[Gate-users] Gate 3.1.2 vs Gate 4.0

[schmidtr at MSKCC.ORG]

Hi Nicolas and Seb,

Thanks for the responses.

The G4 versions are: 4.9.0.p01 and 4.9.1.p03 for v3.1.2 and v4.0
respectively.

So far we've only noticed this difference between runs with a
complicated voxelized phantom.  The runs with standard
cylinder/sphere/box phantoms seem to be fine.  I speak with Hyejoo and
see if she can get me the root files and I'll take a look at the x-y-z
global event distributions.

Thanks,

ross

-----Original Message-----
From: Nicolas Karakatsanis [mailto:knicolas at mail.ntua.gr] 
Sent: Monday, January 12, 2009 8:31 AM
To: Seb
Cc: Schmidtlein, Ross/Medical Physics; gate-users at lists.healthgrid.org
Subject: Re: [Gate-users] Gate 3.1.2 vs Gate 4.0

Hi Ross and Sebastian

could you please specify which Geant4 version you are using for Gate
3.1.2 and 4.0 versions?
Geant4 low energy models have been revised lately.

Best regards
Nicolas

Seb wrote:
> Hi Ross,
>
> First, I would like to say that nothing in the code was changed 
> between version 3.1.2 et 4.0, regarding the physical processes when 
> you use the classical navigator. This is why I can't give you a direct

> and clear solution.
> Could you check the globalPos (X-Y-Z) variables between the 2 versions

> ? Is there a difference in this event distribution ?
>
> Cheers
> Seb
>>
>> I'm working with Hyejoo on this and we have confirmed that we are 
>> using the correct low EM processes. The differences that we have seen

>> in the sinograms for identical models and input parameters are 
>> dramatic (I've attached the figures to this email, sinogramS top and 
>> sinogramTheta bottom).  These differences seem too large to be due to

>> a refinement in the low energy model associated with the new release 
>> of G4.  We are going to perform another GATE run with a simple 
>> asymmetric phantom (I'm thinking an elliptical cylinder phantom with 
>> a 4:1 hot cylinder offset) and run it with both standard GATE shapes 
>> and as a voxelized phantom and see if we can pin down the culprit.
>> Any ideas on your side?
>>
>> Cheers,
>>
>> ross
>>
>
> _______________________________________________
> Gate-users mailing list
> Gate-users at lists.healthgrid.org
> http://lists.healthgrid.org/mailman/listinfo/gate-users
>