Re: [Mesa3d-dev] Doom3 works on R200!
Philipp Klaus Krause wrote: Does anyone have figures for the MOST time consuming parts of software 3D in the Mesa libs? Those would be the logical bits to push into hardware first. But I'm not sure I've ever seen a profile output from X to say which parts are actually most & would benefit more from acceleration than others... Even without any such data you can easily find out the most important stuff: Since It's a 3D graphics _pipeline_ you should always start by accelerating the back end: implement hardware z- and stencil buffer first. Then work your way from the backof the graphics pipeline. Yes, but where the tree branches, you still need to know which branches (Usually) consume the most time/CPU... Admittedly you should be aqble to get this from profiling the X server, but I just wondered if someone had the figures already... --- This SF.Net email is sponsored by: Sybase ASE Linux Express Edition - download now for FREE LinuxWorld Reader's Choice Award Winner for best database on Linux. http://ads.osdn.com/?ad_id=5588&alloc_id=12065&op=click -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Mesa3d-dev] Doom3 works on R200!
Does anyone have figures for the MOST time consuming parts of software 3D in the Mesa libs? Those would be the logical bits to push into hardware first. But I'm not sure I've ever seen a profile output from X to say which parts are actually most & would benefit more from acceleration than others... Even without any such data you can easily find out the most important stuff: Since It's a 3D graphics _pipeline_ you should always start by accelerating the back end: implement hardware z- and stencil buffer first. Then work your way from the backof the graphics pipeline. Philipp --- This SF.Net email is sponsored by: Sybase ASE Linux Express Edition - download now for FREE LinuxWorld Reader's Choice Award Winner for best database on Linux. http://ads.osdn.com/?ad_id=5588&alloc_id=12065&op=click -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Mesa3d-dev] Doom3 works on R200!
Rogelio Serrano wrote: On 2004-10-25 04:10:30 +0800 Vladimir Dergachev <[EMAIL PROTECTED]> wrote: If there weren't all those patents out there we might just try to develop a free graphics chip. I have thought about this (repeatedly - the idea gets very tempting after asking for the docs for the Nth time) and I don't think it is feasible to make an actual chip. By the time we are finished the world will move on. What could work, however, is to make a *board* that is capable of decent 3d. Put lots of memory, lots of bandwidth and several DSP to approximate the same level of raw floating-point power as 3d GPUs. Leave everything else to the software. The problem is getting such a beast under $1000 range. Last time I looked TI DSPs that were up to the task were rather expensive. best Vladimir Dergachev [snipped...] This was discussed in lkml a few days ago. A hardware company is considering building an open fpga based video card. Although the target is mainly 2d accel its a good start. There was a lot of discussion about off screen rendering and support for the new compositing model in xorg. You can see that thread posted on kerneltrap. I was watching that... Does anyone have figures for the MOST time consuming parts of software 3D in the Mesa libs? Those would be the logical bits to push into hardware first. But I'm not sure I've ever seen a profile output from X to say which parts are actually most & would benefit more from acceleration than others... H --- This SF.Net email is sponsored by: Sybase ASE Linux Express Edition - download now for FREE LinuxWorld Reader's Choice Award Winner for best database on Linux. http://ads.osdn.com/?ad_id=5588&alloc_id=12065&op=click -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Mesa3d-dev] Doom3 works on R200!
On Sun, 24 Oct 2004, Mike Mestnik wrote: --- Philipp Klaus Krause <[EMAIL PROTECTED]> wrote: Thank you for doing this work. We really need to get the open-source ATI driver on par with the propretary driver (both feature-wise and performance-wise). But sadly we will NEVER match it. NO SmoothVision, HyperZ docu ever The nonfree xig driver has been developed without HyperZ docs and outperforms fglrx. It's my opinion that xig uses trickery to get there FPS higher then it should be posible to under OpenGL compliant rendering. It's also true in the font and 2d(like lack of xv support) accelerations as well. Uhm... trickery? Like how? And what do you mean by 'should be posible to under OpenGL compliant rendering'? Obviously we do it, so obviously it is possible. If you mean extreme optimization of common code-paths, then you are correct. If you are implying that we somehow 'cheat', well then you are quite incorrect. We evaluate all drivers we produce. If QA finds something amis (like really slow performance in some test that should be faster), then we spend time trying to figure out where the bottleneck is, and if we are successful in determining a cause, and it's a code-path worth optimizing, we deal with it. I can assure you (though I don't think you will wish to believe me), there is no 'lets see if we are doing viewperf, and then cheat' code anywhere in our drivers. I would think playing those kinds of games would be pretty obvious to detect... wouldn't they? Didn't Nvidia get busted playing those games by some 3D testing outfit some time back? Also, I am confused by your statements about lack of Xv, and 'cheating' in 2D tests. We do, of course, support Xv just fine (have since the first Summit release). And how would you 'cheat' in a 2D test? I mean, you can test the demo yourself and see that it does actually draw stuff...? That's what Roland did I believe... I would think if Roland had found something suspicious, he would certainly have voiced them..? [...] PS: I work at XiG. I do not subscribe to this list from there though, which is why I'm sending from home, where I am subscribed. Feel free to flame me mercilessly :) -- Jon Trulsonmailto:[EMAIL PROTECTED] ID: 1A9A2B09, FP: C23F328A721264E7 B6188192EC733962 PGP keys at http://radscan.com/~jon/PGPKeys.txt #include "I am Nomad." -Nomad --- This SF.net email is sponsored by: IT Product Guide on ITManagersJournal Use IT products in your business? Tell us what you think of them. Give us Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more http://productguide.itmanagersjournal.com/guidepromo.tmpl -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Mesa3d-dev] Doom3 works on R200!
