[subexp-daq] Question on UCESB

[Hans Toshihide Törnqvist]

Dear Günter,

Apparently this would lead to major surgery with all the complications 
that can lead to. The surgery would be all over the place in many files 
inside the hbook/ directory. Merging 2 32-bit words is a lot easier, 
until we need to unpack a lot more than just timestamps :)

Cheers,
Håkan & Hans

On 2024-04-10 14:47, Weber, Guenter Dr. wrote:
> Dear Håkan,
> 
> 
> next thing is that we would like to teach the HBOOK part to work with 
> 64-bit uints (to not have the timestamp as two 32-bit words). We noticed 
> there is a pearl script: make_external_struct_sender.pl
> 
> Is this an integral part that we would need to touch?
> 
> 
> 
> 
> Best greetings
> 
> Günter
> 
> 
> ------------------------------------------------------------------------
> *Von:* subexp-daq <subexp-daq-bounces at lists.chalmers.se> im Auftrag von 
> Håkan T Johansson <f96hajo at chalmers.se>
> *Gesendet:* Mittwoch, 10. April 2024 12:57:19
> *An:* Discuss use of Nurdlib, TRLO II, drasi and UCESB.
> *Betreff:* Re: [subexp-daq] Question on UCESB
> 
> On Wed, 10 Apr 2024, Weber, Guenter Dr. wrote:
> 
>> This is really great. The only remaining question for now is if there is a
>> way to get the content of an array (in the above case the counts in the 16
>> scaler channels) without the additional information on the scaler channel
>> numbers? In case of the SIS3316 digitizers a single pulse trace can have a
>> length of tens of thousands of samples. Somehow I do not like the idea that
>> without necessity we always transfer an additional array with just the
>> numbers from 1 to n.
>> 
>> Bottom line: is there an easy way to get rid of the '..._nI' part of the
>> array object in the JSON output?
> 
> Nice that it worked.
> 
> I think the only 'easy' way to avoid the '..._nI' entries is to not use
> zero-suppression for the array.
> 
> However, then it would/should appear fully in all events.  Since the
> scaler module might not always be read, it would then contain the value 0
> when nothing filled the array for that event.  That may (or not) be more
> confusing.
> 
> Cheers,
> Håkan
>