http://GameProgrammer.Com

Programming

GP Mailing List
     Thread Index
     Date Index

ATXGPSIG List
     Thread Index
     Date Index

Google
>

Home

Wise2Food



[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Assembler (was RE:beginner)



Hmm, Does it matter? The instuctions take less than a fraction of a second
to process. hehe :)
----- Original Message -----
From: Neato <nocirckids@nocirc.org>
To: <gameprogrammer@gameprogrammer.com>
Sent: Friday, January 21, 2000 3:55 PM
Subject: Re: Assembler (was RE:beginner)


> I've actually had the opposite experience (speed-wise anyway).  I was
under
> the impression that not only do the instructions on a CISC-based processor
> run faster, but because they are less complex, they can be run even more
> simultaneously than on RISC based processors. It translates to something
> like each instruction on a CISC takes about 1/3 the time than on a RISC.
>
> I was also under the impression that it would be harder to program for
CISC,
> because there's less instructions to do what you want.  One instruction on
a
> RISC (say, a string operation), might take 3 ops on a CISC processor. With
> games, in which the time is spent using very simple instructions (a bunch
of
> mov's to blit to the screen), they would seem to run faster on CISC
> processors.
>
> Anyway, aside from the theoretical aspect, I've noticed first-hand the
> significant difference in speed between the to.  My friend was playing
> tomb-raider on his 180MHz mac at 1024x768.  It was running noticeably
faster
> than quake (a less complex game as i understand it) running at 640x480 on
my
> 233MHz pc.
>
> Some of that might be inaccurate. (minus my personal experience :). please
> correct me if i'm wrong.
>
>
> ----- Original Message -----
> From: Marc Hernandez <marc@ias.jb.com>
> To: <gameprogrammer@gameprogrammer.com>
> Sent: Thursday, January 20, 2000 1:37 PM
> Subject: Re: Assembler (was RE:beginner)
>
>
> > On Thu, 20 Jan 2000, Neato wrote:
> >
> > > > Assembly is so difficult because the x86 standard uses a CISC, or
> > > >Complex Instruction Set, approach.  Motorola uses a RISC, or Reduced
> > > >Instruction Set, in the processors they make for Apple.  CISC is
harder
> to
> > > >use, but more efficient and thus faster.  This is why an Apple, even
if
> it
> > > >is running at the same clock speed as a PC, is slower than the PC but
> also
> > > >easier to program for in Assembly.
> > >
> > > I think you meant that the other way around.
> >
> > No, that is right.  Each clock cycle on the RISC processor does
> > less than each clock cycle on the CISC machine.
> > However the issue is much bigger than clock cycle.  Number of
> > pipes (for integer and floating point), length of pipelines all play
> > important roles.
> > Although it tends to be easier to program assembly for CISCish
> > machines.  VAX and java byte code assembly are examples of this.
> >
> > --
> > Marc Hernandez
> >
> > =================================================================
> > The GameProgrammer.Com mailing list is for the open discussion
> > of any topic related to the art, science, and business of
> > programming games. This list is especially tolerant of beginners.
> > We were all beginners once
> >
> > To SUBSCRIBE or UNSUBSCRIBE please visit:
> > http://gameprogrammer.com/mailinglist.html
> >
>
> =================================================================
> The GameProgrammer.Com mailing list is for the open discussion
> of any topic related to the art, science, and business of
> programming games. This list is especially tolerant of beginners.
> We were all beginners once
>
> To SUBSCRIBE or UNSUBSCRIBE please visit:
> http://gameprogrammer.com/mailinglist.html
>

=================================================================
The GameProgrammer.Com mailing list is for the open discussion
of any topic related to the art, science, and business of
programming games. This list is especially tolerant of beginners.
We were all beginners once

To SUBSCRIBE or UNSUBSCRIBE please visit:
http://gameprogrammer.com/mailinglist.html