Thank you for sharing ! For those who have issues with doing the first one, don't forget to put normals on point AND activate your point into the "add" node !
// project first point vector minpos = minpos(1, @P); @P = minpos; vector currentpos = @P; // set up initial variables vector dir = chv("axis"); int newprim = addprim(0, "polyline"); for( int i=0; i < chi("number_of_pounts"); i++ ) { // check N of closest point to the currently processed point on surface int closept = nearpoint(1, currentpos); vector closeN = point(1, "N" ,closept); // calculate new point position vector newpos = normalize(cross(closeN, dir + curlnoise(currentpos + i * 0.01) * ch("noise"))) * ch("stepsize"); // reproject new point position onto surface vector projectpos = minpos(1, currentpos + newpos); // draw new point with layer offset int newpt = addpoint(0, projectpos + closeN * ch("offset") * 0.00001 * i); addvertex(0, newprim, newpt); // set the new point to be the currentpos for next iteraction currentpos = projectpos; }
Thanks. I got curve wrap to work but its seems to only want to distribute towards the bottom of the object and not to far up unless I mess with the transform translation settings.
I dont understand why but I cannot get this to work, I have copied it exactly (at least I think so) i am getting no errors and the first point is drawn on the surface of my geometry, but nothing else happens, no matter the number of points, stepsize etc, none of it changes anything. I can see more point being added in the spreadsheet but they are all in the same spot. any ideas? was this code posted somewhere by now?
Some experts say that visual programming is the future (nodes) because visually you can understand (cover) a more complex algorithm. Question to the author of the video: Can I use visual scripting in houdini without VEX and other obsolete programming methods at all or is houdini still dependent on old technologies and visual sciping is not yet fully developed and has limitations?
Don't know what experts you are talking about, nooone of programming is "obsolete". Pretty insulting for you to say that. It is much much better for small scripts and such which isn't too complex and in that case it's much more easily understandable. In large and complex stuff it's not the most optimal way, lots of fundamental problems with it. I suppose you can mix both (which is what many people do) to have the best of both worlds. For your question, you can use VOPS which is just vex except with nodes. I think VOPS is pretty much fine for what it is, but again, none of that "you can visually understand a more complex algorithm" stuff, that only applies for people who aren't don't know code I suppose.
@@fastlearner292 If a prorammer see a name of function in alphabetical form - then they can't understand what is inside but if you represent the funbction-sence in the form of icon-node then you will understand more. Function-icon give more sense thah function-name. The visuals are more informative. Icon-designer can represent the whole paragraph of text just in one icon.
this is so vague and non-descriptive that you may as well just put a music track on top and let the timeline fly! If you really want to teach people the code/tool you've written, you should take them step by step through the process and explain what you are doing instead of skipping 99% of the explanation, which makes this video essentially useless and kind of honestly like a flex on others saying. "LOOK AT WHAT I DID". And that is very much 99% of what this industry has become.
![[TH] VCT Pacific Stage 2 - Playoffs Lower Bracket Final // PRX vs DRX](http://i.ytimg.com/vi/5HwO1bQyAx4/mqdefault.jpg)
Thank you for sharing !
For those who have issues with doing the first one, don't forget to put normals on point AND activate your point into the "add" node !
Hey Simon, thanks for the great Asset and Tutorial. But where can i find your asset ? thanks a lot :)
// project first point
vector minpos = minpos(1, @P);
@P = minpos;
vector currentpos = @P;
// set up initial variables
vector dir = chv("axis");
int newprim = addprim(0, "polyline");
for( int i=0; i < chi("number_of_pounts"); i++ ) {
// check N of closest point to the currently processed point on surface
int closept = nearpoint(1, currentpos);
vector closeN = point(1, "N" ,closept);
// calculate new point position
vector newpos = normalize(cross(closeN, dir + curlnoise(currentpos + i * 0.01) * ch("noise"))) * ch("stepsize");
// reproject new point position onto surface
vector projectpos = minpos(1, currentpos + newpos);
// draw new point with layer offset
int newpt = addpoint(0, projectpos + closeN * ch("offset") * 0.00001 * i);
addvertex(0, newprim, newpt);
// set the new point to be the currentpos for next iteraction
currentpos = projectpos;
}
Thank youi!
