Needed information

[arogozhnikov]

Let's gather main points we don't know

• we have some information about time of read-out. Is this useful? Will we have it in real data?
• will we have z-information in real data? If yes, how is it obtained?
• no information from guard and field layers at all?
• shall we expect additional meta-information in new MC (information about track, for instance?)
• how many events we expect? What is the CPU time we are ready to spend on processing of one event?
• [future] reference code to measure quality?
• [future] What is exactly this "95%" quality? What are the penalties for 'losing' track and for 'mis-finding'?

Probably, I missed something, will add later

[ewengillies]

quick idea : we should use the hodoscope trigger as a location to start
looking for a track.

On Fri, Apr 17, 2015 at 12:55 AM, Alex Rogozhnikov <notifications@github.com

wrote:

Let's gather main points we don't know

• we have some information about time of read-out. Is this useful?
  Will we have it in real data?
• will we have z-information in real data? If yes, how is it obtained?
• no information from guard and field layers at all?
• shall we expect additional meta-information in new MC (information
  about track, for instance?)
• [future] reference code to measure quality?
• [future] What is exactly this "95%" quality? What are the penalties
  for 'losing' track and for 'mis-finding'?

Probably, I missed something, will add later

—
Reply to this email directly or view it on GitHub
#2.

[arogozhnikov]

I think, there is no problem of 'where to start', but we can use this information as well. Did you manage to get any information on the questions above?

[ewengillies]

Questions

1. Is the angular position of the 1st wire in each layer (i.e. CdcCell_cellID = 0) at at the end plate the same as written in the TDR? (written phi_EP in the TDR chart of CDC wire geometry)
2. Is the first wire in every sense layer always a sense wire?
3. Is there any timing information available from readout? I've assumed the CDC will read out the global timing of this hit, which can be tranlated to the time of the hit after the beam pulse.
4. Is any z information read out from the CDC? I think not, I suspect this is done all in track fitting, but I just want to be sure before continuing.
5. There is no information from field wires and guard layers at all, correct?

Answers

1. No it's not. There was a misunderstanding in IHEP group's tracking code about the wire configuration. To match with their code I modified my configuration correspondingly. Now in my simulation, the phiEP is 15.867, 0, 0, 0, 0, 14.96, 14.96, 0, 0, ...
2. In the TDR they don't mention. But my understanding is that the first hole (also the fist wire) is a sense wire (in sense wire layer)
3. In my simulation the global time counts from the proton pusle. Every time information recorded meant global time. In the ROOT file I generated, "CdcCell_t" means the global time when the track entered that cell, while "CdcCell_tstart" and "CdcCell_tstop" records the global time of the start point and end point of the corresponding pulse on the sense wire. However I didn't consider the pulse's travelling time in the sense wire and the cables, neither the different clock system on different boards so I didn't try to generate a time information from the end of the readout system.
4. I recorded z information. It's "CdcCell_z". I tried to record every truth information. It's confusing but at this moment I marked all the numbers related with CDC as "CdcCell_xxx". I can find a lot of branches.
5. Correct. I assume that they are blind.