Thank you for all your responses.

@mtucker, yes you are right, it works in karma. my apologies. since your example worked on the viewport, when I did not see my viewport showing the result, I just assumed I did something wrong.

However, I am trying to get the setup running in REDSHIFT, and still having problem.

@Heilieif, thank you for the comment. I have downloaded the OSL, and have it setup. the setup seems to be working (I can at least see the default texture showing up from "RS string user date". however, its not reading the .png files I have setup as an attribute.

NOTE I had to replace .rat files to .png files to get them rendering in REDSHIFT.

also attached the HIP file with redshift setting.

Thank you for your response and the sample file! i had a look and learned few new things!

I apologize if my wording was bit confusing due to terms, still learning the USD terms.

I have also attached a sample HIP file.





so basically I have a situation where each SOP primitives (as in each mesh faces) need to have different texture applied to them.
so each SOP primitive will have s@mytexture = "random_texture.jpg". However, this is a single MESH.

with HOUDINI, I could use the and use material override to get material to read different texture file for different faces.

However, I was having difficulty achieving this with SOLARIS.

Further assistance would be greatly appreciated!!

Hi everyone! been testing out USD on a project.

in SOPs, I was able to have a

1. texture file name on as an attribute on PRIM.
2. and via using material override, I was able to "change" the texture file on the material.



I was trying to replicate this on SOLARIS, but was having a hard time.
In fact did some research on the forum, sounds like it might not be possible. I am running REDSHIFT.



After searching online forums, I tried the above, but not working. not sure if I have done it wrong, or if it is REDSHIFT issue?

thank you in advance for all your help!

Hi everyone, I am aware you can use instance SUBNETWORK.

however, when i try it with 18.5.596, it does not seem to work.

I have made it a and have changed the subnetwork filtering to "everything"

but as soon as i point the instance target to a subnet, nothing happens (the points won't even show up)

i have tried it with simpler scene, but it just don't seem to work.

is this a bug? or am i doing something wrong?

Thanks!

vusta
i modded the code to this cos i find it easier to understand

vector n = normalize ( chv("v") );
float tolerance = ch("tolerance");

int pts [ ] = neighbours ( 0, @ptnum );
foreach ( int pt; pts )
{
    if ( @ptnum < pt )
    {
        vector p = point ( 0, "P", pt );
        vector dir = normalize ( @P - p );
        if ( 1 -  abs(dot ( dir,n )) <= tolerance )
            setedgegroup ( 0, "edges", @ptnum, pt, 1 );
    }
}
}

coz both dir and n are normalised, their dot product will never ever be greater than 1...so 1 - dot is much easier to digest to me.
i'm just a hobbyist tho so don't take me too seriously.

Hi everyone! found this useful!

there is a missing bracket after " if ( 1 - abs(dot ( dir,n )) <= tolerance )"

thought someone might have trouble running the wrangle.

Thanks!

Hi everyone!

i have been using MULTI LAYER EXR for my output, but i read/tested that MULTI PART EXR has a bit more performance gain (and usually easier to use in post-composition).

however, when i am saving out MULTI PART EXR, the name of AOV's do not come across, but only as subimage01, subimage02, subimage03 etc etc.

its not a big deal, but was wondering if there was any way of setting up the naming properly, so they actually reflect the AOVs, or at least manually set the up?

Thanks in advance for any hints!

Hi Mark! thank you for your suggestion.

my memory may be fuzzy (it's about a year ago) but I started the whole HOUDINI journey with redshift, thus my experience of faster loading of HOUDINI was with redshift already installed and licensed properly.

what i find interesting is that it seems to get slower and slower.

again, i have not timed the launch (maybe i should), but it feels noticeably slower than about a month ago.

now i would run Houdini and since it takes a while, will go to youtube for a quick update, and usually gets surprised that it's still not loaded (again, very arbitrary measurement of time).

i do have a few PYTHON scripted tools built as i work on production, but not many (and not too complex) and i only load HDA per project as needed (so i don't have a folder of HDA always loading)

is there any way of seeing a log of Houdini launching for troubleshooting?

Thanks!

Hi everyone, been using Houdini for about a year now.

when I first installed Houdini, was very impressed at its launch time (not file loading time, but launching Houdini).

oddly, after about a year, it's taking close to 30seconds+ to just launch the software and feels like it's taking longer and longer to launch.

again, this is not about opening a file, but just launching the app.

i come from 3ds Max, and usually, with more plugins installed, it takes longer to launch.

however, with Houdini, the reason why I love it is I can just build my own tools rather than relying heavily on 3rd party.

so I do not have anything else installed (other than redshift, which I installed with Houdini about a year ago).

perhaps its some kind of cache file or garbage file that Houdini needs to sort out when launching, which could explain it taking longer and longer more i use it.

any idea? suggestions?

thanks in advance!