On 2004-10-25 15:11:00 +0800 Vladimir Dergachev <[EMAIL PROTECTED]> wrote: [snipped...] I see. Thats a good idea too. We just have to look for a company willing to do that. I have long been convinced that designing the hardware like this is the only way to go. Just for fun here are some numbers: Card BandWidth FillRate TrianglesOps budget for 1GFLOP DSP GB/sGpixel/s Mtriangle/s ops/pixel Radeon 7500 5.3 0.5 62.5 2 Radeon 9000 Pro 16 1.0 62.5 1 Radeon X800 XT 18.4 8.3 520.0 0.12 Note that 1 GFLOP DSP is like Pentium III-1000mhz.. Not quite, but similar. The technology is getting there (for example 14.4 GFLOP board from bittware: http://www.bittware.com/products/PCI/t2pc/t2pc_desc.stm ) but they don't even quote price for such things. Thanks for the info. This is interesting. I think i will go for a pci-x motherboard. -- Blood is thicker then water... And much tastier John Davidorff Pell --- This SF.net email is sponsored by: IT Product Guide on ITManagersJournal Use IT products in your business? Tell us what you think of them. Give us Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more http://productguide.itmanagersjournal.com/guidepromo.tmpl -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Mesa3d-dev] Doom3 works on R200!
Vladimir Dergachev wrote: > What could work, however, is to make a *board* that is capable of decent > 3d. Put lots of memory, lots of bandwidth and several DSP to approximate > the same level of raw floating-point power as 3d GPUs. Leave everything > else to the software. This reminds me of "TIGA" some years ago. They had X servers running mostly on the graphics board, X libraries actually provided merely the interface to the board's API, Martin. -- Unix _IS_ user friendly - it's just selective about who its friends are ! -- --- This SF.net email is sponsored by: IT Product Guide on ITManagersJournal Use IT products in your business? Tell us what you think of them. Give us Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more http://productguide.itmanagersjournal.com/guidepromo.tmpl -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Mesa3d-dev] Doom3 works on R200!
This was discussed in lkml a few days ago. A hardware company is considering building an open fpga based video card. Although the target is mainly 2d accel its a good start. There was a lot of discussion about off screen rendering and support for the new compositing model in xorg. You can see that thread posted on kerneltrap. I was aware of that. However, the proposition was by a company that is known for its 2d video cards (used for air control for example), so I was talking something a bit different - just have a DSP add-on board powerful enough to do decent 3d. Programming DSP is a lot easier than FPGA, the code is more portable, and, besides, there are no FPGAs at the moment that can possibly compete with ASICs that are produced by ATI or NVidia. On the other hand with the movement to programmable everything conventional GPUs are slowly turning into DSPs. It may very well be that a conventional DSP will provide similar perfomance as long as we are using all the whiz-bang features. (as opposed to using as few features as possible - which regular ASICs should still do much better than anything else). best Vladimir Dergachev I see. Thats a good idea too. We just have to look for a company willing to do that. I have long been convinced that designing the hardware like this is the only way to go. Just for fun here are some numbers: Card BandWidth FillRate TrianglesOps budget for 1GFLOP DSP GB/sGpixel/s Mtriangle/s ops/pixel Radeon 7500 5.3 0.5 62.5 2 Radeon 9000 Pro 16 1.0 62.5 1 Radeon X800 XT 18.4 8.3 520.0 0.12 Note that 1 GFLOP DSP is like Pentium III-1000mhz.. Not quite, but similar. The technology is getting there (for example 14.4 GFLOP board from bittware: http://www.bittware.com/products/PCI/t2pc/t2pc_desc.stm ) but they don't even quote price for such things. best Vladimir Dergachev -- Blood is thicker then water... And much tastier John Davidorff Pell --- This SF.net email is sponsored by: IT Product Guide on ITManagersJournal Use IT products in your business? Tell us what you think of them. Give us Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more http://productguide.itmanagersjournal.com/guidepromo.tmpl -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Mesa3d-dev] Doom3 works on R200!
On 2004-10-25 11:11:56 +0800 Vladimir Dergachev <[EMAIL PROTECTED]> wrote: On Mon, 25 Oct 2004, Rogelio Serrano wrote: On 2004-10-25 04:10:30 +0800 Vladimir Dergachev <[EMAIL PROTECTED]> wrote: If there weren't all those patents out there we might just try to develop a free graphics chip. I have thought about this (repeatedly - the idea gets very tempting after asking for the docs for the Nth time) and I don't think it is feasible to make an actual chip. By the time we are finished the world will move on. What could work, however, is to make a *board* that is capable of decent 3d. Put lots of memory, lots of bandwidth and several DSP to approximate the same level of raw floating-point power as 3d GPUs. Leave everything else to the software. The problem is getting such a beast under $1000 range. Last time I looked TI DSPs that were up to the task were rather expensive. best Vladimir Dergachev [snipped...] This was discussed in lkml a few days ago. A hardware company is considering building an open fpga based video card. Although the target is mainly 2d accel its a good start. There was a lot of discussion about off screen rendering and support for the new compositing model in xorg. You can see that thread posted on kerneltrap. I was aware of that. However, the proposition was by a company that is known for its 2d video cards (used for air control for example), so I was talking something a bit different - just have a DSP add-on board powerful enough to do decent 3d. Programming DSP is a lot easier than FPGA, the code is more portable, and, besides, there are no FPGAs at the moment that can possibly compete with ASICs that are produced by ATI or NVidia. On the other hand with the movement to programmable everything conventional GPUs are slowly turning into DSPs. It may very well be that a conventional DSP will provide similar perfomance as long as we are using all the whiz-bang features. (as opposed to using as few features as possible - which regular ASICs should still do much better than anything else). best Vladimir Dergachev I see. Thats a good idea too. We just have to look for a company willing to do that. I have long been convinced that designing the hardware like this is the only way to go. -- Blood is thicker then water... And much tastier John Davidorff Pell --- This SF.net email is sponsored by: IT Product Guide on ITManagersJournal Use IT products in your business? Tell us what you think of them. Give us Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more http://productguide.itmanagersjournal.com/guidepromo.tmpl -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Mesa3d-dev] Doom3 works on R200!
