Windows beta pre-release (incl. VST and 64 bit binaries)

You name it!
typewriter
Advanced User
Posts: 147
Joined: Fri Sep 26, 2008 9:04 am

Re: Windows beta pre-release (incl. VST and 64 bit binaries)

Post by typewriter » Sun Sep 06, 2009 4:46 pm

shuptuu wrote:This is not critical but could we have in cubase only the number of channels as tracks? Or is there a way to modify this default number of tracks? (having 31 tracks when you only need 1 is quite boring...)
This was discussed in forum. Everybody agreed that having 32 channels (16 stereo out) would be the ideal configuration (there is only one). The creation of additional outs depends on the host. I use Reaper and there you are asked after inserting LS if you want 32 channels or not. If you click "no" you'll have only one stereo out.

By the way: you have 32 channels, the first channel is numbered with "0" which is somehow confusing.

shuptuu
Newbie
Posts: 2
Joined: Sun Sep 06, 2009 3:49 pm

Re: Windows beta pre-release (incl. VST and 64 bit binaries)

Post by shuptuu » Wed Sep 09, 2009 2:48 pm

Thanks for your answer,
Ok, I've been confused cause the first track in cubase appear with the name LS1 and the last one LS31 but they are stereo tracks so track named LS31 is channel 30 and channel 31...
If there's no way to have a number of tracks dynamically adjusted to the number of channel in linuxsampler, that's ok. (by the way, In cubase, there's no option to have only one stereo output. When you load linuxsampler, 16 stereo tracks are directly created.)
But, I didn't find the way to use all these tracks: all the channel outputs in linuxsampler are mixed out to the first stereo track. And when I try in fantasia to route the output to an other track, it's not possible, only one stereo output is seen in LS.
Can you help me on that?
For the bug I reported in my last message (sound disappear after a while): I didn't have this problem during my last test. I suspect it was due to a dynamic modification of the buffer size to reduce latency on my soundcard (tascam FW1082) while all apps running (normally cubase support this without problem, but maybe not LinuxSampler...). I will tell you if I can reproduce it.
Thanks
(I didn't mentionned it, but I really appreciate LiuxSampler. It's a very good soft. I'm already using it on linux on an other computer. If now I can use it as a VSTi in cubase, it will be fantastic! The best now would be of course to have a Cubase for linux! )
Olivier

typewriter
Advanced User
Posts: 147
Joined: Fri Sep 26, 2008 9:04 am

Re: Windows beta pre-release (incl. VST and 64 bit binaries)

Post by typewriter » Wed Sep 09, 2009 4:26 pm

shuptuu wrote: But, I didn't find the way to use all these tracks: all the channel outputs in linuxsampler are mixed out to the first stereo track. And when I try in fantasia to route the output to an other track, it's not possible, only one stereo output is seen in LS.
Can you help me on that?
Fire up Fantasia and click on the option button on the panel where your instrument is loaded. If you do that there will be some options among them cr (channel routing) there you should be able to choose the audio output you want. You have to look out: You see only 2 channels at "out", when you click on one of these a drop down with all other available channels will open.

typewriter
Advanced User
Posts: 147
Joined: Fri Sep 26, 2008 9:04 am

Re: Windows beta pre-release (incl. VST and 64 bit binaries)

Post by typewriter » Wed Sep 09, 2009 9:59 pm

Bug report:

I received Win 7 Prof german X64 today.

I installed the actual prerelease of Reaper X64 with the new 32bit bridge.

I installed LS as 32 and 64 versions. Both appear in Reaper - one bridged the other one not.

Here the bug: Inserting the 32bit version opens Fantasia GUI automatically (as it should be) - inserting the 64 bit version does not.

Red Ranger
Newbie
Posts: 2
Joined: Sun Oct 11, 2009 6:10 am

Re: Windows beta pre-release (incl. VST and 64 bit binaries)

Post by Red Ranger » Sun Oct 11, 2009 6:20 am

I'm using LinuxSampler VSTi with Tracktion 1 on WinXP. The plugin loads fine, Fantasia comes up when I need it, and sounds play back and can be routed through the VST effects inside Tracktion. However, when it comes time to render a project, it takes longer than normal and no LinuxSampler sounds in the rendered file. Is anyone having this problem with other hosts?

