Page 1 of 2
SCS 10.3.0 suddenly became really unstable until.... (fixed)
Posted: Mon Jan 05, 2009 3:02 pm
by superspinner
Hi all!
I am a new user. I had been experiencing glitching and read the forum and the instructions to turn off using internal mixer and pre-load all audio cues. This worked.
Everything was really solid for two weeks until last night when I started tweaking some cues. Then scs became very delicate and would easily crash around the last 5 or so cues
in my cue list (it's about 75 cues, all audio and sfx cues). Once I saw an "index array out of bounds". It got so delicate that if I advanced a cue before the previous
one faded (or ran to completion) scs would usually crash. I could also always crash it by trying to advance past the last cue (an audio cue).
crap!
I tried a number of things like turning back on the internal mixing, no pre-loading, but nothing worked. I also tried deleting the audio cue and companion fade
cue that seemed the most delicate and then re-adding them. No joy.
For the heck of it I added after the last cue an sfx cue with an action to stop the playlist at end of current track. To my amazement this fixed it! Now I can't crash
it no matter what I do.
So what happened? Is there a bug such that it's possible to get the count of cues off in a way that crashes scs?
Thanks
Randy
Re: SCS 10.3.0 suddenly became really unstable until....
Posted: Mon Jan 05, 2009 4:30 pm
by Mike Daniell
If you're still in production, leave things the way they are (assuming they remain stable), but when your production has finished can you see if removing that last cue returns you to the unstable state? If so then let me know and I'll email you details on more information that I will need.
Re: SCS 10.3.0 suddenly became really unstable until....
Posted: Mon Jan 05, 2009 5:57 pm
by superspinner
Thanks Mike.
will do.
Re: SCS 10.3.0 suddenly became really unstable until....
Posted: Tue Jan 13, 2009 7:04 am
by superspinner
Crap, it happened again. The first time when I was editing fade-out times and fade-out curve types. No addition or deletion of cues. Just editing existing ones.
I changed all the cues to not pre-load the sounds. This seems to make it stable but then it crashed right around the beginning of the last 5 cues with
absolutely nothing happening during the run through. Fortunately it takes less than 10 seconds to reboot, so we just squeaked by.
But check this out.
I noticed I was able to cause an error by pressing the windows key and either q or a. This caused a box to appear saying
there was some bounds error and asking me whether I wanted to quit the program or not.
This was running 10.3.0. I noticed that there is a 10.3.1 update. I installed the update and after that was unable to reproduce the
windows key q/a error. I then pounded on the program for an hour in every conceivable way I could think of and couldn't crash it.
Tonight I have a run through. I then have 2 dress rehearsals and 5 nights of performances. Tonight is the last chance for running the
program on this laptop. It looks like a software bug but it must be a function of the laptop itself, the wav files I am using (standard 44.1 mono and stereo 16 bit)
and the combination of cues. But all the cues are simple sound cues with some having fade out cues. There is only only level change cue. I have some long fade-outs,
but it's really straightforward stuff. And I have the wireless device disabled, have gone into msconfig and disabled every service (other than microsoft ones), I'm using the
onboard sound, so the machine is running bare metal. I have run dpclat.exe, a latency checker program, and there are no big spikes.
I wish I could have 2 instances of scs running so when one crashes we can just start using the other one.
Wish me luck....
Re: SCS 10.3.0 suddenly became really unstable until....
Posted: Tue Jan 13, 2009 11:19 am
by Mike Daniell
If it does crash again and you have time to do this, please take a screen shot of the error and email it to me (preferably as a jpg file attached to the email). Alternatively, write down the information displayed in the error box and send me this info. I need all the info displayed, not just the type of error.
Re: SCS 10.3.0 suddenly became really unstable until....
Posted: Wed Jan 14, 2009 5:09 am
by superspinner
Mike Daniell wrote:If it does crash again and you have time to do this, please take a screen shot of the error and email it to me (preferably as a jpg file attached to the email). Alternatively, write down the information displayed in the error box and send me this info. I need all the info displayed, not just the type of error.
I am banging on 10.3.1 to try and crash it. So far so good. I perused the event log and found a bunch of scs crashes. I looked at them but there are only 2 different kinds of crashes. Maybe I should have just sent you the event log and maybe just emailed it to you.
Faulting application scs10.exe, version 10.3.0.0, time stamp 0x492b215c, faulting module bass.dll, version 2.4.2.3, time stamp 0x49071b13, exception code 0xc0000005, fault offset 0x00002b78, process id 0x172c, application start time 0x01c96edf944109c3.
Faulting application scs10.exe, version 10.3.0.0, time stamp 0x492b215c, faulting module MSVBVM60.DLL, version 6.0.98.2, time stamp 0x4791a724, exception code 0xc0000005, fault offset 0x00105ce1, process id 0x15d8, application start time 0x01c95781b1021877.
Re: SCS 10.3.0 suddenly became really unstable until....
Posted: Wed Jan 14, 2009 7:46 am
by Mike Daniell
Ah, so the errors weren't the type caught by SCS but errors detected by Windows. Unfortunately the event log won't be much help as the original error would have occurred elsewhere in the program. Errors caught by SCS display the name of the SCS procedure that trapped the error, and sometimes display other information as well, such as values of selected internal variables or pointers. These errors are much easier to track down. Hopefully 10.3.1 will continue to run OK for you. Otherwise I may need you to run a trace version.
