Re: [Edu-sig] Knuth's books.
SNIP This knowledge, what is costly, is hard-won and precious. Nobody who really knows will divide if multiplication is as easy, and the only way to understand why that is true is to understand what the machines do at their core (By the way, this is old- school knowledge -- circa 1975). The current model is memory fetches and conditional code vs. pure ALU work. It is now faster to multiply by 1 or 0 than to test for those cases and blow the pipeline. --Scott David Daniels [EMAIL PROTECTED] Thanks Scott. I'm glad we have talented folk, such as yourself, optimizing at this next-to-the-metal level (as well is in the metal itself). We all benefit from faster, cooler CPUs. Brute force where there could have been an elegant solution is ugly to contemplate when precious resources are at stake. Kirby 4D Solutions ___ Edu-sig mailing list Edu-sig@python.org http://mail.python.org/mailman/listinfo/edu-sig
Re: [Edu-sig] Knuth's books.
El sáb, 05-11-2005 a las 20:55 -0800, Scott David Daniels escribió: They are all tough books, but none gratuitously so. Concrete Mathematics targets (smart) undergraduates, and so might be a more accessible start. It certainly reads faster than the Fascicles, which go at the same rate as the rest of TAoCP. Concrete Mathematics is wonderful! I dropped it around the middle for lack of time (the later chapter do require concentration and memory) - but it's the single best math book I ever read. And it strongly connects to Kirby's ideas about a parallel non-calculus math track. This book develops that kind of math thinking. I tried TAoCP but was dissatisfied because it seems to touch the algorithmic side of programming but ignore the beauty-of-code side. Things like Thou shall not work in assembler ;-). I like the spirit of Structure and Interpretation of Computer Programs much better. -- Not Did anything, just Did. Some things were Done and some things were not Done. And the things that were Done, Igors Did. -- Thief of Time, Terry Pratchett ___ Edu-sig mailing list Edu-sig@python.org http://mail.python.org/mailman/listinfo/edu-sig
[Edu-sig] Knuth's books.
First, I think they are some of the best out there. Hard to get through, but worth keeping as references for the next 30 years. I suggested them to you (rather than generally to students) because you said you felt: My own experience is more toward the learning to program to learn - in my case - mathematical ideas. But ultimately, to get to where I want to get, I realize that basic computational skills are not sufficient - that I need to get somewhat beyond the basics. Knuth was a mathematician by training, and quite a good one at that. Concrete Mathematics is a math book built with an eye to the kind of math that you need the deeper you get into CS. The others, the slowly evolving The Art of Computer Programming (TAoCP) address CS directly. The reason I was suggesting Fascicle 1 (the MMIX machine) was that I think it is an excellent introduction to current machine architectures from the point of view of a programmer, rather than that of an electronics designer. They are all tough books, but none gratuitously so. Concrete Mathematics targets (smart) undergraduates, and so might be a more accessible start. It certainly reads faster than the Fascicles, which go at the same rate as the rest of TAoCP. --Scott David Daniels [EMAIL PROTECTED] ___ Edu-sig mailing list Edu-sig@python.org http://mail.python.org/mailman/listinfo/edu-sig
