Interferance on USB3 connection

Forums Forums Qu Forums Qu troubleshooting Interferance on USB3 connection

This topic contains 31 replies, has 10 voices, and was last updated by Profile photo of  Anonymous 6 years, 3 months ago.

Viewing 15 posts - 16 through 30 (of 32 total)
  • Author
    Posts
  • #64533
    Profile photo of
    Anonymous

    Everyone,

    A&H are not ducking this issue. This has been an ongoing issue that has been troublesome for quite some time it seems with no pinpointed answer.
    In all actuality I was the one who pinpointed the issue regarding the USB.

    Most all PC and MAC nowadays implements Intel’s controller chips. In many setups, depending on your motherboard, if it is a standard or extended and modifiable type…will allow access to bypass this USB3 issue.

    I had originally dug through the Intel protocol and it clearly states that yes, Intel knows about the problem however is not correcting it. It is quite blatant in the documentation from what I had seen. They are the ones who pointed out the issue. I will try to trace back and uncover this for others if you are curious. So I am very happy that A&H are looking forward and trying to implement a fix for this issue. They are correct that this type of problem breaks the entire architecture of the software design, and is quite dependent sometimes on the persons system. I too was an unfortunate one who had this issue and as a troubleshooter by profession had triggered me to look into.

    If you have a specific system issue, I will try and help and see if I can get you a temporary solution to continue on.
    As of now I use an EVGA X99 Classified which provides many overclocking options. For those who do not have higher-end modifiable motherboards yes, this might not expose any quick fix for you.

    As far as MAC, unfortunately I do not have a MAC or access to one to dig further into. I am not a huge MAC fan, especially since the difficulty in exposing modifiable parameters such as this. Again it is unfortunate, as most computers now employ Intel USB controllers.

    If you have any questions or doubts please respond to me, I am a user just like everyone else here in the forum. Yes it is an unfortunate thing to have to deal with…however I do have respect for A&H that they responding to their customers to fix a highly difficult issue.

    -S

    #64534
    Profile photo of
    Anonymous

    I also wanted to mention that A&H are not the only ones who encounter this issue.
    From my research there have been issues as well with Presonus users.
    Also some references from MOTU end-users regarding the 2.0-3.0 conflict.

    I am almost wondering if some of my Focusrite Firmware update issues might have been related.
    I was recommended to use Windows 7 or XP which did work to update the firmware.
    I am not 100% but I think these earlier OS didn’t necessarily support the XHCI or fully force it to be used.

    Again…I am in search of a solution for all until Firmware wise the QU can be modified.

    -s

    #64536
    Profile photo of
    Anonymous

    So at the moment I have no found a solid answer, but some possible things to try.

    I have read Intel NUC issues regarding end-users running into XHCI problems.
    One reference stated that Skylake does not offer any option of EHCI, but I am not sure if this means you can/cannot disable XHCI.

    I am curious what the outcome would be if the actual XHCI drivers in windows were removed. If the controller would maintain use, but not initiate a USB XHCI driver in conjunction with windows. If I get a chance I will retest the issue, and try this.

    #64537
    Profile photo of TritonX
    TritonX
    Participant

    I tried and all the ports became unusable. Blutooth and the built-in camera also.

    #64538
    Profile photo of
    Anonymous

    Thank you for testing Triton,

    I have not been able to get to the studio, been really sick lately and recovering.
    I am trying to find some way to force an alternate driver. Only thing I can think of at the moment.

    -s

    #64539
    Profile photo of
    Anonymous

    For those using MAC,

    I am not sure regarding this, as I said I am not a MAC user.

    Possibly some info for advanced users to somehow reroute their USB???

    https://www.tonymacx86.com/threads/guide-10-11-usb-changes-and-solutions.173616/

    #64540
    Profile photo of
    Anonymous

    For those who are using Intel 6th Generation and beyond, it looks as though Intel has removed EHCI options altogether.
    This might leave with no easy solution.

    #64542
    Profile photo of Dick Rees
    Dick Rees
    Participant

    This is exactly why I prefer NOT to use computers for audio tasks. Yes, I load audio files into a computer for mastering and editing, but that’s all ITB and no “won’t play nice with another device” issues.

    I sympathize with those who come up against such issues. For myself I prefer to simply avoid placing myself at the mercy of Intel, Microsoft, Apple and the lot.

    Disclaimer:

    I use my iPad to control my Qu-16 and even use AirPlay for break music, but Auto Update is disabled and I never do updates until new versions are no longer new…

    #64545
    Profile photo of airickess
    airickess
    Participant

    @Dick, you stated that you use an external hard drive for recording. Do you hook it up via the USB-B port?

