Date: Tue, 03 Jun 1997 09:54:25 -0700
From: Bill Currie <billc AT blackmagic DOT tait DOT co DOT nz>
Subject: Re: Quake vs. Demos
To: Mark T Logan <fwec AT juno DOT com>
Cc: eyal DOT ben-david AT aks DOT com, djgpp AT delorie DOT com
Reply-to: billc AT blackmagic DOT tait DOT co DOT nz
Message-id: <33944C41.2F68@blackmagic.tait.co.nz>
Organization: Tait Electronics NZ
MIME-version: 1.0
Content-type: text/plain; charset=us-ascii
Content-transfer-encoding: 7bit
References: <422564AA DOT 004D330F DOT 00 AT aks DOT com>
 <19970602 DOT 143604 DOT 7119 DOT 3 DOT fwec AT juno DOT com>
Precedence: bulk

Mark T Logan wrote:
> Not necessarily.  It doesn't matter how concise your source code is.  If
> the
> compiler implements virtual functions with a lot of conditional jumps,
> then it

You may have been thinging of switch statements.

> doesn't matter what you use.  Of course, I don't know how compilers
> implement
> virtual functions, so it may be that they are faster.  Anyone care to
> comment?

Sure ;)

Virtual functions are imlemented by table lookup.  While compiling, the 
virtual functions of a class are assigned an index that is used to look 
up the functions pointer in the virtual function table. The table also 
supports an offset of some sort used to handle inheritance issues 
(probably for multiple inheritance as the offset is usually 0 for single 
inheritance).  When a virtual functions is called, gcc generats
something
like 4 or 5 instructions.  Yes, this is slower than a pute functions
call, 
but not a lot as ther are all mov's and add's.

Bill
-- 
Leave others their otherness.