1 |
On Mon, Sep 19, 2011 at 12:23 PM, pk <peterk2@××××××××.se> wrote: |
2 |
> On 2011-09-18 21:52, Michael Mol wrote: |
3 |
> |
4 |
>> The kernel configuration process is actually very nice and very easy. |
5 |
>> You an remove any features you don't want or need. (I'm referring to, |
6 |
>> e.g. menuconfig. I haven't really used genkernel) |
7 |
> |
8 |
> I've never used genkernel and always compile my own kernels... |
9 |
> |
10 |
>> FWIW, PulseAudio predates Windows Vista, Windows 7, even MacOS X. I |
11 |
>> ran it on a 200MHz machine back when it was called Enlightenment Sound |
12 |
>> Daemon. |
13 |
> |
14 |
> Hm... I've used ESD (years and years ago :-) ) in OSS times. Not sure |
15 |
> where the connection between ESD and Pulseaudio is though[1]... Well, |
16 |
> anywho, Gentoo stopped supporting both the ESD and arts server years ago |
17 |
> for security reasons IIRC. |
18 |
> [1] https://secure.wikimedia.org/wikipedia/en/wiki/Enlightened_Sound_Daemon |
19 |
|
20 |
My recollection at the time PA started showing up was that PA was the |
21 |
descendant of ESD. I assumed it was a fork. I may be wrong. |
22 |
|
23 |
>> With as much as I've poked at PulseAudio, I'd have to say I like it |
24 |
>> better than I like the Vista/Win7 implementation of sound daemons. |
25 |
> |
26 |
> I've no experience with Vista/Win7 (I've got an XP machine for gaming). |
27 |
|
28 |
Windows coding is my day job. Workstation is Win7 Ultimate x64. |
29 |
|
30 |
> |
31 |
>> There's probably not much one can do with PA that one couldn't do with |
32 |
>> jackd, which is probably better in terms of latency, but I never got |
33 |
>> around to learning jackd. |
34 |
> |
35 |
> Yes, tried jack a few years ago but couldn't get it working right. Not |
36 |
> that I got "burnt" by it (as dbus etc.) and if the need arise, I'll look |
37 |
> into it again... Thing is, ALSA already have a (simple) sound server |
38 |
> "built" in called dmix so why one would bother with Pulseaudio is beyond |
39 |
> me (but as long as it's not "forced" onto users I don't care much about it). |
40 |
|
41 |
I recall reading about dmix in LinuxJournal years ago, but I don't |
42 |
think I ever got around to setting it up; ALSA was just going through |
43 |
a major API change around 0.9, and I didn't have the resources to stay |
44 |
up-to-date. (Dial-up was a pain. I imagine it's worse today) I did |
45 |
have fantasies about using it to set up a fake sound device to get |
46 |
spatial audio over headphones. I couldn't find the data set I'd needed |
47 |
for calculating delays, though. |
48 |
|
49 |
> |
50 |
>> While I was using PA (I'm not, currently), it was nice for being able |
51 |
>> to monitor and tune the volume levels of individual programs. That can |
52 |
>> be important when trying to manage two different VOIP apps, video |
53 |
>> games and Pandora at the same time. |
54 |
> |
55 |
> If you wish to use it then do so... :-) |
56 |
> |
57 |
> The thing (idea) I was trying to convey but seems to escape most people |
58 |
> is this: |
59 |
> "Cut out the fat!" - "Less is more/do more with less [resources]" etc... |
60 |
> in a general sense. That's what Contiki is doing and what I think |
61 |
> software in general should be doing (yes, in Utopia)... When adding |
62 |
> layer upon layer, we are going in the wrong direction (unless the sum is |
63 |
> less, which it, in my "eyes", seems not to be). |
64 |
|
65 |
Oh, certainly. That's one of the reasons I love Linux's (and |
66 |
especially Gentoo's) modularity so much; there's often a nearly-ideal |
67 |
tool for any given use case. That's part of why I don't like to see |
68 |
things which break that modularity become mandatory. |
69 |
|
70 |
|
71 |
-- |
72 |
:wq |