Re: SCS 10.3.0 suddenly became really unstable until....
Posted: Wed Jan 14, 2009 9:52 am
by superspinner
Ah, too bad.
But I have been running 10.3.1 for about 6 hours now, doing everything I can think of to it and haven't been able to crash it. If it runs without incident at our dress tonight I am going to declare it cured with the update.
BTW, it's great that it starts so quickly even when it does crash. However, I would really love to have another copy running as a hot backup. I guess that's not possible because of some backup files or something in the program files directory?
As soon as I can verify it as rock solid, I want to role it out to another system (another license) and train techs on it.
Thanks
Randy
Re: SCS 10.3.0 suddenly became really unstable until....
Posted: Thu Jan 15, 2009 6:42 pm
by Mike Daniell
superspinner wrote:I would really love to have another copy running as a hot backup. I guess that's not possible because of some backup files or something in the program files directory?
The main issue is that if you edit your cue file in one instance of SCS, then a second instance would need to re-sync itself with the changed file. This could be a problem if you have unsaved changes in both instances.
Re: SCS 10.3.0 suddenly became really unstable until....
Posted: Wed Jan 21, 2009 3:19 am
by superspinner
I just finished a week of rehearsals and a 5 performance run, and.... 10.3.1 was solid as a rock. Not a single glitch. So the solution for me was to update to 10.3.1. scs10 was key for me and allowed me to do things that just would not have been possible to do otherwise. What a great tool! Can't believe I ever got along before without it.
Thanks
Randy
Re: SCS 10.3.0 suddenly became really unstable until....
Posted: Wed Jan 21, 2009 6:12 am
by kharney
So Mike,
This brings up a the backup copy question again for me. I like the idea of two copies of SCS running at the same time. Once we have gotten through all of our editing and have the show set for the runs is there any reason why this would be a bad idea ? Does anyone else have any other ways that they are making sure that a show does not crash if the computer or SCS does. I have never had SCS crash but I like to play it safe. I do not have a budget for another computer. Also wondering how others handle the possibility of the whole computer crashing with no recovery. Do you burn effects to a CD ? so that the CD can be inserted and the show can go on ...
Re: SCS 10.3.0 suddenly became really unstable until....
Posted: Wed Jan 21, 2009 11:03 am
by sound
kharney wrote:So Mike,
This brings up a the backup copy question again for me. I like the idea of two copies of SCS running at the same time. Once we have gotten through all of our editing and have the show set for the runs is there any reason why this would be a bad idea ? Does anyone else have any other ways that they are making sure that a show does not crash if the computer or SCS does. I have never had SCS crash but I like to play it safe. I do not have a budget for another computer. Also wondering how others handle the possibility of the whole computer crashing with no recovery. Do you burn effects to a CD ? so that the CD can be inserted and the show can go on ...
You can do a lot of things to increase your odds...
First step:
http://www.audioforums.com/resources/wi ... ation.html
IMHO its a good idea to have a computer stripped down to the essential:
http://www.nliteos.com/
Don't think that burning effects to a CD is a solution except for a very simple setup.
The next step is to run 2 computers in sync - for the 'money shows' I have 2 computers running in sync connected to seperate input channels on 2 VCA's and if one fails I just have to switch the VCA's ...
Greetings,
Walter
Re: SCS 10.3.0 suddenly became really unstable until....
Posted: Sun Jan 25, 2009 8:12 am
by superspinner
sound wrote:
<SNIP>
The next step is to run 2 computers in sync - for the 'money shows' I have 2 computers running in sync connected to seperate input channels on 2 VCA's and if one fails I just have to switch the VCA's ...
Greetings,
Walter
Hey Walter,
for my show I had a hot backup, but have a couple of massive shows (4 hours+) that are one chance gigs. Never done one of these before. How do you trigger the cues simulataneously? What gear do you use?
Thanks
Randy
Re: SCS 10.3.0 suddenly became really unstable until.... (fixed)
Posted: Mon Jan 26, 2009 11:36 pm
by sound
I have the show running on 2 computers and I trigger the cues with MIDI sent out of the board. So its MIDI Out connected to a MIDI splitter and then to the MIDI In of the computers.
Main computer is a rackmounted PC with an RME Digiface and RME ADI-8 converters, the backup machine is a Lenovo notebook having a MOTU Ultralite connected via Firewire.
Check out
http://www.audioforums.com/resources/wi ... ation.html to optimize your PC for Audio.
I'm also running a stripped down version of XP, see
http://www.nliteos.com/
http://www.msfn.org/board/nLite-f89.html&st=25
http://forum.cockos.com/forumdisplay.php?f=29
for more information.
Greetings,
Walter
Re: SCS 10.3.0 suddenly became really unstable until.... (fixed)
Posted: Tue Jan 27, 2009 3:58 am
by superspinner
sound wrote:I have the show running on 2 computers and I trigger the cues with MIDI sent out of the board. So its MIDI Out connected to a MIDI splitter and then to the MIDI In of the computers.
Main computer is a rackmounted PC with an RME Digiface and RME ADI-8 converters, the backup machine is a Lenovo notebook having a MOTU Ultralite connected via Firewire.
Greetings,
Walter
Thanks Walter! So your board triggers the cue, then the audio comes back into the board and is controlled by two vcas. What board are you using that has vcas?
Cheers
Randy