From: Frank Geurts (geurts_at_rice.edu)
Date: Fri Feb 21 2003 - 16:56:36 EST
Hi John,
it just happens to be that 2002-pp data is being reproduced as we speak
so i peeked into the current log files (log files in STAR are about the
size of raw data, and many people claim you could actually do physics
with them since every thing is dumped in it ...).
The strobe criterium in production has always been based on seeing the
characteristic pVPD TDC signals for at least 5 tubes. Now, i have to
look up what trgWrd 0x40ff exactly means but one thing is clear: we have
strobed events in physics runs while other detector see "physics data".
Of course, i can write some code to check on the scalers and see whether
TRIG=0 and PULSE=1 but i think we can agree that the pVPD peaks at
~1600 (year2 data) is sufficient evidence for strobed signals.
-frank
Frank Geurts wrote:
> John,
> we need you :) ... could you take a look at page 20 of the tofp NIM
> paper and tell me what you think about it.
>
>
> Here's the deal ... we see in physics running mode occasional strobe
> events which we always attributed to events in which we received a star
> L0 but didn't get a pVPD coincidence. In the daq software this situation
> is always covered by the 'ship fake data' mode in which we fill our
> event w/ 0s and ship it to STAR. Remember, the 'fake data' mode is our
> STAR-compliance mode: STAR asks => we deliver.
>
> Now, one would expect that no-TRG means no local trigger => no data =>
> the software has to come up w/ a fake data event (0s all over the
> place). That is not true. Clearly, we see strobe events instead of 0s.
> So, that means that we either strobed ourselves (through the Jorway) or
> we *were* strobed through the TCD 'TRIG CMD 4' ... clearly, we don't
> strobe ourselves, that is only used in the daq emulation mode. Which
> leads to no other conclusion than CMD-4 pulses are send to us *every*
> event. Only a timely busy from "Local BUSY" can prevent the cmd4 to
> common-start and -stop the TDCs and ADCs. In case of a TRG, this will
> set the local busy preventing the CMD4 (or test pulse) to fire PULSE.
>
> So, am I right to conclude that TCD Out-4 a.k.a "TRIG CMD 4" is always
> fired for every event and that the Local Busy prevents
> Common-Start/Stops in case we have a real coincidence? My understanding
> was that "TRIG CMD 4" was only fired in when STAR ran in pedestal mode!
>
>
>
> cheers,
> -frank
>
>
>
>
This archive was generated by hypermail 2.1.4 : Thu Jul 24 2003 - 00:39:35 EDT