Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP Posting-Version: version B 2.10.1 exptools 1/6/84; site ihnss.UUCP Path: utzoo!watmath!clyde!cbosgd!cbdkc1!desoto!cord!hudson!ihnp1!ihnp4!ihnss!knudsen From: knudsen@ihnss.UUCP Newsgroups: net.micro.6809 Subject: More on C floating point vs BASIC09 Message-ID: <2324@ihnss.UUCP> Date: Mon, 3-Dec-84 19:16:10 EST Article-I.D.: ihnss.2324 Posted: Mon Dec 3 19:16:10 1984 Date-Received: Tue, 4-Dec-84 19:47:15 EST Organization: AT&T Bell Labs, Naperville, IL Lines: 14 ((( mixed marriage ]]] but well balanced... My apologies -- I rooted around in the Basic09 manual this weekend and yes,jej, B9 uses 40-bit floats just like RS Basic. That makes C look even slower. Also, I tried to speed up the C programs by redeclaring all the floats to doubles, so as to avoid the conversions. Absolutely no visible change. But I think you other guys on the net have pinpointed the reason-- C insists on doing everything in 64 bits, needed or not. That could quadruple the mult and divide times! Yes, the little white bible sez it's all done in double -- I had mistakenly thought that only function arguments were expanded to doubles. Seems awful stupid, but I guess C saves a little library space by not having to maintain 32-bit operators. Hurts to see something this wrong in my favorite language ... mike k