home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
Amiga Plus 1995 #5 & #6
/
Amiga Plus CD - 1995 - No. 5 and 6.iso
/
pd
/
grafik
/
lightwave
/
lightwave-mar95
/
000391_owner-lightwave-l _Thu Mar 16 12:07:42 1995.msg
< prev
next >
Wrap
Internet Message Format
|
1995-04-09
|
4KB
Return-Path: <owner-lightwave-l>
Received: by netcom15.netcom.com (8.6.10/Netcom)
id LAA18479; Thu, 16 Mar 1995 11:15:59 -0800
Received: from zevs.ifi.unit.no by netcom15.netcom.com (8.6.10/Netcom)
id LAA18457; Thu, 16 Mar 1995 11:15:50 -0800
Received: by zevs.ifi.unit.no id AA20717
(5.67b/IDA-1.4.4 for lightwave-l@netcom.com); Thu, 16 Mar 1995 20:15:19 +0100
Date: Thu, 16 Mar 1995 20:15:18 +0100 (MET)
From: Ole Andre Schistad <Ole.Andre.Schistad@ifi.unit.no>
To: lightwave-l@netcom.com
Subject: Re: Pentium vs Amiga
In-Reply-To: <memo.64031@cix.compulink.co.uk>
Message-Id: <Pine.SUN.3.91.950316192936.12743A-100000@zevs.ifi.unit.no>
Mime-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Sender: owner-lightwave-l@netcom.com
Precedence: bulk
On Wed, 15 Mar 1995, Gary Fenton wrote:
> The Motorola chips are actually clocked doubled so to be fair you can say
> the 50mhz speed was really 100mhz. According to Phase 5's own specs, their
> 060 is about 4x faster than a 4000/040 and their 40mhz 040 is 3x faster
> than a 4000/040. I'm very surprised at your latest info comparing it to a
> Pentuim.
Well , that's strictly not true. What you're probably referring to is the
fact that internal clockspeed on intel CPUs are sometimes doubled , like
the DX-2 66Mhz . A DX-2 will access external hardware at 33MHz , which is
the base speed of the bus , while internal commands are executed at 66
MHz . Motorola CPUs are always the same speed internally and externally,
which may give faster ram accessing .
As for the 060 beating a P90 at LW rendering , allow me to speculate a
little :-) Both CPUs achieve their improced performance through multiple
internal pipelines ; that is , they can execute multiple instructions at
once so if the instructions do not conflict with each other. If two
instructions try to read or write to the same register , they cannot
be executed simultaneously. Now here's the interesting part .. the 680x0
series has more registers than intels x86 series , which COULD account
for the fact that an 060 has better performance per clock cycle than a
pentium. Performance-wise , you should expect a 50MHz 060 to beat a P60
with good margin. It's still slower than a P90 though.
This equation has another side too.. cpu overhead generated by the OS .
When people praise the amiga , it isn't JUST enthousiasm .. it has a hard
core of fact. AmigaOS is probably the tightest and most efficient
operating system available to the public , and all the library routines
are optimalized. In contrast , WindowsNT is a high-end , high-consumption
OS for server machines with endless amounts of ram . Such a contest would
be like telling Ben Johnson to outrun Carl Lewis .. with 20 pounds of led
strapped to his feet.
Lastly , I have a sincere respect Allen Hastings' programming abilities;
may it not be feasible that his 68000 programs are ever so slightly more
efficient than what his compiler spits out ? :-)
>
> I'm NOT complaining! :-))))))
>
Well , I am .. the computer world took an entirely wrong turn somewhere
around 1990 , and it's connected to the words `corporation' and
`lawyers'. I tell thee straight , in 15 years there will be no hobbyist
programmers , and no public domain software. Instead , we'll have big
corporations that cater to our needs .. or their definition of `our
needs' anyway. Sounds like a lotta fun ? :)
>
> Gary F.
>
#include <stddisclaimer.h>
#define REALNAME Ole Andre Schistad
#define EMAIL (schizo@nvg.unit.no || olesc@ifi.unit.no)
#define OCCUPATION Comp.Sci student and 3D animation hobbyist.
main(){ printf("Hello World\n");return(-1);}
/* The world is an illusion. */