On Mon, 25 Oct 2004, Rogelio Serrano wrote: On 2004-10-25 04:10:30 +0800 Vladimir Dergachev <[EMAIL PROTECTED]> wrote: If there weren't all those patents out there we might just try to develop a free graphics chip. I have thought about this (repeatedly - the idea gets very tempting after asking for the docs for the Nth time) and I don't think it is feasible to make an actual chip. By the time we are finished the world will move on. What could work, however, is to make a *board* that is capable of decent 3d. Put lots of memory, lots of bandwidth and several DSP to approximate the same level of raw floating-point power as 3d GPUs. Leave everything else to the software. The problem is getting such a beast under $1000 range. Last time I looked TI DSPs that were up to the task were rather expensive. best Vladimir Dergachev [snipped...] This was discussed in lkml a few days ago. A hardware company is considering building an open fpga based video card. Although the target is mainly 2d accel its a good start. There was a lot of discussion about off screen rendering and support for the new compositing model in xorg. You can see that thread posted on kerneltrap. I was aware of that. However, the proposition was by a company that is known for its 2d video cards (used for air control for example), so I was talking something a bit different - just have a DSP add-on board powerful enough to do decent 3d. Programming DSP is a lot easier than FPGA, the code is more portable, and, besides, there are no FPGAs at the moment that can possibly compete with ASICs that are produced by ATI or NVidia. On the other hand with the movement to programmable everything conventional GPUs are slowly turning into DSPs. It may very well be that a conventional DSP will provide similar perfomance as long as we are using all the whiz-bang features. (as opposed to using as few features as possible - which regular ASICs should still do much better than anything else). best Vladimir Dergachev -- Blood is thicker then water... And much tastier John Davidorff Pell --- This SF.net email is sponsored by: IT Product Guide on ITManagersJournal Use IT products in your business? Tell us what you think of them. Give us Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more http://productguide.itmanagersjournal.com/guidepromo.tmpl -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel --- This SF.net email is sponsored by: IT Product Guide on ITManagersJournal Use IT products in your business? Tell us what you think of them. Give us Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more http://productguide.itmanagersjournal.com/guidepromo.tmpl -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Mesa3d-dev] Doom3 works on R200!
On 2004-10-25 04:10:30 +0800 Vladimir Dergachev <[EMAIL PROTECTED]> wrote: If there weren't all those patents out there we might just try to develop a free graphics chip. I have thought about this (repeatedly - the idea gets very tempting after asking for the docs for the Nth time) and I don't think it is feasible to make an actual chip. By the time we are finished the world will move on. What could work, however, is to make a *board* that is capable of decent 3d. Put lots of memory, lots of bandwidth and several DSP to approximate the same level of raw floating-point power as 3d GPUs. Leave everything else to the software. The problem is getting such a beast under $1000 range. Last time I looked TI DSPs that were up to the task were rather expensive. best Vladimir Dergachev [snipped...] This was discussed in lkml a few days ago. A hardware company is considering building an open fpga based video card. Although the target is mainly 2d accel its a good start. There was a lot of discussion about off screen rendering and support for the new compositing model in xorg. You can see that thread posted on kerneltrap. -- Blood is thicker then water... And much tastier John Davidorff Pell --- This SF.net email is sponsored by: IT Product Guide on ITManagersJournal Use IT products in your business? Tell us what you think of them. Give us Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more http://productguide.itmanagersjournal.com/guidepromo.tmpl -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Mesa3d-dev] Doom3 works on R200!
On Sun, Oct 24, 2004 at 08:10:14PM +0200, Bernardo Innocenti wrote: > > IANAL, but reverse engineering is perfectly legal here in Europe > and probably even in the USA if your goal is achieving > compatibility. Have to be careful - most folks doing reversing do a clean-room implementation (1 person reverses and creates a spec, another person develops based on the spec) to avoid creating some software that might be called a derivative work of the original. > I can freely use the S3TC extension here because it's not (yet) > patentable. Any US developer could write it and even compile it, > as long as he doesn't sell it in his country. Use of a patented algorithm without paying the license fee is a patent infringement. Even if it's your own code on your own machines. Selling or distributing it doesn't even enter into the picture as far as the US legality goes; it only affects the damages which would be awarded in a patent suit. It's also better in general for you not to check whether what you're doing would infringe any patents or not, because damages for willful infringement are usually significantly higher. I think the general rule of thumb regarding patents is to play dumb until you haven't any choice (receive a C&D, or patent is somehow brought to your attention otherwise). -- Ryan Underwood, <[EMAIL PROTECTED]> signature.asc Description: Digital signature
Re: [Mesa3d-dev] Doom3 works on R200!
