hi all..
I notice there is a pcopenlod() new to 11. Does this allow a particle cloud to be accessed in tiled fashion so that it does not have to be completely loaded into RAM?
thanks!
Found 85 posts.
Search results Show results as topic list.
Houdini Indie and Apprentice » pcloud load-on-demand
- david_aiken
- 86 posts
- Offline
Houdini Indie and Apprentice » RSL to VEX questions
- david_aiken
- 86 posts
- Offline
Houdini Indie and Apprentice » RSL to VEX questions
- david_aiken
- 86 posts
- Offline
hi all..
I've got some RSL that computes the aspect ratio, but i'm not sure how to port it to VEX:
uniform point Pcorner0 = transform (“NDC”, “screen”, point(0,0,0));
uniform point Pcorner1 = transform (“NDC”, “screen”, point(1,1,0));
float aratio = (xcomp(Pcorner1)-xcomp(Pcorner0)) /
(ycomp(Pcorner1)-ycomp(Pcorner0));
The docs at http://www.sidefx.com/docs/houdini9.5/news/9/vex [sidefx.com] don't mention space:screen yet, strangely enough, when ported to VEX the shader compiles cleanly and generates the following at runtime:
Warning: can't find object, light or fog named ‘space:screen’ to compute transform
and produces a result of 1.33 for 640x480 resolution. The 3delight renderer produces -1.33. Can i simply ignore the warning or perhaps do this cleanly? I notice there is an iaspect(), but it's for the COPs context.
thanks!
I've got some RSL that computes the aspect ratio, but i'm not sure how to port it to VEX:
uniform point Pcorner0 = transform (“NDC”, “screen”, point(0,0,0));
uniform point Pcorner1 = transform (“NDC”, “screen”, point(1,1,0));
float aratio = (xcomp(Pcorner1)-xcomp(Pcorner0)) /
(ycomp(Pcorner1)-ycomp(Pcorner0));
The docs at http://www.sidefx.com/docs/houdini9.5/news/9/vex [sidefx.com] don't mention space:screen yet, strangely enough, when ported to VEX the shader compiles cleanly and generates the following at runtime:
Warning: can't find object, light or fog named ‘space:screen’ to compute transform
and produces a result of 1.33 for 640x480 resolution. The 3delight renderer produces -1.33. Can i simply ignore the warning or perhaps do this cleanly? I notice there is an iaspect(), but it's for the COPs context.
thanks!
Houdini Indie and Apprentice » Improving render performance
- david_aiken
- 86 posts
- Offline
Thanks to both of you for your help. I went with an i7 930, 12Gb on a GA X58A UD3R board and the performance is significantly better than the Q6600 for both interactive and rendering work. It's currently running at 3.3Ghz but should overclock to 4 if necessary. I looked at the 875K, but the price was roughly equivalent, the turbo boost advantage is apparently not great at higher clock speeds, and upgradeability was a factor.
Houdini Indie and Apprentice » Improving render performance
- david_aiken
- 86 posts
- Offline
hi all..
Currently i'm using a Q6699 at 2.6 Ghz, 4Gb RAM and i would like to speed up the renders.
One possibility is to use Amazon, but unfortunately the highest network bandwidth available in my location (WIMAX) isn't sufficient.
Another option is to improve the CPU. The core i7 920 is highly overclockable, and Cinebench results (C4d renderer) show it can achieve markedly better performance than the Q6600, but based on some earlier posts i'm not sure whether Mantra will make use of the hyperthreads in Houdini 11. The core i5 750 is slower and cheaper since it lacks, among other things, hyperthreading - maybe it is the better choice. The Q6600 may go as high as 4 Ghz on a better motherboard, but the cost is comparable to upgrading to the newer processors with DDR3. The Nehalem Xeons are quite pricey still IMHO. The Apprentice Mantra allows multithreading, but not multiprocessing so this may also influence the decision.
Another option is increasing RAM. I'm not sure this will necessarily increase speed for typical renders but it will allow for larger particle simulations.
