Go Back   nV News Forums > Linux Support Forums > NVIDIA Linux

Newegg Daily Deals

Closed Thread
 
Thread Tools
Old 01-26-04, 08:32 PM   #13
sirexar
Registered User
 
Join Date: Dec 2003
Posts: 20
Default Open-source projects alaways stale? BS

In reply to the open-source projects alaways going stale comment, its not like driver developement would stop here at Nvidia if there drivers were made open-source, they would still be the main developers, only now they would have everyone helping them.
sirexar is offline  
Old 01-29-04, 12:16 AM   #14
delt
Registered User
 
delt's Avatar
 
Join Date: Jan 2004
Location: Montréal, Canada
Posts: 57
Send a message via ICQ to delt
Default


> In reply to the open-source projects alaways going stale comment, its not like driver
> developement would stop here at Nvidia if there drivers were made open-source,
> they would still be the main developers, only now they would have everyone helping them.

My point exactly. What part of this do they NOT understand?
delt is offline  
Old 01-29-04, 04:29 AM   #15
energyman76b
Registered User
 
Join Date: Dec 2002
Location: Clausthal/Germany
Posts: 1,104
Default

Delt:
My point exactly. What part of this do they NOT understand?

Oh, they have understand, but:
What is your problem to understand, that NVIDIA CANNOT open their drivers?
AND it is nice to hear 'we would help you' but in reality only a a handful of people would help.
Millions are using linux. But there are not millions of linux-developers. Millions are using Xfree, but only a few are developing it. Millions are using gcc/glibc...

Face it. Opensource does not mean automatic success or automatic hordes of willing devs.
Additionally drivers are nothing for nOObs and amateurs. You need people with talent, to make good drivers.

And if you want to help nvidia, you can do it NOW. Write bug-reports, help people with problems, and stop discussing a dead topic. Whining around will not open the source nor fastening the development.
energyman76b is offline  
Old 01-29-04, 05:51 AM   #16
Nemesis77
Registered User
 
Join Date: Aug 2002
Posts: 114
Default Re: MGA

Quote:
Originally posted by Inaea Plena
I support open source, I'm a kernel developer. I also support peoples' right to make choices. It's your property, do what you want with it, you know?
really? What kernel-hacking have you done?
Nemesis77 is offline  
Old 01-29-04, 05:53 AM   #17
Nemesis77
Registered User
 
Join Date: Aug 2002
Posts: 114
Default

Quote:
Originally posted by energyman76b
ATI opensource drivers have still problems with Xv, overlays etc...
And that's because Ati doesn't REALLY support open-source drivers. They haven't really given the developers the specs they need, so it's no wonder that it has bugs.
Nemesis77 is offline  
Old 01-29-04, 05:56 AM   #18
Nemesis77
Registered User
 
Join Date: Aug 2002
Posts: 114
Default

Quote:
Originally posted by energyman76b
Delt:
My point exactly. What part of this do they NOT understand?

Oh, they have understand, but:
What is your problem to understand, that NVIDIA CANNOT open their drivers?
I used to believe that, but I'm not so sure anymore. NVIDIA hasn't opened the drivers for their NIC either! Can you REALLY say that their integrated NIC (It's a frigging network-card! How uber-secret that can be?) contains some super-secret IP that must be protected from others? Hell, Intel, Realtek, 3Com have no problems opening their NIC-code, but for some reason NV can't do it.

Luckily we now have reverse-engineered open-source drivers for nForce-NIC.
Nemesis77 is offline  
Old 01-29-04, 11:51 AM   #19
Stu
Registered User
 
Join Date: Jan 2004
Posts: 1
Default No such thing as can't

Quote:
Originally posted by energyman76b
Delt:
My point exactly. What part of this do they NOT understand?

Oh, they have understand, but:
What is your problem to understand, that NVIDIA CANNOT open their drivers?
Of course they can. IP is not an excuse. Methods for not incorporating this in the publicly viewable source have already been suggested. Moreover NVIDIA can just pay to license that IP in a less restricted way. Other companies (e.g. IBM) have done this. They've already put money into Linux by making a driver. Why not go the whole hog?

Another good reason for open sourcing the kernel driver at the minimum is, you get all the kernel gurus able to look at the code. The number of times the nvidia driver has oopsed/panic'd my kernel is off the scale. If the driver was Free then it could go into the mainline kernel just like the other DRI drivers. That would also make the mainline kernel developers hate Nvidia less.


Quote:
AND it is nice to hear 'we would help you' but in reality only a a handful of people would help.
Millions are using linux. But there are not millions of linux-developers. Millions are using Xfree, but only a few are developing it. Millions are using gcc/glibc...
Of course. It is not that the community per se is going to add lots of new features to the driver but it is true that drivers get fixed because people can look at them and say ooo...that looks dodgy what if I twiddle that, or oh no this driver has forgotten to cope with bug X on my motherboard.

It really isn't as simple an equation as you suggest.

Stu.
Stu is offline  
Old 01-29-04, 04:39 PM   #20
meldroc
Registered User
 
Join Date: Apr 2003
Location: Somewhere in Colorado
Posts: 19
Default

Why not provide detailed programming information - as in the instruction set for the graphics processors, the registers, addresses & what not so we can write our own drivers? Don't release the proprietary driver, just release all the necessary information so open source developers can write their own decent driver. There are plenty of people out there willing to contribute if that information is made available. Drivers for hardware where that information is available (such as NICs, motherboard chipsets, sound cards, etc.) are quite decent.

