Hi again,
@Matthias:
updated to HEAD in SVN -- problem persists, see below
@Harald
Problem is independent of yes/no to improve_nmradcc_timing.
It does however depend on the translation function chosen.
The tams decoder does for example *not* react to
SET 1 GA 1 0 1 500
for translation function 3, but it does for translation function 2. However it
seems that the command can be translated by translation routine 3 since a
Littfinski decoder at the same address switches.
If it is a timing (in)tolerance then I seem to hit with translation 3, but not
with translation 2 (however xlate 2 has its own problems -- too many locos
dont stop when their speed is set to zero)
Andre
On Monday 25 January 2010 09:06:31 Harald Barth wrote:
> > All this doesnt happen for the Littfinski decoder programmed at the same
> > addresses. (This seems to indicate that software versions of the decoders
> > and the digital code stream srcpd generates are somehow not compatible)
>
> I suspect tolerances (or non-tolerances) in the timing. Do you run the DCC
> signal with or without the custom timing option? Which DCC conversion
> routine of the three(?) available is active? Any change in behaviour if you
> change any of these options?
>
> > 2. ...no GA command (or another other command) has been
> > given at all in temporal proximity ....
>
> My guess is that if you would plot the bitstream that is output
> continuesly for DCC refresh, say
>
> ....0110010001001000011111100100101111111....
>
> and the one that needs to be output for "unexpected" things to happen
>
> 010000110111001
> ^
>
> that they _almost_ match (like in this randomly thought up example).
> But to know that, you would need to sample the signal and analyze it.
> It is possible even without memory oscilloscope by using a audio in of
> a fast sampling sound card (of course after adjusting "sound" levels).
> But it is a lot of quite boring work.
>
> Harald.
>
>
> ---------------------------------------------------------------------------
>--- Throughout its 18-year history, RSA Conference consistently attracts the
> world's best and brightest in the field, creating opportunities for
> Conference attendees to learn about information security's most important
> issues through interactions with peers, luminaries and emerging and
> established companies. http://p.sf.net/sfu/rsaconf-dev2dev
> _______________________________________________
> Srcpd-devel mailing list
> Srcpd-devel@...
> https://lists.sourceforge.net/lists/listinfo/srcpd-devel
|