Linux-USB device overview

Not logged in
Login to get access to the new bookmark feature!

Username:

Password:



You may register here, if you have no login yet.
 
Details for: DeLorme Publishing, Inc. DeLorme USB EarthmateStatus not supported
( Earthmate USB GPS ) ( Category: HID )
ID0x1163:0x0100
Driverno driver available
Linux-USB linkhttp://N/A
Vendor linkhttp://www.delorme.com/earthmate/overview.asp
CommentMost GPS devices communicate via serial link.. I have NFI why the vendor decided this one should be a HID class device. Looking for a driver....
Submitted byPerrin Aybara ( paybara <at> fyrestorm.com )
Submitted onSat, 13 Dec 2003 22:45:06 +0100
Last edited onSat, 13 Dec 2003 22:45:06 +0100



User commentsPlease login to add comments
Lonnie Mendez
(lmendez19 <at> austin.rr.com)
2005-04-28 09:23:48 Follow up: the linux kernel driver 'cypress_m8' has been included with vanilla kernel distributions (kernel.org) since 2.6.10. It can be found under "USB Serial Converter Support". There is also the EarthMate Userland Library available here: http://emul.berlios.de for those using darwin (MacOSX) and BSD systems (freebsd,openbsd,netbsd).
Lonnie Mendez
(lmendez19 <at> austin.rr.com)
2004-05-16 21:08:04 After researching it a little bit more, I can confirm that the device is a HID, as viewed and communicated with in windows. However, in linux we need to do things a little bit differently. Given the nature of the device, a driver is being developed that will provide a usbserial interface to the earthmate. A functional driver is and has been available for devices using this usb2serial chip, dubbed the "m8". It uses a fixed baud speed of 4800bps and a few other fixed values. When compiling your kernel, disable all other usb2serial modules...otherwise you will receive errors. Either that or modify the modules to adapt the new interrupt additions made in usb-serial.c. http://firstlight.net/~koyama/modules.php?name=Downloads&d_op=viewdownload&cid=1 Kernel 2.6 only for now...if you know how to, port this for an older kernel and post a link here. btw, this works great in gpsd, use this "$gpsd -p /dev/usb/tts/0" - changing path accordingly. On a side note, I'm working with the creator of the driver, Neil Whelchel, so we can have it included into the kernel source. I'm also looking at porting it over to 2.4.x kernels. You can find my project page here: http://geocities.com/i0xox0i.
Stephen Webb
(spidey <at> dodds.net)
2004-04-04 07:31:33 This device now have some support. This site has a patch to gpsd. It is not a HID device. http://vista.civil.northwestern.edu/~jay/
David Tauriainen
(djtavria+linux_usb_qbik_ch <at> iupui.edu)
2004-02-05 02:54:28 A Mac (OSX?) user says that the bluetooth addon allows her serial access through serial-bluetooth. Maybe DeLorme put some HID->CDC software in the bluetooth device?
David Tauriainen
(djtavria+linux_usb_qbik_ch <at> iupui.edu)
2004-02-05 02:43:48 Boy does this look terrible w/o any formatting tags. In RH 9, the device appears as /dev/usb/hiddev0 (for me). If I cat this device, it gives the expected behaviour (has a 1Hz heartbeat of data). http://linux.bbn.com/~grossman has some information about protocols that Mr. Grossman used to make a Visual Basic convertor from the Rockwell format to NMEA. Another URL of some import: http://groups.google.com/groups?hl=en&lr=&ie=UTF-8&oe=UTF-8&threadm=36CC8B54.D2D08CCE%40bbn.com&rnum=8&prev=/groups%3Fq%3Dearthmate%2Blinux%2Bprotocol%26hl%3Den%26lr%3D%26ie%3DUTF-8%26oe%3DUTF-8%26as_drrb%3Db%26as_mind%3D1%26as_minm%3D1%26as_miny%3D1993%26as_maxd%3D4%26as_maxm%3D2%26as_maxy%3D2004%26selm%3D36CC8B54.D2D08CCE%2540bbn.com%26rnum%3D8 It explains that there's a possible typo in the spec: "gotchas: you gotta have the Rockwell Binary spec http://www.acte.no/technical/gps/data/Serial%20IO%20Spec%20Rev%2011.pdf you must notice that the words all come in swapped around ala Motorola you need to know (and here's the one that stumped me the longest) that the spec is incorrect: page 2-25 message 1200 Geodetic Position Initialization, specifies that the resolution for the lat and long is 10**-9 radians. It's not, it's 10**-8 radians just like msg 1000. If it wasn't for msg 1000, I'd blame it on Delorme, tweaking things for it's own use, but there's no reason for the outgoing msg to be different from the incoming, so I suspect a typo. send me an email at rpulsifer@ime.net and I'll send you the output from Bob's Portmon utility - nice rig, though in this case it didn't tell me anything I didn't already know. I take that back, It showed me that there wasn't any secret Delorme proprietry msgs going on, just the standard ones out of the Rockwell manual."
 
 
© 1999-2010 qbik.ch