Most times, bugs of this nature are merely stupid rather than fun. In
this case the problem may be due to a bug in the C function sin() in
whatever C library is being used for the Android port.
On Fri, Mar 23, 2012 at 10:33 AM, Paul Jackson wrote:
> It is fun to see how much I can learn about
It is fun to see how much I can learn about an implementation without
looking at the code. First, function evaluation runs in descending
sequence. Second, 0 appears to be special cased, although that could be in
system supplied code.
Paul
1 o. 0 9
|NaN error
| 1 o.0 9
1 o. 9 0
|NaN er
It occured to me that I should test the various vector lengths. Having done
it, it appears the problem is always associated with the first calculation
in the vector being a problem value.
Bugs =: 3 : 0
r =. i. 0
i =. 0
while. i <: y do.
a =. 0.1 * i: i
try.
b =. 1 o. a
catch.
r =.
Thanks for running the test. According to Sony, I have a
Processor Type NVIDIA Tegra2
which according to NVIDIA is a
Dual-core ARM Cortex-A9 CPU
Paul
On Mar 22, 2012 4:24 PM, "Don Guinn" wrote:
--
For information about
Tried it on my Fire and it did not fail.
ARMv7 Processor rev 3 (v7l) 300.0MHz
generic/blaze/blaze:2.3.4/GINGERBREAD/6.2.2
On Thu, Mar 22, 2012 at 4:45 PM, Paul Jackson wrote:
> I've found a strange set of errors doing trig on fractions. These do not
> occur on my 32 or 64 bit PCs. They also d
I've found a strange set of errors doing trig on fractions. These do not
occur on my 32 or 64 bit PCs. They also don't occur if x is integers.
x=.0.1*i:58
y=.1 o. x
x=.0.1*i:59. NB. this is the first vector length where I saw the error
y=.1 o. x
|NaN error
| y=.1 o.x
x=.0.1*i:143 N