On Sunday 24 October 2004 14:16, Nicolai Haehnle wrote: > On Sunday 24 October 2004 19:38, Bernardo Innocenti wrote: > > Even though I just have a Radeon 9200, I'm very excited about the > > ongoning R300 effort and with there was a similar project for NVidia > > cards too. > > If that "with" above is a "wish" like I think it probably is, you might > want to have a look at Utah-GLX which has rudimentary hw accel support. > Also, somebody, somewhere (possibly in the nv driver in X, but I'm not > sure) figured out how to do DMA. You're probably thinking of the rivatv project, also on sourceforge. I'd be surprised if the open nv driver did DMA. > Of course, what's really needed is the equivalent of glxtest for NVidia and > somebody with NVidia hardware who has a few weeks to spare for long nights > of puzzling over register dumps :) The utah driver isn't totally impenetrable. While the register values are all given as magic numbers, the function names are readable at least. Also, as the closed nvidia driver does not use the DRM, glxtest would be of no use. - ajax > cu, > Nicolai pgpWF7ldQOOtc.pgp Description: PGP signature
Re: [Mesa3d-dev] Doom3 works on R200!
Mike Mestnik wrote: --- Adam K Kirchhoff <[EMAIL PROTECTED]> wrote: Mike Mestnik wrote: --- Philipp Klaus Krause <[EMAIL PROTECTED]> wrote: Thank you for doing this work. We really need to get the open-source ATI driver on par with the propretary driver (both feature-wise and performance-wise). But sadly we will NEVER match it. NO SmoothVision, HyperZ docu ever The nonfree xig driver has been developed without HyperZ docs and outperforms fglrx. It's my opinion that xig uses trickery to get there FPS higher then it should be posible to under OpenGL compliant rendering. It's also true in the font and 2d(like lack of xv support) accelerations as well. What do you mean by the lack of xv support? I often use the XiG driver and have never encountered this limitation. This is when you run (mplayer, xine, ext) and you go full screen. With xv ONE of the things you will see is the video gets streached to fill the whole screen. Withought xv the video stayes the same size and the extra screen is filled with black. Right... And how is this missing with Accelerated-X? xv has worked fine for me with XiG for a very long time. Adam --- This SF.net email is sponsored by: IT Product Guide on ITManagersJournal Use IT products in your business? Tell us what you think of them. Give us Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more http://productguide.itmanagersjournal.com/guidepromo.tmpl -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Mesa3d-dev] Doom3 works on R200!
If there weren't all those patents out there we might just try to develop a free graphics chip. I have thought about this (repeatedly - the idea gets very tempting after asking for the docs for the Nth time) and I don't think it is feasible to make an actual chip. By the time we are finished the world will move on. What could work, however, is to make a *board* that is capable of decent 3d. Put lots of memory, lots of bandwidth and several DSP to approximate the same level of raw floating-point power as 3d GPUs. Leave everything else to the software. The problem is getting such a beast under $1000 range. Last time I looked TI DSPs that were up to the task were rather expensive. best Vladimir Dergachev Philipp --- This SF.net email is sponsored by: IT Product Guide on ITManagersJournal Use IT products in your business? Tell us what you think of them. Give us Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more http://productguide.itmanagersjournal.com/guidepromo.tmpl -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel --- This SF.net email is sponsored by: IT Product Guide on ITManagersJournal Use IT products in your business? Tell us what you think of them. Give us Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more http://productguide.itmanagersjournal.com/guidepromo.tmpl -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Mesa3d-dev] Doom3 works on R200!
--- Adam K Kirchhoff <[EMAIL PROTECTED]> wrote: > Mike Mestnik wrote: > > >--- Philipp Klaus Krause <[EMAIL PROTECTED]> wrote: > > > > > > > Thank you for doing this work. We really need to get the > open-source > ATI driver on par with the propretary driver (both feature-wise and > performance-wise). > > > >>>But sadly we will NEVER match it. > >>> > >>>NO SmoothVision, HyperZ docu ever > >>> > >>> > >>The nonfree xig driver has been developed without HyperZ docs and > >>outperforms fglrx. > >> > >> > >> > >It's my opinion that xig uses trickery to get there FPS higher then it > >should be posible to under OpenGL compliant rendering. It's also true > in > >the font and 2d(like lack of xv support) accelerations as well. > > > > > > What do you mean by the lack of xv support? I often use the XiG driver > and have never encountered this limitation. > This is when you run (mplayer, xine, ext) and you go full screen. With xv ONE of the things you will see is the video gets streached to fill the whole screen. Withought xv the video stayes the same size and the extra screen is filled with black. > Adam > > ___ Do you Yahoo!? Declare Yourself - Register online to vote today! http://vote.yahoo.com --- This SF.net email is sponsored by: IT Product Guide on ITManagersJournal Use IT products in your business? Tell us what you think of them. Give us Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more http://productguide.itmanagersjournal.com/guidepromo.tmpl -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Mesa3d-dev] Doom3 works on R200!
