Hello!
I have tried to run fluidsynth-1.0.1 on my Linux system with an ES1371
sound card (i want to attach a MIDI keyboard to the sound card's midi
input). But it said that there's no device /dev/midi .
There indeed is none, but midi0, midi1, midi2, midi3, midi00, midi01,
midi02, midi03 are present.
I tried:
ln -s /dev/midi0 /dev/midi
but this did not help - fluidsynth still says that no device named
/dev/midi is available.
Is there any way for me to use fluid with my PC's midi input?
--
Sincerely yours, Mikhail Ramendik
On Tue, Apr 08, 2003 at 02:40:40 +0900, Patrick Shirkey wrote:
> Steve Harris wrote:
> >On Mon, Apr 07, 2003 at 11:56:16PM +0900, Patrick Shirkey wrote:
> >
> >>>There will be 3 band compression, linear parametric EQ and a lookahead
> >>>limiter, maybe also a learning limiter.
> >>>
> >>
> >>Is that called time?
> >
> >
> > ?
>
> >>>maybe also a learning limiter.
>
> :)
Ahh, yes time, but also working out a UI for it. Its kinda optional, so I
doubt it will make 0.1.
> >If your willing to work on the UI, then maybe the thing to do is to strip
> >it down and rebuild the UI code. I can stick the current code in a public
> >CVS server somewhere (its currently on a private server), but its pretty
> >scary in its current state.
>
> I have no qualms about learning SDL unless you really think it is a bad
> idea. The thing is *learning* it but I am willing to drop (almost)
> everything and focus on it for a while.
Well, theres almost nothing to SDL (well the parts I used anyway), I just
dont think its appropriate, I ended up writing widget code for the sliders
and stuff, which is why they suck so much. In retrosect just using GTK
would have been easier.
> If you want I can give you ftp access to djcj.org and you can put it
> there in a semi/private place.
The intention wasn't that it be especially private, I just didn't think
anyone else would want to work on it. I originally wanted to get beta bug
reports from a small number of people so I could handle the feedback, but
I dont think that will be a problem now.
> >I have no particular feelings about what toolkit, as long as its C, and
> >preferably has working vertical faders, the ones in GTK-1.x are
> >inconvienient.
>
> I willing to go with the current stuff and work from there. It may be
> that it needs a complete redesign but I think it would be good to get
> something out sooner rather than later.
Thats a good point. The current state is that I'm halfway through
refactoring the UI code to make some of the compressor controls rotary, so
its impossible to tell what its doing at the moment. The question is
wether its quicker to rescue the SDL code or replace it with GTK code or
similar.
Is there anyone else who wants to work on this? If not I'l just leave the
code where it is for now and give Patrick a login on the machine. If there
are a few people I'l have to move it to sourceforge or something, but I
dont really want to do that cos I think theres going to be some major
structural work needs doing.
- Steve
Hello,
I have just got fluidsynth to run on my Linux box. It works fine with
most soundfonts.
But I ran into a strange problem with "Jeux d'orgues 2", the big organ
soundfont which is right at the top of hammersound.net's organ list.
It plays, but produces a click on every release of a key!
Of course I looked at the output of "top" in a different console, and
fluidsynth was only using 10% or so of teh cpu. No other tasks were
running (before I started "top").
I can't check it with an SB Live, so I do not know if it is perhaps a
problem in the soundfont. But somehow I doubt a soundfont woth such an
obvious bug would be up there for so long and the author would not
notice. So, can this be a problem with fluidsynth or my system?
And, is there a way to check, without access to an SB Live? (Perhaps I
should try to install timidity?)
By the way, is there any other good organ soundfont available?
--
Sincerely yours, Mikhail Ramendik
Version 2.1.1 of the score editor NoteEdit for Linux is available:
http://rnvs.informatik.tu-chemnitz.de/~jan/noteedit/noteedit.html
New features:
- staff context (the actual clef and key signature is shown at the
left hand side during replay; configurable)
- (again) KDE2/Qt2 compatible
Please note: I cannot really test this. For a long run I recommend
an KDE3/QT3 update.
- double bar
- Segno issues (see also "example1.not"):
o Segno
o dal Segno
o dal Segno al Fine
o dal Segno al Coda
o Fine
o Coda
Please set these signs in every staff, otherwise ... (don't know ???)
Furthermore, it is your responsibiltity to place the signs
in a reasonable manner. Otherwise the replay behaviour is undefined.
- MUP users: I didn't comprehend some parts of MUP. Thus, I stored
some items in a non-MUP-conform manner. As a result some files
were not MUP-able. Now I changed some of these parts, among them:
o Slurs: I claimed MUP cannot deal with slures which lead over more
than 2 notes. This is not true! NoteEdit-2.1.1 now
produces a MUP-comfrom representation of such slurs.
This holds for the 1st and 2nd voice. From 3rd voice on
the representation remains MUP-incompatible.
o "lonely" trills: I claimed MUP cannot deal with so-called
"lonely" trills (wavy lines without "tr" symbol).
But this is not true. NoteEdit-2.1.1 now produces
a MUP-comfrom representation of such trills.
o Empty lyrics: I claimed MUP cannot deal with partly missing
lyrics in a measure. This is not true.
