Message: Calculation speed in GEANT4 Not Logged In (login)
 Next-in-Thread Next-in-Thread
 Next-in-Forum Next-in-Forum

Sad Calculation speed in GEANT4 

Keywords: calculation speed
Forum: Documentation and Examples
Date: 11 Feb, 2002
From: Yuri Kopatch <>

I didn't find any *really* appropriate group for this message, but as I saw another similar message in this group (, I'm posting it here.

I'm just trying to learn GEANT4, and was quite impressed by the capabilities of this package, but soon my enthusiasm turn into a disappointment. I compared the calculation speed of GEANT4 and GEANT3 on a couple of examples. Namely, I just run the example N01 with /event/verbose 0 and /tracking/verbose 0 and with 20000 events. It takes about 40 seconds on my PC, so 500 events/sec. For comparison I just took my old GEANT3 program simulating gamma response of a Germanium detector, with a similar level of complexity of the geometry. It makes about 10000 events/sec, a factor of 20 faster. I also made a GEANT4 program which has exactly the same geometry as the GEANT3 code, and also does basically the same. It makes 70 events/sec, another order of magnitude down. Example N03 with as input makes about 5 events/second on my PC.

I would like to ask the experts for the reason of such tremendous decrease of calculation speed. Is there any way for optimization/improvement? I would still accept a factor of 2-3 loss of speed for the sake of whatever other advantages which GEANT4 probably gives, but with more than 2 orders of magnitude I will definitely stay with the old GEANT3.

Best regards, Yuri

Inline Depth:
 1 1
 All All
Outline Depth:
 1 1
 2 2
 All All
Add message: (add)

1 None: Re: Calculation speed in GEANT4   (Makoto Asai - 11 Feb, 2002)
 Add Message Add Message
to: "Calculation speed in GEANT4"

 Subscribe Subscribe

This site runs SLAC HyperNews version 1.11-slac-98, derived from the original HyperNews

[ Geant 4 Home | Geant 4 HyperNews | Search | Request New Forum | Feedback ]