bug in phasor~ (version 5)

Sep 18, 2008 at 11:50pm

bug in phasor~ (version 5)

I’ve been getting a bug with phasor where it will respond strangely to input signals.

I think it’s only when driving it with an audio signal. But very often it will not respond with an empty argument, but changing the argument to a floating point number will make it happy again. And also vice versa (well, I suppose that part isn’t a bug, that should be expected behaviour).

It’s been happening very often and seemingly without any reason. So I thought I’d write it up here for notification and to see if anyone else is having the same problems.

I’m using the latest build of version 5, which is running on a new(ish) iMac on 10.5

trashing preferences seems to solve the problem temporarily (or maybe that’s from a series of coincidences), but certainly after a number of trashings the problem always seems to come back.

cheers, Mark.

#39909
Sep 18, 2008 at 11:59pm

On 19 sept. 08, at 01:50, mark wrote:

> I’ve been getting a bug with phasor where it will respond strangely
> to input signals.
>
> I think it’s only when driving it with an audio signal. But very
> often it will not respond with an empty argument, but changing the
> argument to a floating point number will make it happy again. And
> also vice versa (well, I suppose that part isn’t a bug, that should
> be expected behaviour).
>
> It’s been happening very often and seemingly without any reason. So
> I thought I’d write it up here for notification and to see if anyone
> else is having the same problems.
>
> I’m using the latest build of version 5, which is running on a
> new(ish) iMac on 10.5

Hi,

Could you provide an example? Are you sure you’re not sending any
weird frequency?

Best,
ej

#140712
Sep 19, 2008 at 12:10am

definitely not ruling own personal stupidity here, but I’m prety sure I’m not being crazy. The fact that it works half the time and the other half I need to fiddle around. Here’s my kick drum patch where I always seem to have problems…. Quite often the 2 phasors need different arguments even though they’re being driven in very much the same way. The ‘microtone’ subpatch is being fed by integers with points on the end between 0. and 127.

– Pasted Max Patch, click to expand. –
#140713
Sep 23, 2008 at 11:49am

mark schrieb:
> Quite often the 2 phasors need different arguments even though
> they’re being driven in very much the same way.

Could you strip that down to something which actually makes (or should
make) sound? Along with instructions? It is pretty complex and I
wouldn’t bother to check it out, because only you know what its supposed
to do and the likely hood that its just a bug in your patching is way
too high…

If you strip it down, you might actually stumble across the culprit
yourself. If not, it would make it much easier to find out the problem
if its really a bug in phasor~ (which of course is still possible…)

Stefan


Stefan Tiedje————x——-
–_____———–|————–
–(_|_ —-|—–|—–()——-
– _|_)—-|—–()————–
———-()——–www.ccmix.com

#140714
Sep 23, 2008 at 2:20pm

thanks stefan, good advice… probably the wrong idea to assume that it would be obvious what is going on. and yes, putting it into a solution that someone else could understand would be a logical way to isolate the bug.

here’s a simplified version of the patch. I haven’t been able to repeat the bug in this simplified version, but it’s being fed in the same way that the messier patch is, which is where I have run into the problem.

I’m thinking at the moment the most likely cuase for error isn’t a bug. hmm.

– Pasted Max Patch, click to expand. –
#140715
Sep 25, 2008 at 1:31am

I’m still getting odd behaviour here.

I’m using basically the exact method of feeding phasor objects that I linked above in a number of sub patches and I still have to jiggle the phasor arguments in random places every so often.

#140716
Sep 25, 2008 at 8:13am

mark schrieb:
> here’s a simplified version of the patch. I haven’t been able to
> repeat the bug in this simplified version, but it’s being fed in the
> same way that the messier patch is, which is where I have run into
> the problem.
>
> I’m thinking at the moment the most likely cuase for error isn’t a
> bug. hmm.

