[Gate-users] Cluster version of GATE

[sebnem erturk]

Hello All,

In references , it says that distributed computer of GATE depends on
time-domain decomposition, which splits the full length of the experiment in
equally long smaller experiments. This approach does not change important
experiment parameters such as singles, randoms, scatter rates and
deadtime.

My question, if we think that we split our job into only two projections and
we rotated the scanner 90 degree (0-45 and 45-90). Then , if we decide that
we will split the same job more smaller experiments (6 time slices )  and of
course this means that our rotating angles will be much more smaller (0-15,
15-30 ,30-45,45-60, 60-75, 75-90).

For example for the 1st run (0-45, 45-90 degree) number of singles, randoms
, scatter rates and dead time is corresponds to 45 degree will be the same
as with second run  which again the corresponds to 45 degree in the second
run (6 time slices) or they will change?

Between these two different run of the same macro , which  parameters will
be change  (because random seeds will be change ) or my  1st run and second
give me exactly the same results?


Thank you very much for any kind of information in advance .


Best regards,
-- 
Sebnem