On Tue, 2011-07-19 at 09:15 -0400, Luke Szczepaniak wrote: > Hi everyone, > > Perhaps a way to manually set the start time after the begging of the > task would allow XCSoar to continue with the appropriate calculations > after a crash has occurred. Better yet, every time XCS detects a > start it writes a temp file containing the start time height and speed > as shown on the screen, if a crash happens the pilot can choose to > load the task start data from the file, this file should never grow as > it is overwritten each time a task is started/restarted. A simple > button "load last start" could be placed somewhere in the Nav/Task > menu. > I have a somewhat extended version of the same proposal to put forward:
XCSoar should keep a checkpoint in memory that contains details of each waypoint it has passed during a task (so this would include the start). The data cost would be minimal, as it would need to hold little more than the task ID together details of each waypoint that has been passed. Each waypoint's details need only include the waypoint's ID and the content of the IGC record that marks the point when the program switched to the next leg. As part of the process of ending one leg and starting the next XCSoar should write the checkpoint out, overwriting any previous checkpoint file, and flush any unwritten IGC points to the log. After the finish line has been crossed the checkpoint details should be erased. Any existing checkpoint file would also be erased when a task is loaded so that the checkpoint file, when it exists, can only apply to the current task. This would allow a single menu button to reinstate the current task as it was at the last checkpoint, bring it fully up to date by finding the last recorded point in the log (which would in general be beyond the last checkpointed position) and then reconstitute the task state at that point before continuing normal operation. It should also be possible for XCSoar to do this automatically if it was restarted in the air. This is easy to recognise because its the only situation when a checkpoint file exists which applies to the current task. I'm suggesting this because the other day I managed to fat-finger LK8000 into a non-task state and then found that the process of reloading and restarting the task followed by stepping forward along the waypoint list so the program knew it was on the correct leg took more head down and button tapping time than I cared for. This was while flying a self-appointed task. I'd care for it even less if it had been a competition task. Martin ------------------------------------------------------------------------------ Magic Quadrant for Content-Aware Data Loss Prevention Research study explores the data loss prevention market. Includes in-depth analysis on the changes within the DLP market, and the criteria used to evaluate the strengths and weaknesses of these DLP solutions. http://www.accelacomm.com/jaw/sfnl/114/51385063/ _______________________________________________ Xcsoar-user mailing list Xcsoar-user@lists.sourceforge.net https://lists.sourceforge.net/lists/listinfo/xcsoar-user