Linked by Thom Holwerda on Thu 26th Nov 2009 19:41 UTC, submitted by Gabor
FreeBSD Astute readers probably already saw this one waiting in our backend, but since there was no official announcement yet, I decided to wait. Now that it's officially here, let's rejoice: the FreeBSD team has released version 8.0 of their operating system, packed with new features and improvements.
Thread beginning with comment 396768
To view parent comment, click here.
To read all comments associated with this story, please click here.
phoenix
Member since:
2005-07-11

Well, I never had performance problems on my desktop. Sometimes, I read stories about Linux users complaining that their audio is skipping or that video playback gets desynchonized from audio. I cannot imagine that. It's 2009. Computing power is much more than some years ago. One of my first systems was a 150 MHz Pentium (1 - one, to emphasize this) with 64 (later 128) MB EDO RAM. I could compile the kernel, download an ISO via FTP, burn a CD, browse the web with a responsive Opera and have XMMS play MP3 files AT THE SAME TIME and WITHOUT SKIPPING AUDIO. It's true!


That's one thing I've always like about FreeBSD: sound just works! Their modified version of OSSv3 that supports in-kernel mixing, auto-device cloning (apps just open /dev/dsp, the kernel auto-maps that to either a hardware channel (/dev/dsp0, /dev/dsp1) or a software channel (/dev/dsp0.0, /dev/dsp0.1) and everything just works and sounds nice. Been that way since the 4.x days (possibly earlier, but I didn't use FreeBSD as a desktop until 4.0, 3.x was just on servers).

It still boggles me why the Linux devs went the ALSA route instead of just fixing OSSv3 like the rest of the Unix world did. Talk about cutting yourself off, and creating lock-in.

A lot of my Linux-using aquaintances don't believe me when I tell them that FreeBSD uses OSS and has all the features that ALSA was supposed to bring, and still hasn't delivered (even with PulseAudio). ;)

Reply Parent Score: 3

Doc Pain Member since:
2006-10-08

That's one thing I've always like about FreeBSD: sound just works! Their modified version of OSSv3 that supports in-kernel mixing, auto-device cloning (apps just open /dev/dsp, the kernel auto-maps that to either a hardware channel (/dev/dsp0, /dev/dsp1) or a software channel (/dev/dsp0.0, /dev/dsp0.1) and everything just works and sounds nice. Been that way since the 4.x days (possibly earlier, but I didn't use FreeBSD as a desktop until 4.0, 3.x was just on servers).


In the 4.x days, sound was a bit complicated because it required you to build a custom kernel, as far as I remember, because the KLD infrastructure wasn't so comfortable those days. Took 24 hours on my 150 MHz P1. :-) Today, you can just kldload the proper module and sound runs.

But all imaginable sound devices worked, starting with an ISA SoundBlaster, and today I'm using a PCI sound card that supports the cmi driver - I try to avoid using the AC'97 kind "sound emulator through CPU". Works like a charm, and worked for many years. It's interesting that playing audio doesn't create massive CPU load on FreeBSD systems.

One problem I see is the already present idea to moving all stuff to USB; I'm talking about "USB sound cards", USB microphones, USB headphones and other stuff. Support for them may lack.

But as I said earlier: I don't own it - I don't care for it. :-)

Reply Parent Score: 2