering polygons but not the accelerated geometry setup or T&L engines under Linux, *BSD, etc...? I don't think so, but several companies still seem to think this way.
  • There's a good review here. [eunuchs.org]

  • [...] XFree4 has so many new features to debug:

    * SilkenMouse

    What's that?

    * Multihead __&__ Xinerama

    Will this work with dual-head cards like the Matrox G400?

    * DRI/DRM - 3D

    Direct Rendering Interface, right? Woohoo! No more running Q3A as root! :-)

    * XAA and DGA completly rewritten

    What advantage does this give us?

    * New modular server.. everything is dynamic load

    Again, why? What will we need loaded dynamically?

    * Integration of new X sample implementation code

    What's this?
    Will there [ever] be support for font anti-aliasing or hinting? Is it even possible with backward compatibility? I really miss the anti-aliased fonts from Acorn's RISC OS [riscos.com] machines, they made low resolutions usable and high resolutions a joy. (Note: don't equate this with Windows' "Font Smoothing" - basically a gaussian blur - because Acorn's system included hints and skeletons in the font file format in order to add information to the display, not take it away.)
  • Yes, albeit dated 10/1998. I hope you get ahold of the new XiG software so you can do an update.
  • Have you ever tried to get X86 Solaris to work on a laptop and use the best graphic modes? If so, then you'll appreciate XiG drivers.

    There's money to be made in supporting graphics cards and writing x-servers for intel unix platforms where business is done. Maybe not so many linux developers or gamers are interested in buying non-free x servers, but you can bet that there are enough people interested to keep them profitable.

    I wouldn't be surprised to see some bundling agreements like redhat had with metro-X a while back...

    It's a matter of time before we see PC software that comes on a bootable cd running linux that autodetects the hardware and just runs without the need to install or configure. I'd bet that a game developer trying to use this model would want to have the best/fastest drivers for graphics and would be interested in striking a deal with XiG.

  • It's not about the $99. It's the fact that I'd be willing to bet that all these cards shipped with Win9x drivers, and if they didn't I bet it's an easy, free download.

    But if we want to use Linux, we're expected to pay $99 for the privilege of using the hardware that we own ? Ummmm... no.

    I am careful to buy hardware that supports Linux because that's what I have to do right now. But in a perfect, or even a reasonable, world, I would buy the hardware that did the job for the best price, and just EXPECT that it supports Linux.

    Right now, hardware manufacturers are only beginning to start hearing the message from their customers... 'We want to have a choice in operating systems, and fuck you if you're going to penalize us for not choosing to run Windows'. I look forward to the time when those $5 video cards that you can pick up at discount computer shows all have XF86 compatible drivers on their little CD.
  • i may sound like a closed source goon, but if they don't want to release there specs to open source programmers its there damn business, remember? sorry, but they have the right to warry about open source, we have to convince them on other merrits than gaining the linux market, because frankly i bet most of them don't care for this sliver of the market for another 2x's the work.

    alough your other points about xf86 are very valid, xf86 *is* the standard, everyone knows how to deal with it, knock some points for open source.
  • by TLL (130099)
    I am fine with Xi's product, in fact, I called and had mine shipped early. 3D Accelerated-X uses a kernel module that supports a few standard GART interfaces (Intel, VIA, ...) and uses a character device (/dev/xsvc i think). The product gives me better frame rates in Q3A than the latest S3 reference drivers for my Savage4 GT/Pro. It uses SGI Licensed OpenGL 1.1.1, unlike the Mesa Project's OGL-like API. Trust me, when it comes right down to it, Mesa is still crap. Also, the $250 price is /extremely/ cheap for a product of this quality, the companies I have worked with pay $12k for their SGI workstation implementations. The 3D Accelerated-X would probably cost this much if they had paid for the OpenGL conformance tests. Xi cannot be expected to support every card on the market, they will have a licensing problem. The nvidia, 3dfx, matrox, and ati drivers are all GPL/open source. If they release open source and closed source drivers in their product, they will slaughter that beautiful code. And for all of you kiddies saying stuff about linux going big and companies releasing commercial software, check out Xi's page, they've been around for a while.
    I have been a linux user since RHL 1.0 in 1991, and I really hate to see linux go down the drain. With such hateful users it _will_ do this, so please hold down the flames, linux is gonna keep goin without all the immature children sending bitch letters to Mr. Dvorak... ;)
  • Uh, 'scuse me? There are a few of us losers out here who'd like to play games, too. Saying that only 3dfx, nVidia, and Matrox are important because they've gotten the most attention is like saying only Windows is important because it gets 99% of the attention. I've got a Diamond FireGL 1000 Pro with a 3dlabs Permedia 2, which I would like to use to play Q3A in Linux. But I have to reboot into Windows, which annoys me.
  • by Kyobu (12511)
    The V770 is supported, including gamma. http://www.xig.com/Pages/CardMfgrDiam ond.html [xig.com] has it right at the top.
  • What I don't get is why Diamond, Creative, etc. don't just release their own closed-source drivers (X servers). If they did, they would lose nothing except a little money to pay the programmers, and they would gain a lot of customers. I get why they don't want to release OSS drivers - competitors could reverse-engineer the hardware from the software. But if they would just include Linux drivers, they would get lots of fans. Then they would succeed or fail depending upon their hardware's merits instead of whether there is any support from your and my favorite OS. RMS types still wouldn't be happy, but oh well.
  • The G200 isn't supported by 3D Accelerated-X because of the lack of documentation for the G200. Some 3D docs are freely available from Matrox. However, one of the main performance gains for 3D hardware acceleration is in a bit of documentation not disclosed by Matrox, even under a Non-Disclosure Agreement.

