Vienna Symphonic Library Forum
Forum Statistics

186,538 users have contributed to 42,463 threads and 255,857 posts.

In the past 24 hours, we have 3 new thread(s), 6 new post(s) and 43 new user(s).

  • Crackling and dropouts with Vienna Ensemble and logic. Are 4 VE instances too much?

    Hello.  I'm running into some weirdness with Vienna Ensemble and Logic.  I wanted to use Vienna Ensemble (instead of the Vienna Instrument) as plugins in logic but I'm wondering if I'm expecting too much by having 4 instances of Vienna Ensemble at a time.  I've been trying to build up a template and I can't get past having more than 3 instances before digital crackling and dropout occurs. This is with a very low number of samples loaded (one or two instruments per instance).  I've tried playing with the buffer bumping it up to 528 with no difference and I'm using FW 800.  I'm currently on a:

    MacPro 8core 2.8GHz with 18G of Ram. OSX 10.5.6

    Apogee Symphony DA/AD16x

    Does anybody know if I'm expecting too much by loading 4 instances of Ensemble2?  If not, any suggestions?

    I'm in a situation where it makes more sense for my workflow to use Ensemble (instead of the VI) but obviously it doesn't make much sense if it's killing the computer. 

    Any advice would be appreciated. Thanks!


  • I can't answer your question directly, but in XP64 I have 5 instances running with 17GB loaded, so your computer should be able to handle 4 instances. However, do check the memory usage when crackles occur, in case this has something to do with the poor performance you're getting.


  • Hello AG,

    errors like this can be linked to your hard disk drive too. Playing too many samples from a hard disk drive which is too slow/ too full or too fragmented can result in errors like drop out´s etc.

    Which hard disk model do you have?


  • Thanks for your replies.   When testing it, I'm only using 1 solo pizz instrument on each track and I'm getting this issue.  I'm using a 2TB raid drive (western digital) FW 800.  I'll have a look at my disk and if it's not a fragmentation issue, maybe I'll move the library onto an internal SATA drive and see what that does. I shall report back. 

    Thanks again, 


  • Hello again.  So I've loaded everything onto an internal SATA drive with lots of space and I'm still getting the same glitches.  It sounds like a sync issue but I use this system every day with lots of virtual instruments (both as plugins as well as stand alone instruments synced to logic).  I just tried switching from 44.1 to 48k but it doesn't seem to help.  Does anyone else have this? It sounds like the machine is grinding down to a halt with crackles and pops and dropouts as soon as I introduce a 4rth instrument.  I'm noticing fluctuations or delays in timing as well.  This is only happening with Vienna Ensemble.  I've got lots of other things like BFD and PLAY and Omnisphere working without a hitch so I have to have some sort of setting issue going on with Vienna Ensemble.  Anyone have any advice?



  • I have the same issue with Logic. As soon, as 4 instances of VE play, it crackles. It doesn't matter how much instruments are playing. If each VE instance has only a single instrument, and they all play a sustained note, it will not sound properly. If the same 'composition' is played from a single instance it plays perfectly.

  • Same issue on an 8 core Mac Pro, OS 10.5.6, 10GB RAM running VE3 with Digital Peformer 6.0.2.

    Always goes snap crackle and pop with the 4th instance (even when the 4th instance is reduced to just one instrument).

    Looking at the Activity Monitor there is loads of RAM free and the crackles and dropouts do not align with any spikes on the Disk IO graph. Peculiarly when the single (simple) instrument in the 4th instance began to play with the other instruments the CPU activity went up to about 85% on all CPU meters in parallel and stayed there. Muting that instance gets the CPU down to around 25-30% on all meters and each meter varies somewhat independently rather than in parallel. Could this be an issue with the way in which instructions are assigned to the multi-core processors? Maybe something that's fixed in Snow Leopard / Grand Central?, and/or in VE Pro?

    I'm having to work around by freezing tracks to ensure no more than 3 instances, which is far from ideal.

    One thing I don't understand is why the disk speed matters for VE3 at playback time (not load time). Aren't the samples loaded into RAM and accessed from there?

    My previous project was much bigger, and required more samples, so I had to work with the DAW on Mac, and the VE3 host on 64 bit Windows XP. This 64 bit Windows instance was running inside VMWare Fusion which was running on the Mac Pro described above. Even with this setup, it was possible to have 4 instances running with a dense piece and no crackles or dropouts. I didn't think it was necessary for this project as the RAM requirements are much lower.

    Any thoughts anyone?



  • last edited
    last edited

    @grahamhadfield said:

    One thing I don't understand is why the disk speed matters for VE3 at playback time (not load time). Aren't the samples loaded into RAM and accessed from there?


    No. If all the samples were loaded into RAM, I'd need a 700GB template. Even if I had that much RAM, I wouldn't want to wait for that to load. [:(]


  • This problem could be linked with the form of Inter-Process Communication we are using in Vienna Ensemble, which is "system V messaging". System V messaging has some limitations.

    For Vienna Ensemble 3.1, shared memory will be implemented, which is another form of IPC. Here, separate programs running on the same computer are able to communicate with each other by reading from and writing to a commonly allocated memory block without the limitations of system V messages. I´ll discuss this with our developers and will get back to you ASAP.


  • Cool, thanks! Looking forward to it.

    Cheers Maya.

  • Thank you Maya.

    Thanks everyone else for chiming in.  It's reassuring to know that my technical problem is common and not an isolated issue. 


  • Hi all,

    @ AG: the problems you reported with version Vi/VE r3452 are presently being investigated. As soon as I have further details for you I´ll contact you directly via e-mail.

    However there are quite different reasons for "crackling or audio drop outs", it can be caused by the hard disk drive (too slow, too fragmented or too many FW devices are sharing the same controller, streaming and recording to the same drive....), a buffer size problem and so on, in most cases it is not caused by a software bug.