i am hitting same issues. However the above recommendations did not work.
on windows 10
launching 17.0.416 command prompt says EnvControl: HOUDINI_USER+PREF_DIR missing __HVER__, ignored.
Then a pane comes up - the Houdini License administrator
saying: Error installing key: LICENSE Genereric Houdini-Master-NonCommercial 17.0 1 15-jan-2019 blah blah ….
Thats it. No houdini …
Found 35 posts.
Search results Show results as topic list.
Houdini Indie and Apprentice » error installing key
- oblongowl
- 40 posts
- Offline
Technical Discussion » Pops - document for to explain targetv , implicit drag - how and when to use them
- oblongowl
- 40 posts
- Offline
I cannot find any documentation on targetv and reading the popsolver notes there seems to be a definite science to how multiple attributes: drag, spindrag, targetv, airresist spinresist etc etc interact. There is no actual explanation of what targetv is?
Technical Discussion » Using VEX with variables to construct commands
- oblongowl
- 40 posts
- Offline
Never mind resolved it:
————————-
//assign points to groups by rand:
int i = int(rand(@ptnum)*10);
string grp = sprintf(“group_%d”, i);
setpointgroup( geoself(), grp, @ptnum, 1, “set”);
————————-
//assign points to groups by rand:
int i = int(rand(@ptnum)*10);
string grp = sprintf(“group_%d”, i);
setpointgroup( geoself(), grp, @ptnum, 1, “set”);
Technical Discussion » Using VEX with variables to construct commands
- oblongowl
- 40 posts
- Offline
Hello
I was curious about trying to make a procedurally created set of point groups
using vex wrangle - but i cannot find a way to execute the last command as a legal entry:
i@test = int(rand(@id)*10);//creates int test with values 0-9 randomly on your points.
//this is the pseudo code :
i@group_group_@test=1; // to make all the groups in one line.
Im assuming there may be a way via an addattribute ? + a string concat to sprintf the group creation with the attribute suffix…
I was curious about trying to make a procedurally created set of point groups
using vex wrangle - but i cannot find a way to execute the last command as a legal entry:
i@test = int(rand(@id)*10);//creates int test with values 0-9 randomly on your points.
//this is the pseudo code :
i@group_group_@test=1; // to make all the groups in one line.
Im assuming there may be a way via an addattribute ? + a string concat to sprintf the group creation with the attribute suffix…
Technical Discussion » Reduce the collision effect by a static deforming object
- oblongowl
- 40 posts
- Offline
Is there a way to reduce the explosive effect of a fast moving pre-animated object hitting my rbd fractured packed objects. Currently, once they hit, they just fly out really quick.
Pop drag is not good as it makes a very unrealistic result, when i really just want to see a reduce impact force upon my geometry, without seeing much drag.
Pop drag is not good as it makes a very unrealistic result, when i really just want to see a reduce impact force upon my geometry, without seeing much drag.
Technical Discussion » packed deforming static - constraints not working
- oblongowl
- 40 posts
- Offline
Technical Discussion » packed deforming static - constraints not working
- oblongowl
- 40 posts
- Offline
I am still trying to make sense of the new packed prim feature to allow deforming or animated objects work with standard active objects.
I will attach a very simple example to demonstrate my dilemma - that neither glue nor hard constraints will move with the animating static primitive.
The glue and hard constraints clearly work for the active piece, but the constraint does not see the movement of the static one.
Any ideas or suggestions why this does not work appreciated. thanks.
I will attach a very simple example to demonstrate my dilemma - that neither glue nor hard constraints will move with the animating static primitive.
The glue and hard constraints clearly work for the active piece, but the constraint does not see the movement of the static one.
Any ideas or suggestions why this does not work appreciated. thanks.
Technical Discussion » Packed Prims and Edit Sop
- oblongowl
- 40 posts
- Offline
thats odd
I tried again and still same result.
I have no control on the version im using.
You say it works - as in - it drops down an edit sop and moves objects or it doesnt allow you to drop down an edit sop ?
I tried again and still same result.
I have no control on the version im using.
You say it works - as in - it drops down an edit sop and moves objects or it doesnt allow you to drop down an edit sop ?
Technical Discussion » Packed Prims and Edit Sop
- oblongowl
- 40 posts
- Offline
When i select a packed prim in the viewport ( click on it ) , a handle displays and an edit sop is auto created in sops. But modifying the handle has no effect.
Feels like a bug.
Im using add to selection - Primitives mode and 14.0.335
Feels like a bug.
Im using add to selection - Primitives mode and 14.0.335
Technical Discussion » Packed Prims and Edit Sop
- oblongowl
- 40 posts
- Offline
Trying to translate/rotate some packed prims in the viewport - there is a bug.
Houdini drops down an edit sop, but the EDIT sop will not do anything to the selected points OR prims.
Itried to resort to a transform sop and manipulate everything using the ladder , however, every object that is made to rotate, maintains its original orientation - probably due to the complexity of packed prims and the intrinsics.