    #64546
    Profile photo of Dick Rees
    Dick Rees
    Participant

    Yes, USBb port. It’s a 500 gig Seagate I got years ago. 7200rpm.

    I off-load the files directly after the job and re-format just before the next job.

    #64552
    Profile photo of BCEly
    BCEly
    Participant

    Hi Folks,

    This is all very interesting and it’s good to know that A&H are looking at it whilst they haven’t said that publicly or in any of the emails I’ve had with their support team – And there’s been a few !

    Recording to USB isn’t an issue, I did a concert on Friday with an 11 piece band and three singers and there wasn’t a hitch.

    It’s USB playback from a laptop that’s the issue. It’s very impractical to convert all files I’m given by a client to .WAV, rename to suit A&Hs naming convention and play in the order required. The lack of any real disk/track management in the firmware or even simple play-list indicates that this isn’t what the facility is intended for. It’s great for a virtual sound check and saves time in that respect but I have to playback from a laptop.

    My laptop is 6th generation and I came the the same realisation as ‘Sephult’ in so far as the EHCI options have been removed. There is nowhere in the BIOS (or software) to revert the USB3 back to a USB2 and switching out the drivers just disables other devices.

    #64553
    Profile photo of
    Anonymous

    Yep BCEly,

    Sucks the situation, really such an awesome board though.
    Hopefully A&H will at least respond sometime soon with some sort of announcement.
    I am just glad they are now aware of where the issue stems from, this is at least a first step.
    (Kinda wish Ethernet audio Tx Rx also worked..lol)

    Unfortunately the software was written and tested around the isochronous transmission…which I believe is where the untested Intel problem lies (This is as deep as I got…maybe something else an issue as well?). Sucks they know about the possibility of problems and didn’t fix. You would really think especially since they are negating EHCI that they would ensure XHCI was solid to support.

    Really several layers deep, but I have some faith that there will be a resolution.
    I agree though sometime soon A&H needs to officially respond.

    -s

    #64572
    Profile photo of TritonX
    TritonX
    Participant

    I got this from them this morning after a few email back and forth,
    , I guess it is proper to share with you so you know what to expect:

    From A&H support:


    Yes… Well…. It should be backwards compatible of course. My point is that there is nothing wrong with your Qu, but I am trying to help as best I can.

    We have been able to reproduce the issue here, the problem is that it is not a reliable reproduce.
    For example, if you monitor the USB bus, the issue can resolve itself.
    It can appear for seemingly no reason, then disappear without reason too.
    We could soak a unit for a week and have no problem, then suddenly it appears again in general use.

    Then the answer from Intel support(all in the same day, wow)


    It is much appreciated that you reported this matter with us.

    At this point, we are aware of this and the current status would be No plan to fix it.

    Sorry for any inconveniece this may cause.

    Please double check this link https://intel.ly/2uOHDon to find the Errata where this is mentioned (Page 15 Bullet #1).

    Make up what you want of those “official” answer.

    In the Intel link they sent there is even a solution(good practice lesson) to their incompetence, how nice of them.


    Isochronous traffic is periodic and cannot be retried thus it is considered good
    practice for software to schedule isochronous transactions to start at the beginning
    of a microframe. Known software solutions follow this practice.

    • To sensitize the system to the issue additional traffic such as other isochronous
    transactions or retries of asynchronous transactions would be required to push the
    inbound isochronous transaction to the end of the microframe.

    … I don’t expect a fix anytime soon :/

    #64573
    Profile photo of Dick Rees
    Dick Rees
    Participant

    Thanks to all for your persistence in digging up the dirt and for posting the results.

    This is why I consider “upgrade” to be a one-word oxymoron. If I have something that actually functions as advertized, I cling to it like grim death and only upgrade, update or replace with time-tested options.

    #65176
    Profile photo of TritonX
    TritonX
    Participant

    Follow up from Intel

    “I have gotten the research done to find out what is required to create the logs for this matter.

    It would be required to purchase hardware devices and download the software form the vendor to perform the validation test to get the log files. Customer can consultant vendor for further information.

    For hardware required please visit: https://teledynelecroy.com/protocolanalyzer/protocolstandard.aspx?standardid=4&capid=103&mid=511

    Since the current OS is Windows 10*, and the USB 2.0 driver is native driver which means it comes from Microsoft, so if this is an driver issue, customer needs to get support from MS for a solution.

    I understand it may be tricky to proceed with the tests, but in case you are able to, it would be great.

    I’m not buying a hardware that cost more than my laptop without billing them…

    Admit that it is a good way for them to dodge the issue and put the blame on me…

    BOYCOTT INTEL

Viewing 15 posts - 16 through 30 (of 32 total)

You must be logged in to reply to this topic.