Game Development Reference
In-Depth Information
Scenario A
Scenario B
Scenario C
col
ori
col
ori
col
ori
Low k r
60 Hz
0.250
0.268
0.280
0.306
0.247
0.338
40 Hz
0.234
0.276
0.281
0.302
0.152
0.178
30 Hz
0.190
0.240
0.239
0.314
0.151
0.168
15 Hz
0.461
0.240
0.197
0.215
1.112
0.361
Medium k r
60 Hz
0.079
0.111
0.086
0.107
0.067
0.091
40 Hz
0.097
0.122
0.067
0.071
0.121
0.077
30 Hz
57.94
53.36
51.41
48.90
49.69
46.59
15 Hz
71.94
68.60
66.06
64.08
58.14
56.92
High k r
60 Hz
0.101
0.189
2.351
30.40
0.200
0.186
40 Hz
106.5
102.3
113.1
111.3
82.23
81.21
30 Hz
108.5
107.9
114.4
114.2
86.95
85.80
15 Hz
102.5
101.2
105.7
103.3
85.73
85.73
Ta b l e 6 . 1 . Stability for three scenarios, measured as the average v r of all particles over
a period of one second for both the original (ori) and the collapsed (col) SPH algorithm.
Scenario A models 8,000 particles with g r = 100 . Scenario B models 4,000 particles with
g r = 100 . Scenario C models 8,000 particles with g r =33 . All scenarios use a low,
medium and high pressure constant of k r =35 , k r = 250 ,and k r = 750 , respectively,
except for scenario B, which allows for a high pressure of k r =1 , 100 . None of the
scenarios incorporate viscosity. Unstable simulation results are marked in bold. All tests
were performed on a single core of an Intel Xeon W3520.
Figure 6.5. Execution of scenario A from Table 6.1. On the left, the simulation with
k r = 100 . On the right, the simulation with k r = 250 . Particle colors range from blue
to green to red to denote areas of low, medium, and high density, respectively (see Color
Plate I).
influence stability much, but bear in mind that the advantage of decreased gravity
is a decreased number of neighbors as the fluid simulation expands, increasing
performance. This allows us to obtain more or less the same fluid behavior with a
smaller number of fluid particles.
Search Nedrilad ::




Custom Search