So that was a no-way forward solution either.
And i dont want to unpack … I imagine SESI just hasnt gotten to this , yet ?
Houdini drops down an edit sop, but the EDIT sop will not do anything to the selected points OR prims.
Itried to resort to a transform sop and manipulate everything using the ladder , however, every object that is made to rotate, maintains its original orientation - probably due to the complexity of packed prims and the intrinsics.
So that was a no-way forward solution either.
And i dont want to unpack … I imagine SESI just hasnt gotten to this , yet ?
Technical Discussion » constraints for packed prims
- oblongowl
- 40 posts
- Offline
I see that - thanks again for the tip - i'm adopting this now as standard - the packed objects have a name, and my constraint networks have the dop objects name prefixed to make a more specific name inside dops for the constraints.
Im now hitting issues with trying to constrain to animated geo.
An object that comes out from Dops, and brought into a Dopnetwork again, as a packed static OR packed static animating - works:
ie: a constraint network can attach this object to another and we can see it moving.
However anything else i supply to Dops , ie: anything not coming out from a dop network, will not work - that is to say - the constraint is created and i can see it at the start frame, but the object does not move. Not as static, static animated or static deforming…
What is Dops doing/ supplying to the geometry, that allows it to move in another dopnetwork as a static object? Is there some kind of invisible transform matrix going on here - i dont see it in any of the attributes or intrinsics.?
Cheers.
Im now hitting issues with trying to constrain to animated geo.
An object that comes out from Dops, and brought into a Dopnetwork again, as a packed static OR packed static animating - works:
ie: a constraint network can attach this object to another and we can see it moving.
However anything else i supply to Dops , ie: anything not coming out from a dop network, will not work - that is to say - the constraint is created and i can see it at the start frame, but the object does not move. Not as static, static animated or static deforming…
What is Dops doing/ supplying to the geometry, that allows it to move in another dopnetwork as a static object? Is there some kind of invisible transform matrix going on here - i dont see it in any of the attributes or intrinsics.?
Cheers.
Technical Discussion » constraints for packed prims
- oblongowl
- 40 posts
- Offline
Hey - thanks for the heads up - didnt occur to me that the docs would sneak in a new example for packed prims
I managed to get it working myself, but this helped a lot since i set it up with the more complex conetwist constraint.
For anyone reading this the trick to make your packed primitive obey the actual point locations is to make the name attributes literally ‘objectname’/'packedprim name'
eg: object/packedname_02
allowing the constraint t opinpoint the actual packed prim which is buried inside a DOP rbd object name visible in the details pane.
I managed to get it working myself, but this helped a lot since i set it up with the more complex conetwist constraint.
For anyone reading this the trick to make your packed primitive obey the actual point locations is to make the name attributes literally ‘objectname’/'packedprim name'
eg: object/packedname_02
allowing the constraint t opinpoint the actual packed prim which is buried inside a DOP rbd object name visible in the details pane.
The Orbolt Smart 3D Asset Store » Remove orbolt from system
- oblongowl
- 40 posts
- Offline
I have tried to use Orbolt twice on a Mac OSX through Safari and on a Irix fedora core system.
Neither times have i ever managed to download anything successfully.
It is incredibly hard to penetrate and try to identify why it will not work as the interface is so plug in and friendly (NOT) .
Is there some kind of tutorial or demo or key points that perhaps I am missing ??
Neither times have i ever managed to download anything successfully.
It is incredibly hard to penetrate and try to identify why it will not work as the interface is so plug in and friendly (NOT) .
Is there some kind of tutorial or demo or key points that perhaps I am missing ??
Technical Discussion » delayed load geometry offset sequences
- oblongowl
- 40 posts
- Offline
Thanks - some good info
In my setup my ‘instance obj’ was referencing another object which was using the load delay procedural shader and using full point instancing, however, turning it to fast point instancing and not referencing an external object works just as fast , however to rotate and scale (in x y or z) each unique instance is a bigger technical setup - passing transform matrices.
Something i will revisit for sure.
Im passing point attributes from my grid points to ‘transform’ each instance, via point() expressions in conjunction with instancepoint() and getting the same result.
As ever houdini wins in its flexibility. cheers.
In my setup my ‘instance obj’ was referencing another object which was using the load delay procedural shader and using full point instancing, however, turning it to fast point instancing and not referencing an external object works just as fast , however to rotate and scale (in x y or z) each unique instance is a bigger technical setup - passing transform matrices.
Something i will revisit for sure.
Im passing point attributes from my grid points to ‘transform’ each instance, via point() expressions in conjunction with instancepoint() and getting the same result.
As ever houdini wins in its flexibility. cheers.
Technical Discussion » delayed load geometry offset sequences
- oblongowl
- 40 posts
- Offline
Let me try to be a bit clearer, i will try to drop a hip file if i can cut it down for portability.
