Hi,
will a 64bit Pd solve the problem of numerical inaccuracy when indexing larger audio tables (compare B16.long-varispeed.pd)?
thanks and cheersz, P
- IOhannes m zmoelnig zmoelnig@iem.at [2022-03-01 12:03]:
On 3/1/22 11:43, Peter P. wrote:
Hi,
will a 64bit Pd
what's a "64bit Pd"?
Ah yes, the old confusion. I was at first thinking of a "64 bit executable" or "64 bit application" as in http://msp.ucsd.edu/software.html or as in $ file /usr/bin/puredata /usr/bin/puredata: ELF 64-bit LSB pie executable [...]
but now I figure my question would be about double-precision Pd, that is a Pd compiled with ./configure CPPFLAGS="-DPD_FLOATSIZE=64"
"64-bit $APP" traditionally means: $APP compiled for a 64-bit architecture. In that sense, Pd has already been 64-bit for a couple of yours, unless you're running a 32-bit OS or use 32-bit legacy builds.
Now, I guess what you actually mean is single-precision vs. double-precision, but this is completely unrelated to CPU architeture.
64-bit Pd -> Pd compiled for a 64-bit architecture.
double precision Pd -> Pd that internally uses double precision floating point numbers for audio computation and messages.
will a 64bit Pd solve the problem of numerical inaccuracy when indexing
no, but double precision Pd does!
Christof
On 01.03.2022 11:43, Peter P. wrote:
Hi,
will a 64bit Pd solve the problem of numerical inaccuracy when indexing larger audio tables (compare B16.long-varispeed.pd)?
thanks and cheersz, P
Pd-list@lists.iem.at mailing list UNSUBSCRIBE and account-management -> https://lists.puredata.info/listinfo/pd-list
Thanks Christof and IOhannes for your replies! I also just found this nice article from Katja https://www.katjaas.nl/doubleprecision/doubleprecision.html
best, P
-
Christof Ressi -
IOhannes m zmoelnig -
Peter P.