Is there a problem with this? I remember a discussion about it a while ago, but I don't think there was ever a conclusion.
m.
-----Original Message----- From: Marc Lavallée [mailto:odradek@videotron.ca] Sent: Wednesday, July 23, 2003 12:20 PM To: pd-list@iem.kug.ac.at Subject: Re: [PD] PD n00b: counter kludge
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Le 23 Juillet 2003 14:11, Matthew Allen a écrit :
They both do the same thing, I think the second one is a little cleaner and easier to read/figure out what is going on.
Of course I don't agree, since I prefer my patch... One problem with Frank's patch: the f object is indefinitely incremented.
Marc -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux)
iD8DBQE/Ht/SQdzoeKQ0PccRAgknAKC+SM8N8qI0QiFEEw4awX7MCSkAKgCcCsB6 u2ElADPZOaVniwqO1foJTug= =2St8 -----END PGP SIGNATURE-----
PD-list mailing list PD-list@iem.at http://iem.at/cgi-bin/mailman/listinfo/pd-list
Scanned on 23 Jul 2003 19:37:18 Scanning by http://erado.com
values in pd are floating* point values. floating point numbers can be very large or very small, but their accuracy is limited depending on how large or small they are. generally the accuracy is best with values close to +/- 1.
the upshot of this is that there is a large floating point number (lets call it x), such that x + 1 = x (as far as the computer is concerned), because the accuracy is less that 1. similarly there is some very small floating point number (lets call it y) so small that 1 + y = 1.
the "very large" and "very small" numbers depend on the floating point implementation on the machine running pd.
so, if you keep incrementing some value, eventually (possibly after a few minutes or a few years depending on the situation) the value will stop incrementing and appear to "stick".
- guenter's strongarm port of pd uses fixed point math i think, but this
only makes the situation worse for purposes of this discussion ;)
pix.
On Wed, 23 Jul 2003 12:38:55 -0700 Matthew Allen matthew@lith.com wrote:
Is there a problem with this? I remember a discussion about it a while ago, but I don't think there was ever a conclusion.
m.
-----Original Message----- From: Marc Lavallée [mailto:odradek@videotron.ca] Sent: Wednesday, July 23, 2003 12:20 PM To: pd-list@iem.kug.ac.at Subject: Re: [PD] PD n00b: counter kludge
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
Le 23 Juillet 2003 14:11, Matthew Allen a écrit :
They both do the same thing, I think the second one is a little cleaner and easier to read/figure out what is going on.
Of course I don't agree, since I prefer my patch... One problem with Frank's patch: the f object is indefinitely incremented.
Marc -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.2.1 (GNU/Linux)
iD8DBQE/Ht/SQdzoeKQ0PccRAgknAKC+SM8N8qI0QiFEEw4awX7MCSkAKgCcCsB6 u2ElADPZOaVniwqO1foJTug= =2St8 -----END PGP SIGNATURE-----
PD-list mailing list PD-list@iem.at http://iem.at/cgi-bin/mailman/listinfo/pd-list
Scanned on 23 Jul 2003 19:37:18 Scanning by http://erado.com
PD-list mailing list PD-list@iem.at http://iem.at/cgi-bin/mailman/listinfo/pd-list
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
pix wrote: | ... such that x + 1 = x ...
i'm a complete n00b, too, but this is a obvios problem, but could be solved in this case with ease. just reset the counter based on the event when select 0 sends a bang. but ... then ... is the mod thing necessary? uhm ... i can send you my modification if anyone is interested ;)
harald
-
Harald Schilly -
Matthew Allen -
pix