RE: Floating point

--===============0427392043136300721==
Content-Type: multipart/alternative;
 boundary="=_f74c3a8229c57f62b03381893d9b841e"

--=_f74c3a8229c57f62b03381893d9b841e
Content-Transfer-Encoding: quoted-printable
Content-Type: text/plain;
 charset=utf-8

In any case, yes, I see the test does not work. Its 1735 (from the Pascal-P=
5 test catalog):
=20
{
PRT test 1735: For trunc(x), the value of trunc(x) is such that if x is
 positive or zero then 0 < x-trunc(x) < 1; otherwise 1 <x-
 trunc(x) < 0. It is an error if such a value does not exist.
 ISO 7185 reference: 6.6.5.3
 }
 program iso7185prt1735;
 var a: integer;
 b: real;
 begin
 { assign maximum value of integer }
 b :=3D maxint;
 { now move it completely out of range in floating point only }
 b :=3D b+10.0;
 { now the assignment is invalid }
 a :=3D trunc(b)
 end.
 I'm open to suggestions. ISO 7185 does not have a "maxreal" or similar.
 I guess b :=3D maxint; b :=3D b*2 would do it. I was trying for something =
more subtle
 (which clearly didn't work).
 Regards,
 Scott Franco
=20
--------- Original Message --------- Subject: RE: Floating point
From: "scott andrew franco" <[email protected]>
Date: 8/31/20 8:06 pm
To: "Waldek Hebisch" <[email protected]>, "[email protected]" <[email protected]>

 Waldek,
=20
Sure, 48 bits vs 64 bits. Why didn't it truncate the mantissa on conversion=
 to float, ie, b :=3D maxint?
I would have expected something like zeros on the right side.
=20
Thanks,
=20
Scott
=20
--------- Original Message --------- Subject: Floating point
From: "Waldek Hebisch" <[email protected]>
Date: 8/31/20 7:58 pm
To: "[email protected]" <[email protected]>

On Mon, Aug 31, 2020 at 01:07:42PM -0700, scott andrew franco wrote:
 > samiam@samiam-home-pc:~/projects/pascal/pascal-p5$ cat test.pas
 > program test(output);
 > var b,c: real;
 > begin
 > b :=3D maxint;
 > c :=3D 10.0;
 > b :=3D b+c;
 > writeln(b:40:1);
 > writeln(maxint:38);
 > end.
 > samiam@samiam-home-pc:~/projects/pascal/pascal-p5$ gpc -o test test.pas
 > samiam@samiam-home-pc:~/projects/pascal/pascal-p5$ ./test
 > 9223372036854775808.0
 > 9223372036854775807
=20
 Note that floating point results are approximate and on
 64-bit machines real accuracy is lower than integer
 accuracy.
=20
 --=20
 Waldek Hebisch
=20
 _______________________________________________
 Gpc mailing list
 [email protected]
 https://www.g-n-u.de/mailman/listinfo/gpc
 _______________________________________________ Gpc mailing list Gpc@gnu=
=2Ede https://www.g-n-u.de/mailman/listinfo/gpc

--=_f74c3a8229c57f62b03381893d9b841e
Content-Transfer-Encoding: quoted-printable
Content-Type: text/html;
 charset=utf-8