On Sun, 24 Oct 2004, Nicolai Haehnle wrote: On Sunday 24 October 2004 19:38, Bernardo Innocenti wrote: Even though I just have a Radeon 9200, I'm very excited about the ongoning R300 effort and with there was a similar project for NVidia cards too. If that "with" above is a "wish" like I think it probably is, you might want to have a look at Utah-GLX which has rudimentary hw accel support. Also, somebody, somewhere (possibly in the nv driver in X, but I'm not sure) figured out how to do DMA. Of course, what's really needed is the equivalent of glxtest for NVidia and somebody with NVidia hardware who has a few weeks to spare for long nights of puzzling over register dumps :) I just want to point out that glxtest is not that hardware specific. Only the register pretty-printer is. So, for starters, it would be curious to look at the output of glxtest with NVidia driver - what kind of maps it uses, how they change during drawing, etc. So volunteers with hardware and spare time are needed :) best Vladimir Dergachev cu, Nicolai --- This SF.net email is sponsored by: IT Product Guide on ITManagersJournal Use IT products in your business? Tell us what you think of them. Give us Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more http://productguide.itmanagersjournal.com/guidepromo.tmpl -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Mesa3d-dev] Doom3 works on R200!
Mike Mestnik wrote: --- Philipp Klaus Krause <[EMAIL PROTECTED]> wrote: Thank you for doing this work. We really need to get the open-source ATI driver on par with the propretary driver (both feature-wise and performance-wise). But sadly we will NEVER match it. NO SmoothVision, HyperZ docu ever The nonfree xig driver has been developed without HyperZ docs and outperforms fglrx. It's my opinion that xig uses trickery to get there FPS higher then it should be posible to under OpenGL compliant rendering. It's also true in the font and 2d(like lack of xv support) accelerations as well. What do you mean by the lack of xv support? I often use the XiG driver and have never encountered this limitation. Adam --- This SF.net email is sponsored by: IT Product Guide on ITManagersJournal Use IT products in your business? Tell us what you think of them. Give us Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more http://productguide.itmanagersjournal.com/guidepromo.tmpl -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Mesa3d-dev] Doom3 works on R200!
--- Philipp Klaus Krause <[EMAIL PROTECTED]> wrote: > Dieter Nützel schrieb: > > >> > >>The nonfree xig driver has been developed without HyperZ docs and > >>outperforms fglrx. > > > > > > Are you sure. > > I thought Xig had it all before. > > > > Do you have actual numbers? > > Haven't looked at them for very long time, but I bought the first > version of > > there X server 1994 (?) for mga. > > > > Roland made a comparison a year ago: > http://homepage.hispeed.ch/rscheidegger/atilinux_oct03/ati_linux_comp_oct03.html > Xig even outperforms the ATI windows drivers. > This has allot more todo with OSs then video drivers. Take the original doom and quake for example. Both had software rendering directly to the HW, via DOS. It's also true that on a 386(33mhz) doom was playable on linux(about 23 FPS) and unplayable in DOS(about 18 FPS) about just by changing OSs. For quake the numbers where less devistating on a 486(32 FPS) but in windows emulated dos(with IP/UDP support) got only 27 with quake. > Philipp > > > --- > This SF.net email is sponsored by: IT Product Guide on ITManagersJournal > Use IT products in your business? Tell us what you think of them. Give > us > Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out > more > http://productguide.itmanagersjournal.com/guidepromo.tmpl > -- > ___ > Dri-devel mailing list > [EMAIL PROTECTED] > https://lists.sourceforge.net/lists/listinfo/dri-devel > ___ Do you Yahoo!? Declare Yourself - Register online to vote today! http://vote.yahoo.com --- This SF.net email is sponsored by: IT Product Guide on ITManagersJournal Use IT products in your business? Tell us what you think of them. Give us Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more http://productguide.itmanagersjournal.com/guidepromo.tmpl -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Mesa3d-dev] Doom3 works on R200!
--- Philipp Klaus Krause <[EMAIL PROTECTED]> wrote: > > >> > >>Thank you for doing this work. We really need to get the open-source > >>ATI driver on par with the propretary driver (both feature-wise and > >>performance-wise). > > > > > > But sadly we will NEVER match it. > > > > NO SmoothVision, HyperZ docu ever > > The nonfree xig driver has been developed without HyperZ docs and > outperforms fglrx. > It's my opinion that xig uses trickery to get there FPS higher then it should be posible to under OpenGL compliant rendering. It's also true in the font and 2d(like lack of xv support) accelerations as well. Do you mean "fglrx" or "opengl.dll -> radeon.drv"? > > > > > >>Even though I just have a Radeon 9200, I'm very excited about the > >>ongoning R300 effort and with there was a similar project for NVidia > >>cards too. > > > > > > Above applies here, too. - Sorry. > > > > The situation seems to be much worse in the future. > > Bad IP (TRIPS, etc.) madness due to USA-"law". > > > > True. ATI no longer releases docs. 3dlabs no longer does. Nvidia never > did. Intel requires an NDA now. > If there weren't all those patents out there we might just try to > develop a free graphics chip. > > Philipp > > > --- > This SF.net email is sponsored by: IT Product Guide on ITManagersJournal > Use IT products in your business? Tell us what you think of them. Give > us > Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out > more > http://productguide.itmanagersjournal.com/guidepromo.tmpl > -- > ___ > Dri-devel mailing list > [EMAIL PROTECTED] > https://lists.sourceforge.net/lists/listinfo/dri-devel > ___ Do you Yahoo!? Declare Yourself - Register online to vote today! http://vote.yahoo.com --- This SF.net email is sponsored by: IT Product Guide on ITManagersJournal Use IT products in your business? Tell us what you think of them. Give us Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more http://productguide.itmanagersjournal.com/guidepromo.tmpl -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Mesa3d-dev] Doom3 works on R200!