    Free software developers and users will find that a G200 with some 3D hardware acceleration is faster than a G200 with no hardware acceleration, but Xi Graphics has always aimed for the maximum hardware acceleration. With missing information, this isn't possible and would make both the G200 and 3D Accelerated-X look relatively pathetic, and probably attract justifiable criticism if we charged full price and delivered a small fraction of the cards possible speed under Windows. That is clearly not a problem for free software.

    Cheers, JeremyC.

  • The cards supported represent (most of) those chip makers that have provided databooks for their 3D engines.

    There are no databooks available for the 3Dfx 3D hardware; there is a 2D databook, so Xi Graphics has been able to offer support for some 3dfx hardware in 2D/X mode. AFAIK, XFree86 relies on using a binary library provided by 3Dfx (GLIDE), which is not open source/GPL/whatever. The IP remains in the hands of 3Dfx and free software offers an interface to the real binary proprietary driver.

    There is no support for the Matrox boards because of missing critical data for the 3D hardware. Some of the 3D hardware is openly documented by Matrox, but certain portions that give useful speed increases are not described. If 3D Accelerated-X claimed support for the Matrox boards using partial documentation, the performance would be lower than should be available, users would be disappointed. This is not a problem for free software developers; they can offer suboptimal software and people are grateful to have their card supported in any way at all. It is probably unacceptable for us to claim support and then deliver a small fraction of the speed that the card should deliver.

    Similar properties apply to nVidia. The "free" X Server has been developed by an nVidia employee. Although the source code is available for XFree86, the data book is not available. Consequently if there is a bug or if someone wants to extend or improve the nVidia X Server, e.g. for higher speed or to fix a bug, there is no reasonable way to do so. The availability of source does not mean that you control the software; the IP remains in the hands of those with the databook.

    Cheers, JeremyC.

  • Xi Graphics was born as X Inside in 1994 in Denver Colorado because that's where I was living then. I have never been part of XFree86, though I cooperated with Steve Matewski(sp) to produce an early version of an XFree86 X Configuration utility as part of the Prime time freeware collection, some time around 1990.

    Of the other two co-founders, one was never involved in any software development at all and is not a programmer.

    The technologically significant co-founder was also never a member of XFree86, but he contributed the sources that XFree86 used. Thomas Roell was the developer of X386, the free software X Server before XFree86. He remains the primary developer for the company and has developed most of the ideas that form the basis for XFree86 4.0, several years ago.

    We have not hired four XFree86 developers at any time in the company history. AFAIK, we have interviewed several XFree86 developers over the five years we've been running, but we've never hired any. Our X Server/graphics chips hires are usually made from hardware driver developers that we train to understand graphics chips. If anything, we add to the pool of graphics hardware expertise.

    Cheers, JeremyC.

  • I didn't mean to imply that they were the only ones that are important, but they are the most popular.

    PERSONALLY, I would only use either a nVidia, Matrox (oversight in my original post) or 3dfx card for gaming, but that's just me. In terms of performance, they seem to have won the game, as far as I can see.

    In a perfect world all cards would be supported, but if you've got to pick and choose, pick the most popular, best and most recent cards.
  • Freedom to change the software? that is a bogus argument but is does precisely demonstrate why Accelerated-X is a product. You should also be aware that XFree86 is not GPL'ed code, though it is freely available.

    I am an ex-academic. I appreciate that rms' ideas might encourage diversity and novelty in creating algorithmic solutions. However, most of an X Server these days isn't novel solutions that can be patented, but technowledgery. It is knowing which registers to poke on a graphics chip, in which order, at which time intervals, to get a correct and clear image. This is not IP in the traditional sense, and I don't think the propogation of this knowledge would do much for any other field of software development.

    X Servers and graphics chips are intrinsically pretty boring. You have to follow someone else's fifteen year old ideas about what graphics should be like and marry it to an insanely complex new chip that was developed a few weeks ago with inadequate documentation.