<div>In any case, yes, I see the test does not work. Its 1735 (from the Pas=
cal-P5 test catalog):</div>
<div>&nbsp;</div>
<div>{</div>
<div>
<p>PRT test 1735: For trunc(x), the value of trunc(x) is such that if x is<=
br /> positive or zero then 0 &lt; x-trunc(x) &lt; 1; otherwise 1 &lt;x-<br=
 /> trunc(x) &lt; 0. It is an error if such a value does not exist.</p>
<p>ISO 7185 reference: 6.6.5.3</p>
<p>}</p>
<p>program iso7185prt1735;</p>
<p>var a: integer;<br /> b: real;</p>
<p>begin</p>
<p>{ assign maximum value of integer }<br /> b :=3D maxint;<br /> { now mov=
e it completely out of range in floating point only }<br /> b :=3D b+10.0;<=
br /> { now the assignment is invalid }<br /> a :=3D trunc(b)</p>
<p>end.</p>
<p>I'm open to suggestions. ISO 7185 does not have a "maxreal" or similar=
=2E</p>
<p>I guess b :=3D maxint; b :=3D b*2 would do it. I was trying for somethin=
g more subtle</p>
<p>(which clearly didn't work).</p>
<p>Regards,</p>
<p>Scott Franco</p>
</div>
<blockquote class=3D"threadBlockQuote" style=3D"border-left: 2px solid #C2C=
2C2; padding-left: 3px; margin-left: 4px;">--------- Original Message -----=
----
<div>Subject: RE: Floating point<br />From: "scott andrew franco" &lt;samia=
[email protected]&gt;<br />Date: 8/31/20 8:06 pm<br />To: "Waldek Hebisch" &lt=
;[email protected]&gt;, "[email protected]" &lt;[email protected]&gt;<br /><br />
<div>Waldek,</div>
<div>&nbsp;</div>
<div>Sure, 48 bits vs 64 bits.&nbsp;Why didn't it truncate the mantissa on =
conversion to float, ie, b :=3D maxint?</div>
<div>I would have expected something like zeros on the right side.</div>
<div>&nbsp;</div>
<div>Thanks,</div>
<div>&nbsp;</div>
<div>Scott</div>
<div>&nbsp;</div>
<blockquote class=3D"threadBlockQuote" style=3D"border-left: 2px solid #C2C=
2C2; padding-left: 3px; margin-left: 4px;">--------- Original Message -----=
----
<div>Subject: Floating point<br />From: "Waldek Hebisch" &lt;hebisch@math=
=2Euni.wroc.pl&gt;<br />Date: 8/31/20 7:58 pm<br />To: "[email protected]" &lt;gpc=
@gnu.de&gt;<br /><br />On Mon, Aug 31, 2020 at 01:07:42PM -0700, scott andr=
ew franco wrote:<br /> &gt; samiam@samiam-home-pc:~/projects/pascal/pascal-=
p5$ cat test.pas<br /> &gt; program test(output);<br /> &gt; var b,c: real;=
<br /> &gt; begin<br /> &gt; b :=3D maxint;<br /> &gt; c :=3D 10.0;<br /> &=
gt; b :=3D b+c;<br /> &gt; writeln(b:40:1);<br /> &gt; writeln(maxint:38);<=
br /> &gt; end.<br /> &gt; samiam@samiam-home-pc:~/projects/pascal/pascal-p=
5$ gpc -o test test.pas<br /> &gt; samiam@samiam-home-pc:~/projects/pascal/=
pascal-p5$ ./test<br /> &gt; 9223372036854775808.0<br /> &gt; 9223372036854=
775807<br /> <br /> Note that floating point results are approximate and on=
<br /> 64-bit machines real accuracy is lower than integer<br /> accuracy=
=2E<br /> <br /> -- <br /> Waldek Hebisch<br /> <br /> ____________________=
___________________________<br /> Gpc mailing list<br /> [email protected]<br /> h=
ttps://www.g-n-u.de/mailman/listinfo/gpc</div>
</blockquote>
_______________________________________________ Gpc mailing list [email protected]=
 https://www.g-n-u.de/mailman/listinfo/gpc</div>
</blockquote>

--=_f74c3a8229c57f62b03381893d9b841e--


--===============0427392043136300721==
Content-Type: text/plain; charset="us-ascii"
MIME-Version: 1.0
Content-Transfer-Encoding: 7bit
Content-Disposition: inline

_______________________________________________
Gpc mailing list
[email protected]
https://www.g-n-u.de/mailman/listinfo/gpc

--===============0427392043136300721==--