Dieter Nützel wrote: The asm output looks quite readable: you can see symbol names and accesses to PCI registers (base ptr + offset). A "bad" original for DRI;-) This information should only be used to write a header file describing the registers. Of course I'm not talking about cutting & pasting asm code into the open-source DRI module ;-) I'm not familiar with 3D hardware, but my rough guess is that you could easily guess what the registers if you know what the GL extensions are supposed to do and see what values are written in registers. Some are on it for ages, but. Perhaps I'm underestimating the complexity behind that code... I seemed to me that there was not too much glue code between the module API and the hardware registers. IANAL, but reverse engineering is perfectly legal here in Europe and probably even in the USA if your goal is achieving compatibility. If we didn't get another IP right ("software patents", which we didn't have today, even if the EPÜ/EPC did it falsely the US-way) in Europe. The new patent law is still being discussed. The current convention explicitly disallows patenting computer programs, mathematical methods and the like: http://www.european-patent-office.org/legal/epc/e/ar52.html BTW Which is the official Italian standpoint. European Commision Draft or Parliament (later is against)? Italy has always been vaguely against the new software patent law. AFAIK, the strongest supporter of this regulation is Irland (where Microsoft's European HQ is, along with many other big corporations). The law will be discussed again in the Parliament... let's hope not too many politicians will already have been bought by that time. I can freely use the S3TC extension here because it's not (yet) patentable. Yes, but IS falsely by the EPÜ/EPC... ...and "solved" with Roland's work;-) Who I'm very grateful to for his clever hack. Let's hope the distributors can license that code from VIA. Microsoft did so for DX9. Any US developer could write it and even compile it, as long as he doesn't sell it in his country. Somewhat to simple I think. Well, WANL (We Are Not Lawyers)... -- // Bernardo Innocenti - Develer S.r.l., R&D dept. \X/ http://www.develer.com/ --- This SF.net email is sponsored by: IT Product Guide on ITManagersJournal Use IT products in your business? Tell us what you think of them. Give us Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more http://productguide.itmanagersjournal.com/guidepromo.tmpl -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Mesa3d-dev] Doom3 works on R200!
Dieter Nützel wrote: Thank you for doing this work. We really need to get the open-source ATI driver on par with the propretary driver (both feature-wise and performance-wise). But sadly we will NEVER match it. NO SmoothVision, HyperZ docu ever Do you really need the datasheet to get these to work? Some time ago I disassembled ATI's fglrx kernel module and their DRI module. The asm output looks quite readable: you can see symbol names and accesses to PCI registers (base ptr + offset). I'm not familiar with 3D hardware, but my rough guess is that you could easily guess what the registers if you know what the GL extensions are supposed to do and see what values are written in registers. IANAL, but reverse engineering is perfectly legal here in Europe and probably even in the USA if your goal is achieving compatibility. Even though I just have a Radeon 9200, I'm very excited about the ongoning R300 effort and with there was a similar project for NVidia cards too. Above applies here, too. - Sorry. The situation seems to be much worse in the future. Bad IP (TRIPS, etc.) madness due to USA-"law". This is certaily bad, but not as bad as being unable to develop the driver at all. You may implement patented algorithms and restrict its use in some countries. I can freely use the S3TC extension here because it's not (yet) patentable. Any US developer could write it and even compile it, as long as he doesn't sell it in his country. -- // Bernardo Innocenti - Develer S.r.l., R&D dept. \X/ http://www.develer.com/ --- This SF.net email is sponsored by: IT Product Guide on ITManagersJournal Use IT products in your business? Tell us what you think of them. Give us Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more http://productguide.itmanagersjournal.com/guidepromo.tmpl -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Mesa3d-dev] Doom3 works on R200!
Am Sonntag, 24. Oktober 2004 20:45 schrieb Philipp Klaus Krause: > Dieter Nützel schrieb: > >>The nonfree xig driver has been developed without HyperZ docs and > >>outperforms fglrx. > > > > Are you sure. > > I thought Xig had it all before. > > > > Do you have actual numbers? > > Haven't looked at them for very long time, but I bought the first version > > of there X server 1994 (?) for mga. > > Roland made a comparison a year ago: > http://homepage.hispeed.ch/rscheidegger/atilinux_oct03/ati_linux_comp_oct03 >.html Xig even outperforms the ATI windows drivers. Ah, OK I know that ones, but know the side effects, too. No standard xv, etc. -Dieter --- This SF.net email is sponsored by: IT Product Guide on ITManagersJournal Use IT products in your business? Tell us what you think of them. Give us Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more http://productguide.itmanagersjournal.com/guidepromo.tmpl -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Mesa3d-dev] Doom3 works on R200!
Dieter Nützel schrieb: The nonfree xig driver has been developed without HyperZ docs and outperforms fglrx. Are you sure. I thought Xig had it all before. Do you have actual numbers? Haven't looked at them for very long time, but I bought the first version of there X server 1994 (?) for mga. Roland made a comparison a year ago: http://homepage.hispeed.ch/rscheidegger/atilinux_oct03/ati_linux_comp_oct03.html Xig even outperforms the ATI windows drivers. Philipp --- This SF.net email is sponsored by: IT Product Guide on ITManagersJournal Use IT products in your business? Tell us what you think of them. Give us Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more http://productguide.itmanagersjournal.com/guidepromo.tmpl -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Mesa3d-dev] Doom3 works on R200!
