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=0.4 required=5.0 tests=BAYES_00,FORGED_MUA_MOZILLA
	autolearn=no autolearn_force=no version=3.4.4
X-Google-Thread: 103376,7e15102eb14c0020
X-Google-NewGroupId: yes
X-Google-Attributes: gida07f3367d7,domainid0,public,usenet
X-Google-Language: ENGLISH,ASCII-7-bit
Received: by 10.216.193.131 with SMTP id k3mr868908wen.4.1349078877226;
        Mon, 01 Oct 2012 01:07:57 -0700 (PDT)
Path: 
 q10ni47419980wif.0!nntp.google.com!feeder1-2.proxad.net!proxad.net!feeder1-1.proxad.net!217.188.199.168.MISMATCH!takemy.news.telefonica.de!telefonica.de!newsfeed.arcor.de!newsspool1.arcor-online.net!news.arcor.de.POSTED!not-for-mail
Date: Mon, 01 Oct 2012 10:07:33 +0200
From: Georg Bauhaus <rm.dash-bauhaus@futureapps.de>
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.7;
 rv:15.0) Gecko/20120907 Thunderbird/15.0.1
MIME-Version: 1.0
Newsgroups: comp.lang.ada
Subject: Re: highest bit, statically determined
References: <50673111$0$9505$9b4e6d93@newsspool1.arcor-online.net>
 <38594cbf-b52d-4002-8db9-410f55126a09@googlegroups.com>
In-Reply-To: <38594cbf-b52d-4002-8db9-410f55126a09@googlegroups.com>
Message-ID: <50694f43$0$6577$9b4e6d93@newsspool3.arcor-online.net>
Organization: Arcor
NNTP-Posting-Date: 01 Oct 2012 10:07:31 CEST
NNTP-Posting-Host: 8892f5a3.newsspool3.arcor-online.net
X-Trace: 
 DXC=g^g30;]_28`^Y=RbYBPl4`McF=Q^Z^V3h4Fo<]lROoRa8kF<OcfhCOk80[U>_ljHilPCY\c7>ejVhA=7823L\`akY\]5eQ>3Mhm
X-Complaints-To: usenet-abuse@arcor.de
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Date: 2012-10-01T10:07:31+02:00
List-Id: <comp.lang.ada>

On 30.09.12 17:39, Anatoly Chernyshev wrote:
> Ouch...
> with ada.numerics.elementary_functions;
> use ada.numerics.elementary_functions;
> Highest_Bit_In_Octet:=natural(float'truncation(log(float(N),2.0)));
>
> I didn't check it for speed though. Pros that it doesn't depend on the size.

I did. And this time I had the program actually call the randomizing
Initialize I had written for the test data. .-/ Results have changed
in favor of the recursive first_1_bit_a. The test is running on a laptop,
so it likely says little about results on a microcontroller.

x an array of 12_000 octets, 10_000 iterations for each test.

First row : Count(f(x_{k}) = f(x_{k + 1})),
second row : time in seconds

    68580000 2.635991000    -- f is first_1_bit
    68580000 2.036651000    -- f is first_1_bit_a
    68580000 27.735934000   -- f is first_1_bit_log

with inline expansion:

    68580000 1.687923000
    68580000 1.447859000
    68580000 24.481472000

Slightly faster in both cases when translation suppresses checks,
the integer versions more than the FPT version.