From mboxrd@z Thu Jan  1 00:00:00 1970
X-Spam-Checker-Version: SpamAssassin 3.4.4 (2020-01-24) on polar.synack.me
X-Spam-Level: 
X-Spam-Status: No, score=-1.3 required=5.0 tests=BAYES_00,INVALID_MSGID
	autolearn=no autolearn_force=no version=3.4.4
X-Google-Language: ENGLISH,ASCII-7-bit
X-Google-Thread: fac41,b87849933931bc93
X-Google-Attributes: gidfac41,public
X-Google-Thread: f43e6,b87849933931bc93
X-Google-Attributes: gidf43e6,public
X-Google-Thread: 109fba,b87849933931bc93
X-Google-Attributes: gid109fba,public
X-Google-Thread: 103376,b87849933931bc93
X-Google-Attributes: gid103376,public
X-Google-Thread: 1108a1,b87849933931bc93
X-Google-Attributes: gid1108a1,public
X-Google-Thread: 114809,b87849933931bc93
X-Google-Attributes: gid114809,public
From: pcg@aber.ac.uk (Piercarlo Grandi)
Subject: Re: What is wrong with OO ?
Date: 1997/01/12
Message-ID: <vwjk9pi4iz3.fsf@osfb.aber.ac.uk>#1/1
X-Deja-AN: 209673065
sender: pcg@osfb.aber.ac.uk
references: <5a0niaINNlda@topdog.cs.umbc.edu> <32C43AC8.24E2@sn.no>
 <32C557F6.532C@rase.com> <5aa0eo$thd@krusty.irvine.com>
 <5aadbr$ad8@masters0.InterNex.Net> <32D64433.41C6@wi.leidenuniv.nl>
organization: Prifysgol Cymru, Aberystwyth
newsgroups: 
 comp.lang.c++,comp.lang.smalltalk,comp.lang.eiffel,comp.lang.ada,comp.object,comp.software-eng
Date: 1997-01-12T00:00:00+00:00
List-Id: <comp.lang.ada>


>>> "bsamwel" == Bart Samwel <bsamwel@wi.leidenuniv.nl> writes:

>> Assembler will ALWAYS be the most efficient language.  The more you
>> abstract the problem, the more you "generalize" the solution to a
>> given problem, the more you necessarily give up in efficiency.  C
>> takes a minimal 3x hit on integer arithmetic, 10x on more complex
>> stuff; and OO, because of the overhead interpreting where
>> to send things, is 10x on top of that.

bsamwel> IMO this number is only a valid approximation for Smalltalk and
bsamwel> other dynamically typed OO languages. Statically typed
bsamwel> languages like C++ (hybrid, but supporting OO) and Eiffel (pure
bsamwel> OO) achieve a much lower impact on performance than 10x. I
bsamwel> won't give any numbers because I don't have them, but I know
bsamwel> most number-crunching usually runs at very acceptable speed in
bsamwel> these languages; a bit, but not much slower than C.

By using clever implementation tricks (such as dynamic inlining!) many
dynamically typed languages can be just as fast as. Self is claimed to
run at just half the speed of C, and two to three times the speed of
most compiled ``Smalltalk'' implementations (to the point that the
Smalltalk-80 implementation written in Self by Mario Wolzcko usually
runs twice as fast as native commercial ``Smalltalk'' implementations).
Then some Lisp implementations are amazingly fast as well; CMU CL with
the Python compiler can be quite impressive. The NeXT and GNU
Objective-C iplementations have been carefully tuned, and result in
programs usually about as fast as those written in C; for dynamic
resolution is quite fast and is not the dominant overhead. ``Eiffel''
compilers also offer default dynamic resolution which is well optimized.

In many cases high performance is also achieved by analyzing the program
and inferring where dynamic overload resolution is actually not needed...