Re: A86: 86 Benchmarking


[Prev][Index][Thread]

Re: A86: 86 Benchmarking




in assembly, you can disable interrupts.  if you're gonna count actual
processor cycles, interrupts would eat a lot of them.

one way to determine clock speed without openning the calc would be to
flip the link port on & off as fast as possible & examine the output on a
digital oscilloscope or something.

but usually when you want to benchmark something, the actual processor
speed isn't important, it's the relative speed.  what i would recommend
is to time the calc graphing some complex function, probably resetting
between tests.

-josh


On Fri, 20 Oct 2000 15:38:16 EDT ComAsYuAre@aol.com writes:
> 
> The speed of the processor is in fact a lot closer to 4.5 mhz due to 
> the 
> direct memory access used by the screen port.  Also, the speed of 
> the 
> processor varies with battery power.  Besides, the tokenized BASIC 
> program I 
> proposed is translated into the same short assembly code and 
> executed over 
> and over again, so you get the same net result with much less 
> overhead.
> 
> 
> In a message dated 10/20/00 12:46:24 AM Eastern Daylight Time, 
> mperley@nmt.edu writes:
> 
> > I don't really think that will work. you need to know the exact 
> number of 
> > loops
> >  in a second, and assembly is much more suited for that. Let's 
> see... the
> >  processor speed is what, 6 MHz? That's 6,000,000 cycles. so make 
> a small 
> > looping
> >  program that just counts 6,000,000 cycles and stop it. store the 
> number to 


________________________________________________________________
YOU'RE PAYING TOO MUCH FOR THE INTERNET!
Juno now offers FREE Internet Access!
Try it today - there's no risk!  For your FREE software, visit:
http://dl.www.juno.com/get/tagj.