    If the sources for Accelerated-X were made available, it wouldn't significantly add to the IP value of free software. The registers to drive a Matrox chip faster can't be applied to make an ATI or any other graphics chip faster, much less a non-graphics chip. They can, however, be applied to make an XFree86 implementation of an X Server for a Matrox chip yet faster.

    All that would be accomplished by making Xi Graphics source code available would be to give a transient speed increase to XFree86 at the expense of the loss of a few dozen jobs. At the end of the process, you'd still have an XFree86 team that generates slower X Servers that fail the X test Suite and that are less robust than Accelerated-X.

    What you'd be missing is a source for an X Server implementation that makes some effort to be faster, better and offers timely support.

    I do believe that making novel algorithms available can improve other efforts. For example, there is little doubt that public inspection of crypto code is valuable. However, public availability of which register to poke is not, in my mind, of the same order.

    It is however, valuable. If your machine crashes because some programmer can't be bothered to correct a bug, you may lose hours of work. If you can't correctly display an image of some project on which you are working, you may cause millions of dollars of lost productivity when a pipe or wire is sent the wrong way because it was drawn incorrectly because a programmer ignored the X specs or thought that their opinion about the significance of meeting the spec was important.

    I do believe that open inspection of security code and OS code can improve it. I do not believe, with practical evidence, that open inspection of complex register poking code for graphics, has much to recommend it. I don't think rms' model for software improvement applies usefully to improving register poking on graphics chips.

    XFree86 already demonstrates that the principle isn't working. I can crash XFree86 far faster and more easily than I can crash Accelerated-X. I can make XFree86 display incorrectly far more easily than I can make Accelerated-X display incorrectly. These are not the result of algorithmic improvements unique to Accelerated-X. They are the result of significant practical effort to find the best set of registers to poke and of a QA process intended to find errors before a user sees them. IMO, this is entirely not the point of GPL.

    AFAIK, the point of GPL was to find a better algorithm, not a better register. The GPL makes no claim about the quality of the code it is to produce; that has been inferred and is the subject of religious adherence to creed rather than repeatable practical measurement. In any case, as I started off pointing out, XFree86 is not GPL'ed... you probably have to apply a different argument for non-GPL bt freely available sources ;-)

    FWIW, I can contribute anecdotal evidence about Linux' stability versus Windows. I can contribute practical, measurable, repeatable evidence about XFree86 versus Accelerated-X stability and correctness. Two different types of data. I'd welcome a practical, repeatable and measurable way to demonstrate Linux reliability versus Windows, rather than refutable anecdotal evidence. I'd love to see an independent group test the verifiable behavior of XFree86 and Accelerated-X Servers for stability and correctness.

    Cheers, JeremyC.

  • You can find it at http://linux.3dfx.com/open_source/drivers.htm Its open source, and 2D performance is the best I have seen so far. I am still evaluating 3D performance, please post you comments.
  • The web site is put together by David Methvin, President. methvin@xig.com

    BTW, while David intends a friendly tone with the "This'n'that" type abbrevs (ever heard of Toys'R'Us, who even got the R backwards?) most of his typos are intended. He does also reply to email sent to the webmaster and he puts up a prominent mailto: for the webmaster on the front page, just below the spot you discuss. It is not exactly a secret.

    I think that coming clean about the quality of the code should be applauded. I've bought hardware products where after three months I'm still using beta drivers. If I'd known, I'd have been unlikely to buy the product at that point. As it is, I'm now unlikely to ever buy anything from that company again, because they weren't upfront about their code quality.

    Cheers, JeremyC.

  • Things like top aren't too smart about how much RAM things like X takes. It includes all mapped physical address spaces, as well was regular RAM... That means if you have a 32MB video card, you're looking at most programs reporting you have an X server that's bloated to more than 32MB. For example, in my case:

    PID USER PRI NI SIZE RSS SHARE STAT LIB %CPU %MEM TIME COMMAND
    4252 root 10 0 44732 2975 872 R 0 3.3 9.2 8:30 XFCom_Rage128

    There's no way X is taking 44MB of RAM on my system, in fact, it's probably taking much less than even 12MB

    The only thing X can safely be accused of is being slow for graphic intensive screen redraws. Sending everything through pipes has some real disadvantages for things like this, which is why these glx projects are pioneering "direct rendering" and by-passing the X server altogether.

  • >>>Currently, the utah GLX project (glx.on.openprojects.net) has GLX support from Matrox G200 and G400 cards, ATI RagePro (and derivatieves), and some (not in the main source t ree) support for S3 virge cards.

    >>>The speed of the project is amazing - after gX00 support was implemented, they got the ATI driver running in under a month

    A lot of this is thanks to John Carmack( of Id fame.) He has put in a lot of time getting the ATI drivers working and tracking down bugs in the G200/G400 code. He has provided a lot of design advice and code.


Pause for storage relocation.

Working...