The original NR file writing system had it's own buffer included to allow for
communication between the real time acquisition thread and the slower
file-writing thread. Discrepancies between the two could lead to 'buffer
overrun' errors. Since there is another buffering system in place used by the
closed loop code, why should we have two buffers? Also, the DatSrv buffer has
a user-defined length, so it would allow for tweaking to remove some of the
'buffer overrun' errors.
Original issue reported on code.google.com by RZellerT...@gmail.com on 19 Jun 2012 at 6:39
Original issue reported on code.google.com by
RZellerT...@gmail.com
on 19 Jun 2012 at 6:39