raspberry jengine will be updated in the next beta. Thanks On Tue, Jan 26, 2021, 12:49 PM Joey K Tuttle <[email protected]> wrote:
> I can verify the exact same behavior (crashes and results) reported by > Devon in - > > JVERSION > Engine: j903/j64avx2/darwin > Beta-a: commercial/2020-12-17T10:25:30 > Library: 9.03.01 > Platform: Darwin 64 > Installer: J903 install > InstallPath: /applications/j903 > Contact: www.jsoftware.com > > and also in J902 then moving to the Pi (ARM) > > JVERSION > Engine: j902/j32/raspberry > Release-a: commercial/2020-12-08T14:21:56 > Library: 9.02.08 > Platform: Linux 32 > Installer: J902 install > InstallPath: /usr/local/lib/j902 > Contact: www.jsoftware.com > > and > > JVERSION > Engine: j902/j32/raspberry > Release-a: commercial/2020-12-08T14:21:56 > Library: 9.03.01 > Platform: Linux 32 > Installer: J903 install > InstallPath: /usr/local/lib/j903 > Contact: www.jsoftware.com > > with the added complaint that the jEngine is the same in 902 and 903.... > No crashes in j901 in the same environments. > > > > On 2021Jan 25, at 16:37, Devon McCormick <[email protected]> wrote: > > > > I can verify what Paul says for (on Windows 10): > > JVERSION > > Engine: j902/j64avx/windows > > Release-a: commercial/2020-12-05T13:40:50 > > Library: 9.02.08 > > Platform: Win 64 > > Installer: J902 install > > InstallPath: c:/pgm/j902 > > Contact: www.jsoftware.com > > > > These all crash: > > 2 4 6 %. 1 2 3 ^/ i.2x > > 2 4 6 %. 1 2 3 ^/ x: i.2 > > 2 4 6 %. 1 2 3x ^/ i.2 > > 2 4 6 %. 1 2 3 ^/ 0x 1x > > 2 4 6 %. 1 2 3x ^/ x:i.2 > > > > But these work: > > 2 4 6x %. 1 2 3 ^/ x:i.2 > > 2 4 6x %. 1 2 3x ^/ i.2 > > returning 0 2 and this one returns a floating point answer: > > 2 4 6x %. 1 2 3 ^/ i.2 > > > > > > On Mon, Jan 25, 2021 at 7:25 PM Paul Jackson <[email protected]> wrote: > > > >> I tried the expression in the associated report on several machines. > >> 2 4 6 %. 1 2 3 ^/ i.2 > >> > >> While I found varience in the numbers between releases, it varied less > on > >> 64 bit Windows and 32 bit Android machines. > >> > >> I tried the following as a mistaken approach to create extended numbers. > >> 2 4 6 %. 1 2 3x ^/ i.2 NB. J abends > >> > >> On every Android machine I tried, J abends when only the one argument is > >> extended, regardless of the > >> ---------------------------------------------------------------------- > >> For information about J forums see http://www.jsoftware.com/forums.htm > >> > > ---------------------------------------------------------------------- > For information about J forums see http://www.jsoftware.com/forums.htm > ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
