The existance of Java and .NET seem to me to imply that a significant watershed has been reached in computing.
Once you used assembler for speed, then computers became fast enough for the extreme convenience of compilers to be worthwhile. Now, people are finding the conveniences of managed code, GC etc to be worth using VMs. For a huge number of applications the conveniences are obviously worth the penalties. Pity those of us for whom computers are still not fast enough. When we worked out what we could do with a couple of hours on the supercomputer if we could get 1000 processors, and the numbers were still depressingly limited. When does a 1THz computing come do you suppose? With Delphi you can harness other peoples cycles with BOINC but for clusters and supercomputers it seems that C and fortran are the only options . And with the advances in computing, the community (and thus the market) for high performance computing seems to have shrunk. A binding for MPI with Kylix is real possibility but lack of 64bit is stumbling block. To escape C and Fortran, I think Freepascal seems the best hope. ---------------------------------------------------------- Phil Scadden, Institute of Geological and Nuclear Sciences 764 Cumberland St, Private Bag 1930, Dunedin, New Zealand Ph +64 3 4799663, fax +64 3 477 5232 _______________________________________________ Offtopic mailing list [email protected] http://ns3.123.co.nz/mailman/listinfo/offtopic
