Vienna Symphonic Library Forum
Forum Statistics

201,249 users have contributed to 43,236 threads and 259,222 posts.

In the past 24 hours, we have 2 new thread(s), 9 new post(s) and 63 new user(s).

  1. Home
  2. Sample Libraries
  3. Vienna Instruments Libraries (legacy)
  4. Vienna Ensemble-better performance through host than stand alone

  • Vienna Ensemble-better performance through host than stand alone

    C
    C
    Help me...can anyone answer this. I have been troubleshooting for days now and have come to the conclusion that you can get better performance running Vienna Ensemble through a host than as a standalone. I set up a template through VE standalone and found that a test sequence I had made (kind of torture test for the software) was getting audio pops and hiccups when it was played. It seemed to me that the sequence was tough but I felt that the computer should have been able to handle it. (G5 dual 2.7, 8 gig ram-I have another one so I have some experience with what it can and can't handle) anyway, I decided to take the same Vienna pallet of sounds and set it up inside a VI host. (Both Digital Performer as VI host and V-stack) and the same exact sequence played without any audio glitches. It was clearly running close to the limit of the machine but was playing glitch free. The buffer settings were always the same. I actually would expect the VI host idea to tax the computer more because to run the same pallette of sounds inside a host you had to run two instance of VE to accomodate all the sounds. Hopefully this is clearly written out. Does anyone have any idea what is causing this discrepancy in performance?
  • cmC
    VSL Team
    cmC

    just a shot into the blue ... it is caused by the display resp. the GUI needing more resources in stand-alone mode than as plugin.

    be it because a larger part of the VE GUI is visible as stand-alone or be it that communication between the GUI and aqua needs more CPU ...

    does anything change if you minimize the window?

    christian

    and remember: only a CRAY can run an endless loop in just three seconds.
  • marnixM
    VSL Team
    marnixM

     Hi,

     

    You mention a test sequence - did you play back this sequence in DP adressing Vienna Ensemble as standalone through the IAC Bus or did the sequence come from another computer?

     

    Almost always, audio pops are in some way related to the communication between the audio hardware drivers and Core Audio - except for simple CPU overflow and harddisc bus issues, which both do not seem to apply in your case.

    Please give us more information about your Audio Interface/Drivers and the exact buffer settings you used in either Vienna Ensemble standalone/Host and Vienna Ensemble Plugin/Host.

     

    Also, have you altered settings in Vienna Ensembles Plugin Latency Configure utility?

     

    Best, Marnix 
  • C
    C
    update to my original post: Man, life can be complicated. So, I thought I was getting better performance running VE through Digital Performer or V-stack. Well, it only appeared this way. I ran some tests where I recorded audio from a test sequence to evaluate actual timing of events as they are played through the different hosts with different buffer settings. (I did this cuz I knew something wasn't right...although I should have been working!!) Well what I found was very interesting.... While it appeared DP and V-stack were giving me better performance when set to a particular buffer setting, 256 for example, they were actually playing quite a bit later than VE at the same 256 buffer setting. So when I was getting better performance out of the two VI hosts I think it was because in fact they were working with a larger latency than indicated by the buffer setting. DP was worse than V-stack (for mac) but V-stack was still a good bit behind the actual performance when compared to using VE stand alone. VE as standalone gave very good and consitent results. Surprisingly-running VE standalone @ 512 played things sooner than V-stack @ 256. It didn't matter if I was hosting VE or the individual VI's-the results were still behind VE stand alone. I have had great succes with V-stack on the PC but V-stack for Mac is clearly not as robust. Now, what's up with DP here? I don't know. I ran a similar test using DP on my main mac (which is where I use it as my sequencer-the test above is just in reference to using DP as a VI host only on a slave mac) and the results were spot on. Clearly, the latency compensationw when playing and sequencing is working well, so if you are using DP as the sequencer and host on your main mac you are probably not experiencing any of these issues.