hey Erik,
heres the updated shader, ive included an example vector displacement map also. Im not quite sure what the scale settings should be have a play around with your own displacement tests and see if you can match them 100%
Cheers
Nick
Found 90 posts.
Search results Show results as topic list.
Houdini Indie and Apprentice » Vector Displacement??
- niko3d
- 338 posts
- Offline
Houdini Indie and Apprentice » Vector Displacement??
- niko3d
- 338 posts
- Offline
Sure no problem. I will clean up the file and post it on here at some point today.
Cheers
Nick
Cheers
Nick
Houdini Indie and Apprentice » Vector Displacement??
- niko3d
- 338 posts
- Offline
Afew more tweaks and I will post the shader update, im currently trying to figure out the best approach when using exr maps that are black = 0 and maps that use 50% grey as 0.
Cant quite get the exact values i need from it, its close but not quite right. From what i read in the 3dcoat forum, the 50% grey = 0 image maps are normalised and will make it trickier to get the displacement amount. these need the usual offsets as well.
but the black = 0 maps aren't normalised so should just need scaled down …. but doesnt seem the case with my modo test displacement files. These exr files support negative values and dont need offset like the 50% grey image maps.
Almost there.
Nick
Cant quite get the exact values i need from it, its close but not quite right. From what i read in the 3dcoat forum, the 50% grey = 0 image maps are normalised and will make it trickier to get the displacement amount. these need the usual offsets as well.
but the black = 0 maps aren't normalised so should just need scaled down …. but doesnt seem the case with my modo test displacement files. These exr files support negative values and dont need offset like the 50% grey image maps.
Almost there.
Nick
Houdini Indie and Apprentice » Vector Displacement??
- niko3d
- 338 posts
- Offline
Hey guys
Looks like the negate of the normals sorts the normal direction problem I will do a proper vector displacement test and make sure the displacements are correctly offset and theres nothing odd is happening due to this, but my up down texture test works so far.
Cheers
Nick
Looks like the negate of the normals sorts the normal direction problem I will do a proper vector displacement test and make sure the displacements are correctly offset and theres nothing odd is happening due to this, but my up down texture test works so far.
Cheers
Nick
Houdini Indie and Apprentice » Vector Displacement??
- niko3d
- 338 posts
- Offline
I have been suggested by a colleague to try and negate the normal direction if the blue channel is a negative value. If this works as we expect I will post a new shader here. Still unsure why we would need todo this.
Cheers
Nick
Cheers
Nick
Houdini Indie and Apprentice » Vector Displacement??
- niko3d
- 338 posts
- Offline
Houdini Indie and Apprentice » Vector Displacement??
- niko3d
- 338 posts
- Offline
The problem isnt quite todo with remapping/fitting our image to between -1 and 1 etc.
For example
If I use just the blue channel to displace with the displacement direction is correct i.e it works positively and negatively , but as soon as we use the normal direction defined by the the rgb values in tangent space the displacement direction is always positive.
The results of the displacement look correct for the “positive” direction which is why the examples shown work, the only problem is the inverse direction of the down displacement creating more positive displacement.
This looks more like a normal direction issue rather then a simple texture remap that defines the amount of displacement.
Just to be sure I tested this with that same up down set up as in the “simple displacement material” and there's no change in the direction.
Any ideas why the tangent space normal direction would be doing this? Im still trying to figure this one out. If I get any further I will lets you know.
Cheers
Nick
For example
If I use just the blue channel to displace with the displacement direction is correct i.e it works positively and negatively , but as soon as we use the normal direction defined by the the rgb values in tangent space the displacement direction is always positive.
The results of the displacement look correct for the “positive” direction which is why the examples shown work, the only problem is the inverse direction of the down displacement creating more positive displacement.
This looks more like a normal direction issue rather then a simple texture remap that defines the amount of displacement.
Just to be sure I tested this with that same up down set up as in the “simple displacement material” and there's no change in the direction.
Any ideas why the tangent space normal direction would be doing this? Im still trying to figure this one out. If I get any further I will lets you know.
Cheers
Nick
Houdini Indie and Apprentice » Vector Displacement??