thanks
thanks for this tutorial, but i cannot make it for last two cases. how can i make that curves avoid each other..
whats after curl noise? its cut off
vector minpos = minpos(1, @P);
@P = minpos;
vector currentpos = @P;
vector dir = chv("axis");
int newprim = addprim(0, "polyline");
for( int i=0; i < chi("number_of_points"); i++)
{
int closept = nearpoint(1, currentpos);
vector closeN = point(1, "N", closept);
vector newpos = normalize(cross(closeN, dir + curlnoise(currentpos + i * 0.01) * ch("noise"))) * ch("stepsize");
vector projectpos = minpos(1, currentpos + newpos);
int newpt = addpoint(0, projectpos + closeN * ch("offset") * 0.00001 * i);
addvertex(0, newprim, newpt);
currentpos = projectpos;
}
!!! Before feeding geometry to the wrangler add Normal node and set "Add normals to: Points"
its working for me thanks bro
Where can I find the Noodle doodle sop?
hey can you tell me how to added geometry to the line for the first one??
Thanks. I got curve wrap to work but its seems to only want to distribute towards the bottom of the object and not to far up unless I mess with the transform translation settings.
Can i get help from you? how did you made it to work?
Can you send me a screen shot on discord or something
@@user-hl5zx1qh7s hey mate do you have a email? i need help with this.
cool
I dont understand why but I cannot get this to work, I have copied it exactly (at least I think so) i am getting no errors and the first point is drawn on the surface of my geometry, but nothing else happens, no matter the number of points, stepsize etc, none of it changes anything. I can see more point being added in the spreadsheet but they are all in the same spot. any ideas? was this code posted somewhere by now?
You target geo needs point normals, you can add a normal node and set it to points
after geometry create attribwrangle @N=@N;
@@sergeyrybkin1331 tried adding the normal node and attribute wrangle both. checked the code, no visible errors. nothing shows. what am i not doing?
I"m not trying to be dodgy or smug, but you may want to consider taking your question to forums.odforce.net/ then link your thread back here.
@@felipe_ai Can i get help from you? how did you made it to work?
Can you send me a screen shot on discord or something
True love
Thank you
hey can you tell me how to added geometry to the line for the first one??
@@Ali-pi6mn did you get it to work?
@@keanodaley512 yes I used poly wire
@@Ali-pi6mn do you have Instagram or something? I think I need to help bud.
It's working for me,thanks. In a weird way though :/
Can i get help from you? how did you made it to work?
Can you send me a screen shot on discord or something
same issue for me.
after geometry create attribwrangle @N=@N;
God
grüsse aus der schweiz
Some experts say that visual programming is the future (nodes) because visually you can understand (cover) a more complex algorithm. Question to the author of the video: Can I use visual scripting in houdini without VEX and other obsolete programming methods at all or is houdini still dependent on old technologies and visual sciping is not yet fully developed and has limitations?
Don't know what experts you are talking about, nooone of programming is "obsolete". Pretty insulting for you to say that. It is much much better for small scripts and such which isn't too complex and in that case it's much more easily understandable. In large and complex stuff it's not the most optimal way, lots of fundamental problems with it. I suppose you can mix both (which is what many people do) to have the best of both worlds. For your question, you can use VOPS which is just vex except with nodes. I think VOPS is pretty much fine for what it is, but again, none of that "you can visually understand a more complex algorithm" stuff, that only applies for people who aren't don't know code I suppose.
@@fastlearner292 If a prorammer see a name of function in alphabetical form - then they can't understand what is inside but if you represent the funbction-sence in the form of icon-node then you will understand more. Function-icon give more sense thah function-name. The visuals are more informative. Icon-designer can represent the whole paragraph of text just in one icon.
this is so vague and non-descriptive that you may as well just put a music track on top and let the timeline fly! If you really want to teach people the code/tool you've written, you should take them step by step through the process and explain what you are doing instead of skipping 99% of the explanation, which makes this video essentially useless and kind of honestly like a flex on others saying. "LOOK AT WHAT I DID". And that is very much 99% of what this industry has become.
cool