"... In PAGE 5: ... It can be seen that the amount of time spent in the ProperHITEC phase is a small fraction of the overall execution time and it helps in improving the e ciency of the overall ATPG process. Table2 shows the performance of ProperGATEST1 on eight processors. It can be seen that the results remain unchanged from the uniprocessor run and that execution times have been reduced signi cantly.... ..."

"... In PAGE 5: ... It can be seen that the results remain unchanged from the uniprocessor run and that execution times have been reduced signi cantly. As a comparison, Table3 shows the results for the same circuits when a parallel deterministic algorithm such as ProperHITEC is used in isolation[9]. One must look at the total execution time, the overall test set size in terms of the number of vectors, the fault coverage and the fault e ciency when comparing dif- ferent algorithms.... ..."

"... In PAGE 5: ... This algorithm takes longer to execute than ProperGATEST1 but provides better quality results, in terms of the fault coverage obtained, more often than not, when compared to ProperGATEST1. The algorithm ProperGATEST3, whose perfor- mance can be seen in Table5 provides excellent ex- ecution times, but the test set sizes are larger than the previous algorithms. The quality of the result in terms of the fault coverage and e ciency is usually worse than that given by ProperGATEST2 and bet- ter than ProperGATEST1 in most cases.... ..."

"... In PAGE 30: ... A more detailed discussion of the purpose and utilities of these calculations is given in conjunction with the discussion of the results, below. The large scale substitution calculations that were conducted are summarized in Table4 . Each of these involved changes only to anthropogenic emissions, with anthropogenic VOC emissions either removed or reduced, and varying amounts of ethane added.... In PAGE 69: ... This requires further investigation, which is beyond the scope of this project. Results of Large-Scale Substitution Calculations As indicated on Table4 , above, the large scale substitution calculations carried out for this project consisted of removing all anthropogenic VOC emissions and then replacing these with varying amounts of ethane, to represent effects of large scale substitutions of current emissions with slowly ... In PAGE 74: ... Ethane Substitution Calculations The ethane substitution calculations carried out for this project consisted of adding back ethane to replace the removed anthropogenic VOCs by varying amounts relative to the amount of anthropogenic VOCs removed. As indicated on Table4 , above, these consisted of, in order of amount of ethane added, no ethane added, ethane added on an approximately equal mole basis, ethane added on an equal carbon basis, and ethane added on an approximately equal reactivity basis. The equal carbon basis is based on the total number of carbons in the carbon bond model species used to represent the anthropogenic VOCs that are removed (as given on Table 1, above), and the molar and the reactivity-based factors are derived from this and the relative numbers of moles or ozone impact per carbon.... ..."