Tools
Sorted by:
Try your query at:
 |
 |
 |
 |
 |
 |
Results 1 - 10
of
8,063
Table 1. Effect of Increased Light Truck Market Share on Light-Duty Vehicle Fuel Economy
"... In PAGE 17: ...5 to 2.0 MPG higher than it actually was in 1996 ( Table1... In PAGE 21: ... Had light truck MPG improved as much as pa ssenger car MPG, the impact of the shift to light trucks would have been half as large, about 0.8 MPG ( Table1 ). The difference in stringency of passenger car and light truck standards is attributable to the fact that passenger car targets were set in the law itself, whereas the light truck standards were left to be determined by rulemakings of the Department of Transportation.... ..."
Table 2.2: Mass spectrum of B mesons predicted by the Relativistic Quark Model in comparison with experimental data. n denotes the radial quantum number and jq = Sq +L is the angular momentum of the light degrees of freedom. The notation used here is di erent from the 2S+1LJ notation because in the heavy quark limit the spin and the orbital momentum of the light degrees of freedom are combined rst, whereas in the standard spectroscopic notation rst the spins of the quarks are added up and then combined with the orbital momentum (LS coupling). The experimental results presented for the radial excitations should be taken with great care as they have not been con rmed by other experiments so far.
Table 11 shows us optimum values are attained for the intermediate load with 20 EAI threads. For the light load test, the 1*1*10, TPS optimum is also attained for 20 EAI threads, whereas mean optimum is for 25 EAI threads. For the heavy load test, the 5*5*1000 test, mean optimums were attained for 25 and 30 EAI threads, and TPS optimums for 20 and 30 threads. In the overall, the optimum number of EAI threads to use thus is 20 when sleep times between test runs are 100 ms.
2003
"... In PAGE 79: ... We now use the results of the test runs in the previous section to try to find the optimal number of EAI threads to use in a given configuration with a given sleep time, by putting the Mean results of the Mean and TPS fields into tables 11, 12 and 13. Table11 compares the variation of the amount of threads for initial sleep times (100 ms.), whereas table 12 compares the variation of the amount of threads for decremented sleep times, and table 13 the variation of the amount of threads for increased sleep times (2000 ms.... ..."
Table 1: Timings from our test scenes. The number of VPLs is 256 in each case. All timings are in milliseconds. The columns from left to right: name of the test scene; resolution; time taken by direct illumination including shadow map rendering and percentage-closer filtering; rendering the G-buffer; common scene-independent operations (see Table 2 for breakdown); ren- dering shadow maps for new VPLs in our method; rendering shadow maps for all VPLs in the comparison method; frames per second for comparison method; frames per second for our method; speedup in indirect illumination calculations; speedup in total frame time. All results are averaged over 1000 frames where both the light source and the camera are moving. Note that the total frame time is the same for spot and omnidirectional lights for our method, since the VPL management is done on CPU in parallel with GPU operations.
"... In PAGE 8: ... The huge difference is explained by the need for 3D Delaunay tetrahedralization with omnidirectional lights, whereas 2D triangulation is sufficient for spot lights. Table1 summarizes the timings from our test runs. The timings for each individual GPU task were obtained by flushing the GPU pipeline before and after each task, whereas FPS measurements were made without any addi- tional synchronization.... ..."
Table 1: Identi cation of three-particle distribution amplitudes with projections onto di erent light-cone components of the nonlocal operators. For example, ?? refers to e G ? ? 5 .
"... In PAGE 15: ...13), the distribution amplitudes and are antisymmetric under the exchange of d and u, whereas e and e are symmetric. The distributions e , correspond to the light-cone projection G (see Table1 ) and have the conformal expansion e ( ) = 120 u d g h e 00 + e 10(3 g ? 1) + : : : i ; ( ) = 120 u d g h 0 + 10( d ? u) + : : : i ; (4.12) respectively.... ..."
Table 5. Results on PIE database (Lighting) Lighting
2006
Cited by 1
Table 1: Reduction of light beams / light shafts.
"... In PAGE 6: ... Be- cause many point-in-beam queries often occur within the same 3D region, the localized behavior of caching the more recently requested portions of the hierarchy allows for efficient memory management. Table1 shows the impact on rendering time of reduc- ing the memory requirements relative to keeping all struc- tures (100%) in memory. We used different image resolu- tions of the sharks scene (Fig.... ..."
Table 1: The spectral characteristics of the SEVIRI channels including the in-flight measured radiometric noise to be compared to the on-ground characterisation results, as measured at satellite level. The in-flight value of NIR1.6 is lower than the on-ground results due to stray light seen around midnight (outage case being 15o around the Earth acquisition window, corresponding to an outage duration of about 14 nominal repeat cycles). There are no significant differences between on-ground and in-flight measurements. Any of the differences shown are due to testing accuracies. ***Note that the noise specification refers to End of Life conditions whereas the actual noise results are for Beginning of Life.
Table 1: Light
Results 1 - 10
of
8,063