This all assumes that Houdini 11 Apprentice will not allow hyperthreading, interfacing with external renderers, or built-in GPU-based rendering.. any of which would be nice.
thanks.. any ideas would be appreciated!
Currently i'm using a Q6699 at 2.6 Ghz, 4Gb RAM and i would like to speed up the renders.
One possibility is to use Amazon, but unfortunately the highest network bandwidth available in my location (WIMAX) isn't sufficient.
Another option is to improve the CPU. The core i7 920 is highly overclockable, and Cinebench results (C4d renderer) show it can achieve markedly better performance than the Q6600, but based on some earlier posts i'm not sure whether Mantra will make use of the hyperthreads in Houdini 11. The core i5 750 is slower and cheaper since it lacks, among other things, hyperthreading - maybe it is the better choice. The Q6600 may go as high as 4 Ghz on a better motherboard, but the cost is comparable to upgrading to the newer processors with DDR3. The Nehalem Xeons are quite pricey still IMHO. The Apprentice Mantra allows multithreading, but not multiprocessing so this may also influence the decision.
Another option is increasing RAM. I'm not sure this will necessarily increase speed for typical renders but it will allow for larger particle simulations.
This all assumes that Houdini 11 Apprentice will not allow hyperthreading, interfacing with external renderers, or built-in GPU-based rendering.. any of which would be nice.
thanks.. any ideas would be appreciated!
Houdini Indie and Apprentice » custom VEX multithreading
- david_aiken
- 86 posts
- Offline
Thanks - i wasn't aware of the batching. A 10x100 grid shows the proper behaviour.
I'm interested in threading because i want to transfer the geometry data to a GPU. Is there a good way to do this?
I'm interested in threading because i want to transfer the geometry data to a GPU. Is there a good way to do this?
Houdini Indie and Apprentice » custom VEX multithreading
- david_aiken
- 86 posts
- Offline
I tried using the standard VEX_Ops.C example using hcustom and had the same behaviour. Perhaps due to some misunderstanding on my part, but i've logged a bug. If anyone has this working please let me know.
Houdini Indie and Apprentice » custom VEX multithreading
- david_aiken
- 86 posts
- Offline
hmm.. perhaps i'm doing something wrong. I print the point number from within the VEX SOP c++ code and even when the number of threads is set to 4 or 1 per processor the point numbers are printed out in order each time. I'm running on ubuntu 9.10 64-bit, Houdini Apprentice 10.0.595.
btw - thanks for clearing that up Mark.. i posted at the same time
btw - thanks for clearing that up Mark.. i posted at the same time
Houdini Indie and Apprentice » custom VEX multithreading
- david_aiken
- 86 posts
- Offline
Got it - thanks .
I was expecting a multi-threaded/single-threaded toggle in the HDK since this SOP should always run multi-threaded and benefits from as many processors as are available (it doesn't place a heavy load on them.. it just runs better). Is there a way to do this rather than setting a specific number of processors for each instance of this type on each box?
As a kludge i was thinking of changing the default number of processors in the type properties to 99, but it doesn't let me edit the channel settings. I'm using the “File:New Operator Type..” method to create the SOP. Maybe it wouldn't work anyway..
Out of idle curiosity, what happens if i create a hip on a box with 2 processors and throw it on boxes in the Amazon farm with 4 processors or 1 processor? In the former case i'm hoping that Houdini creates 2 processors .
I was expecting a multi-threaded/single-threaded toggle in the HDK since this SOP should always run multi-threaded and benefits from as many processors as are available (it doesn't place a heavy load on them.. it just runs better). Is there a way to do this rather than setting a specific number of processors for each instance of this type on each box?
As a kludge i was thinking of changing the default number of processors in the type properties to 99, but it doesn't let me edit the channel settings. I'm using the “File:New Operator Type..” method to create the SOP. Maybe it wouldn't work anyway..
