craigaalders@hotmail.com

Hello, I've been working on a midi sequencing patch for quite some time. Just the other day it has suddenly become very unstable. Just trying to open the file crashed Max. After opening an earlier version, I realized that things now weren't working the same as they had been prior to the crash. Since then, I've been trying to continue to work on my patch and it continues to crash suddenly or certain aspects of it will work one time and then not another, which makes it impossible to try to to figure out what's wrong. Has this happend to anyone else, and does anyone have any general info that I could try to get things back to normal? I'm running 4.5.7 on OS 10.4.

patch-suddenly-very-unstable

My situation might have been for different reasons, but I had a similar 

experience. My patch started crashing all the time. I figured out what 

was wrong by accident. I deleted an object that seemed to have caused 

the problem, then put it back in just like before and everything worked 

My situation might have been for different reasons, but I had a similar 
experience. My patch started crashing all the time. I figured out what 
was wrong by accident. I deleted an object that seemed to have caused 
the problem, then put it back in just like before and everything worked 
fine again.

this happened to me recently after i updated to osX 10.4.6. this could be your problem. if you search the forum there is a link to the "combo updapter" which somehow fixes things (even though i already had 10.4.6 installed...).

the only other time when a patch has suddenyl started to malfunction in my experience was when a deeply nested subpatcher began attempting to divide by zero. which is never a good idea.

this happened to me recently after i updated to osX 10.4.6.  this could be your problem.  if you search the forum there is a link to the "combo updapter" which somehow fixes things (even though i already had 10.4.6 installed...).

the only other time when a patch has suddenyl started to malfunction in my experience was when a deeply nested subpatcher began attempting to divide by zero.  which is never a good idea.

Hi Joe, I think you may have answered my problem. I noticed a warning: divide by zero message when my patch was running. Is this simply a math calculation where I've asked it to divide by zero, or is it implying something else? Also is there a way to find where the source of this is coming from without having to start looking through every subpatch.

Yes, this is simply a situation in which you've asked the patch to divide by zero. This causes the patch to crash. In my sequencing patch, this came up when I was doing something like dividing the time of a loop or sample across n-steps. The simplest way to avoid divide by zero would be to just set a lower limit that is greater than zero (.01?) in the inspector of the integer/float box in question. Unfortunately you will probably have to find this manually. I imagine it must be a "/" or "expr" object.

Yes, this is simply a situation in which you've asked the patch to divide by zero.  This causes the patch to crash.  In my sequencing patch, this came up when I was doing something like dividing the time of a loop or sample across n-steps.  The simplest way to avoid divide by zero would be to just set a lower limit that is greater than zero (.01?) in the inspector of the integer/float box in question.  Unfortunately you will probably have to find this manually.  I imagine it must be a "/" or "expr" object.

On 4 May 2006, at 17:20, joseph stewart wrote:

> The simplest way to avoid divide by zero would be to just set a 

> lower limit that is greater than zero (.01?) in the inspector of 

>   The simplest way to avoid divide by zero would be to just set a  
> lower limit that is greater than zero (.01?) in the inspector of  
> the integer/float box in question.

I didn't think that divide by zero errors were

generally fatal, but in case this is the case, and

assuming you find the source, you might want to have a

I didn't think that divide by zero errors were
generally fatal, but in case this is the case, and
assuming you find the source, you might want to have a
look at the bitsafe~ object,
cheers
Roger


Patch suddenly very unstable