- niko3d
- 338 posts
- Offline
Hey guys
Has anyone noticed that the shader setup doesnt do negative displacement correctly, it always displaces in a positive direction? Im trying to solve this issue just now but i cant see where its going wrong.
Related to this very useful shader, Im also working on a vector displacement baker from within houdini. Theres no tracing involved so its pretty quick and should be accurate to.
Once we have a fully working shader and I get my baking tool finished up, I think it be a great addition to add this to the houdini exchange.
Cheers
Nick
Has anyone noticed that the shader setup doesnt do negative displacement correctly, it always displaces in a positive direction? Im trying to solve this issue just now but i cant see where its going wrong.
Related to this very useful shader, Im also working on a vector displacement baker from within houdini. Theres no tracing involved so its pretty quick and should be accurate to.
Once we have a fully working shader and I get my baking tool finished up, I think it be a great addition to add this to the houdini exchange.
Cheers
Nick
Technical Discussion » Delayed Load IFD Bounds
- niko3d
- 338 posts
- Offline
Hi Georg, mark.
Unrelated to any instancing we actually had occasions in the past that caused delayed load bounds not to cook correctly or not at all when generating ifds and rendering these. This seemed to be fixed with a build update. Unfortunately it looks to have come back again. Neither the sop based bounds nor the explicit bounds seem stable with DLAs when generating IFDs with hbatch, this could be due to a cook problem in expressions etc but the errors do not occur when in Houdini itself only when generated by hbatch and rendered through mantra on the renderfarm . I will be submitting a bug to support on this tomorrow when I get back into work.
A related bug to instances and delayed loads I have found recently is that when instances do not compute procedurals correctly if there are any expressions in geometry procedural parm. For an example an if statement in that parm would not work and the sop bounding box would render instead of the geometry file. When the if statement was removed it rendered correctly. I will be submitting this bug to sidefx support along side the other.
Thanks
Nick
Unrelated to any instancing we actually had occasions in the past that caused delayed load bounds not to cook correctly or not at all when generating ifds and rendering these. This seemed to be fixed with a build update. Unfortunately it looks to have come back again. Neither the sop based bounds nor the explicit bounds seem stable with DLAs when generating IFDs with hbatch, this could be due to a cook problem in expressions etc but the errors do not occur when in Houdini itself only when generated by hbatch and rendered through mantra on the renderfarm . I will be submitting a bug to support on this tomorrow when I get back into work.
A related bug to instances and delayed loads I have found recently is that when instances do not compute procedurals correctly if there are any expressions in geometry procedural parm. For an example an if statement in that parm would not work and the sop bounding box would render instead of the geometry file. When the if statement was removed it rendered correctly. I will be submitting this bug to sidefx support along side the other.
Thanks
Nick
Technical Discussion » hda trigger a python script with a button..
- niko3d
- 338 posts
- Offline
Rdg is basicaly there with what hes saying. Though If memory serves me correctly, i also asked this awhile back. I was told easiest way to grab the button name directly was to use
kwargs
so if the this was your defintion
def printButtonName(buttonName):
print buttonName
in your python call back place
testMyButton.printButtonName( kwargs)
Worked for me when using multiparms and generating folders based on a multiparm button and multiparm file path parameter
Hope this helps
Nick
kwargs
so if the this was your defintion
def printButtonName(buttonName):
print buttonName
in your python call back place
testMyButton.printButtonName( kwargs)
Worked for me when using multiparms and generating folders based on a multiparm button and multiparm file path parameter
Hope this helps
Nick
Technical Discussion » Mantra Vray Transperency
- niko3d
- 338 posts
- Offline
Hey guys
I noticed you have been mentioning a vray procedural, can i ask exactly what does the vray procedural do?
Below is an email from out render manager and it seems that the libvray.dll is failing when we try to render specfic frames. We arent sure whats causing the problem… Any help identifing the issue would be great.
Thanks
Nick
I noticed you have been mentioning a vray procedural, can i ask exactly what does the vray procedural do?
Below is an email from out render manager and it seems that the libvray.dll is failing when we try to render specfic frames. We arent sure whats causing the problem… Any help identifing the issue would be great.
