lists@lowfrequency.or

Simple question, maybe someone can ignore their hangover for a minute 

I want to use jit.matrix with the "exprfill" message to replace a 

jit.bfg in my patch, but I can't seem to get the right arguments to 

exprfill to duplicate the jit.bfg object's output. Please see the 

patch below - in particular, the "scale" message in exprfill doesn't 

seem to change the output in any noticeable way, perhaps I'm not 

using it correctly. I can't seem to find any docs that explain what 

the input range of jit.bfg should be, but I assume that since it 

takes float32 data they are something like 1- to 1, or snorm[0] (in 

Simple question, maybe someone can ignore their hangover for a minute  
and post me a reply.

I want to use jit.matrix with the "exprfill" message to replace a  
jit.bfg in my patch, but I can't seem to get the right arguments to  
exprfill to duplicate the jit.bfg object's output.  Please see the  
patch below - in particular, the "scale" message in exprfill doesn't  
seem to change the output in any noticeable way, perhaps I'm not  
using it correctly.  I can't seem to find any docs that explain what  
the input range of jit.bfg should be, but I assume that since it  
takes float32 data they are something like 1- to 1, or snorm[0] (in  
jit.expr-speak)?

simple-exprfill-question

Check out the patch below. You have to supply scaled and offset

coordinates to the functions instead of using the attrbiutes.

Hi Evan,
Check out the patch below.  You have to supply scaled and offset
coordinates to the functions instead of using the attrbiutes.

It would be great if this were in the documentation, especially the 

docs for for jit.expr, because the helpfile for it gives the 

impression that you can set these attributes using "@ATTRIBUTE VALUE" 

like I tried in my original patch - the helpfile uses a messagebox 

with: expr " noise.gradient(norm[0]*2, norm[1]*2, @seed 313)"

On Jan 1, 2007, at 6:49 PM, Wesley Smith wrote:

> Check out the patch below. You have to supply scaled and offset

> coordinates to the functions instead of using the attrbiutes.

It would be great if this were in the documentation, especially the  
docs for for jit.expr, because the helpfile for it gives the  
impression that you can set these attributes using "@ATTRIBUTE VALUE"  
like I tried in my original patch  - the helpfile uses a messagebox  
with: expr " noise.gradient(norm[0]*2, norm[1]*2, @seed 313)"

> Hi Evan,
> Check out the patch below.  You have to supply scaled and offset
> coordinates to the functions instead of using the attrbiutes.
>
> wes
>
> 


Something I thought to mention after my previous is that calling

noise.gradient is calling a functor and not calling jit.bfg. You can

pass in attributes to these functions like @seed because the functor

supports these attributes. If you look in the gnoise subpatch in

jit.bfg.help, you will see what messages/attributes noise.gradient can

handle. Notice that scale and offset aren't in there. These are

global to jit.bfg which is why they didn't have any effect in your

original patch. Hope this clarifies things a bit.

> It would be great if this were in the documentation, especially the

> docs for for jit.expr, because the helpfile for it gives the

> impression that you can set these attributes using "@ATTRIBUTE VALUE"

> like I tried in my original patch - the helpfile uses a messagebox

> with: expr " noise.gradient(norm[0]*2, norm[1]*2, @seed 313)"

> On Jan 1, 2007, at 6:49 PM, Wesley Smith wrote:

> > Check out the patch below. You have to supply scaled and offset

> > coordinates to the functions instead of using the attrbiutes.

> > #P comment 732 186 100 196617 you have to do it with you args to the

> > #P comment 218 372 140 196617 compare this to your original;

> > #P user jit.pwindow 437 411 66 66 0 0 0 0 1 0;

> > #P newex 412 362 392 196617 jit.bfg 1 char 128 128 128 @basis

> > noise.gradient @offset -1. -1. -1. @scale 2. 2. 2.;

> > #P message 508 149 314 196617 exprfill "noise.gradient(snorm[0]\,

> > snorm[1]\, snorm[2]\, @scale 1. 1. 1.\, @offset -2.25 0.24

> > #P message 350 278 105 196617 weight 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.

> > 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.;

> > #P newex 350 253 62 196617 prepend set;

> > #P newex 154 184 72 196617 sprintf get%s;

> > #P newex 215 319 78 196617 prepend append;

> > #P newex 215 274 46 196617 t l clear;

> > #P newex 215 251 87 196617 route attributes;

> > #P user ubumenu 156 160 93 196617 0 1 1 0;

Something I thought to mention after my previous is that calling
noise.gradient is calling a functor and not calling jit.bfg.  You can
pass in attributes to these functions like @seed because the functor
supports these attributes.  If you look in the gnoise subpatch in
jit.bfg.help, you will see what messages/attributes noise.gradient can
handle.  Notice that scale and offset aren't in there.  These are
global to jit.bfg which is why they didn't have any effect in your
original patch.  Hope this clarifies things a bit.

