AO2D: extend writing of tracks and particles #6691
Conversation
|
@noferini we can start from here |
njacazio
requested review from
a team,
iarsene,
jgrosseo and
noferini
as code owners
sawenzel
left a comment
sawenzel
left a comment
There was a problem hiding this comment.
Fine for me (although I can't judge correctness)
|
Hi @njacazio. Looks fine for me too |
| } | ||
| const float mass = o2::constants::physics::MassPionCharged; // default pid = pion | ||
| if (tofInt.getTOF(o2::track::PID::Pion) > 0.f) { | ||
| const float expBeta = (intLen / tofInt.getTOF(o2::track::PID::Pion) / cSpeed); |
There was a problem hiding this comment.
one division less with const float expBeta = (intLen / (tofInt.getTOF(o2::track::PID::Pion) * cSpeed))
There was a problem hiding this comment.
Hi Ruben, indeed, I missed that thanks!
| const float limit = 1e-4; | ||
| const float mom = TMath::Sqrt(pX * pX + pY * pY + pZ * pZ); | ||
| const float eta = 0.5f * TMath::Log((mom + pZ) / (mom - pZ)); | ||
| if (TMath::Abs(eta) > 0.9) { | ||
| if (TMath::Abs((mom - pZ) / pZ) <= limit) { | ||
| pX = truncateFloatFraction(TMath::Sqrt((1.f + limit) * (1.f + limit) - 1.f) * pZ * 0.5, mMcParticleMom); | ||
| pY = truncateFloatFraction(TMath::Sqrt((1.f + limit) * (1.f + limit) - 1.f) * pZ * 0.5, mMcParticleMom); | ||
| } | ||
| if (TMath::Abs(energy - pZ) < limit) { | ||
| energy = truncateFloatFraction(pZ + limit, mMcParticleMom); | ||
| } |
There was a problem hiding this comment.
could you elaborate what produces FPE? In any case, if you want to approximate the
pT^2 = ( (1+limit)^2 - 1) pZ^2
then one should have
pX = truncateFloatFraction(TMath::Sqrt((1.f + limit) * (1.f + limit) - 1.f) * pZ / sqrt(2), mMcParticleMom);
...
There was a problem hiding this comment.
Here the FPE is observed when computing rapidity/pseudorapidity for particles that are in the beampipe. This hack is needed to bypass a limitation that will be solved in the future when we will be able to handle conditions in Gandiva for the computation of the expression columns.
There was a problem hiding this comment.
OK, but if you want to make (mom - pZ) / pZ) = limit, this means pT^2 = ( (1+limit)^2 - 1) pZ^2 and with pX=pY=pT/sqrt(2) this gives pX=pY=truncateFloatFraction(TMath::Sqrt((1.f + limit) * (1.f + limit) - 1.f) * pZ / sqrt(2), mMcParticleMom); (note the factor 1/sqrt(2) instead of 0.5)
There was a problem hiding this comment.
that's true, thanks for spotting this!
There was a problem hiding this comment.
nonetheless, the difference should be small i.e. *0.7 vs *0.5
|
@shahor02 thanks for the useful comments! Everything should be taken care of, could we go on? |
* Fix the way TOF is written * Add PDG codes of heavier particles * Add shift in momentum energy hack * Add more fixes * Add dummy TOF EventTime maker * Resolve extra division * Fix for sqrt(2) in FPE protection hack
* Fix the way TOF is written * Add PDG codes of heavier particles * Add shift in momentum energy hack * Add more fixes * Add dummy TOF EventTime maker * Resolve extra division * Fix for sqrt(2) in FPE protection hack
5 participants
Uh oh!
There was an error while loading. Please reload this page.