NEW Thread For: TI Detune problems (when used in USB mode)

  • Hi,

    I just thought I'd post a new thread to hopefully combine all the other 'detune bug' threads into one. Maybe we can look for possible causes, rather than waiting for Access support to forward the many problems to the engineers and find no solutions after months.

    PROBLEM: When the Virus TI is used in USB mode, there is the tendency for the pitch to oscillate up and down slowly. (Aparrently this problem does not occur with Macs.)

    MY THEORY: I believe the PLL used to keep the TI's internal clock in sync with the USB port may be buggy or poorly designed. It's just a theory, but after researching a little into electronics, I believe it could be one possible area for the engineers to look at.

    The reason I have come to this conclusion is that nothing else could possibly be causing a slow oscillation of the pitch. When the TI is used on its own, the pitch is 100% stable - only when the TI is connected to and has to maintain sync with a host does it exhibit this problem.

    [ One workaround in Cubase is to keep changing the ASIO buffer size from any value to any other until you hear the pitch is holding steady. Then you can continue working on your project until it decides to start oscillating the pitch again. ]

    EDIT: Video of the problem:

  • The same problem occurs on Mac, at least on my Mac Pro.
    On Mac internal USB ports ( with nothing else connected) or using a PCIe USB extension card. Doesn't fluctuate when using analog outs ( pitch only change a bit for each note, but during the note duration, no pitch oscillation).
    Of course, my system is clean, nothing else on any other USB port. Only audio interface on firewire port. Been in contact with Access support, they were able to replicate this problem, but no solution unfortunately. This is a bit embarrassing...

  • This does happen on Macs. Usually only happens for a few seconds then returns to normal however I remember
    one patch was just permanently detuned and I could not get it back to normal. Seems that most of the problems happen
    when I use TI plugin as a multi-instrument in Logic.

  • I thought this one was fixed way back in some update

    At least I read it in some patch notes.

    What tool did you use to display the tuning? I want to have a try on reproducing this issue

  • I thought this one was fixed way back in some update

    At least I read it in some patch notes.

    What tool did you use to display the tuning? I want to have a try on reproducing this issue

    It's only since OS4 that I have started to have the detune problem. I did hear they fixed some detune problem in an OS 3 patch, but seems I got unlucky.

    I sent the main output from Cubase into another soundcard then used spectral view in Adobe Audition 3.0 to view it.

  • I've had the same thing here on my Ti2 in usb mode. Stabilizes after unplugging usb and using analog outs.

    It runs perfectly stable for me through the analog outputs too.

    This is why I believe it is a badly designed PLL loop filter and that the global clock of the TI is modulating up and down as it tries to lock to the USB host clock. *hint hint to the engineers*

    EDIT: I tried rolling back to the older OS's (3.0. . .) and still have the detune issue. Maybe I didn't notice it before. I did think it was an analog emulation algorithm that was impossible to disable.

  • Any news on when the detune problem will be fixed or how it is possible to get rid of it? It's driving me crazy at the moment and I'm having to bounce tracks down several times in order to get even half of a decent take.

  • Yeah is this problem going to be solve in OS 5 thats is the reason why you guys are going to add new filter poles, and new envelopes, Hope this gets fix because it does seem to be noticeable in Hypersaw.

  • I think I have a simmilar or the same issue. Usb connection only.
    On a spectrogram If I have a pure sine at 440 you will see harmonic waves propigate out and then roll back into the primary. Sounds like a half volume slow lfo zipper. I've found that if I put the virus to sleep and wake it back up it is usally ok again for awhile. Ill try and get a video up later.
    Live 8.x on OSX Lion.

  • Ill try and capture it later today, yesterday it didnt rear its head.
    Its defintly not intentionaly, on normal occasions the sine sits fairly pure on the graph, when the detune starts happening it looks like half size waves propigating out and then back into the key freq.
    Just using the standard spectrograph in ableton.
    Running an I7 Imac quad 12GB ram on a focusrite 8/6
    Defintly turned off all effects, reproduced by going to init patch and stripping out the filters and second osc.

  • Bumping this post, just incase anyone's forgotten about it.

    I stopped producing music for a long while due to problems with the Virus TI (my main synth.) Now I am back and trying to make music, but this bug keeps annoying me. It is almost like having an analog synth the way it goes out of tune.

    Here's my checklist up to now:
    1. Disabled all power saving optimisations in BIOS, Windows and everywhere I can find.
    2. Using dedicated PCI-e USB2.0 controller card with a short USB cable to the Virus TI.
    3. Invested in power distribution and filtering system (Samson Powerbrite Pro), plus surge protected sockets.
    4. Checked the Virus TI with Windows XP 32bit and Windows 7 64bit. Same on both.
    5. Checked it with/without USB connection. Without USB connection it is perfect.

    Video again:
    In the video, it makes no difference whether multi-processing is switched on or off. It occurs with both. Simply switching it off and on again causes the Virus TI to become stable for another few seconds - enough time to bounce down a short segment of audio before it starts going out of tune again. I have also checked with only one CPU core active in BIOS, so there's no multi processing or hyperthreading possible. It exhibits the same behaviour.

    Thanks, Jim.

  • What TIOS are you currently using?

    Hi. I'm using I have had many problems with previous versions, but this one seems the most stable and has sorted out the other issues I was having. I'd attempt an upgrade to 5, but have heard a lot of people were having trouble with the beta. Think I'll wait until it's in a release state before trying to upgrade.

    EDIT1: My current system and settings are:
    Intel Core i7 3.33GHz Extreme // 12GB RAM // Gigabyte GA-X58A-UD7 (disabled all the CPU power saving as well as TurboBoost®)
    MOTU 24i/o (Tested at 44.1kHz, 48kHz and tested at 16, 32, 64, 128, 256, 512, 1024, 2048 sample buffers.)
    Windows XP 32bit (tweaked down to 13 background processes, disabled selective suspend, power saving, enabled Windows 3GB switch, etc.)
    From Cubase 5.5.1 to Cubase 6.5.4, the same issue is evident.

  • Same problem, unstable pitch using usb outs with OS beta.

    The sample rate change fix the problem but for one part of virus..if you want another you have to do it again.

    Analog outs are stable.

    any news with this?