From: Sean <seanlkml@xxxxxxxxxxxx> > You're just wrong. The simple fact is that if you want to use open > source drivers there are better options than those provided by nVidia. And better than ATI, than Matrox, etc... I can play the game of DRI/UtahGLX support for NV0x/NV1x (nVidia TNT-GeForce2) too. I think people like yourself don't stop to realize they are applying "double standards." I can also "play the game" of nVidia's better MIT X11 "nv" support than ATI's MIT X11 "radeon" support when it comes to 2D, especially regarding video in/video out (VIVO) support. So at what point do you actually stop to recognize that not many vendors are really supporting "open source" other than maybe Intel (possibly 3DLabs? I haven't checked) and that's _more_ than just nVidia? This is why _you_ are fixated on branding, _not_ me. > No, the problem you've had is in failing to recognize that the brand > doesn't matter. Excuse me? That's _exactly_ what I have been saying and you have not. > All i'm saying is buy the best card you can that runs open source > drivers. > You haven't said once that nVidia open source drivers are superior to > other offerings. I have not made _any_ claims, other than the fact that demonizing anyone who uses nVidia's Standardware (open standard, proprietary source) is a double-standard compared to ATI, Matrox and other companies. Especially companies like Matrox, who's DRI/UtahGLX drivers are for the same generation as those for nVidia's DRI/UtahGLX support (circa 2000- 2001), and _not_ newer either. ATI went a bit longer, through the R200 series (circa 2002-2003), before closing up the specs. And as someone else pointed out, it wasn't ATI who necessarily "supported" open source, other than releasing _some_ specifications. Which leaves little more than Intel (I have not checked 3DLabs personally). I'll not only applaud Intel, but I'll even buy their products when I just need minimal GLX support. But the performance and features are very sub-par, even on the i9xx series -- not enough to run many CAM/EDA application. But I neither advocated, nor will ever advocate use nVidia over other vendors for general usage. I stayed clear out of the threads _until_ it reached a point where there were assumptions, half-truths and other non-sense -- such as the junk you are throwing around. You do _not_ fight a marketing agenda with another marketing agenda, especially one that is _not_ factual. That's has _always_ been my "problem." > That's because cards from other vendors have _better_ open source > drivers. To what point? I have some NV0x/NV1x cards with UtahGLX that do some things. I have some Intel cards. I even have a few ATI R200 series. On many I use the available, open source GLX. On many others, I don't even bother with GLX, and just use the MIT 2D. No issues/arguments there. But when I have a CAM/EDA application that doesn't run on those lesser GLX solutions, I'm left with 4 options -- each of which may or may not be available: 1. Use ATI and/or nVidia's Standardware drivers 2. Use a 3rd party GLX product like X/Accelerated 3. Use a non-Linux POSIX/GLX platform like Irix, Solaris, etc... 4. Use the Win32/DirectX port if the vendor has one According to you, I should be chastized for going with #1, along with ATI and nVidia for supporting this marketing when they have absolutely no legal way to do it with open source. That's _all_ I've ever said! > Please try to understand this once and for all so we can stop this > thread. If you haven't noticed, you're keeping this thread going on right along with me. > What you want is the last word. Oh, don't even try to be that hypocritical. > You should have just let it die when Rahul asked it to die. You're right, I should have. And maybe you should consider the same? But no, you have repeatedly stated *1* company name based on your assumptions and half-truths. *1* company over and over. You are _no_better_ than the people who expect the Fedora Project to support nVidia (let alone ATI, Matrox, etc...) Standardware drivers. > Wrong again. I'm just dealing with the reality that the best open > source supported cards don't come from nVidia today. I'd be just > as happy as anyone else if that changed. You are fixated on nVidia, get over it. You keep asserting things that are _not_ true, that's my problem as an engineer. > You've been decidedly intolerant of anyone who stands up and says, > use open source solutions instead of binary only solutions. This is what I'm talking about, you continue to demonize _anyone_ who even sees a use for the Standardware drivers. All I _ever_ stated were the reasons why people use the Standardware drivers, not that I approve of them! That's all! When are you going to stop and recognize what I really said, and not what agenda you think I am spreading (and trying to counter with your own)? > All you hear is an attack on nVidia instead of a support for open > source solutions. There are some very good graphic cards available > today that are supported by open source. Unfortunately they don't > come from your favorite supplier. Any chance you're going to stop demonizing whatever I say to what fits you? All I have ever said is that nVidia was the company that offered a solution for CAM/EDA applications on Linux when _no_one_else_ would -- that's all! I would _love_ to see Freedomware (Open Standard, Open Source) drivers be able to viably drive many CAM/EDA applications on Linux. And when I don't need that capability, I don't use such a solution. Why you continue to paint me as a nVidia "cheerleader," I don't know. I guess you just want something to argue about. In fact, that's _exactly_ why I started responding to you in the first place, because you were going around chatizing people as bad as others complaining about lack of Fedora support. Again, damn me for trying to explain a "middle ground." > That's because it is only nVidia cheerleaders who are constantly > telling people (usually without reservation or caution) to embrace > binary only drivers. > If more of you would tone down your support for binary only > solutions, you might get a different reaction. Please point out where I have ... and let's be explicit ... "constantly telling people (usually without reservation or caution) to embrace binary only drivers" That's the "bigotry" and other non-sense I'm talking about! You demonize some of our statements for the sake of arguing! > Well i'm glad to hear that; If you'd stop and actually _read_ what I say, instead of demonizing what I say for the sake of arguing, maybe, just maybe you'd realize what we are actually saying! > you should really spend more time telling people that open source > solutions are preferable because you seem to spend way more > energy justifying binary drivers. I do! I have stated time and time again that people should not only not expect the Fedora Project to support those Standardware drivers, but that there is little reason to run with them for many applications. I haven't had to do much of that on this board, people like Alan, Rahul, etc... have done an excellent job. But I alos do it _across_the_board_ on _all_vendors_, not just get fixated on 1. That's where your "bigotry" comes in, and then you "demonize" anyone who doesn't agree with your "vendor list" -- which seems to be everyone but nVidia. People ignore the previous XFree 3.3.x releases, the UtahGLX work for NV0x/1x and the continued 2D "nv" drivers. Stop and listen to what I'm actually saying, not what you think I'm saying or what agenda you think I'm pushing. I don't prefer nVidia's products for many applications, and I typically just stick with Intel's integrated i8xx/i9xx solutions for the majority of non-engineering desktops. But when it comes to many CAM/EDA solutions, I have the option of these 4: 1. Use ATI and/or nVidia's Standardware drivers 2. Use a 3rd party GLX product like X/Accelerated 3. Use a non-Linux POSIX/GLX platform like Irix, Solaris, etc... 4. Use the Win32/DirectX port if the vendor has one I don't see "open source" even listed there, so I can either choose one of them, or just let a sale/integration go to another company -- possibly with #3 or, quite often, #4. -- Bryan J. Smith mailto:b.j.smith@xxxxxxxx -- fedora-devel-list mailing list fedora-devel-list@xxxxxxxxxx http://www.redhat.com/mailman/listinfo/fedora-devel-list