Out of idle curiosity, what happens if i create a hip on a box with 2 processors and throw it on boxes in the Amazon farm with 4 processors or 1 processor? In the former case i'm hoping that Houdini creates 2 processors .
Houdini Indie and Apprentice » custom VEX multithreading
- david_aiken
- 86 posts
- Offline
hi all..
I'm pretty sure this is a FAQ, but my custom VEX SOP is called on the same thread for each point. Is there a way to inform Houdini that the SOP can be called on multiple threads?
thanks!
I'm pretty sure this is a FAQ, but my custom VEX SOP is called on the same thread for each point. Is there a way to inform Houdini that the SOP can be called on multiple threads?
thanks!
Houdini Indie and Apprentice » VEX_Ops example
- david_aiken
- 86 posts
- Offline
Fixed thanks to mrice. There is a need for a VEXdso file as documented in $HFS/houdini/vex/dso/VEX_VexOp.html. This information doesn't seem to be indexed by google.
It would be nice if the VEX_Ops sample included a VEXdso file or a readme.. but really this method is cumbersome. Houdini is probing for all of the hook methods in each .so on startup (most of which fail and are marked “(fatal)” in the LD_DEBUG log). It might be easier/quicker/simpler if there was a standard capabilities method exported by each .so which would allow Houdini to determine which methods are required, probe only for those methods, and report any problems.
It would be nice if the VEX_Ops sample included a VEXdso file or a readme.. but really this method is cumbersome. Houdini is probing for all of the hook methods in each .so on startup (most of which fail and are marked “(fatal)” in the LD_DEBUG log). It might be easier/quicker/simpler if there was a standard capabilities method exported by each .so which would allow Houdini to determine which methods are required, probe only for those methods, and report any problems.
Houdini Indie and Apprentice » VEX_Ops example
- david_aiken
- 86 posts
- Offline
Houdini Indie and Apprentice » VEX_Ops example
- david_aiken
- 86 posts
- Offline
hi all..
I've compiled the VEX_Ops.C example code using hcustom and it happily places the .so into $HFS/dso/VEX_Ops.so. On starting Houdini 10.0.595 it complained about missing version information in the .so so i added an include of UT_DSOVersion.h and recompiled. On restarting Houdini there is no complaint.. but i'm not sure how to invoke the getpid() VEX function.
The “vcc -X surface” doesn't list this function and if i create a new VEX surface shader operator and modify the code to call getpid() it complains “Undefined instruction ‘getpid’”. To see whether the newVEXOp() was getting called i put some debug code at the start of the function, but it doesn't seem to be called. Is there something i'm doing wrong?
thanks!
I've compiled the VEX_Ops.C example code using hcustom and it happily places the .so into $HFS/dso/VEX_Ops.so. On starting Houdini 10.0.595 it complained about missing version information in the .so so i added an include of UT_DSOVersion.h and recompiled. On restarting Houdini there is no complaint.. but i'm not sure how to invoke the getpid() VEX function.
The “vcc -X surface” doesn't list this function and if i create a new VEX surface shader operator and modify the code to call getpid() it complains “Undefined instruction ‘getpid’”. To see whether the newVEXOp() was getting called i put some debug code at the start of the function, but it doesn't seem to be called. Is there something i'm doing wrong?
thanks!
Houdini Indie and Apprentice » what a script create dialog 'Create clip dialog box'?
- david_aiken
- 86 posts
- Offline
Are you asking where the “.ui” file is for the “Create Clip…” dialog? If so, a grep turns up PSI2_CreateClipDlg.ui, but it is in a binary format.
Вы спрашиваете, где “.ui” файл на “Создание клипа …” диалог? Если это так, Grep поворачивает вверх PSI2_CreateClipDlg.ui, но это в двоичном формате.
Google Translate является вашим другом
Вы спрашиваете, где “.ui” файл на “Создание клипа …” диалог? Если это так, Grep поворачивает вверх PSI2_CreateClipDlg.ui, но это в двоичном формате.
