[Halld-offline] TOF Software Request

[Matthew Shepherd]

Hi Beni,

In your revitalized role as TOF software guru, I have a request.  Currently the TOF software utilizes the old standard of doing "digitization" or whatever you want to call it in the DTOFMCResponse_factory.  The problem is that in our current method of operation this is typically done at run-time when one does analysis. This means for a given file that has been generated and gone through mcsmear the TOF hits fluctuate around due to random digitization effects if a single analysis program is run multiple times on the same file.

Kei has been doing some studies and he has found out that this effect is somewhat noticeable because, as I understand it, the TOF is actually used in track fitting.  The bottom line is that the momenta of foward going tracks that hit the TOF actually giggle around a bit each time the code is run.  There are even more surprising (but rare) effects:  some tracks get lost or found depending on these TOF hits.  The track multiplicity distributions actually change slightly with each run of the code.

So, the request is this:  can you move the TOF digitization into the mcsmear package.  We have now done this for just about every detector system.  Then you can safely delete the DTOFMCResponse.  Hopefully once this is done, the results of track finding on an MC sample won't change when the analysis program is rerun.

Cheers,

Matt