[net.micro.cbm] Benchmarks,

geoffs@brl-tgr.ARPA (Geoffrey Sauerborn ) (02/13/85)

[this line intentionally left to be a helpless  sacrifice to a ruthless bug]

>...  In the meantime if there are people
>out there using compilers or psuedo compilers lets
>hear how good they are -- or how bad they are.
>
>
>rich


	I too would like to here about a few languages on the commodore(s).
Sorry but this really isn't a review per-say but I would like to hear
some reviews myself - especially on compilers.
		      -----------------------
	Brian Hilchie posted some bench marks for the C64 a while back.
To my memory they went something like this. 

		assembly 	- 15 sec.
		his C compiler 	- 26 sec.
		Basic 		- On the order of 3000 seconds.

The program used was the Eratosthenes - "Sieve" algorithm to which he posted
a listing written in C. (For the sake of reproducibility, I used the 
version found in August 84 ed. of BYTE - page 403. And for those 
interested I will mail you copies of the code in Forth, Basic, Pascal, C,
or Fortran )

	Correct me if I'm wrong, but here are some results I got.

	6502:
		Basic 2.0	- 3939 seconds [1:05:39] (VIC-20)
		VIC-Forth 	-  404 sec. (VIC-20)


questions:
	1.) I knew Forth would beat Basic by a factor of 10, but
		Brian Hilchie's C beat Forth by a factor of 10!
				       -----
		Does this mean this is a slow version of Forth or
		a Very fast version of C.

	2.) Let's here what other languages can do?
		Interpreters:
			Simon Basic
			Comal
		Compilers:
			Compiled (or almost compiled) Basic
			C64-Forth (Is any faster (slower?) )
			Promal
			Pascal - (various brands)
			C -  ( 2 that I know of )
			Fortran - (1 that I know of )
others.....	CPM:		
			name it.

					Geoff Sauerborn
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - 
...and by the way,
I know these are not micros, but I just couldn't resist:
	sys = actual cpu time
	user = time the program was somewhere in the system
	real = time that I am staring at the CRT

	PDP-11:
	 PDP-11 Basic-Plus	- real 2:49	user 2:26	sys 2.1
		Pascal		- real 5.0	user 4.4	sys 0.2
		Fortran77	-  ** error ** file too big! ???
		Fortran66 (f4p)	- real 5.0	user 5.1	sys 0.3
		C		- real 2.0	user 1.5	sys 0.1

	VAX-11/780:
		Pascal	  (pc)	- real 7.6 	user 7.2	sys 0.1
		Fortran77 (f77)	- real 2.4	user 1.5	sys 0.1
		C		- real 1.6	user 1.4 	sys 0.0
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -