"... In PAGE 7: ...op.type= ash, Top.bits=4, Top.conv=5 S, Top.opamp=OpAmp2 Table5 : Models from the analog component library Circuit Parameter Model Model CPU Time SPICE SPICE CPU Time (1) S amp;H DC Gain 2.1 0.... ..."

"... In PAGE 2: ....1. Specification phase The statistical behavioral modeling of A/D converters was covered in [1] and will not be further discussed in this paper. Using this model, the targeted specifications as listed in Table2 later on, can be explored and determined on system level. 3.... In PAGE 5: ... 105 30 35 40 45 50 55 60 106 107 108 f [Hz] in SF D R / S N R [ dB ] Figure 10: Measured dynamic performance: SFDR gt;50dB and SNR gt; 43 dB. The measured performance is summarized and compared to the specified values in Table2 . All results are comparable to what had been predicted during the sizing.... ..."

"... In PAGE 2: ... A fth photomultiplier mounted on an o set guider is used to monitor a comparison star. A log of observations is shown in Table1 . The cycle num- bers quoted are with respect to the ephemeris in Schwope et al.... In PAGE 2: ... The data were calibrated using the standard star observations. The mean out-of-eclipse magni- tudes in each waveband for the three nights are listed in Table1 ; these were converted into spectral irradiance using the values quoted in Allen (1973) to enable comparisons between eclipses obtained on di erent nights and between features in di erent bandpasses. 3 RESULTS 3.... ..."

"... In PAGE 10: ...For the high-speed uniprocessors in Table2 , the crossover in the HINT graphs (Figure 6) is profound. The curve shown is for a vectorized version of HINT; an unvectorized version runs 7 times slower.... ..."

"... In PAGE 2: ....e. the VVoIP system. These earlier works focused on many aspects of network performance bottleneck issues such as frame rate, packet delay, jitter, loss and reordering mainly in the purview of voice traffic and in some cases in the purview of video traffic. While addressing video traffic behavior, these studies have limited their scope to low-speed interactive audiovisual streams ( lt;384 Kbps) and have failed to verify these findings to be applicable to high-speed interactive multimedia data streams shown in Table1 . Since high-speed VVoIP applications using technologies in Table 1 are beginning to become widespread in use [22], it has become necessary to study performance bottlenecks relating to these applications by extending earlier principles and methodologies to better understand high-speed VVoIP system bottlenecks.... In PAGE 2: ... While addressing video traffic behavior, these studies have limited their scope to low-speed interactive audiovisual streams ( lt;384 Kbps) and have failed to verify these findings to be applicable to high-speed interactive multimedia data streams shown in Table 1. Since high-speed VVoIP applications using technologies in Table1 are beginning to become widespread in use [22], it has become necessary to study performance bottlenecks relating to these applications by extending earlier principles and methodologies to better understand high-speed VVoIP system bottlenecks. ... ..."

"... In PAGE 9: ...Table 5: High-speed network: Performance measures versus In Table5 , for the high-speed network, we compare the performance measures obtained by the uid rst order ap- proximation with the performance measures obtained by simulation of the discrete-space model with c = 0. We verify that the analytical results obtained with the rst order approximation are accurate when d, i.... In PAGE 9: ...rst order approximation are accurate when d, i.e., 0:04 s. In particular, the values in Table5 are consis- tent with the fact that Qmax is inversely proportional to and X ? decreases as increases. 4 Conclusion It appears that uid modeling holds great promise in ana- lyzing network congestion and ow control problems.... ..."

"... In PAGE 6: ... HINT for high-performance computers. For the high-speed uniprocessors in Table2 , the crossover in the HINT graphs (Figure 6) is profound. The curve shown is for a vectorized version of HINT; an un- vectorized version runs 7 times slower.... ..."