Windows detects USB GPS as Serial Ballpoint - please suggest permanent modification to prevent any serial mouse detection

This behavior is so similar to the behavior of Windows with respect to devices attached to  "ordinary serial ports" that I think it must be essentially (or even exactly) the same problem.  I have not heard of this problem on non-Windows computers either

• so unless I've missed this in one of the posts.

As best I can tell from these posts and others I've read, the problem occurs on USB ports that are emulating a serial device, so my thought at this point is that its not really USB enumeration that is the problem is the serial port enumeration that is the problem.  Obviously, this is a conjecture on my part.

Au contraire - this is a MICROSOFT problem - not a general USB problem.

I have been dealing with this problem for a long time - NSIL Mike, I hope you will not take these remarks as offensive, since your remarks have some useful insight in them...

 

First, this problem occurs on ordinary, on board serial ports.  If you attach many different models of Weigh Scales which communicate via RS232, the computer MAY determine that this is a mouse.  This has been a problem since Windows 2000, but prior to Vista, Microsoft released a patch called "ComDisable" which cause the Operating System (Windows), not to detect the device as a Mouse.

Secondly, with respect to USB specifications - Microsoft is major player at the USB organization - they would have provided and approved the specs for determining that a device is a mouse.

Third - and this relates to point one - GPS, weigh scales and a number of other serial devices are mistaken by the OS as specific types of mice (Ballpoint mice, early model MS Mice) - NOT ORDINARY MICE!  Also, the weigh scales that we work with do not communicate bi-directionally - they just send data to the PC.

Fourth:  The outputs from serial weigh scales considerably PRE-DATE Microsoft Mouse outputs!

1. No, I don't think this is a general USB problem either.

My only point is that USB enumeration is the likely culprit here... whether that be an OS issue, or device issue.

2. My comments do not apply to 'ordinary on board serial ports' by the way... only USB and not RS232.
3.  They may have participated in the enumeration specs for a generic mouse, but that doesn't relate to the USB issue here, where a functionally different device (and again I am talking only about the GPS devices on USB) is somehow messed up.
4.  Makes my point about the difference between generic and specific USB drivers.  The driver defines the capability of the device, what commands it may reply to, and how it provides data.

I certainly do not take your comments as offensive.

You're right, this does NOT seem device specific - I'm experiencing a similar problem on a different system!

 

Upon the bootup, my system opens the COM port for 'actual' serial communication with other MPU-based devices.

However the system sometimes pops up a COM port error message, and I see this 'microsoft serial ballpoint' from my device manager - I did tried to disable it, but it comes back.

 

What I found so far is, if I force to transmit '0000' to the COM port while it is booting up, the Windows will detect this 'microsoft serial ballpoint'. Haven't found any way to prevent this though..

I also tried playing with 'EWF' options which did not help.

 

Have you discevered any fixes that works from your experience?

Usually, disabling the Mouse works -  however, it seems a bit quirky - you can not always find the Mouse in device manager to disable it, etc.  Some brands of serial ports have special software to disable this - try Digi.com - seems to work.  We are gradually switching to using Serial to Ethernet converters - kind of an expensive way to solve a problem, but acceptable for the industrial environment we work in.

If you have questions, you can phone me at Pacific Industrial Scale Co. Ltd. in Richmond, BC, Canada - ask for Bill - I'll be away from the office the week of Jan 16....

Au contraire - this is a MICROSOFT problem - not a general USB problem.

I have been dealing with this problem for a long time - NSIL Mike, I hope you will not take these remarks as offensive, since your remarks have some useful insight in them...

First, this problem occurs on ordinary, on board serial ports.  If you attach many different models of Weigh Scales which communicate via RS232, the computer MAY determine that this is a mouse.  This has been a problem since Windows 2000, but prior to Vista, Microsoft released a patch called "ComDisable" which cause the Operating System (Windows), not to detect the device as a Mouse.

Secondly, with respect to USB specifications - Microsoft is major player at the USB organization - they would have provided and approved the specs for determining that a device is a mouse.

Third - and this relates to point one - GPS, weigh scales and a number of other serial devices are mistaken by the OS as specific types of mice (Ballpoint mice, early model MS Mice) - NOT ORDINARY MICE!  Also, the weigh scales that we work with do not communicate bi-directionally - they just send data to the PC.

Fourth:  The outputs from serial weigh scales considerably PRE-DATE Microsoft Mouse outputs!

Folks- I don't know if there is a windows bug on this or not... Certainly lasting this long it would seem to be more complex.  Here's a little bit of how USB works which may help...

When you plug a new USB device into a PC with windows, there is a standard enumeration process that is supposed to happen (a standard defined by USB.org, not by MS.)   There are several 'universal' device profiles, that windows can install if the detected device fits that profile (an example- you plug in a USB memory stick, and windows recognizes it and installs the universal mass storage USB Driver... and your device works.)

If there is a bug, it would seem to be in how these GPS devices are responding to the enumeration commands from the USB master (the PC) or in how windows is reading the response from the device.  So, yes, it could be a windows bug, or it could be a device bug...  although an issue arising from a windows update is certainly suspicious!

Now, part 2:  I don't know, but suspect there is no standard GPS USB profile (most if not all of these gadgets are probably unique as to their functionality,) thus the supplier of the GPS device would typically supply an install disk with their proprietary USB device driver.  I did not see any mention of anyone installing these...