Linear Collider Forum



Home » Software Tools » Marlin et al » Event display - a combined solution?
Re: Event display - a combined solution? [message #1028 is a reply to message #1012] Wed, 29 August 2007 09:28 Go to previous messageGo to previous message
ralf
Messages: 43
Registered: August 2007
Location: DESY Hamburg
Here's some update on this topic:
Outside this forum I got some responses and everybody seems to be not 100% happy with the existing solutions. So, many people liked the idea to work together on a better one.

Here some opinions that until now seem to be shared by most (all?) people:
  • for histograms in the display ROOT seems the way to go (you get whole lot of functionality)
  • the displayed objects do have to be clickable and show the properties of the lcio object
  • graphics hardware acceleration is mandatory for the 3D view (OpenGL), else it slows down to much when displaying complicated structures
  • there should be the possibility to read in geometry from GEAR and data from LCCD
  • one should be able to switch on/off the displaying of the different parts/objects (detector and signal)



And to maybe start the discussion, here some of my opinions and questions:

1. The possibility to project data on a plane should be included (like on the anode in case of the TPC).
And it should be possible to have different projections and maybe the 3D view displaying at once for better overview. Maybe by including several instances of the display in a GUI. For speed it might even be good to have different modules displaying 2D and 3D?

2. To elaborate the thought above: I think a good way to go would be a system where into some GUI several modules could be plugged in. Every project has its own needs what to see and examine of an event and so people could build up their own specific display by using existing (generic) modules.

3. I think WIRED already offers a lot of the features one wants to see in an event display. In my case, when looking at simple structures it was fast enough, but an event in a detailed LDC detector was slowing it down painfully. This did not change after switching on hardware graphics acceleration under linux. I have heard the rumor that the hardware acceleration under windows works: has anybody more experience with this speed issue and worked with WIRED under windows?
If this speed issue cannot be solved it seems unusable to me.

4. HepRep seems to me like a good solution for the client server approach. It has the advantage that there are already existing solutions (WIRED, FRED). So if people do not like one frontend, they could exchange it with another one (that they are used too?!?)...
Has anybody experience with its speed? Could this become a bottleneck or is this no problem?

5. A color coding like a palette from bright to dark red, blue, green etc. would be very useful when displaying objects that contain a charge/signal height information, e.g. raw TPC data. It would be great to have this with online adjustable threshold and adjustable range. This is IMHO very useful when looking at measured data from a prototype, especially to identify and evaluate noise from different sources during commissioning.


 
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Read Message
Previous Topic:gear file for LDC00_02Sc?
Next Topic:Problem installing Marlin 0.9.9
Goto Forum:
  


Current Time: Thu Feb 20 19:37:21 Pacific Standard Time 2020
.:: Contact :: Home ::.

Powered by: FUDforum 3.0.1.
Copyright ©2001-2010 FUDforum Bulletin Board Software