>
>
> Konstantin Olchanski wrote:
>
> > >
> > > One thing about your update: If you have sequences with negative
> > > adc-values, i.e. if you for some reason do not run the PreProcessor o=
> r
> > > similar, then you may end up with an fMeanPad that is outside ( minPa=
> d <=3D
> > > fMeanPad || fMeanPad <=3DmaxPad ) causing your first assert() in
> > > BrTPCCluster->Build() to fail. This caused me some headaches just now=
> =2E.. I
> > > think maybe this should be changed, since neg. adc-values may well be=
> a
> > > feature of unprocessed sequences we might want to work with (with all=
> the
> > > problems that gives...). What do you think?
> > >
> >
> > Methinks negative adc values are a feature of the raw data.
> >
> > The ADC values you see are ("real ADC values from the TPC" - "pedestal"=
> ).
> > The pedestal is determined as an average adc value, so non-compressed
> > raw data will have half of the ADC values negative (below pedestal).
> >
> > Because the tpc data compression keeps two adc time bins before and aft=
> er
> > each "peak", by definition these extra time bins will be below the
> > compression threshold and could well be below the pedestal, that is, ne=
> gative.
> >
>
> What Konstatin says is true, and I have talked to Peter about this. I gu=
> ess the
> problem is with
> the assert function which causes (as I understand it) the program to st=
> op
> execution (crash, if you
> will) when the condition in assert's argument is false. I think that the=
> asserts
> should be removed and
> replaced with something else which allows the program to continue to the =
> next
> cluster, or event...
> Cheers,
> Ian
>
> > --
> > Konstantin Olchanski
> > Physics Department, Brookhaven National Laboratory, Long Island, New Yo=
> rk
> > olchansk@bnl.gov
>
> --
> Ian--------------------------------------------------------------------
> | I.G. Bearden |
> | Niels Bohr Institute Tb 3 email: bearden@nbi.dk |
> | Blegdamsvej 17 phone: (+45) 35 32 53 23 |
> | K=F8benhavn =D8 FAX: (+45) 31 42 10 16 =
> |
> | Danmark |
> ----------------------------------------------------------------------
>
> --------------FEA80B3FE0F182A126BDBF08
> Content-Type: text/html; charset=us-ascii
> Content-Transfer-Encoding: 7bit
>
> <!doctype html public "-//w3c//dtd html 4.0 transitional//en">
> <html>
> <body text="#000000" bgcolor="#FFFFFF" link="#33CC00" vlink="#000000" alink="#000088">
> Konstantin Olchanski wrote:
> <blockquote TYPE=CITE>>
> <br>> One thing about your update: If you have sequences with negative
> <br>> adc-values, i.e. if you for some reason do not run the PreProcessor
> or
> <br>> similar, then you may end up with an fMeanPad that is outside ( minPad
> <=
> <br>> fMeanPad || fMeanPad <=maxPad ) causing your first assert() in
> <br>> BrTPCCluster->Build() to fail. This caused me some headaches just
> now... I
> <br>> think maybe this should be changed, since neg. adc-values may well
> be a
> <br>> feature of unprocessed sequences we might want to work with (with
> all the
> <br>> problems that gives...). What do you think?
> <br>>
> <p>Methinks negative adc values are a feature of the raw data.
> <p>The ADC values you see are ("real ADC values from the TPC" - "pedestal").
> <br>The pedestal is determined as an average adc value, so non-compressed
> <br>raw data will have half of the ADC values negative (below pedestal).
> <p>Because the tpc data compression keeps two adc time bins before and
> after
> <br>each "peak", by definition these extra time bins will be below the
> <br>compression threshold and could well be below the pedestal, that is,
> negative.
> <br> </blockquote>
> What Konstatin says is true, and I have talked to Peter about this.
> I guess the problem is with
> <br>the <i>assert </i>function which causes (as I understand
> it) the program to stop execution (crash, if you
> <br>will) when the condition in <i>assert'</i>s argument is false.
> I think that the asserts should be removed and
> <br>replaced with something else which allows the program to continue to
> the next cluster, or event...
> <br>Cheers,
> <br>Ian
> <blockquote TYPE=CITE>--
> <br>Konstantin Olchanski
> <br>Physics Department, Brookhaven National Laboratory, Long Island, New
> York
> <br>olchansk@bnl.gov</blockquote>
>
> <pre>--
> Ian--------------------------------------------------------------------
> | I.G. Bearden |
> | Niels Bohr Institute Tb 3 email: bearden@nbi.dk |
> | Blegdamsvej 17 phone: (+45) 35 32 53 23 |
> | København Ø FAX: (+45) 31 42 10 16 |
> | Danmark |
> ----------------------------------------------------------------------</pre>
>
> </body>
> </html>
>
> --------------FEA80B3FE0F182A126BDBF08--
--
Ian--------------------------------------------------------------------
| I.G. Bearden |
| Niels Bohr Institute Tb 3 email: bearden@nbi.dk |
| Blegdamsvej 17 phone: (+45) 35 32 53 23 |
| København Ø FAX: (+45) 31 42 10 16 |
| Danmark |
----------------------------------------------------------------------
This archive was generated by hypermail 2b29 : Thu Oct 05 2000 - 05:26:05 EDT