### Table 1 CPU times required to compute the solution at time t = 5 Second-order upwind scheme Particle method

2006

"... In PAGE 9: ...olume counterparts, presented in Fig. 9. Another advantage of the particle method is its self adaptivity (due to the mesh-free nature) and computational efficiency. To demonstrate the latter, we compare the CPU times required to obtain the solution by both the second-order upwind scheme and the particle method at time t = 5, see Table1 . The fin... ..."

### Table I. TVD Limits for Some Existing Implicit Schemes

### Table 3: Rate Constants for Reduction of Thiopyridinyl-Cysteine Derivatives by DTTa

1999

"... In PAGE 4: ... In the case of R69C (Figure 4c), P71C, L68C, and L72C (Figure 4a), the grouping was primarily based on the reactivity in the light (see below). Table3 gives the rate constants divided by the DTT concentrations used in the measurement. The rate constants of the mutants in the highly reactive group (0.... ..."

Cited by 1

### Table 1 A Second Order Method Based on SDC.

Cited by 1

### Table 6: Second-Order Approximation Using Idiosyncratic Data

"... In PAGE 27: ...8. Table6 presents the results when the corresponding experiment is performed in my model under the baseline set of parameter values, and under several alternative parametric... In PAGE 31: ... These are eminently testable propositions.18 Given the results of Table6 , it even seems worthwhile to attempt to estimate an equation of the form of the second-order approximation to the Euler equation (but only if idiosyncratic data are used). The point of the earlier discussion of Table 6 was that the coe cient on Ei;t 2 i;t+1 didnotyieldanunbiasedestimateof .... In PAGE 31: ...arameter values). These are eminently testable propositions.18 Given the results of Table 6, it even seems worthwhile to attempt to estimate an equation of the form of the second-order approximation to the Euler equation (but only if idiosyncratic data are used). The point of the earlier discussion of Table6 was that the coe cient on Ei;t 2 i;t+1 didnotyieldanunbiasedestimateof . From a less structural point of view, however, the lesson of the table is that for any tested set of parameter values the model implies a hugely statistically signi cant relationship between consumption growth and Ei;t 2 i;t+1.... ..."

### Table 1 Computational results for Test Problem 1 h Fourth-order scheme Second-order scheme

2003

"... In PAGE 9: ... The test results are tabulated in Table 1. It is clear from Table1 that the fourth-order compact difference scheme computes more accurate solution than the second-order central difference scheme does. Multigrid method with both schemes exhibits grid independent convergence rate.... ..."

### Table 1: Second order statistics of RGB, HSI, KL and YCbCr images.

1999

"... In PAGE 4: ....1.4 Comparison of Di#0Berent Color Spaces To determine which color space is the best for joint- and separate-component coding methods, respectively,we haveevaluated the second-order statistics of the above color spaces. The results are given in Table1 . It can be seen that with the rawRGB coordinate, there is a strong correlation between anytwo color components, and that the energies of the three components are about equal.... In PAGE 5: ...3.1 FC2VQ in the KL-transformed Color Space #28KL-FC2VQ#29 As shown in Table1 , the variance of C 1 is much higher than C 2 and C 3 . The variance of the error between the original C 1 and the quantized C 1 in our study is also much higher than the variances of the C 2 and C 3 .... In PAGE 6: ...FC2VQ in Y-Cb-Cr Color Space #28YCbCr-FC2VQ#29 As shown in Table1 , the variances of Cb and Cr are small compared to the variance of Y. Recognizing that the human eyes are less sensitive to the chrominance than to the luminance and that color variation in a photo ID picture is limited, only the Y image is compressed using the FC2VQ algorithm directly; the Cb and Cr images are quantized independently in one-stage only.... ..."

Cited by 3

### Table 6: Comparison of serial CPU times required for pointwisemethods and waveform methods on RS6000/590using second-order BDF.

1993

"... In PAGE 22: ...32 64 128 256 512 1024 2048 1 2 4 8 16 Execution Time (Wall Clock Seconds Number of Processors quot;GMRES quot; quot;CSOR quot; quot;WRN quot; quot;WGMRES quot; quot;WCGS quot; Figure 9: IBM SP-2 execution times as a function of number of processors for pointwise and waveform methods for karD example, using second-order BDF integration. the order of integration, Table6 shows serial CPU times required for solutionof the eight examples using second-order BDF on an IBM RS/6000Model590 workstation. As withfirst-order BDF, the convolutionSOR results are competitive... ..."

Cited by 3

### Table 6: Comparison of serial CPU times required for pointwisemethods and waveform methods on RS6000/590using second-order BDF.

"... In PAGE 22: ...32 64 128 256 512 1024 2048 1 2 4 8 16 Execution Time (Wall Clock Seconds Number of Processors quot;GMRES quot; quot;CSOR quot; quot;WRN quot; quot;WGMRES quot; quot;WCGS quot; Figure 9: IBM SP-2 execution times as a function of number of processors for pointwise and waveform methods for karD example, using second-order BDF integration. the order of integration, Table6 shows serial CPU times required for solutionof the eight examples using second-order BDF on an IBM RS/6000Model 590 workstation. As withfirst-order BDF, the convolutionSOR results are competitive... ..."