From owner-freebsd-current@FreeBSD.ORG Tue Dec 18 15:02:19 2007 Return-Path: Delivered-To: freebsd-current@freebsd.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id 906D116A419 for ; Tue, 18 Dec 2007 15:02:19 +0000 (UTC) (envelope-from rainer@ultra-secure.de) Received: from bsd.ultra-secure.de (bsd.ultra-secure.de [62.146.20.26]) by mx1.freebsd.org (Postfix) with ESMTP id 039F113C468 for ; Tue, 18 Dec 2007 15:02:18 +0000 (UTC) (envelope-from rainer@ultra-secure.de) Received: (qmail 10949 invoked by uid 89); 18 Dec 2007 15:02:17 -0000 Received: by simscan 1.1.0 ppid: 10928, pid: 10930, t: 4.2276s scanners: attach: 1.1.0 clamav: 0.88.7/m:44/d:4673 spam: 3.1.7 X-Spam-Checker-Version: SpamAssassin 3.1.7 (2006-10-05) on bsd.ultra-secure.de X-Spam-Level: X-Spam-Status: No, score=-2.4 required=5.0 tests=AWL,BAYES_00 autolearn=ham version=3.1.7 Received: from unknown (HELO ?212.71.117.70?) (rainer@ultra-secure.de@212.71.117.70) by bsd.ultra-secure.de with (DHE-RSA-AES256-SHA encrypted) SMTP; 18 Dec 2007 15:02:13 -0000 Message-ID: <4767E0F4.2060402@ultra-secure.de> Date: Tue, 18 Dec 2007 16:02:12 +0100 From: Rainer Duffner User-Agent: Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.8.0.12) Gecko/20060911 SUSE/1.5.0.12-3.4 Thunderbird/1.5.0.12 Mnenhy/0.7.5.0 MIME-Version: 1.0 To: Marian Hettwer References: <86ve6wmgq4.fsf@ds4.des.no> <4a37b1e529bab094e2c50b7deca7df87@127.0.0.1> In-Reply-To: <4a37b1e529bab094e2c50b7deca7df87@127.0.0.1> Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Cc: freebsd-current@freebsd.org Subject: Re: amd64 NVIDIA support in FreeBSD 7 X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.5 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Tue, 18 Dec 2007 15:02:19 -0000 Marian Hettwer wrote: > On Tue, 18 Dec 2007 11:37:07 +0100, Dag-Erling Smørgrav wrote: > >> "Sam Fourman Jr." writes: >> >>> I wonder if having a petition signed by a bunch of people would help >>> this along, because I believe that amd64 3D accel on nvidia, is life >>> or death to PC-BSD in a year or so. >>> >> What, exactly, do you think a petition would achieve? Can a petition >> write code? Can a petition pay somebody to write code? Petitions may >> work in a democracy, but neither nVidia nor FreeBSD is a democracy. >> >> > While this is true (by the way, what is FreeBSD? nVidia is a corporation, therefor not a democracy, but what is FreeBSD?), at least a petition could show how many users would like to have amd64 nvidia support for FreeBSD. > These numbers could (!) be interesting for nvidia. That's what you usually call a "market need" in captilsm speak. > Although I do have my doubts wether we could show nvidia that our need as that big that nvidia would think "hej, wow, what a huge market, let's get em" ;-) > I believe a better way would be just asking nvidia "Hej, how much money do you need to deliver and probably maintain a amd64 version of your driver for FreeBSD". > That's up to interpretation (like everything in accounting...). Currently, the FreeBSD-driver is probably shipped "as-is", which can be done with little effort at small additional cost. But when you factor in support etc. I suppose SUN could tell you more about this, the Solaris-support for NVIDIA is probably more than a hobby - for both sides. Let's face it, the only reason the NVIDIA-driver actually exists for Linux is because of the workstation-gfx-market, where Linux is a noticable blipp on the radar. Though that market is relatively small, it's high-margin - so there's a profit to be made. The fact that the drivers also work for consumer-cards (or can be made to work with little effort) is probably just coincidence. > Instead of signing a petition, users could donate... > The only usefull donation in this case is code. Otherwise, you'd have to pay the same someone regularly, for keeping the source buildable on CURRENT. The problem is: people who can code are usually OK with the features the nv(4) driver delivers (coding is 2D, mostly). On the other hand, the people in need of accelerated 3d-graphics usually are mostly gamers and can't code too well (there's a 3rd camp, with people like me who can't code too well and don't play games). Also, as Macs "just work", more and more developers are moving to MacOS (just count the Apple-laptops at any BSD-conference). So the pressure just isn't there anymore like in earlier days. cheers, Rainer