We were recently trying to render a character pass which quite a number of frames. On our test sequence we're experiencing some crashes that occur on the same frame each time. So it's a reproducible error. We are using the win64 build 9.1.293 since we started the project using that build and to update at this point could alter some key assets etc.
In a bid to resolve these mantra crashes can you maybe shed some light on what is causing Mantra to crash giving this error report.
Faulting application mantra.exe, version 9.1.0.293, faulting module libvray.dll, version 9.1.0.293, fault address 0x00000000000b1e5b.
Thanks
Nick
Technical Discussion » python hscriptExpression not evaluating in python shell?
- niko3d
- 338 posts
- Offline
Hey guys,
Has anyone had problems with python hscript expressions not working? Ive tried this a with a few expressions that point to a cop and nothing works in the python shell. But strangely they do work in parms?
For example add a ramp into a cop net and run this in the python shell
hou.hscriptExpression('pic(“/img/comp1/ramp1”, .5, .5, D_CR)')
it returns 0.0
I tried this with the res expression also. Not sure if this is a bug with cops and python?
This works in the hscript shell and on any parm on node perfectly fine.
Any ideas?
windows 64 any version of houdini 9.1/9.5
Thanks
Nick
Has anyone had problems with python hscript expressions not working? Ive tried this a with a few expressions that point to a cop and nothing works in the python shell. But strangely they do work in parms?
For example add a ramp into a cop net and run this in the python shell
hou.hscriptExpression('pic(“/img/comp1/ramp1”, .5, .5, D_CR)')
it returns 0.0
I tried this with the res expression also. Not sure if this is a bug with cops and python?
This works in the hscript shell and on any parm on node perfectly fine.
Any ideas?
windows 64 any version of houdini 9.1/9.5
Thanks
Nick
Technical Discussion » The ever increasing hip files size :¬( BUG????
- niko3d
- 338 posts
- Offline
Glad this has been sorted now, have to say the new build saved our butts :¬) meant redoing a little work rather then fresh builds of scenes.
I hope to never see this again.
:wink:
Cheers
Nick
I hope to never see this again.
:wink:
Cheers
Nick
Technical Discussion » The ever increasing hip files size :¬( BUG????
- niko3d
- 338 posts
- Offline
Already have :¬)
Strangely one of our hip files that was bloated to 530mb compressed down to 600KB Sounds like a alot of empty lines that probbins mentioned.
Lets hope theres a nice solution to this.
Cheers
Nick
Strangely one of our hip files that was bloated to 530mb compressed down to 600KB Sounds like a alot of empty lines that probbins mentioned.
Lets hope theres a nice solution to this.
Cheers
Nick
Technical Discussion » The ever increasing hip files size :¬( BUG????
- niko3d
- 338 posts
- Offline
Ok i found the source of the problem and the reason why this problem is shared across the houdini scenes.
The corrupted data is from the userDisplayOptions.h file in the hip file. Its full of “junk” data. I can delete this and hcollapse the scenes hopefully this will work.
Does anyone know how to fix this from within houdini? And why this could happen?
Thanks
Nick
The corrupted data is from the userDisplayOptions.h file in the hip file. Its full of “junk” data. I can delete this and hcollapse the scenes hopefully this will work.
Does anyone know how to fix this from within houdini? And why this could happen?
Thanks
Nick
Technical Discussion » The ever increasing hip files size :¬( BUG????
- niko3d
- 338 posts
- Offline
Its ok, someone pointed me to my error :¬D I was passing a full path to my hip file when using hexpand. Just needed a simple cd to the files directory so im guessing i was still in my houdini filestructure which makes sense .
doh!!!
cd $filefolder
hexpand $filename
This does the job as expected, nothing complicated there. If i want todo this on windows do I still need cygwin installed to run hexpand?
Anyway if this allows me to fix my file then I will follow this up.
Thanks for the help guys.
nick
doh!!!
cd $filefolder
hexpand $filename
This does the job as expected, nothing complicated there. If i want todo this on windows do I still need cygwin installed to run hexpand?
