[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: How to force single precision (32bit) math ? (Octave crash on mixed
Re: How to force single precision (32bit) math ? (Octave crash on mixed 32/64bit math)
Mon, 26 Dec 2016 15:53:11 -0800 (PST)
Well, I am not sure yet, which of them. I use toolboxes
(communications-1.2.1, signal, control) and manual math for filtering and
passband (|H(f)|) analysis. Most variables were not specifically declared
for the sake of scripting time. But some vectors, like time signal and
filter coefficients, when converted to single(), cause "warning: matrix
singular to machine precision" in quite straightforward
sums/multiplications, or functions like cheby2. Of course I use point".* .+"
operators, so I wonder, where can that influence the math.
But in a loop it usually results in unexpected termination of Octave after
10-20 steps. It gives an impression of memory leakage in single(),
plus(single) or times(single).
So, my idea was to initialize all the core declarations and operations in
View this message in context:
Sent from the Octave - General mailing list archive at Nabble.com.