Thank you for your response jsmack!

yep that's the sense i got, USD used as a final "assembly" pipeline tool.

but due to USD being "universal", feels like there is quite a bit of workaround of USD specific approach that's required to get it working with normal (at least how i use HOUDINI) workflow.

i love the concept of USD, and definitely can see a huge benefit for a larger team, but right now, again as a SOLO artist, feels compromises are greater than the benefit.

that being said, i have been trying to immigrate my existing project into USD, so i can get a real-world project experience with it.

but think USD is something i need to commit from the start and set up the workflow around it, rather than using it as an extra tool here and there.

figured out a solution, kind of brute math solution, would be awesome to know if there are other ways to approach this.

for my current solution, basically get the resolution of the bigger image, subtract the resolution of the smaller image, divide the result by 2, you get the offset you need from LEFT and BOTTOM corner

attached the image for solution.

again, i am finding HOUDINI's way of overlaying two different resolution image odd, not something i am used to coming from FUSION.

if there is better way of approaching this, i would greatly appreciate it!

thanks!

Hi everyone!

i have a base image of 1000x1000pixel image. and i am compositing 500x500px image "over" it using "over" composite node.

i come from FUSION, and when i do this, usually 500x500px image will sit on top of 1000x1000px image centered.

however with HOUDINI COP, the image is aligned to BOTTOM LEFT corner. i am assuming its using UV to place the LEFT BOTTOM CORNER to UV 0,0?

however this makes it quite difficult to center the TOP image to the middle of background image.

i could do some math to calculate the offset, but thought there might be easier way of doing this.

i have attached the current HOUDINI behaviour.

any help will be apprecaited!

Hi everyone, learning SOLARIS here.

i do a lot of automobile-related visualization work.

before testing SOLARIS, i would usually have a GEO node with most of my mesh prepared.
so i would have the wheels, chassis, engine prepared here.

and when i need to rig them or animated them, i would create a new SUBNET on /OBJ level, and inside it have a NULL and use them as a rigging structure. so i would have main NULL, which connects to BODY, WHEELS ENGINE etc, and the NULLs are animated. (this way i can separate the animation from the mesh date, so eash to manage projects)

but basically any kind of simple rig with combination or parenting, some expression, and some keyframes.

however, i am trying to understand what the proper workflow is for USD to do something like this.

it feels like SOLARIS/USD workflow is to prepare MESH assets in /OBJ level, but do a placement, materials on /STATE USD context?

if placement happens on USD/SOLARIS context, shouldn't animation happen here as well? but feels like USD is not built as an animation/rigging tool, but rather a "staging tool"

if this is the case, should i be doing all the assembly, placement, animation in /OBJ level and bring in the whole animated mesh into SOLARIS/USD?


thank you in advance!

Hi mtucker for the information.


your explanation on "render in single process" makes good sense. i am using Deadline as my manager, so ideal workflow would be to chunk out a sequence of animation (0-100, 101-200, 201-300...) and get those rendered on each GPU with "render in single process"

i have tested this out with Deadline GPU affinity, and so far seems to work. however, i am still on my test scene, so the animations are very basic, some keyframe, some expression.

but by doing this, i am getting the expected performance, except the very first frame, where i am still seeing about 20-30 second overhead.

-- viewport render

that's the weird thing, on viewport both redshift and karma renders instantaneously, and any USD changes are reflected realtime.

-- karma

i have tested animation sequence with karma, same simple scene, but i am still getting same overhead per frame (not just the first frame, but for every frame.

all my test scene is stored in my main project NVME SSD, and i am testing locally, so don't think harddisk space / network is an issue here.

is there any way to debug the process?

i am seeing verbosity level but i am not seeing any log output displayed, assuming its being saved to a log somewhere?

-- probably not deadline issue

as this occurs even if i just click "render to disk" from houdini

--- summary

from my testing, i get the feeling its launching of HUSK? i am still learning solaris, but how is HOUDINI rendering its viewport? is it running another copy of HUSK? i don't see it in the process manager.

from my understanding it seems difference between rendering on viewport and rendering to image is in invoking HUSK to do the rendering.


RENDER INITIATE - HUSK LAUNCH - HUSK USD READ - CONVERSION TO REDSHIFT - RENDER REDSHIFT - FRAME WRITE - HUSK CLOSE? - HUSK RE-LAUNCH? - HUSK USD RE-read?

is above the kind of process HOUDINI takes to render image sequence? does it actually close and re-launch HUSK per frame?
or does HUSK runs once, and it keeps on re-preparing USD?

just trying to pinpoint where the overhead is occuring is all.

Thanks!

