Page 4 of 5 FirstFirst 12345 LastLast
Results 31 to 40 of 42

Thread: FB3 not being recognized by my laptop

  1. #31
    Join Date
    Apr 2006
    Location
    Orlando, FL - USA
    Posts
    1,770

    Default

    Quote Originally Posted by heroic View Post
    Don't get me wrong, I really want to use the pango stuff; it's by far the best application software.
    Well I am happy to hear that you want to hang onto it and that you think it's the best application software. We think so too ;-)

    And don't get ME wrong either. If there are legitimate problems, we want to hear about them and solve them. But I think you and I have distinctly different methods of communication. Within the walls of Pangolin, we communicate in a very "visual" way. If you send me a direct email, I'll send you back examples of how we communicate, which personally I think is the best way to handle things.



    Quote Originally Posted by heroic View Post
    I just wish the lower levels of it worked better and more reliably.
    Yes well, fortunately we can comfortably say that not too many people are having problems in their general uses. As I mentioned in another post, when QuickShow came out, it created such a demand for FB3 systems that it was impossible for us to build enough of them to satisfy demand over the last eight months. We haven't updated the front page of our web site in a while, but currently we're close to 15,000 total Pangolin users.

    I personally have installed QS on dozens of computers in China, while traveling around to client's sites. These computers are far less controlled than the computers we find elsewhere. Frankly, much to my surprise, QS worked on all of them -- even build 128 which was our first public build.

    Internally we're up to build 238 which handles the FB3s even better, especially in the way it handles multiple threads and FB3 communication. Also, internally (and distributed to only around 10 or so people outside) we have completely different firmware in the FB3. I am not sure if you have this firmware or not.

    In any event, based on your experience and reports here, I might be able to agree that there are some problems somewhere -- of which we are completely unaware because of the very low incidence of people having problems. And we're definitely interested in pushing all problems out of our products -- as long as we can see the problems with our own eyes first, so we can completely understand them. That requires higher-quality communication than has occurred up to this point.

    It's not like we're trying to aviod fixing problems, and it's not even that we are claiming there are zero problems. We actually enjoy fixing problems because we know once a problem is solved, it is behind us. And all developers at Pangolin are mature enough to know that all software has bugs (as determined years ago in a study by IBM). You don't write 450,000 lines of code without a few of them being wrong... Even software in the space shuttle (which I've heard is the most expensive software ever written) has bugs.

    We'd just like to understand -- with higher resolution and higher quality -- what the problems are, so we can investigate and then solve the problems.


    Quote Originally Posted by heroic View Post
    The major problem is that the fb3 usb implementation was seemingly written by someone with little knowledge of how descriptors are supposed to be used; they change multiple times in ways that are not allowed in the usb state model, and critical fields like device serial number are never filled in (despite the fact tht fb3s have electronic serials!). This means that they tend to become unmoored from the driver; since windows can only support one combo of endpoints for any given VID/DID tuple, this confuses things and causes disconnections which are not well handled by the application (LivePro).
    Well I thank you for providing the additional information so we can look into this. It is Ante, on our team, who has written the firmware for the FB3 and -- within Pangolin -- is certainly the smartest person when it comes to USB and descriptors. He also uses the USB bus analyzer. I'll point him to your discussion above and get his feedback on it. If he believes you are correct, then he'll certainly take action to correct things. But it could be that there will be questions and that this will require some back-and-forth emails to get completely resolved.

    Since your interest is in LivePRO, it is likely that Alexey will also need to be involved in the discussions too. Please send us a direct email so that we can get into a kind of "party line". We can come up with alternative FB3 firmware, which can be easily loaded into the FB3 with a special part of the program, and do some experiments that way.

    Best regards,

    William Benner
    Last edited by Pangolin; 08-19-2010 at 07:15.

  2. #32
    Join Date
    Apr 2006
    Location
    Orlando, FL - USA
    Posts
    1,770

    Default

    @ Heroic,

    By the way, we also have the FB3-QS which likely will be seen at Photon Lexicon for the first time in public. This is a completely different design, having no removable memory and six color channels and capable of 120K sample rates.

    The USB communication with this device is also completely different because of the additional color channels and level of control. Much testing and work has been done in the area of USB for this device while we have investigated USB 2.0 and it's use on older USB 1.1 ports as well.

    Since you seem to have critical hardware and additional test equipment, we'd like to have you test these as well, so we can make sure there are no problems before circulation increases.

    Best regards,

    William Benner

  3. #33
    Join Date
    May 2008
    Location
    nerdtown, USA
    Posts
    1,165

    Default

    Quote Originally Posted by Pangolin View Post
    @ Heroic,

    By the way, we also have the FB3-QS which likely will be seen at Photon Lexicon for the first time in public. This is a completely different design, having no removable memory and six color channels and capable of 120K sample rates.

    The USB communication with this device is also completely different because of the additional color channels and level of control. Much testing and work has been done in the area of USB for this device while we have investigated USB 2.0 and it's use on older USB 1.1 ports as well.

    Since you seem to have critical hardware and additional test equipment, we'd like to have you test these as well, so we can make sure there are no problems before circulation increases.