NoteEdit-2.1.1 now produces a MUP-comfrom
representation of such lyrics.
I apologize to the MUP developers for these wrong statements.
Before asking this: Of cource, NoteEdit can still read the old
representation. So if you want a MUP conform
version of some non-MUP-conform "*.not" file
simply load the the file an save it again.
--
J.Anders, Chemnitz, GERMANY (ja(a)informatik.tu-chemnitz.de)
hey, all.
I'm a lurker, currently confined to recording my albums on a 200 MHz Pentium II
with 32 megs RAM running Win95 and Cakewalk 8. I'm sure I'll be happy with
whatever latency I happen to get when I buy the new computer and put linux on it.
:)
In the meantime, I wanted to mention indexing as another nice feature to have in
a mastering/burning application, and wanted to mention Wavelab as a program I
absolutely love mixing and "mastering" (I'm a geek with a PC - I don't know how
to master :) with - it's the only program ever built with that "stuff the thing
full of features till there's no screen left" mentality that I've actually liked.
So we might glean some ideas from them. (Or not - it can be frustrating to have
the pointer change shape and function every 50 pixels or so...)
I look forward to contributing more as I can - just got my first full-time desk
job, so I get to spend the next few months picking out the perfect motherboard,
the perfect case, etc... in the meantime, I'll be
Lurking,
dan
Rev. Dan Easley
Burnt Possum Poets
Harrisonburg, VA, USA
Hi everybody,
Simple question:
I want to use my Evolution USB-keyboard for input with rosegarden. I
don't seem to get any midi-i/o, nor any sound comming out of the
speakers. (Yes, checked the mixersettings)
How can I solve these 2 probs?
When I compose the notes manually, I see the small VU-meter next to the
channel is moving when playing the track, but in no way I get any
soundoutput...
Any hints?
Karel (alias Moeflon)
|
-----> the perfect midi-newbie :-p
Someone asked recently about a sound visualisation app and got no
response IIRC. I just received a link for the alsa applications page
http://www.maysl.at/
Fische provides standalone sound visualisation under linux using the
ALSA and SDL libraries
Looks pretty trippy.
There are also:
bomb
synaesthesia
various xmms plugins
--
Patrick Shirkey - Boost Hardware Ltd.
Http://www.boosthardware.comHttp://www.djcj.org - The Linux Audio Users guide
========================================
Being on stage with the band in front of crowds shouting, "Get off! No!
We want normal music!", I think that was more like acting than anything
I've ever done.
Goldie, 8 Nov, 2002
The Scotsman
philicorda wrote:
>>would be nice to open up the possibility of some FFT stuff and spectral
>>mangling a la Chris Penrose's 'PVNation' or NI Spektral Delay... which i
>>use rather a lot in Live - im wanting to move my Live stuff to linux...
>
>
> Check out freqtweak. It's like spectral delay and has some cool pitch shifting
> stuff too.
>
> http://freqtweak.sourceforge.net/
ahh cheers yes i know freqtweak, very yummy...
i was kinda thinking along the lines of plugins tho... afaik you cant
use freq. as a plugin as yet?... the nicest thing about spektral delay
is that i can use it as a plugin in live, so while i have all that other
stuff going i can mangle my sounds to my hearts content... dunno that i
could use freq. all of its own for a set. im looking at moving to using
spiralloops tho instead of live, as it uses ladspa plugins...
bw
m~
>
>
--
|\ _,,,---,,_
ZZZzz /,`.-'`' -. ;-;;,_ HTTP 503: Too Busy
|,4- ) )-,_. ,\ ( `'-'
'---''(_/--' `-'\_) fL
.::. www.iriXx.org .::. www.copyleftmedia.org.uk .::.
gnupg key ID: AEB7A31E
http://plugin.org.uk/releases/0.4.0/
This one is a very experimental release, hence the .0, its even more
likly to cause pain and death than other releases :)
I've attached a list of the plugins against what state I think there in,
for the people who want to help by testing, it should save you some
time by weeding out the very simple and definatly broken ones. Please test
against this version, not previous ones.
It includes rough support for gettext, so, in theory the plugins can be
localised without support from hosts. If people want to provide
translations, look in the po directory, I think theres everything you
need, but I've never used gettext before, so I may have messed up.
I've done some significant work on the compressors, most notably SC4,
(stereo, no sidechain) which has a continuously variable RMS/peak control
and a magick 'fast' mode: when the attack time is set to very low it
behaves more like a limiter. This feature seems to be present on a lot of
hardware compressors.
SC4 now has ID 1882, so you should move away or delete the old sc4_1434.so
file. It wasn't strictly neccesary to rename the file, but it would be
confusing (to me) if I didn't.
The others all have slightly faster responses and less damping after I
found out that you were /supposed/ to get pumping with fast envelopes :)
I ran through them with demolition
(http://www.ecs.soton.ac.uk/~njl98r/code/ladspa/) and fixed a lot of crash
bugs, so things should be more stable once the major code changes settle
down.
There is still no support for the beta FFTW (version 3), or OSX. I wanted
to get this release out this week, they should be supported in the next
version.
Summary:
please test, especially the compressors
translations welcome
- Steve