Re: Chris W's WFD Problems (fwd)
Chris Walter (walter@cithe504.cithep.caltech.edu)
Tue, 12 Mar 1996 12:22:07 -0800 (PST)
>
> Ash is of course right. I use the same code Chris W does to decode
> the WFD buffers and it deccodes the times correctly (it doesn't assume
> that the first two times will be separated by greater than 4 ticks). So
> I assume that there is some bug in the text output above only, and that
> this was just a bad example of some problem Chris has found with this
> channel. In the past we have replaced channels that have incorrect
> timing like this in the MIDDLE of data.
I am very confused by what you guys are syaing. It sounds to me like
Ash has explained why this happens. It is something we noticed a long
time ago. But what is this stuff about the "some bug in the text
output"? As I pointed out in the note I normally correct for this and
never look at the first four samples. For this run I turned that
feature in the software off so I could see the stop words and that is
why this is visible in this case. It happens something like 5% of the
time I guess but the software handles it seamlessly.
-Chris