Is this audio from the previous patch that cut out or audio from the current patch you just switched to?
It will cut off audio from patches you switch away from if your cpu load is too high. You might be able to decrease cpu load quite a bit by changing all of the plugins and the input to be mono (especially if you are using a guitar, this probably is what you want anyway).
I'm having exactly the same problem. I'm using a MacBook Pro Unibody and an Apogee Duet. Clicking on patches sometimes works as expected, but other times the audio becomes silent as soon as I select another patch. Clicking on a different patch usually fixes it (and often then going back to the silent patch then works normally). It's a 2.4GHz MBP, 4GB RAM, and none of the meters are more than about a third across.
I have the same issue on occasion. In my case, I've discovered that just flicking the expression wheel on the faulty channel strip snaps the sound level right back.
I've tried muting cc 11 to the strip, but it doesn't seem to make a difference. The only thing that I can say might relate is the fact that the pedal off of my controller (an S90ES) is sending CC 11. But, there is no predictability to a patch getting muted (or quieting in volume). 3rd party PI's (VI's) seem to be more easily effected, as does a channel strip used for the S90, an external one re-routing MIDI data back to the S90. Local control is off.
Now I need to restate, it is rare that a channel strip gets muted. Not quite as rare when the volume from the S90 is lessened. But in both cases, a flick of the Expression dial on the channel strip always corrects the problem. However, it always displays full level, or at least 70 on a scale of 0-127.
Now, in my case at any rate, where is it getting the CC11 info from, and more importantly, why the discrepancy between the readout and the level?
btw, I'm using a MOTU Ultralight for the interface, and the S90ES is going in via USB.
Your device might be sending out CC11 (expression) randomly. I saw this once at a friend's house... we worked around this by creating a knob in the layout that was assigned to expression, and set the "midi thru" to "do not pass through." This intercepted the expression message that was being sent, and stopped the problem from happening. I guess the other way to fix the issue is to see if you can keep expression from being send in these random cases, but MainStage is easier to set up then trying to figure out a cryptic series of steps from a controller's/synth's manual!
I too am experiencing this same issue. To be clear, let me explain my exact scenario.
First, the setup is as follows: 24" aluminum iMac 2.4GHz, 2 Gb RAM, OS X 10.5.8, Mainstage 2.0.1, M-Audio Axiom 25, FastTrak Pro, Behringer FCB 1010.
Yesterday, I created a new guitar rig concert just to play around with the sounds and get my foot controller set up to control stompboxes, patch changes, etc. I used the factory template, 'single amp with patches.' So, this was not a custom concert made from scratch with a ton of resource hogging plug-ins. I got everything assigned and up and running. Then, I noticed that when changing patches, occasionally (about 50% of the time I would approximate) there is no output from the new patch. The new patches' input meter shows signal, but the output does not. This occurs whether I change patches from my foot controller, the arrow keys on the keyboard, or via a mouse click on the patch list. This also occurs when switching to a new patch via jumping to a new set by any of the above means.
I have restarted Mainstage several times which did not solve the issue. I restarted the computer also, but this didn't help either. The CPU and memory meters are not in the red. I thought at first, it may be that I don't have enough RAM, but seeing that this is happening to others with more than enough verifies to me the meters are relatively accurate and this is not the issue. I am going to continue troubleshooting this issue by retrying this scenario on my 13" Macbook (with 4 Gb RAM) to reconfirm that it is not a shortage of RAM on my part.
I have not extensively used Mainstage 2 since it's release, but I have been playing with it enough to know this is a new issue. Maybe related to the 2.0.1 update? I got the software shortly after release, and played quite a bit with the new features. I never had this issue before.
I have found somewhat of a work-around, but I don't know how well it will hold up. While troubleshooting, I came to notice that when I came to patch that did not produce output, if I backed up to the previous patch then switched to the new one again, it worked fine. It seems almost as if previously unused patches in a concert are somehow inactive. But once they are "woken up," they stay awake and work fine. My final test yesterday was to open the concert and then simply hold down the arrow key to scroll through the entire patch list quickly. Then, I could switch to any patch or set and it worked as expected. Now, I have not tested this work-around for long periods of time, such as that which would be required in a live situation. I don't know yet if the patches will "go back to sleep" after a certain period of time.
I realize that I am coming into this discussion thread late. I hope those who have posted here have not abandoned the issue. If necessary I will post this as a new question. I was just wondering if anyone has learned anything more about this issue. I am unsure what to do next to try and solve the problem. One question for those affected- What version of OS X is everyone using? Is anyone using Snow Leopard? I just wonder if this is occurring to those who are using SL also.
Any advice or suggestions is greatly appreciated!
What you are reporting is exactly the same experience I have had, albeit not as frequently. My OS is SL, which I have now upgraded to 10.6.2 - keeping an eye on things to see if that's made any difference.
Did you report this to Apple via the Mainstage feedback? If not, please do, as it'll help give them a bit more to go on (and convince them it really is a bug!).
I have reported the issue vie the feedback option in Mainstage. I will keep my eye on this thread and update should I learn anything helpful. I hope for your sake the 10.6.2 update fixes the issue. Please let us know if that is the case. I am still holding off on the update to SL due to issues with other apps.
Well, hello again - if you're still watching, here's another update.
I thought 10.6.2 had cured the 'silent patch' issue, but I'm afraid I have to report another occurence of it in rehearsal last night. Admittedly, this is the first in quite a long time, but it still makes me very nervous about switching patch mid-song during a performance.
Once again the issue was resolved by quickly switching to another patch then back again, but you wouldn't want to have to do that at the top of your big solo.
Message was edited by: littleeden
Hi, I noticed that some patches when switched to or it could be just a change on anything which causes a spike which then freezes up. The problem existed in MS1. Re-initializing the program. It is possible the setting at the end of the patch being played before the patch being switched to isn't set right. Or the patch you are switching to may be set wrong. There are choices of how a patch changes make sure they are setup right for what you want. It is always good practice to check your patches and settings before your ready to go. It may be something as simple as adding or changing a setting. Best of Luck! Always re-initialize when you make changes if it doesn't work the 1st time. Saving your work may save you some time narrowing it down.
Well, here we go again. Same problem at my New Year's Eve gig last night, switching patch for a solo and ... silence! From now on, I'm going to ditch all mid-song patch changes and set up everything with stomp boxes.
Another interesting one was a silent channel - I opened up a concert, and there was no audio on my vocal channel in the first song (ok in all the others). There was audio in the guitar channel, just not in the vox. The input was set correctly, nothing muted, just no level on the meter.
I checked again this morning just using my MB's internal in/out, and the problem was still there. However, I closed that concert, had a look at a few more, then came back to the original one, and waddya know, it was working fine again.
Can anyone cast any light on this?