    Best regards,

    William Benner
    Hi Bill,

    Since this is different hardware, what are you proposing? I think it might be best if you set up a test rig with the following hardware:

    - PC or Mac with recent Intel chipset (Nehalem or late-model Core2 Duo)
    - Windows 7 and XP dual booted
    - At least two FB3s with projectors attached (we usually run three)
    - LivePro and QuickShow

    We find that the most troublesome part is entering and leaving the projection zones setup. I believe this is because the FB3 changes its descriptors and, on recent Intel hardware, this causes a re-enumeration of the ports. (When a USB device is detected on a port on these motherboards, it is either connected to an OHCI or an EHCI depending on its speed; since the FB3 does not fill the iSerialNumber field in the root descriptor, Windows uses the location in the bus topology of the machine to determine which FB3 is which- and this obviously changes during the process, confusing the device driver.)

    I believe this will be enough information to reproduce our problems, but if not, installing the software inside VMware is almost guaranteed to show it, as the virtualized USB ports will always re-number themselves on disconnect & reconnect events.

    Sorry I can't go into more detail but I have a broken wrist right now and typing with one hand is very tiring.

  4. #34
    Join Date
    Apr 2006
    Location
    Orlando, FL - USA
    Posts
    1,770

    Default

    Quote Originally Posted by heroic View Post
    Since this is different hardware, what are you proposing?
    I guess I'm proposing that we send you a few, you try them and report back your results.



    Quote Originally Posted by heroic View Post
    - PC or Mac with recent Intel chipset (Nehalem or late-model Core2 Duo)
    Ante and Dieter both have pretty new Mac Book Pro. And we have a brand new HP with a few builds of Win7 at the office.


    Quote Originally Posted by heroic View Post
    We find that the most troublesome part is entering and leaving the projection zones setup.
    Yes, I forwarded Ante to your post. He has some opinions, but would like to maximize bandwidth with direct communications (i.e. direct emails and not forum posts).

    He and Alexey will make a change to descriptors on Monday, and I've already had Jason re-do the drivers and driver installation process and test it on a variety of PCs including Windows 7 64-bit and 32-bit. So, we've got something new but will need a direct email so we can send it.


    Quote Originally Posted by heroic View Post
    I believe this is because the FB3 changes its descriptors and, on recent Intel hardware, this causes a re-enumeration of the ports.
    Yes, from your description, Ante confirmed this theory. But it is not us that is conciously doing this. He thinks that this might only be a symptom of a different problem, which might be unique to your setup. We'll have to dig into this further to find out why you are having problems when others generally are not. Even myself, I use multiple FB3 at tradeshows with no problems.

    Quote Originally Posted by heroic View Post
    (When a USB device is detected on a port on these motherboards, it is either connected to an OHCI or an EHCI depending on its speed; since the FB3 does not fill the iSerialNumber field in the root descriptor, Windows uses the location in the bus topology of the machine to determine which FB3 is which- and this obviously changes during the process, confusing the device driver.)
    Ante says that, up until now, he purposely did not use the serial number field as it is not a required field and also because it is possible that someone can use multiple FB3s (with different serial numbers) and the customer experience would be that each time they plug a FB3 with a different serial number, the PC would ask for the driver (since a particular USB port would have never seen that particular FB3 with that particular serial number before). But we might have worked a way around that problem this weekend with a new way to install the drivers, particularly on Windows 7. In any event, probably on Monday Ante will have updated firmware for the FB3 that fills in the serial number field.

    Quote Originally Posted by heroic View Post
    installing the software inside VMware is almost guaranteed to show it, as the virtualized USB ports will always re-number themselves on disconnect & reconnect events.
    Ante and Dieter both have Mac Book Pro. I know Dieter uses Parallels, but I don't know about Ante. We'll delve into this more over the coming days...

    Bill

  5. #35
    Join Date
    May 2008
    Location
    nerdtown, USA
    Posts
    1,165

    Default

    Well... with the latest drivers, very latest QuickShow, latest LiveQ etc., on a vanilla fully-patched Windows XP machine, I got it all working well with my new FB3-XE. Buoyed with enthusiasm, I plugged in the projector that has the Neo-Neon card in it and ran LiveQ. It did a firmware upgrade to v7.20, and now when I try to run LiveQ it complains that the hardware is not supported and exits.

    It doesn't run QuickShow either, but I expected that. The only thing that works is Showrider. I can't get TraceIT to work with the old Neo card, or any of the other LAStudio apps except Showrider.

    Question: will LivePro work? I don't have the license on this machine.

  6. #36
    Join Date
    Aug 2008
    Location
    UK
    Posts
    5,704

    Default

    You need updated versions of the files, contact Pangolin.

  7. #37
    Join Date
    May 2008
    Location
    nerdtown, USA
    Posts
    1,165

    Default

    Quote Originally Posted by White-Light View Post
    You need updated versions of the files, contact Pangolin.
    I got the latest updates from the pangolin site, but it didn't help; I've put in a support email and we'll see if they can fix it.

  8. #38
    Join Date
    Aug 2008
    Location
    UK
    Posts
    5,704

    Default

    I probably shouldn't post build numbers but as its the only way of helping you:

    You want the 25 11 2010 versions of the files.

  9. #39
    Join Date
    May 2008
    Location
    nerdtown, USA
    Posts
    1,165

    Default

    Quote Originally Posted by White-Light View Post
    I probably shouldn't post build numbers but as its the only way of helping you:

    You want the 25 11 2010 versions of the files.
    That is the version I have, I think.

  10. #40
    Join Date
    Aug 2008
    Location
    UK
    Posts
    5,704

    Default

    You'll know if it is because there's no build number just the date instead.

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •