#11596 NORM Not Tri: Distance reports very wrong distances on XO-1.75

[Zarro Boogs per Child]

#11596: Distance reports very wrong distances on XO-1.75
-------------------------------------------------+--------------------------
           Reporter:  carrott                    |       Owner:  bemasc       
               Type:  defect                     |      Status:  new          
           Priority:  normal                     |   Milestone:  Not Triaged  
          Component:  acoustic-measure-activity  |     Version:  not specified
         Resolution:                             |    Keywords:               
        Next_action:  never set                  |    Verified:  0            
Deployment_affected:                             |   Blockedby:               
           Blocking:                             |  
-------------------------------------------------+--------------------------
Changes (by sridhar):

 * cc: sridhar (added)


Comment:

 Comment from Barry Vercoe:

 Such variance can be caused by echoes (the audio signal bouncing between 2
 screens that are strictly parallel).  I've seen this problem with ultra-
 sound, for instance.  The variations won't be strict multiples, since
 there are audio I/O buffer times (in and out), along with processing times
 in the loop.  Presuming I/O buffer times X and Y respectively (possibly
 different) and processing time P, the code will likely subtract X+Y+P from
 the single perceived round trip.  For echo bounces, this subtraction is
 not appropriate, giving 2D - (X+Y+P), 4D -(X+Y+P), etc.

 I've not looked at the code, but a quick glance should make all apparent.
 Of course, with a faster processor, the P time should be made smaller.  I
 wonder if that's the source of problems for the 1.5?  This will be
 exacerbated in the 1.75 and XO 3.

-- 
Ticket URL: <http://dev.laptop.org/ticket/11596#comment:1>
One Laptop Per Child <http://laptop.org/>
OLPC bug tracking system