The results look pretty good to me now. Thanks!
On 9 January 2018 at 20:44, Juergen Sauermann wrote:
> Hi Jay,
>
> I've done a complete rework of the residue function for real arguments.
> The implementation follows the ISO standard almost literally. *SVN 1037*.
>
> /// Jürgen
>
>
> On 01/08/201
Hi Jay,
I've done a complete rework of the residue function for real
arguments.
The implementation follows the ISO standard almost literally. SVN
1037.
/// Jürgen
On 01/08/2018 02:19 PM, Jay Foad wrote:
Yes, th
Sure: I would expect that the values I've highlighted in red below should
all be zero. The other results are all OK as far as I know, though I
haven't scrutinised them carefully. I'm just saying that GNU APL doesn't
seem to be following this particular rule from the ISO standard: "If Z is A
, retur
It's a waste of time to make modulo exact for inexact floating point numbers.
People should expect round off errors in those cases. Until GNU APL have
multi-precision numbers, it is pointless to make modulo exact for inexact
numbers.
Asking for exact result with something like
(1E200+⍳
Hi,
Probably this could be of help:
http://www.softelectro.ru/ieee754_en.html
Elias Mårtenson writes:
> I can't easily find the document online without having to pay for it, but
> doesn't the Wikipedia
> page contain all the information you need?
> https://en.m.wikipedia.org/wiki/IEEE_754
>
Hi Elias,
the ISO APL standard says:
"Note: The implementation-algorithm P modulo Q provides an
exact modulo operation for realnumbers
P and Q. It evaluates R←B-(×B)×|A×⌊|B÷A and return
R if (×A)=×B or R+A otherwise.
The definition of
I can't easily find the document online without having to pay for it, but
doesn't the Wikipedia page contain all the information you need?
https://en.m.wikipedia.org/wiki/IEEE_754
On 9 Jan 2018 12:14 am, "Juergen Sauermann"
wrote:
> Hi Jay,
>
> I am still puzzled by the ISO description (and can'
Hi Jay,
I am still puzzled by the ISO description (and can't find the
"IEEE standard for Binary Floating-Point Arithmetic (754)"
referenced in the standard.
Would you be able to provide the expected expected output of your
example below?
Yes, thanks! Now, when ⎕CT=0 there are some odd results:
⎕CT←0
A←(-⌽A),0,A←1e¯200 1e¯100 1 1e100 1e200
A∘.|A
0E0 ¯1E100 ¯1E0 ¯1E¯100 ¯1E¯200 0 0E00E0¯1E200 0E0 0E0
0E00E0 ¯1E0 ¯1E¯100 ¯1E¯200 0 0E00E0¯1E100 0E0 0E0
0E00E00E0 ¯1E¯100 ¯1
Hi Jay,
maybe SVN 1036 works better.
/// Jürgen
On 01/08/2018 01:02 PM, Jay Foad wrote:
Thanks. With r1035 I get:
A←(-⌽A),0,A←1e¯200 1e¯100 1 1e100 1e200
A∘.|A
Thanks. With r1035 I get:
A←(-⌽A),0,A←1e¯200 1e¯100 1 1e100 1e200
A∘.|A
0E00E00 0E0 0E00 0E00E00 0E0 0E0
0E00E00 0E0 0E00 0E00E00 0E0 0E0
0E00E00 0E0 0E00 0E00E00 0E0 0E0
0E00E00 0E0 0E0
Hi Jay,
thanks, fixed in SVN 1035.
BTW tryapl.com gives this:
A←1E¯200 1E200 ¯1E¯200 ¯1E200
A ∘.∣ A
0 0 0 0
0 0 0 0
0 0 0 0
0 0 0 0
/// Jürgen
TOn 01/08/2018 10:29 AM, Jay Foad wrote:
Thanks. At r1034 I ge
Thanks. At r1034 I get:
A←(-⌽A),0,A←1e¯200 1e¯100 1 1e100 1e200
A∘.|A
DOMAIN ERROR
And here's one of the cases that fails:
1e¯200|1e200
DOMAIN ERROR
This still seems wrong to me, since the ISO standard for Residue says
"Implementations should avoid signalling limit-error in re
Hi,
thanks, fixed in SVN 1029.
/// Jürgen
On 01/05/2018 04:37 PM, Jay Foad wrote:
Yes, that _expression_ hangs on my Linux box too. It
gets stuck here:
FloatCell::bif_residue (this=0x55
Yes, that expression hangs on my Linux box too. It gets stuck here:
FloatCell::bif_residue (this=0x55ae13a8, Z=0x55ae24f8,
A=0x55ae11d8) at FloatCell.cc:643
643 while (z < 0.0)z = z + a;
(gdb) p z
$1 = -inf
(gdb) p a
$2 = 9.9998e-201
Jay.
On 5 January
1e¯200|1e200 hangs on my mac.
> On Jan 5, 2018, at 6:57 AM, Juergen Sauermann
> wrote:
>
> Hi Jay,
>
> hmm, interesting. I am getting this:
>
> A←(-⌽A),0,A←1e¯200 1e¯100 1 1e100 1e200
> A∘.|A
> 0E00E00 0E0 0E00 0E00E00 0E0 0E0
> 0E00E00 0E0
Hi Jay,
hmm, interesting. I am getting this:
A←(-⌽A),0,A←1e¯200 1e¯100 1 1e100 1e200
A∘.|A
0E0 0E0 0 0E0 0E0 0 0E0 0E0 0 0E0 0E0
0E0 0E0 0 0E0 0E0 0 0E0 0E0 0 0E0 0
At svn r1028 on Linux I get:
A←(-⌽A),0,A←1e¯200 1e¯100 1 1e100 1e200
A∘.|A
(hangs)
Jay.
18 matches
Mail list logo