I paid for and own the video card, I think I have the right to know how it works.
meldroc is offline  

Old 01-29-04, 05:46 PM   #21
The Baron
Guest
 
Posts: n/a
Default

it's not Andy's decision. nothing you say or do is going to convince NVIDIA that open-sourcing drivers is a good business decision, at least at this point in time. NICs don't cost three grand--a top-of-the-line Quadro card does, so driver quality must be absolutely assured. while I or others may feel that drivers with some sort of open license could provide that, and even if Andy were to agree, his bosses and his bosses' bosses would not. even if they did, there is probably information about a plethora of other chips besides NVIDIA's in those drivers. scream about IP being immoral and yadda yadda yadda all you like--it's the law.

so: in essence, leave the guy alone about open-source drivers. he's doing what he can to improve the drivers--don't give him grief about what he can't control. he comes in here, responds personally to problems, and does his absolute best to get things straightened out. what more do you want?
 
Old 01-30-04, 03:22 AM   #22
Nemesis77
Registered User
 
Join Date: Aug 2002
Posts: 114
Default

Quote:
Originally posted by The Baron NICs don't cost three grand--a top-of-the-line Quadro card does, so driver quality must be absolutely assured.
I'm not sure I'm following you here. Drivers for the NIC don't need to be open since NICs are cheap. Drivers for the vid-card don't need to be open, since vid-cards are so expensive?

And I fail to see how opening the drivers would lower the quality of the drivers! It's not like everyone and their uncle would be releasing half-assed NV-drivers. People would be providing patches, yes. But the primary source for the drivers would still be NVIDIA.

Good example of this is the OS-Matrox-drivers. Alan Cox noticed that the card had poor performance in Linux, so he took a look at the drivers source-code. Now, vid-card is supposed to keep most used textures in local vid-RAM while least used textures reside in system-RAM (Accessed via AGP). Matrox-drivers did just the opposite. He fixed it, sent patches to Matrox, and the problem was solved!

While NV's Linux-drivers ARE good and I appreciate their support of the platform, there ARE bugs there that could be fixed if there were enough eyes looking at the code (switching between X and console for example. Or adding framebuffer-support). Also, having the drivers open would improve the quality of the overall system. Closed-source drivers taint the kernel. The kernel-developers have no way of knowing whetever some problem is caused by the kernel, by X or by the NV-drivers, if the NV-drivers are closed. So, instead of wasting their time trying to fix something they cannot fix, they ignore such errors. If the drivers were open, tehy could look at each possible cause and fix the problem.

Quote:
so: in essence, leave the guy alone about open-source drivers. he's doing what he can to improve the drivers--don't give him grief about what he can't control. he comes in here, responds personally to problems, and does his absolute best to get things straightened out. what more do you want?
Well, I haven't been here in a while, so I don't know what has been going on in here. I'm well aware that he's not the one who can decide whether to open or close the drivers. And it's not like I whined and pestered him about the drivers. I simply stated an opinion.
Nemesis77 is offline  
Old 02-01-04, 06:39 PM   #23
AndyCap
Registered User
 
Join Date: Apr 2003
Location: Oslo, Norway
Posts: 7
Default Re: MGA

[quote]Originally posted by Inaea Plena
[b]>Is there any company which develops 3D accelerated graphics cards and has open source drivers for it ?

>Yes. Matrox does, for example. Fairly nice ones too, though their cards aren't nearly as >fast as Nvidia's and ATI's. ATI also puts some work into the open source drivers. I don't >blame Nvidia for their decision to keep their drivers closed (and as it was stated above, >they do have other peoples' IP to protect). They make good drivers... none of the OSS >ones even come close performance-wise.

Well, actually I think ATI, Nvidia and Matrox are equally bad in this regard. Matrox drivers require a binary Hal if you want to use the DVI output. ATI has binary only cores for their later cards, and Nvidia has binary modules for theirs. But at least nvidia does publish official drivers. Some support invested into the XFree86 driver would also be nice though.

As for whining about nvidia opensourcing their drivers I try not to do it _too_much. The ultimate solution is as always "Vote with your wallet" I don't really care for people who complain but ultimately won't stop buying the products. (Same goes for MS btw.)
AndyCap is offline  
Old 02-02-04, 12:30 AM   #24
delt
Registered User
 
delt's Avatar
 
Join Date: Jan 2004
Location: Montréal, Canada
Posts: 57
Send a message via ICQ to delt
Default


> Some support invested into the XFree86 driver would also be nice though.

That would be possible if we at least had the hardware's specs. It's sh¤tty enough from nvidia not to open-source the drivers, and to completely ignore anyone who asks why the hell they don't. but keeping hardware specifications under a tight lock.......... wtf????
delt is offline  
Closed Thread


Thread Tools

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Forum Jump


Similar Threads
Thread Thread Starter Forum Replies Last Post
NVIDIA Drivers Receive Windows 8 Certification News Archived News Items 0 06-01-12 05:30 AM
Radeon 9700 not all that? sancheuz Other Desktop Graphics Cards 200 10-12-02 09:31 PM
Nvidia Stereo Drivers Soudontsay NVIDIA Windows Graphics Drivers 2 08-26-02 10:48 AM
nvidia drivers in a motherboard with AGP 1.0 (motherboard MVP3+) knocker NVIDIA Linux 1 08-19-02 01:57 AM
NVIDIA 2960 Drivers & RH 7.3 W/2.4.18-5 XASCompuGuy NVIDIA Linux 6 08-02-02 11:53 AM

All times are GMT -5. The time now is 11:32 AM.


Powered by vBulletin® Version 3.7.1
Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Copyright ©1998 - 2014, nV News.