Anyway if this allows me to fix my file then I will follow this up.
Thanks for the help guys.
nick
Technical Discussion » The ever increasing hip files size :¬( BUG????
- niko3d
- 338 posts
- Offline
Yeah there is some locked nodes in the scene but even i remove them it doesnt make a diffrence.
hexpand sounds like an idea that could work. Ive just tried to get hexpand to work on a linux box but no luck. Not sure how im supposed todo this? Anyone got some pointers?
Any help would be great
Thanks
Nick
hexpand sounds like an idea that could work. Ive just tried to get hexpand to work on a linux box but no luck. Not sure how im supposed todo this? Anyone got some pointers?
Any help would be great
Thanks
Nick
Technical Discussion » The ever increasing hip files size :¬( BUG????
- niko3d
- 338 posts
- Offline
Yeah we spotted some of these also. I dont know the source of the problem but i do know how it is spreading to other scenes. Seems like the encountered bad data/memory stays with houdini across new scenes/newly loaded scenes.
I.e If i load a “bad” scene into houdini, the corrupted data sticks around for the whole session and gets saved to what ever scene it encounters. Like a virus i suppose ;¬) then doubles on every save.
I got around it by restarting houdini and loading my good scene into a new the new session and its been saving fine since. I still have bad scenes that i cant fix tho, so will have to rebuild those from a good source.
Thanks
Nick
I.e If i load a “bad” scene into houdini, the corrupted data sticks around for the whole session and gets saved to what ever scene it encounters. Like a virus i suppose ;¬) then doubles on every save.
I got around it by restarting houdini and loading my good scene into a new the new session and its been saving fine since. I still have bad scenes that i cant fix tho, so will have to rebuild those from a good source.
Thanks
Nick
Technical Discussion » The ever increasing hip files size :¬( BUG????
- niko3d
- 338 posts
- Offline
Howdy people,
We seem to have hit a huge problems with a bunch of our scenes at work. I really cant figure out how to solve this or what is causing it.
I have some large production hip files that normaly sit at 100mb. I have been working with these files for a good bit of time, and had no problem until a few days ago where my hip files where causing our renderfarm hbatch to choke. When saving, my .hip file bloats to over 600 mb, then resaving that file cause the hip file to bloat to 1 gb.
If i merge into a new scene i get a 1.6gb hip file. Nothing seems to work at all to save these scenes from becoming corrupted. When we compare the decent sized file to the news one there looks to be huge amounts of binary data flooding it.
Has anyone ever had problem saving hip files? It must be due to an external file as it effects all the old files also. The wierd thing is even if i delete all contents for the scene it and save it, it still sits at 600mb???????
I tried calling support but I think its canada day.
Thanks
Nick
We seem to have hit a huge problems with a bunch of our scenes at work. I really cant figure out how to solve this or what is causing it.
I have some large production hip files that normaly sit at 100mb. I have been working with these files for a good bit of time, and had no problem until a few days ago where my hip files where causing our renderfarm hbatch to choke. When saving, my .hip file bloats to over 600 mb, then resaving that file cause the hip file to bloat to 1 gb.
If i merge into a new scene i get a 1.6gb hip file. Nothing seems to work at all to save these scenes from becoming corrupted. When we compare the decent sized file to the news one there looks to be huge amounts of binary data flooding it.
Has anyone ever had problem saving hip files? It must be due to an external file as it effects all the old files also. The wierd thing is even if i delete all contents for the scene it and save it, it still sits at 600mb???????
I tried calling support but I think its canada day.
Thanks
Nick
Technical Discussion » Image Planes and Channel Names
- niko3d
- 338 posts
- Offline
“It appears I need to add ”C“ in the channel name to get the VEX variable data into the rgb channel of the respective exr…. hmm obvious really.”
This Should be defaulted at rendertime if no name has been set and your outputs are all diffrent files :¬) Solves all this running about looking for solutions if it worked as people expected.
Nicely found John
This Should be defaulted at rendertime if no name has been set and your outputs are all diffrent files :¬) Solves all this running about looking for solutions if it worked as people expected.
Nicely found John
-
- Quick Links