Am Sonntag, 24. Oktober 2004 20:10 schrieb Bernardo Innocenti: CC trimmed. > Dieter Nützel wrote: > >>Thank you for doing this work. We really need to get the open-source > >>ATI driver on par with the propretary driver (both feature-wise and > >>performance-wise). > > > > But sadly we will NEVER match it. > > > > NO SmoothVision, HyperZ docu ever > > Do you really need the datasheet to get these to work? Some > time ago I disassembled ATI's fglrx kernel module and their > DRI module. > > The asm output looks quite readable: you can see symbol names > and accesses to PCI registers (base ptr + offset). A "bad" original for DRI;-) > I'm not familiar with 3D hardware, but my rough guess is that > you could easily guess what the registers if you know what the > GL extensions are supposed to do and see what values are > written in registers. Some are on it for ages, but. > IANAL, but reverse engineering is perfectly legal here in Europe > and probably even in the USA if your goal is achieving > compatibility. If we didn't get another IP right ("software patents", which we didn't have today, even if the EPÜ/EPC did it falsely the US-way) in Europe. BTW Which is the official Italian standpoint. European Commision Draft or Parliament (later is against)? > >>Even though I just have a Radeon 9200, I'm very excited about the > >>ongoning R300 effort and with there was a similar project for NVidia > >>cards too. > > > > Above applies here, too. - Sorry. > > > > The situation seems to be much worse in the future. > > Bad IP (TRIPS, etc.) madness due to USA-"law". > > This is certaily bad, but not as bad as being unable to develop > the driver at all. You may implement patented algorithms and > restrict its use in some countries. > > I can freely use the S3TC extension here because it's not (yet) > patentable. Yes, but IS falsely by the EPÜ/EPC... ...and "solved" with Roland's work;-) > Any US developer could write it and even compile it, > as long as he doesn't sell it in his country. Somewhat to simple I think. Greetings, Dieter PS Fight for the European Parliament's draft! PPS We had Thursday some very good results in our parlament. http://www.heise.de/newsticker/meldung/52417 --- This SF.net email is sponsored by: IT Product Guide on ITManagersJournal Use IT products in your business? Tell us what you think of them. Give us Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more http://productguide.itmanagersjournal.com/guidepromo.tmpl -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Mesa3d-dev] Doom3 works on R200!
Philipp Klaus Krause wrote: Thank you for doing this work. We really need to get the open-source ATI driver on par with the propretary driver (both feature-wise and performance-wise). But sadly we will NEVER match it. NO SmoothVision, HyperZ docu ever The nonfree xig driver has been developed without HyperZ docs and outperforms fglrx. Even though I just have a Radeon 9200, I'm very excited about the ongoning R300 effort and with there was a similar project for NVidia cards too. Above applies here, too. - Sorry. The situation seems to be much worse in the future. Bad IP (TRIPS, etc.) madness due to USA-"law". True. ATI no longer releases docs Not true. They have released r300 documents to XiG under an NDA recently. Adam --- This SF.net email is sponsored by: IT Product Guide on ITManagersJournal Use IT products in your business? Tell us what you think of them. Give us Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more http://productguide.itmanagersjournal.com/guidepromo.tmpl -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Mesa3d-dev] Doom3 works on R200!
Am Sonntag, 24. Oktober 2004 20:20 schrieb Philipp Klaus Krause: > >>Thank you for doing this work. We really need to get the open-source > >>ATI driver on par with the propretary driver (both feature-wise and > >>performance-wise). > > > > But sadly we will NEVER match it. > > > > NO SmoothVision, HyperZ docu ever > > The nonfree xig driver has been developed without HyperZ docs and > outperforms fglrx. Are you sure. I thought Xig had it all before. Do you have actual numbers? Haven't looked at them for very long time, but I bought the first version of there X server 1994 (?) for mga. > >>Even though I just have a Radeon 9200, I'm very excited about the > >>ongoning R300 effort and with there was a similar project for NVidia > >>cards too. > > > > Above applies here, too. - Sorry. > > > > The situation seems to be much worse in the future. > > Bad IP (TRIPS, etc.) madness due to USA-"law". > > True. ATI no longer releases docs. 3dlabs no longer does. Nvidia never > did. Intel requires an NDA now. > If there weren't all those patents out there we might just try to > develop a free graphics chip. ;-) Greetings, Dieter --- This SF.net email is sponsored by: IT Product Guide on ITManagersJournal Use IT products in your business? Tell us what you think of them. Give us Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more http://productguide.itmanagersjournal.com/guidepromo.tmpl -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Mesa3d-dev] Doom3 works on R200!
Thank you for doing this work. We really need to get the open-source ATI driver on par with the propretary driver (both feature-wise and performance-wise). But sadly we will NEVER match it. NO SmoothVision, HyperZ docu ever The nonfree xig driver has been developed without HyperZ docs and outperforms fglrx. Even though I just have a Radeon 9200, I'm very excited about the ongoning R300 effort and with there was a similar project for NVidia cards too. Above applies here, too. - Sorry. The situation seems to be much worse in the future. Bad IP (TRIPS, etc.) madness due to USA-"law". True. ATI no longer releases docs. 3dlabs no longer does. Nvidia never did. Intel requires an NDA now. If there weren't all those patents out there we might just try to develop a free graphics chip. Philipp --- This SF.net email is sponsored by: IT Product Guide on ITManagersJournal Use IT products in your business? Tell us what you think of them. Give us Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more http://productguide.itmanagersjournal.com/guidepromo.tmpl -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Mesa3d-dev] Doom3 works on R200!
