waddingt@umn-cs.UUCP (Jake Waddington ) (01/31/86)
In article <891@h-sc1.UUCP> breuel@h-sc1.UUCP (thomas breuel) writes: >| One of the assumptions I made before purchasing the Mac >|was that the 68000 could whiz through numerical calculations >|and impress my friends with its speed. >|I used Aztec C's 1.00D compiler, purportedly one >|of the faster version of 'C' for the Mac (I'd hate to use a >|slower one). At least the code was portable, didn't have to >|change a line! the reply was: > Also note that Aztec 'C' *might* be using 80 bit >FP ('SANE') everywhere (as compared to 32bit floats and 64bit doubles on >the PDP). I don't know; that is something you should check before >comparing the two. Finally, your compiler for the PDP might be vastly >superior to the Aztec compiler. I am working a some floating point bench marks for Aztec "C", MegaMac "C" and TML Pascal. I will post full results next week. But sence the question came up I will say that: Aztec "C" is SLOW. Yes, I have seen bench marks that say the that is is one of the fastest but a I feel that most of the bench marks are done by people with no interest in floating point math. Further Aztec does not use the SANE 80 bit floating point routines. It has it's own and this seems to be why it is slow. MegaMax does use the SANE routines and is 4 time faster doing floating point. TML Pascal is by far the fastest! It also uses the SANE interface and you can get se the full 80 bits, you cann't in "C". Note that Aztec is the most compatable and their are bugs in Megamax. As I said I will post full details next week after I run some more tests. The bottom line is the Mac can hold it's own doing floating point but with out hardware floating point its not a speed king. And the right compiler can make a big difference. Paul Fink University of Minnesota Cosmic Ray Lab
jimb@amdcad.UUCP (Jim Budler) (02/01/86)
In article <870@umn-cs.UUCP> waddingt@umn-cs.UUCP (Jake Waddington ) writes: >... >> Also note that Aztec 'C' *might* be using 80 bit >>FP ('SANE') everywhere (as compared to 32bit floats and 64bit doubles on >>the PDP). I don't know; that is something you should check before >>comparing the two. Finally, your compiler for the PDP might be vastly >>superior to the Aztec compiler. > > >I am working a some floating point bench marks for Aztec "C", MegaMac "C" >and TML Pascal. >I will post full results next week. But sence the question came up I will >say that: >... > point math. Further Aztec does not use the SANE 80 bit > floating point routines. It has it's own and this seems > to be why it is slow. > > MegaMax does use the SANE routines and is 4 time faster > doing floating point. > > TML Pascal is by far the fastest! It also uses the SANE > interface and you can get se the full 80 bits, you cann't > in "C". [ in the C's you have, see below ] Some small points: SANE is not the fastest implementation, but it is accurate. Consulair Mac C uses SANE 80 for all internal FP calculations. 80 bit FP *is* available in Consulair Mac C. -- Jim Budler Advanced Micro Devices, Inc. (408) 749-5806 Usenet: {ucbvax,decwrl,ihnp4,allegra,intelca}!amdcad!jimb Compuserve: 72415,1200
berry@tolerant.UUCP (David Berry) (02/03/86)
> In article <891@h-sc1.UUCP> breuel@h-sc1.UUCP (thomas breuel) writes: > TML Pascal is by far the fastest! It also uses the SANE > interface and you can get se the full 80 bits, you cann't > in "C". Sure you can. If you use Consulair C. It adds an 'ext' data type for 80 bit floating point numbers. -- David W. Berry dwb@well.UUCP Delphi: dwb {ucbvax,pyramid,idsvax,bene,oliveb}!tolerant!berry I'm only here for the beer.