Using the instance points and doing a $OFFSET is fine.
Should work good for most things.
My question (and the solution i now have) is faaar faster - by using the delay load procedural CVEX shader for geometry.
i can now adjust the scale of each instanced volume (bgeo) at every point, as well as supply a unique frame offset to every point instanced volume.
Renders as fast as rendering one volume (and I now have a hundred voumes in the scene, all offset and uniquely scaled). Works like a dream.
The main issue is with the fact i cannot make the load delay procedural shader ‘procedural’.
If you need more clarification i can email you a hip.
Thanks.
Using the instance points and doing a $OFFSET is fine.
Should work good for most things.
My question (and the solution i now have) is faaar faster - by using the delay load procedural CVEX shader for geometry.
i can now adjust the scale of each instanced volume (bgeo) at every point, as well as supply a unique frame offset to every point instanced volume.
Renders as fast as rendering one volume (and I now have a hundred voumes in the scene, all offset and uniquely scaled). Works like a dream.
The main issue is with the fact i cannot make the load delay procedural shader ‘procedural’.
If you need more clarification i can email you a hip.
Thanks.
Technical Discussion » delayed load geometry offset sequences
- oblongowl
- 40 posts
- Offline
Thanks for that , however I need this to be an ‘at render’ delayed load, for speed and memory issues and this method doesnt seem to work well for this (im loading animated bgeo volumes)
The ‘Delayed load procedural’ seems to be the way to go
I have a solution - using the delay load procedural, - i can use the instancepoints() expression to return me th point number of an instanced point, so i am using this to lookup one of many many instance geometries that are all the same - they are named eg; instance_0 to instance_99 and they each have the same load procedural with the bgeo offset to its respective opdigits number. This renders super fast.
Its just really lame - having to create all the objects - no procedural way i can see to do this , especially no surface level control.
The ‘Delayed load procedural’ seems to be the way to go
I have a solution - using the delay load procedural, - i can use the instancepoints() expression to return me th point number of an instanced point, so i am using this to lookup one of many many instance geometries that are all the same - they are named eg; instance_0 to instance_99 and they each have the same load procedural with the bgeo offset to its respective opdigits number. This renders super fast.
Its just really lame - having to create all the objects - no procedural way i can see to do this , especially no surface level control.
Edited by - May 3, 2013 01:47:54
Technical Discussion » delayed load geometry offset sequences
- oblongowl
- 40 posts
- Offline
Hi - Im sure this must have been covered somewhere, as its a pretty standard requirement that I cant find a solution to:
Want to do a delayed load sequence of geometry (an animation) and have the animation offset (and scaled) at every point instance.
I could provide a file but the geo is too heavy.
The CVEX procedural shaders is very confusing and none of the examples have animating geo, and Im not sure if this is the way to proceed .
Want to do a delayed load sequence of geometry (an animation) and have the animation offset (and scaled) at every point instance.
I could provide a file but the geo is too heavy.
The CVEX procedural shaders is very confusing and none of the examples have animating geo, and Im not sure if this is the way to proceed .
Technical Discussion » Glue an active object to a static or non active object
- oblongowl
- 40 posts
- Offline
Im trying to glue active objects to a presimmed object, that is being brought into Dops.
The presimmed object obviously is a sequence and I have tried using RBD fractured object set to non active - deforming as this sim can not be affected by this post dop sim.
The active object that is glued, does not move with the deforming object.
It acts as though it is glued in space, but does not move.
The only other way i can think is to use pin constraints, but i would prefer glue or some other way that is ‘procedural’ since the number of objects i will glue will vary from scene to scene.
The presimmed object obviously is a sequence and I have tried using RBD fractured object set to non active - deforming as this sim can not be affected by this post dop sim.
The active object that is glued, does not move with the deforming object.
It acts as though it is glued in space, but does not move.
The only other way i can think is to use pin constraints, but i would prefer glue or some other way that is ‘procedural’ since the number of objects i will glue will vary from scene to scene.
Technical Discussion » Foreach sop again
- oblongowl
- 40 posts
- Offline
Technical Discussion » Bullet solver breaking glue update
- oblongowl
- 40 posts
- Offline
Very cool. I may use this at some point in my current project.
I wanted to ask you, if you have access to the bullet solver code, do you know a way to stop the bullet solver from its inclination towards massive repulsion on fractured groups that are initially touching (or close to)?
The more polygons you stack against each other at the start frame, the bigger the initial repulsion and there is probably a way to soften this down over the first few frames of a bullet sim if you have access to the code I would imagine.
I wanted to ask you, if you have access to the bullet solver code, do you know a way to stop the bullet solver from its inclination towards massive repulsion on fractured groups that are initially touching (or close to)?
The more polygons you stack against each other at the start frame, the bigger the initial repulsion and there is probably a way to soften this down over the first few frames of a bullet sim if you have access to the code I would imagine.
-
- Quick Links