Percolate JC_rev~ bug???

Nov 12, 2008 at 5:42pm

Percolate JC_rev~ bug???

Hi

I just found what seems to be (is) a bug in Percolate’s JC_rev~
object (took me quite some time to track it !!)

_When you input some sound (audio) in JC_rev~ AND then cut this sound
(in the case of my patch I create short (1-5 seconds “bursts” of
sound), some time later (a few seconds, maybe one minute) the JC_rev~
suddenly makes the CPU go much higher. And then higher again. And
higher….

_in my real patch, which starts at a mean 20-30%, i end up quickly at
more than 100% (in the test patch this behaviour seems not to be
present – it just jumps once. In my “real patch” there is about 50 or
so “bursts”)

_when i input another burst, the cpu goes back. One more burst, it
climbs again…. and again

_I work on a macBook pro 2×2.5, osX.5.4, max4.6.3 (maybe percolate
does not support max4 anymore?????!!!!)

In any case it is not normal, I think

many thanks

kasper

#P toggle 368 612 15 0;
#P window setfont “Sans Serif” 9.;
#P window linecount 1;
#P message 389 585 14 196617 1;
#P newex 372 642 64 196617 metro 100;
#P newex 372 664 64 196617 adstatus cpu;
#P message 785 206 37 196617 $1 10;
#P user gain~ 576 474 24 100 158 0 1.071519 7.94321 10.;
#P toggle 788 118 79 0;
#P newex 632 232 39 196617 noise~;
#P user gain~ 474 474 24 100 158 0 1.071519 7.94321 10.;
#P newex 476 446 55 196617 cycle~ 56;
#P user multiSlider 239 895 883 87 0. 101. 1 3433 15 0 0 2 0 0 0;
#M frgb 0 0 0;
#M brgb 255 255 255;
#M rgb2 127 127 127;
#M rgb3 0 0 0;
#M rgb4 37 52 91;
#M rgb5 74 105 182;
#M rgb6 112 158 18;
#M rgb7 149 211 110;
#M rgb8 187 9 201;
#M rgb9 224 62 37;
#M rgb10 7 114 128;
#P window setfont “Sans Serif” 36.;
#P number 391 832 107 36 0 0 0 3 0 0 0 221 221 221 222 222 222 0 0 0;
#P window setfont “Sans Serif” 9.;
#P newex 381 777 53 196617 maximum;
#P window setfont “Sans Serif” 24.;
#P number 383 707 63 24 0 0 0 3 0 0 0 221 221 221 222 222 222 0 0 0;
#P user ezdac~ 496 634 540 667 0;
#P window setfont “Sans Serif” 9.;
#P message 550 277 54 196617 mix 4 0.4;
#P newex 598 443 51 196617 JC_rev~;
#P newex 514 194 48 196617 loadbang;
#P newex 785 224 30 196617 line~;
#P newex 637 285 158 196617 *~ 0.;
#B color 5;
#P connect 1 0 0 1;
#P connect 15 0 1 0;
#P connect 13 0 15 0;
#P connect 12 0 0 0;
#P fasten 0 0 3 0 642 422 603 422;
#P fasten 4 0 3 0 555 367 603 367;
#P connect 3 0 14 0;
#P connect 2 0 18 0;
#P connect 2 0 4 0;
#P connect 11 0 5 0;
#P connect 11 0 5 1;
#P connect 14 0 5 1;
#P connect 14 0 5 0;
#P connect 10 0 11 0;
#P hidden connect 8 0 7 1;
#P connect 7 0 8 0;
#P connect 16 0 9 0;
#P connect 16 0 6 0;
#P connect 6 0 7 0;
#P connect 17 0 16 0;
#P connect 19 0 17 0;
#P connect 18 0 19 0;
#P window clipboard copycount 20;

Kasper T. Toeplitz
noise, composition, bass, computer

http://www.sleazeArt.com

http://www.myspace.com/sleazeart

#40802
Nov 13, 2008 at 7:02am

On 12 nov. 08, at 18:42, Kasper T Toeplitz wrote:

> _When you input some sound (audio) in JC_rev~ AND then cut this
> sound (in the case of my patch I create short (1-5 seconds “bursts”
> of sound), some time later (a few seconds, maybe one minute) the
> JC_rev~ suddenly makes the CPU go much higher. And then higher
> again. And higher….

< ...>

> _I work on a macBook pro 2×2.5, osX.5.4, max4.6.3 (maybe percolate
> does not support max4 anymore?????!!!!)

Time to use your good old G4 PB again… This sounds like a denormal
problem, which is intel only. Let’s hope it will be corrected.

p

#144715
Nov 14, 2008 at 8:03am

Patrick Delges schrieb:
> Time to use your good old G4 PB again… This sounds like a denormal
> problem, which is intel only. Let’s hope it will be corrected.

One simple way to never let a denormal problematic audio object get into
denormals, is to feed it with a noise of -132 dB, which is inaudible
even with the best converters…

The example below is only -126 dB, but might be sufficient already, call
it denormal~ and just connect it to the input of your reverb…

I can’t test it, as I am on PPC. Let us know if it works…

And if you are at it, you could also try a simple dc offset with

[sig~ 0.000001] (Don’t know if it would work if there are high pass
filters involved)

– Pasted Max Patch, click to expand. –


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

#144716
Nov 14, 2008 at 9:34am

hi

i don’t have max5 so i could not try your patch

but I found another, very easy, workaround

I just throw away this JC_rev~ and used a good old freeverb~ instead
- works great!!! ;-)

best

kasper

>Patrick Delges schrieb:
>>Time to use your good old G4 PB again… This sounds like a
>>denormal problem, which is intel only. Let’s hope it will be
>>corrected.
>
>One simple way to never let a denormal problematic audio object get
>into denormals, is to feed it with a noise of -132 dB, which is
>inaudible even with the best converters…
>
>The example below is only -126 dB, but might be sufficient already,
>call it denormal~ and just connect it to the input of your reverb…
>
>I can’t test it, as I am on PPC. Let us know if it works…
>
>And if you are at it, you could also try a simple dc offset with
>
>[sig~ 0.000001] (Don’t know if it would work if there are high pass
>filters involved)
>
>———-begin_max5_patcher———-

This post seems to contain a broken pasted patch. This can happen if you don’t copy the entire “———-begin_max5_patcher———-” or “———–end_max5_patcher———–” for each and every patch you paste; perhaps you missed a preceding or trailing “-” when you copied the patch? Please notify the Cycling 74 web team if you think it is a bug.

#144717
Nov 14, 2008 at 12:14pm

Kasper T Toeplitz schrieb:
> i don’t have max5 so i could not try your patch

it was nothing more than a [noise~] > [*~ 0.000001]. feed that into the
reverb. I just wonder if it would help, might come in handy with other
objects as well…


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

#144718

You must be logged in to reply to this topic.