On 1/1/07, evan.raskob [lists]  wrote:
> Hi Wes,
>
> Thanks, that does it!
>
> It would be great if this were in the documentation, especially the
> docs for for jit.expr, because the helpfile for it gives the
> impression that you can set these attributes using "@ATTRIBUTE VALUE"
> like I tried in my original patch  - the helpfile uses a messagebox
> with: expr " noise.gradient(norm[0]*2, norm[1]*2, @seed 313)"
>
> Best
> Evan
>
>
> On Jan 1, 2007, at 6:49 PM, Wesley Smith wrote:
>
> > Hi Evan,
> > Check out the patch below.  You have to supply scaled and offset
> > coordinates to the functions instead of using the attrbiutes.
> >
> > wes
> >
> > #P window setfont "Sans Serif" 9.;
> > #P window linecount 4;
> > #P comment 732 186 100 196617 you have to do it with you args to the
> > function , not atributes.;
> > #P window linecount 1;
> > #P comment 218 372 140 196617 compare this to your original;
> > #P button 363 371 15 0;
> > #P user jit.pwindow 437 411 66 66 0 0 0 0 1 0;
> > #P newex 412 362 392 196617 jit.bfg 1 char 128 128 128 @basis
> > noise.gradient @offset -1. -1. -1. @scale 2. 2. 2.;
> > #P window linecount 2;
> > #P message 508 149 314 196617 exprfill "noise.gradient(snorm[0]\,
> > snorm[1]\, snorm[2]\, @scale 1. 1. 1.\, @offset -2.25 0.24
> > 1.2\, @seed -1138)";
> > #P window linecount 5;
> > #P message 350 278 105 196617 weight 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.
> > 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1. 1.;
> > #P window linecount 1;
> > #P newex 350 253 62 196617 prepend set;
> > #P newex 154 184 72 196617 sprintf get%s;
> > #P newex 215 319 78 196617 prepend append;
> > #P newex 215 297 27 196617 iter;
> > #P newex 215 274 46 196617 t l clear;
> > #P newex 215 251 87 196617 route attributes;
> > #P user ubumenu 156 160 93 196617 0 1 1 0;
> > #X add outputmode;
> > #X add adapt;
> > #X add out_name;
> > #X add type;
> > #X add dim;
> > #X add planecount;
> > #X add align;
> > #X add autocenter;
> > #X add basis;
> > #X add classname;
> > #X add offset;
> > #X add origin;
> > #X add precision;
> > #X add rotation;
> > #X add scale;
> > #X add seed;
> > #X add weight;
> > #X prefix_set 0 0  0;
> > #X pattrmode 1;
> > #P message 79 161 69 196617 getattributes;
> > #P newex 268 276 38 196617 print;
> > #P message 79 183 52 196617 summary;
> > #P button 470 199 15 0;
> > #P button 84 140 15 0;
> > #P user jit.pwindow 491 248 66 66 0 0 0 0 1 0;
> > #P window linecount 2;
> > #P message 508 187 217 196617 exprfill
> > "noise.gradient(5.77*norm[0]-2.25\, 5.77*norm[1]+0.24\,
> > 5.77*norm[2]+1.2)";
> > #P window linecount 1;
> > #P newex 492 223 191 196617 jit.matrix spherish 1 char 128 128 128;
> > #P user jit.pwindow 95 256 66 66 0 0 0 0 1 0;
> > #P window linecount 2;
> > #P newex 70 207 249 196617 jit.bfg 1 char 128 128 128 @basis
> > noise.gradient @scale 5.77 5.77 5.77 @offset -2.25 0.24 1.2;
> > #P comment 169 101 184 196617 how do you get these to be the same?
> > (jit.bfg --> exprfill??);
> > #P window linecount 4;
> > #P comment 405 115 100 196617 how come changing the scale argument
> > doesn't seem to make a difference?;
> > #P connect 23 0 21 0;
> > #P fasten 21 0 22 0 417 404 443 404;
> > #P connect 4 0 6 0;
> > #P fasten 5 0 4 0 513 218 497 218;
> > #P connect 8 0 4 0;
> > #P fasten 20 0 4 0 513 180 497 180;
> > #P connect 18 0 19 0;
> > #P connect 13 1 18 0;
> > #P fasten 13 1 10 0 297 272 273 272;
> > #P connect 15 0 16 0;
> > #P connect 14 0 15 0;
> > #P connect 13 0 14 0;
> > #P fasten 2 1 13 0 314 243 220 243;
> > #P fasten 14 1 12 0 256 298 339 298 339 153 161 153;
> > #P fasten 16 0 12 0 220 342 339 342 339 153 161 153;
> > #P fasten 12 1 17 0 202 181 159 181;
> > #P fasten 2 0 3 0 75 249 101 249;
> > #P fasten 7 0 2 0 89 158 75 158;
> > #P fasten 9 0 2 0 84 201 75 201;
> > #P fasten 11 0 2 0 84 179 75 179;
> > #P fasten 17 0 2 0 159 204 75 204;
> > #P window clipboard copycount 26;
>
>


simple exprfill question