Date: Mon, 7 Sep 2026 21:11:53 +1000
MIME-Version: 1.0
User-Agent: Mozilla Thunderbird
Subject: Re: Getting the exponent of an x87 float
Newsgroups: comp.lang.forth
References: <115qthh$3pnrg$1@dont-email.me> <116rnis$27njv$1@dont-email.me> <6a9170ca$1@news.ausics.net> <2026Aug28.161745@mips.complang.tuwien.ac.at> <nnd$270a0ef9$1e25b41c@c1212810329f950c> <116slt9$2k0en$1@dont-email.me> <116snc1$2k0en$2@dont-email.me> <116so74$2k0en$3@dont-email.me> <116ubjb$34u1v$1@dont-email.me> <2026Aug29.123526@mips.complang.tuwien.ac.at> <116ul9s$mj0h$1@dont-email.me> <116vd6n$3gogv$1@dont-email.me> <116vdq6$3gogv$2@dont-email.me> <116viqp$mvdi$1@dont-email.me> <6a98c5d1$1@news.ausics.net> <117bmre$3lgr0$1@dont-email.me> <6a996e70$1@news.ausics.net> <117bulh$3oe72$1@dont-email.me> <6a9a1f89$1@news.ausics.net> <117f3th$tbos$1@dont-email.me> <6a9b7cf3$1@news.ausics.net> <117h78t$1hpi9$2@dont-email.me> <6a9c4832$1@news.ausics.net> <117i84b$1tflm$1@dont-email.me> <6a9cd297$1@news.ausics.net> <6a9d2ae0$1@news.ausics.net> <117jn2q$2boon$2@dont-email.me> <6a9e0cc0$1@news.ausics.net> <117l6pe$2sjp8$1@dont-email.me> <117lmhg$2e74o$1@dont-email.me>
Content-Language: en-GB
From: dxf <dxforth@gmail.com>
In-Reply-To: <117lmhg$2e74o$1@dont-email.me>
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
NNTP-Posting-Host: news.ausics.net
Message-ID: <6a9e9bfa@news.ausics.net>
Organization: Ausics - https://newsgroups.ausics.net
Lines: 29
X-Complaints: abuse@ausics.net
Path: csiph.com!news.bbs.nz!news.ausics.net!not-for-mail
Xref: csiph.com comp.lang.forth:135591

On 7/09/2026 4:45 pm, marcel hendrix wrote:
> On 9/7/2026 4:16 AM, Krishna Myneni wrote:
>> On 9/6/26 20:00, dxf wrote:
> [..]>>    s" 9.999999999999999D" >dd ddfs. cr
>>>    +9.9999999999999990000000000000003 dd 0
>>> While the correct values are now printed, the formatting is a bit off in
>>> some cases.  It doesn't seem occur with mine (for whatever reason).
>
> While experimenting/fooling  around, I noticed something strange with my own implementation:
> [...]
> It looks like >DD leaves something small for the float having the LSBs, instead of 0e. This must be related to a problem with the FPU stack not
> obeying 53bits! for some reason:

Try replacing:

           ddprecision 0 ?DO  peelDigit shiftBy10  LOOP
           peelDigit
           ddprecision 0 ?DO cnvrt LOOP  FDEC @ HOLD  cnvrt

with:
           ddprecision 0 ?DO  peelDigit shiftBy10  LOOP
           peelDigit
           ddprecision 0 ?DO
             DUP 0< IF  10 +  SWAP 1- SWAP  THEN  cnvrt
           LOOP  FDEC @ HOLD  cnvrt

I found 'peelDigit' could return weird values and these messed up the output.
The above attempts to workaround those values.