Google Translate является вашим другом
Houdini Indie and Apprentice » custom UI
- david_aiken
- 86 posts
- Offline
hmm.. there is another thread with your question as the topic so i'll reply there, though unfortunately i can't be too much help.
Houdini Indie and Apprentice » custom UI
- david_aiken
- 86 posts
- Offline
Thanks.. do you know if it is possible to create a gadget to display an updatable image, such as a histogram? I would actually like to be able to include such a gadget in an HDA parameter interface to provide a preview of the effects of parameter changes, or, alternatively, create a pane to display an Osipa-style animation view. Perhaps too ambitious :>).
Houdini Indie and Apprentice » custom UI
- david_aiken
- 86 posts
- Offline
hi all..
The “Python Tips and Tricks and Upcoming Features in 10.5” presentation says it is possible to create custom Houdini native UI interfaces in 10.0 given sufficient bravery. By “native” i'm assuming this allows direct extensions to the Houdini UI rather than the wxWidgets/pyQT approach. If so, are there any details available?
thanks
The “Python Tips and Tricks and Upcoming Features in 10.5” presentation says it is possible to create custom Houdini native UI interfaces in 10.0 given sufficient bravery. By “native” i'm assuming this allows direct extensions to the Houdini UI rather than the wxWidgets/pyQT approach. If so, are there any details available?
thanks
Work in Progress » Sci-fi hi detail space ship
- david_aiken
- 86 posts
- Offline
Maybe this is useful if you want a weathered look:
http://forums.cgsociety.org/showthread.php?t=373024 [forums.cgsociety.org]
http://mos.futurenet.com/pdf/3dworld/TDW88_t_cyborg.pdf [mos.futurenet.com]
http://www.thegnomonworkshop.com/store/product/427/ [thegnomonworkshop.com] might give you some ideas. 3D Coat has made inroads into Bodypaint 3D's market since the DVD was created.
I like Headus for UV creation, but there are lots of others. The best UV tool in the world won't help if your model isn't built from clean quads/tris, though.
hope it helps..
http://forums.cgsociety.org/showthread.php?t=373024 [forums.cgsociety.org]
http://mos.futurenet.com/pdf/3dworld/TDW88_t_cyborg.pdf [mos.futurenet.com]
http://www.thegnomonworkshop.com/store/product/427/ [thegnomonworkshop.com] might give you some ideas. 3D Coat has made inroads into Bodypaint 3D's market since the DVD was created.
I like Headus for UV creation, but there are lots of others. The best UV tool in the world won't help if your model isn't built from clean quads/tris, though.
hope it helps..
Houdini Indie and Apprentice » VEX vs CUDA
- david_aiken
- 86 posts
- Offline
There is at least one limitation. If you are using Amazon EC2 via HQueue you cannot currently use a GPU setup (or HDK nodes) without further work from Amazon. Neither of their standard racks currently support GPUs.
Houdini Indie and Apprentice » HQueue questions
- david_aiken
- 86 posts
- Offline
There's a bit more info in the HQueue cloud guide FAQ section pg. 14. Only two Amazon Machine Images (AMIs) are supported (32- and 64-bit) and neither supports a GPU. You cannot create your own custom AMI without working with Sidefx/Amazon to resolve the licensing and hardware/software setup. I don't believe that any of the Amazon AMIs currently support a GPU, though other cloud providers, like Media Temple, do.
Haven't tried this, but if you have custom HDK nodes you may be able to cache the results out (i.e. they are part of the geometry processing) and send the files as part of the render pre-flight.
I think you're right.. for a local HQueue setup HDK nodes using GPU are probably ok. As i understand it, fluid sims are only supported on a local farm out-of-the-box (pg.3 of the guide).
Haven't tried this, but if you have custom HDK nodes you may be able to cache the results out (i.e. they are part of the geometry processing) and send the files as part of the render pre-flight.
I think you're right.. for a local HQueue setup HDK nodes using GPU are probably ok. As i understand it, fluid sims are only supported on a local farm out-of-the-box (pg.3 of the guide).
-
- Quick Links