J807 beta-g is now available for window/linux/macos.
This is a relatively small release that fixes one particular problem that
was discovered by beta users. Thanks for the feedback. Please have another
go.
--
For information about
For the fun of it.
d2=: 3 : 0
NB. t : total to add to
NB. m : maximum number allowed
NB. c : count of numbers to add up
't m c'=:y
NB. Reduce problem to start from zero to <:t
t=:t-c
NB. Upper Limit
ul=:{:<.t-+/
NB. Lower limit
ll=:[:>.(t-+/)%c-#
NB. numbers from [ to ]
x2y=:[+i.@>:@-~
NB.
I haven't checked it all out, but I suspect the problem is that the
Stirling formula is asymmptotic, and has a larger error near 1.
On 2018-07-02 12:20 PM, Bill Heagy wrote:
First four runs on 32bit linux:
1j7.26
1j8.53
1j_3.14
1j_2.67
64bit machine gives different imaginary parts, but rea
Henry fixed a beta-f bug reported by another user. It also fixes yours,
which is nice. We'll get out a beta-g as soon as we can.
On Mon, Jul 2, 2018 at 12:33 PM, Eric Iverson
wrote:
> Thanks for the crash report. We will work on this.
>
> On Sun, Jul 1, 2018 at 8:11 AM, David Mitchell
> wrote:
Thanks for the crash report. We will work on this.
On Sun, Jul 1, 2018 at 8:11 AM, David Mitchell
wrote:
> I installed beta-f on Windows10-64. The following steps crash J in jhs:
>
> load'jd'
> jdwelcome_jd_
> jdrt''
> jdrt 'csv'
> jdrt 'bus_lic'
> {ctl+.}
> Copyright 2017, Jsoftware Inc. All
First four runs on 32bit linux:
1j7.26
1j8.53
1j_3.14
1j_2.67
64bit machine gives different imaginary parts, but real part=1. Looks
like 1j(whatever) may be causing the problem.
On 2018-07-02 07:45 AM, Henry Rich wrote:
When this fails, what is the value in z that gives the error?
BTW,
When this fails, what is the value in z that gives the error?
BTW, Bill, that error with gcc that you reported a few weeks back turned
out to be a program error that showed up only on that compiler. If you
switch back to gcc now it should be gone.
Henry Rich
On 6/29/2018 9:59 PM, Bill Heagy
