Error in dashboard spec

[Don Reid]

I believe I have found an error in the dashboard spec.
The file dashboard-specification.pdf, dated 1.5.2005 says (on page 4) that
the "extended" packet is:

...
    Byte06 USER BYTE 2
    Byte07 USER BYTE 3
    Byte08 CTRL_A
    Byte09 USER BYTE 5
    Byte10 CTRL_B
    Byte11 USER BYTE 5
    Byte12 CTRL_C
    Byte13 USER BYTE 6
    Byte14 PACKET NUMBER
    Byte15 USER BYTE 6
    ...

My experiments have shown that byte07 is not USER BYTE 3, but that byte09 is, and that bytes 4 & 5 are in byte11 and byte13, like this:

...
    Byte06 USER BYTE 2
    Byte07 ?????????
    Byte08 CTRL_A
    Byte09 USER BYTE 3
    Byte10 CTRL_B
    Byte11 USER BYTE 4
    Byte12 CTRL_C
    Byte13 USER BYTE 5
    Byte14 PACKET NUMBER
    Byte15 USER BYTE 6
    ...

[Jared Russell]

Can anyone else confirm this?

[seanwitte]

How did you determine that the packet spec is wrong? If it's wrong I'd like to patch the software I've released, but I don't have access to a robot controller to verify. Can someone please confirm or deny the claim?

[Don Reid]

Originally Posted by seanwitte

How did you determine that the packet spec is wrong? If it's wrong I'd like to patch the software I've released, but I don't have access to a robot controller to verify. Can someone please confirm or deny the claim?

First I wrote a small c program to decode the packets based on the spec, I printed each of the duplicate entries shown. Then I hacked a copy of the robot code to output unique and changing values to the User_Byte variables. Looking at the reaults showed what was
happening.

I'll attach my updated dump code, It requires that the serail port already be set to 19200 baud etc. I write this for gcc on Linux, results on other platforms may vary.

[seanwitte]

Originally Posted by Don Reid

First I wrote a small c program to decode the packets based on the spec, I printed each of the duplicate entries shown. Then I hacked a copy of the robot code to output unique and changing values to the User_Byte variables. Looking at the reaults showed what was
happening.

I'll attach my updated dump code, It requires that the serail port already be set to 19200 baud etc. I write this for gcc on Linux, results on other platforms may vary.

Did you validate the results with the IFI dashboard? Just curious whether theirs shows the correct values.

[Don Reid]

Originally Posted by seanwitte

Did you validate the results with the IFI dashboard? Just curious whether theirs shows the correct values.

I posted the same basenote to their forum. No responce yet.

[Don Reid]

IFI has now posted http://www.ifirobotics.com/forum/vie....php?p=579#579

"Answer: You are correct. We have updated the dashboard-specification.pdf. Thanks for the feedback."