[Coco] the upshot from the optimizing BASIC code mail and a followon question

[theother_bob]

In practical programming, speed gained by compiled vs. interpreted Basic would be significantly dependent on the actual function of the program as well as how well optimized the code is (both interpreted and post-compiled optimization matter.) Programs that would gain the most from being compiled would be those using simpler Basic commands whereas programs that use a lot of more advanced functions (graphics commands that already rely on longer ML loops for example) would not see as much gain. The simpler the function, the more "overhead per instruction" used by an interpreter.

I seem to recall (but could be imagining) that Basic compilers for the CoCo claimed "up to hundreds of times faster" speed increase... always with the key phrase "up to,"

Bob

 
----- Original Message ----
From: John <johnguin at hotmail.com>
To: CoCoList for Color Computer Enthusiasts <coco at maltedmedia.com>
Sent: Thursday, November 6, 2008 6:54:55 PM
Subject: [Coco] the upshot from the optimizing BASIC code mail and a followon question

Say all,

First, a shameless plug: I got a good blog article out of the BASIC
optimization question I asked earlier.  The full article is here:
http://blogs.msdn.com/johnguin/archive/2008/11/06/how-my-old-microsoft-8-bit
-basic-coding-skills-caused-me-to-fail-a-code-review.aspx.

The summary is modern coding conventions tend to instantiate variables
immediately before their use rather than at the beginning of a program in
BASIC.  Following the "old" guidelines is verboten.

The follow on question.

Does anyone know about how much time was consumed by the interpretation
process relative to actually executing the code?  I seem to remember
compiled BASIC programs running 200% or more faster than interpreted BASIC,
but I'm wondering if anyone ever quantified the timing.

Thanks all,
John



--
Coco mailing list
Coco at maltedmedia.com
http://five.pairlist.net/mailman/listinfo/coco