Much better! Now if you’d had some values for the adsrs by just adding a
preset object and creating some presets which expose the problem, it
would be even easier (I have no idea what settings you are using… ;-)

Now describe what you expect to hear (I have no idea what you are after,
or what you want to hear), and what you hear instead if its failing. It
might be that the bug is only happening on your machine, that’s why its
important to also add what kind of computer/soundcard/OS/Max Version you
are running this…

All the best,

Stefan


Stefan Tiedje————x——-
–_____———–|————–
–(_|_ —-|—–|—–()——-
– _|_)—-|—–()————–
———-()——–www.ccmix.com

#140717
Sep 27, 2008 at 2:40pm

Quote: Stefan Tiedje wrote on Thu, 25 September 2008 02:13
—————————————————-
> Much better! Now if you’d had some values for the adsrs by just adding a
> preset object and creating some presets which expose the problem, it
> would be even easier (I have no idea what settings you are using… ;-)
>
> Now describe what you expect to hear (I have no idea what you are after,
> or what you want to hear), and what you hear instead if its failing. It
> might be that the bug is only happening on your machine, that’s why its
> important to also add what kind of computer/soundcard/OS/Max Version you
> are running this…
>
> All the best,
>
> Stefan

Hmm well, the ADSR values are not so important, as the bug is not at all dependant on their values.

The unexpected behvaiour is that a phasor fed in this way sometimes has no output whatsoever (a flat line in scope~) until it’s argument is changed to a nonlogical value (i.e, it happens to me where I feed it with a signal, but it won’t accept the signal until I tell it to accept a floating point). This occurs seemingly at random (i.e. not everytime)… Mostly it looks for signals when it is told, but quite often this comes up.

I listed my details in my first post. (though I didn’t mention which soundcard… in built soundcard by the way).

#140718
Sep 28, 2008 at 9:43am

mark schrieb:
> Hmm well, the ADSR values are not so important, as the bug is not at
> all dependant on their values.
>
> The unexpected behvaiour is that a phasor fed in this way sometimes
> has no output whatsoever (a flat line in scope~) until it’s argument
> is changed to a nonlogical value (i.e, it happens to me where I feed
> it with a signal, but it won’t accept the signal until I tell it to
> accept a floating point). This occurs seemingly at random (i.e. not
> everytime)… Mostly it looks for signals when it is told, but quite
> often this comes up.
>
> I listed my details in my first post. (though I didn’t mention which
> soundcard… in built soundcard by the way).

The question is not if there is a dependency on certain values for the
bug itself, its that nobody but you knows what that patch should do. If
I don’t get it, I immediately stop going on, close the patch and let you
find out the problem…

Get the picture? You need to provide more, best a very stripped down
version which exposes the problem (including the mentioned scope~). Or
something somebody wants to look into. You need to attract interest in
your problem…
Even if you get responses like “on my system everything is running
properly” you will get more information about your problem…
I won’t be the person which could fix a bug of phasor~, but those who
are, need to believe there is a problem. So far I don’t see the evidence…

Stefan


Stefan Tiedje————x——-
–_____———–|————–
–(_|_ —-|—–|—–()——-
– _|_)—-|—–()————–
———-()——–www.ccmix.com

#140719
Nov 10, 2008 at 3:08am

Quote: Stefan Tiedje wrote on Sun, 28 September 2008 03:43
—————————————————-
> The question is not if there is a dependency on certain values for the
> bug itself, its that nobody but you knows what that patch should do. If
> I don’t get it, I immediately stop going on, close the patch and let you
> find out the problem…
>
> Get the picture? You need to provide more, best a very stripped down
> version which exposes the problem (including the mentioned scope~). Or
> something somebody wants to look into. You need to attract interest in
> your problem…
> Even if you get responses like “on my system everything is running
> properly” you will get more information about your problem…
> I won’t be the person which could fix a bug of phasor~, but those who
> are, need to believe there is a problem. So far I don’t see the evidence…
>
> Stefan

Crossed wires here slightly.

Thanks for your interest in helping me try to solve the problem. I listed all I could, but this environment is definitely not my forte, and I was not able to expose the problem in any definative manner. But certainly there are things that I could bear in mind for next time should I wish to report a problem I am having.

I’m getting closer to the root of the problem (it seems to be somehow tied to poly~ usage), and if I am able to condense it into something useful I will post here. it is not occurring with the same regularity now with the 5.05 update, but perhaps that is coincidence only.

#140720

You must be logged in to reply to this topic.