On Sunday 24 October 2004 19:38, Bernardo Innocenti wrote: > Even though I just have a Radeon 9200, I'm very excited about the > ongoning R300 effort and with there was a similar project for NVidia > cards too. If that "with" above is a "wish" like I think it probably is, you might want to have a look at Utah-GLX which has rudimentary hw accel support. Also, somebody, somewhere (possibly in the nv driver in X, but I'm not sure) figured out how to do DMA. Of course, what's really needed is the equivalent of glxtest for NVidia and somebody with NVidia hardware who has a few weeks to spare for long nights of puzzling over register dumps :) cu, Nicolai pgppeOhHYFYMY.pgp Description: PGP signature
Re: [Mesa3d-dev] Doom3 works on R200!
Am Sonntag, 24. Oktober 2004 19:38 schrieb Bernardo Innocenti: > Dave Airlie wrote: > >>r200 render path looks really A LOT better, unfortunately the open-source > >>driver doesn't implement the required extensions (some bits of > >> documentation are missing afaik, and even if not (I have no idea what's > >> in the documentation or not) it would probably quite a bit of work as > >> core mesa doesn't support them neither (mostly > >> ATI_(text_)fragment_shader). > > > > Well I've started it, but it'll take me a while to finish off the > > software implementation, Eric thinks he can do the hardware side (I think > > I can probably do it as well.. but I'll try and get the software side > > working first hopefully...) > > Thank you for doing this work. We really need to get the open-source > ATI driver on par with the propretary driver (both feature-wise and > performance-wise). But sadly we will NEVER match it. NO SmoothVision, HyperZ docu ever > Even though I just have a Radeon 9200, I'm very excited about the > ongoning R300 effort and with there was a similar project for NVidia > cards too. Above applies here, too. - Sorry. The situation seems to be much worse in the future. Bad IP (TRIPS, etc.) madness due to USA-"law". -Dieter --- This SF.net email is sponsored by: IT Product Guide on ITManagersJournal Use IT products in your business? Tell us what you think of them. Give us Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more http://productguide.itmanagersjournal.com/guidepromo.tmpl -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Mesa3d-dev] Doom3 works on R200!
Dave Airlie wrote: r200 render path looks really A LOT better, unfortunately the open-source driver doesn't implement the required extensions (some bits of documentation are missing afaik, and even if not (I have no idea what's in the documentation or not) it would probably quite a bit of work as core mesa doesn't support them neither (mostly ATI_(text_)fragment_shader). Well I've started it, but it'll take me a while to finish off the software implementation, Eric thinks he can do the hardware side (I think I can probably do it as well.. but I'll try and get the software side working first hopefully...) Thank you for doing this work. We really need to get the open-source ATI driver on par with the propretary driver (both feature-wise and performance-wise). Even though I just have a Radeon 9200, I'm very excited about the ongoning R300 effort and with there was a similar project for NVidia cards too. -- // Bernardo Innocenti - Develer S.r.l., R&D dept. \X/ http://www.develer.com/ --- This SF.net email is sponsored by: IT Product Guide on ITManagersJournal Use IT products in your business? Tell us what you think of them. Give us Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more http://productguide.itmanagersjournal.com/guidepromo.tmpl -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Mesa3d-dev] Doom3 works on R200!
Bernardo Innocenti wrote: The only thing I'm complaining about is the light torch: the aura looks good, but the projected light circle is invisible most of the times. Other lightning effects look fine, including dangling lights in ceilings. Actually, those rendering errors are pretty bad in some parts of the game (makes it unplayable for some parts). You can get a correct rendering (as far as I can tell) if you switch off tcl (tcl_mode=0, or use driconf). And I hate to say it, but IMHO the arb path of doomIII really looks ugly (it looks just as ugly if you use the windows driver and force the arb path, the r200 render path looks really A LOT better, unfortunately the open-source driver doesn't implement the required extensions (some bits of documentation are missing afaik, and even if not (I have no idea what's in the documentation or not) it would probably quite a bit of work as core mesa doesn't support them neither (mostly ATI_(text_)fragment_shader). I also noticed that wine + Doom3.exe doesn't work any more (blue window complaining about OPENGL32.DLL missing). I'm not sure if it's caused by new Mesa or new Wine, but I could investigate. Looks more like a wine bug to me (there have been quite some changes afaik lately in that area). Roland --- This SF.net email is sponsored by: IT Product Guide on ITManagersJournal Use IT products in your business? Tell us what you think of them. Give us Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more http://productguide.itmanagersjournal.com/guidepromo.tmpl -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel
Re: [Mesa3d-dev] Doom3 works on R200!
> r200 render path looks really A LOT better, unfortunately the open-source > driver doesn't implement the required extensions (some bits of documentation > are missing afaik, and even if not (I have no idea what's in the documentation > or not) it would probably quite a bit of work as core mesa doesn't support > them neither (mostly ATI_(text_)fragment_shader). Well I've started it, but it'll take me a while to finish off the software implementation, Eric thinks he can do the hardware side (I think I can probably do it as well.. but I'll try and get the software side working first hopefully...) Dave. -- David Airlie, Software Engineer http://www.skynet.ie/~airlied / airlied at skynet.ie pam_smb / Linux DECstation / Linux VAX / ILUG person --- This SF.net email is sponsored by: IT Product Guide on ITManagersJournal Use IT products in your business? Tell us what you think of them. Give us Your Opinions, Get Free ThinkGeek Gift Certificates! Click to find out more http://productguide.itmanagersjournal.com/guidepromo.tmpl -- ___ Dri-devel mailing list [EMAIL PROTECTED] https://lists.sourceforge.net/lists/listinfo/dri-devel