Also, running LS standalone, when I try to close the backend (sometimes qsampler/fantasia asks and sometimes I have to close it manually in task manager) my computer freezes.

typewriter
Advanced User
Posts: 147
Joined: Fri Sep 26, 2008 9:04 am

Re: Windows beta pre-release (incl. VST and 64 bit binaries)

Post by typewriter » Sun Oct 11, 2009 4:02 pm

Red Ranger wrote:However, when it comes time to render a project, it takes longer than normal and no LinuxSampler sounds in the rendered file. Is anyone having this problem with other hosts?
I had this with Reapers stem rendering, too.

typewriter
Advanced User
Posts: 147
Joined: Fri Sep 26, 2008 9:04 am

Re: Windows beta pre-release (incl. VST and 64 bit binaries)

Post by typewriter » Sun Oct 11, 2009 4:58 pm

I got vienna ensemble pro today. Unfortunatelly LS vsti does not work perfectly on this audio over lan solution.

In the latest ensemble pro release LS vsti dosen't show any panel. It is impossible to select 16 channels. Only the stereo version appears.

User avatar
cuse
Developer
Posts: 366
Joined: Wed Jan 23, 2008 10:07 pm
Location: Germany

Re: Windows beta pre-release (incl. VST and 64 bit binaries)

Post by cuse » Fri Oct 16, 2009 1:55 pm

typewriter wrote:
Red Ranger wrote:However, when it comes time to render a project, it takes longer than normal and no LinuxSampler sounds in the rendered file. Is anyone having this problem with other hosts?
I had this with Reapers stem rendering, too.
I think that's because in those *freeze modes* or *offline rendering* or however you want to call it, the sequencers tell the LinuxSampler VST to run with a huge VST audio buffer size. And LinuxSampler probably can't handle huge buffer sizes yet. Remember that LinuxSampler was til recently only used for realtime rendering, which means small buffer sizes. I think Andreas already suggested a hack to catch those large buffer sizes ordered by sequencers in LinuxSampler's VST plugin class, and splitting the offline rendering call of the VST sequencer in a loop by smaller, realtime typical period sizes.

Red Ranger
Newbie
Posts: 2
Joined: Sun Oct 11, 2009 6:10 am

Re: Windows beta pre-release (incl. VST and 64 bit binaries)

Post by Red Ranger » Wed Oct 21, 2009 6:05 am

cuse wrote:Remember that LinuxSampler was til recently only used for realtime rendering, which means small buffer sizes. I think Andreas already suggested a hack to catch those large buffer sizes ordered by sequencers in LinuxSampler's VST plugin class, and splitting the offline rendering call of the VST sequencer in a loop by smaller, realtime typical period sizes.
Where can I read more on how to perform this workaround?

Andreas
Developer
Posts: 214
Joined: Sun Feb 03, 2008 8:33 am

Re: Windows beta pre-release (incl. VST and 64 bit binaries)

Post by Andreas » Wed Oct 21, 2009 6:43 pm

Red Ranger wrote:
cuse wrote:Remember that LinuxSampler was til recently only used for realtime rendering, which means small buffer sizes. I think Andreas already suggested a hack to catch those large buffer sizes ordered by sequencers in LinuxSampler's VST plugin class, and splitting the offline rendering call of the VST sequencer in a loop by smaller, realtime typical period sizes.
Where can I read more on how to perform this workaround?
I have only talked about this with Christian once in a private chat, so you can't read about it anywhere. I wouldn't call it a workaround, it's a change in the code. The DSSI plugin already has this type of functionality, look in PluginDssi.cpp, RunMultipleSynths - there you can see that audioDevice->Render is only called with buffers with max length 128. If SampleCount is bigger than 128, the Render function is called multiple times. The LV2 plugin also has this, maybe it's easier to read.

I'm not sure if this is enough to fix offline rendering though. If the rendering is done faster than real time, I think we need to implement a special mode in LS, where the audio render call waits for the disk thread to fill the buffers. As it is now, if a secuencer calls the plugin faster than real time in order to render audio to disk as fast as possible, the LS disk threads will not keep up.

Post Reply