USB Connection Lost - still!

    Zitat von Marc

    ...they are outside the TI's scope...


    True, but we should still try to get the poor guy's virus working properly, although it is clear that he is using (in one way or another) hardware that is not really up to the job...


    jim_griff - what about the USB cable? Are you using the one that came with, have you tried using a different one? Probably not the source of the problem, but simple enough to check - you don't want to go through all this nonsense over a duff cable...

    Zitat von AtonyB

    True, but we should still try to get the poor guy's virus working properly, although it is clear that he is using (in one way or another) hardware that is not really up to the job...


    jim_griff - what about the USB cable? Are you using the one that came with, have you tried using a different one? Probably not the source of the problem, but simple enough to check - you don't want to go through all this nonsense over a duff cable...

    Thanks very much for all your help. It does seem pretty straight forward - all the things I should've done in the first place, but didn't think of doing.


    I have tried many different USB cables, but they are all the same.


    The last thing to sort (before selling the Virus TI) is the power, but they are still trying to get hold of discontinued stock that is lying around somewhere - Samson Powerbrite PB10 Pro. It acts as a surge protector, a filter and conditioner all in one, as well as having Volt and Amp displays to show any problems with the power supply.


    If all is still bad, I'll probably sell the Virus and invest in another synthesizer of similar calibre - even though there aren't many out there with the same functionality in the same price range!


    Thanks, Jim.

    Zitat von AtonyB

    Have you tried taking it to a mate's house and on their computer? Preferably a mate who isn't on the same block or side of town you are...

    I have had the Virus TI at two different houses with three different computer systems/specs, each with the same result.


    It's actually been running perfectly for the past few hours. I just did an all-nighter, but I suppose people aren't switching things on and off during the night which would cause the problem to occur.


    I think the biggest culprit is noisy power lines that will be sorted as soon as my Power Conditioner arrives tomorrow (just been dispatched next day delivery.) Middlewich has a lot of factories down the road using a LOT of power, so I think that's what the problem has been all this time. *crosses fingers this final solution works*!

    never the less - I think the results of testing it on a pc not in such an area might be useful - that is, if you are waiting a long time for the conditioner... Obviously seeing it run smoothly will give evidence that it is a power problem and not a faulty Virus or PSU.

    Zitat von AtonyB

    never the less - I think the results of testing it on a pc not in such an area might be useful - that is, if you are waiting a long time for the conditioner... Obviously seeing it run smoothly will give evidence that it is a power problem and not a faulty Virus or PSU.

    Update:


    Power conditioner arrived. I still need to create a kettle-cord connector so I can plug my Virus TI into it. It's actually been running pretty well since I installed the PCI-e USB card, but I have been running it at night when the power supply is more constant. Hopefully it'll be perfect when plugged into the power conditioner.

    It won't stay in tune and keeps wobbling up and down by a few cents. I've had to fiddle with the ASIO sample buffer to get it to play in tune with everything else, but it still goes out after a couple of minutes of staying in tune.


    I originally thought this was an analog emulation algorithm but have recently trawled the forums and found it is a serious bug that PC users have when using it through USB.


    Here's a screenshot and a test file to demonstrate how bad it can get. I reduced the ASIO sample buffer to 32 samples running at 96kHz to exaggerate what it usually sounds like.


    unstable pitch.mp3 - unstable pitch.jpg


    My workaround:
    1. When pitch is unstable (most of the time), I mess around with the ASIO buffer size until it decides to hold pitch.
    2. I bounce down to an audio track quickly before it goes out of pitch again.
    3. IF 'UNSTABLE' THEN GOTO 1, ELSE GOTO 4
    4. END

    I'm not really bothered about help with this one. Just thought I'd post it here in case anyone else was wondering about the 'analog emulation bug' and the workaround that might be effective for other users.

    1. PCI-e USB 2.0 card installed on a PCI-e x1 slot that is NOT on a shared buss. Virus TI connected to it.


    2. Power Conditioner installed and fully functioning. Line voltage measures between 240 and 260 volts depending on time of day. Amperage is a constant 1.6 amps, showing no current spikes which would indicate a fault in a piece of equipment (it only jumps up by about 0.2 amps (~50 watts) when the CPU or graphics card are under high load). All of these pieces of hardware currently connected to it: PC, TR-8 Monitors, MOTU 24 i/o, Roland XV-2020, Roland SH-32, Virus TI, Dell LCD display.


    3. Tested thoroughly and still getting "USB connection has been lost" errors occasionally.
    -- No current/voltage spikes are shown on the display of the power conditioner when this happens, showing that it is not my electric supply - unless there are very transient power spikes that may not be measured, but these high frequency spikes would be filtered by the coils in the conditioner itself.

    I was having the "USB connection Lost" problem. I tried many different things, and in the end what helped me was a smaller USB cable. I have just tested it a couple of times and it seems it did the trick. I just thought about sharing this. I know how frustrating it is to troubleshoot instead of making music, hopefully this will work for someone else. Good luck to all. :thumbup:

    Here are the steps I've taken, and it has worked quite well:


    1. Installed PCI-e USB 2.0 adapter card for the Virus TI.
    2. Bought a power conditioner and am using that for the main studio gear connected to the computer and the computer itself.
    3. Replaced standard IEC1363 strip sockets with surge protected ones, making sure to connect everything to the power conditioner so there's no possibility of ground loops. (Audio equipment is usually not grounded to Earth anyway, but may aswell make sure.)
    4. Cleaned the USB cable connectors with lighter fluid (or can be done with any other solvent) to remove any grease.
    5. Disabled 'Selective Suspend' by using a registry hack.
    6. Used an SSD drive for System paging file.
    7. Disabled ALL CPU power saving optimisations in BIOS. C1E state, C3, C6, C7, EIST Function, Intel TurboBoost®, Thermal Function, etc etc etc.
    8. Double checked voltages fed to motherboard components with multimeter and increased the voltage where necessary (DDR3 Memory +0.04V).
    9. **Downgraded the Virus TI back to OS 4.0.5.01 because anything after that causes the arpeggio to play out of time by -11ms no matter what I do. I will NOT upgrade to the latest version until it is fixed. See THIS THREAD // SAMPLE MP3 // SAMPLE PROJECT (with extras)

    I hope this chunk of information helps others who have had similar problems.