From owner-freebsd-multimedia Sun Aug 10 20:53:06 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id UAA06699 for multimedia-outgoing; Sun, 10 Aug 1997 20:53:06 -0700 (PDT) Received: from labs.usn.blaze.net.au (mail@labs.usn.blaze.net.au [203.17.53.30]) by hub.freebsd.org (8.8.5/8.8.5) with SMTP id UAA06694 for ; Sun, 10 Aug 1997 20:52:59 -0700 (PDT) Received: from labs.usn.blaze.net.au [127.0.0.1] (davidn) by labs.usn.blaze.net.au with esmtp (Exim 1.62 #1) id 0wxlXD-0000Dg-00 (Debian); Mon, 11 Aug 1997 13:52:47 +1000 X-Mailer: exmh version 2.0zeta 7/24/97 To: Amancio Hasty cc: multimedia@freebsd.org Subject: Re: xquake + sound?? In-reply-to: Your message of "Thu, 07 Aug 1997 11:13:36 MST." <199708071813.LAA04746@rah.star-gate.com> X-Face: (W@z~5kg?"+5?!2kHP)+l369.~a@oTl^8l87|/s8"EH?Uk~P#N+Ec~Z&@;'LL!;3?y Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Date: Mon, 11 Aug 1997 13:52:47 +1000 From: David Nugent Message-Id: Sender: owner-freebsd-multimedia@freebsd.org X-Loop: FreeBSD.org Precedence: bulk > Has anyone managed to get xquake with sound to work with guspnp15? YES! Both xquake and xf86quake work fine with it. (YAY!) But I have a few tales to tell. :-) Firstly, my configuration: Cyrix 6x86L running at 150mh (6x86L/200+) S3 Trio64V+ (for quake, configured with a 300x200 modeline) 3.0-current (as of today) SB16 Pnp, A220 I5 D1 H7 P330 T6, aka: sb0 at 0x220 irq 5 drq 1 on isa SoundBlaster 16 4.13> at 0x220 irq 5 dma 1 sbxvi0 at ? drq 7 on isa SoundBlaster 16 4.13> dma 7 sbmidi0 at 0x300 on isa SoundBlaster MPU-401> at 0x300 opl0 at 0x388 on isa Yamaha OPL3 FM> at 0x388 Xfree86 3.3 Linux emulation compiled into the kernel (not module) First, to get the kernel compile I had to cp soundcard.h to sys/i386/include. Otherwise, compilation falls over in linux_ioctl.h which doesn't see the new sound ioctls. Now, I just changed motherboards (yet again, 5 in two weeks isn't too bad, eh?) to one that actually works reliably with a IBM/Cyrix 6x86L at this speed. Previously, I had a VXPro chipset motherboard(s) with the Award PNP BIOS extensions. I had no idea what the original SB configuration dropped out at via the BIOS, but under DOS, Linux, FreeBSD and OS/2 I had to reconfigure the thing anyway since it would not work without it at the default settings, and not having Win95 at all I didn't even know what they were, but they were NOT as above. For Linux, I used isapnptools, and for FreeBSD I used Sujal's pnp patches. Now, with this new motherboard, Sujal's pnp driver no longer worked. :-( I have no idea what it was doing, but after running it - and it seemed to work correctly by identifying the card and saying it was configured - the sb and all associated devices just "disappeared". Anyway, after figuring out with a DOS help program the default settings were actually the same as what I wanted anyway, I removed the PnP driver and suddenly the sound driver configured correctly. Go figure. I tried to investigate why this was happening (after numerous reboots and fiddling with parameters etc, dropping back to the default -current kernel to try to figure out the problem), I couldn't. Initially, I thought perhaps the system was just too fast, and some delays were required between inb() and outb() statements [well, it worked a few years ago on AT's :-)], but after making some adjustments and recompiling, still no go. Either the driver isn't doing something "right", that worked previously, or this motherboard has a problem. I don't *think* it is the latter, since the driver configures with AND without isapnptools under Linux, and works fine under DOS with and without Creative's CTCM pnp configuration tool. BTW, in spite of what the README.GUSPNP said, I did *NOT* have to define BROKENSB to get xquake/xf86quake to work. Either that, or it is already defined somewhere in the source code. :-) I didn't investiate, but I tried it first with a default compilation, and only realised later that I had forgotten to do that. xquake/xf86quake work ok, except that now (unlike before when I ran it without sound), they won't exit correctly and require a kill from another session (telnet, since c-a-Fkey don't work while quake is running). This doesn't happen under Linux, so it may be sound driver related (Linux kernel 2.0.47, if anyone is interested). Only one problem remains in playing x86quake, and this happens under Linux XFeee86 too, so it isn't related to the OS or sound drivers - the mouse keeps on creeping upwards during play. It doesn't happen in Linux squake, nor DOS quake, so that rules out mouse, hardware and serial driver problems. It seems to be XFree86 specific. Perhaps I need to reconfigure it somehow, but I'm at a loss as to exactly how. The mouse (actually, any mouse I try has the same problem) works in mousesystems mode, and I tried having XFree use moused/sysmouse and direct to port, but it makes no difference. Any fellow quake players out there who use the mouse? Sorry, but I just can't play it any other way... keyboard players suck in multiplayer quake, not being able to turn and hook (in ctf games) fast enough. :-) Also, does anyone have any idea what it would take to get svgalib ported and working under FreeBSD? Yeah, I realise the library is a real mess in many respects, but it is pretty much a standard for console-based games for i386 UNIX and clones. Having conversed with him a couple of times, I'm reasonably sure that Dive Kirsch could be convinced to compile a FreeBSD native version of squake, or even running the Linux version would be cool too. :) I gather we need a few hooks in syscons. Regards, David