Hi Mark, thank you for your reponse. if you have the chance, would be awesome if you could compare render time between just regular houdini render vs image sequence render.


also wanted to leave an updated note here. i have managed to fix the issue with toggling on "RENDER ALL FRAMES USING SINGLE PROCESS"

the problem was i had the output setup as $HIP/render/test/test.$F4.exr, as per normal houdini output setup.
however, someone on the redshift forum kindly advised me to replace $F4 to <F4>, which is USD way of doing things i am guessing.

with that change in the output path, i have managed to render out the image sequence, and it actually improved the render speed significantly (as in no overhead).

there is still the 30 second overhead at the start of the rendering sequence, but once the frames start rendering, consecutive frames were rendering at their normal speed.

i have animated geometry + camera and i both animation seems to render out just fine, so not exactly sure the CONS on toggling "RENDER ALL FRAMES USING SINGLE PROCESS".

keen to know the downside of enabling "RENDER ALL FRAMES USING SINGLE PROCESS" toggle, if anybody has any idea, would love to hear them.

Thanks!

Thank you for response!

thing is i am just testing SOLARIS out, so my test scene literally has a SPHERE and a box. no instancing.
so with very basic USD stage, still has about 20-30second overhead.

i have also downloaded a scene from VERAMIX's "SOLARIS WITH 3RD PARTY RENDERER" tutorial, which is more "complicated" scene.
this scene also seems to have about 20-30 second overhead, so don't think its the USD complexity.

my guess is the time is mostly due to HUSK initiation.

what i don't understand why it has this overhead every frame? do SOLARIS shut down HUSK at end of render frame, and re-launch?

again i am very new to SOLARIS, so i am hoping i am missing something.
as i really love the concept of USD, and can see huge benefits, but for the work i do, probably not worth the overhead.

@Mark, when you render image sequence from ROP, does your render at "expected speed"?

thank you in advance!

Hi everyone, just wanted to update the post with more findings.

i have managed to setup the scene with KARMA as well, and ran some animation render test.

i have setup the render so the actual render it self takes about 10 seconds per frame.

however, when i render the same scene via KARMA_ROP to render out the image sequence, similar "time overhead" occurs.

so each frame will take about 40 second on average to render out. so similar to REDSHIFT test, there seems to be about 20-30 second overhead per frame.

i am assuming this is the USD translation or interpretation phase?

my question is as per previous post

1. is this an expected behaviour? anything i can change to speed things up?

2. i don't mind the overhead on HUSK launch and USD interpretation, but the overhead on every frame really adds up, is there anything i can do to save this time inbetween frames in animation?

Thank you all in advance!

Hi everyone! i am using HOUDINI 18.0 + RESHIFT 3+ as the main setup.

been testing SOLARIS/LOP/USD with REDSHIFT.

i have a very simple scene (with some spheres) setup in SOLARIS, managed to set up everything and seeing the result rendered in REDSHIFT in the viewport at expected speed (fast).

however, once i setup a ROP to render the images out, rendering is taking about 30seconds per frame on my RTX3090, where i would guess it should take about few seconds per frame to render.

i am guessing this is an overhead from "translating/preparing" USD to render in REDSHIFT?

however, the beauty of REDSHIFT is its very fast render times with optimized settings, and having 20+ seconds overhead per frame is quite a lot.

also, i would assume once the USD is "translated" the subsequent frames should render faster, but my test shows that each frame is taking about 30 seconds per frame.

was wondering if it is something i have done wrong?

is there anything i can do to speed up the process?

i have tried "render all frames with a single process", but getting below error. will this help render animation faster? also, is there something i am missing?
<b style='color: red;'> Error: Output file 'C:/Users/alter/Desktop/test/test.0001.exr' should have variables</b>

an argument could be that USD is really for complex scenes, where 20 second overhead could mean little, but my USD setup is very simple (few spheres, single light).

thank you in advance for any help!

hey guys! found an answer to my questions and thought i share it here for anybody else.

again, i am very new to HOUDINI, so this might not be the best method, so if anybody has a better approach, please let me know


basically i thought the best approach for FOR EACH would be to set as many attributes to the POINT.
this way, i have all the driving attributes on a consolidated place, and any copied geometries are transformed/modified based on those attributes.


to add CHANNEL data (animated) to each points, i could use

CHANNEL

node, where i was able to map different channel to an attribute in a POINT. and bind that attribute to a transform in incoming geometry.

thanks!

Hi everyone, new to houdini, long time user of 3ds max

i have

1. CHOP setup with chan1 ~ chan99 (100 channels setup with random noises) | done
2. forEach system copying a torus to 10 points | done

i am trying to setup a system where each copied torus is assigned to a random channel from CHOP.

i have managed to add a node after the torus and added expression to .

chop(“../chopchop/_out/inChan1”)

but i can't seem to be able to assign random channels for each copy.

attached current setup

i have tried——

i thought i could build out some kind of random number driven access to CHOP. for example,

chop(“../chopchop/_out/inChan(fit01(rand(@seed),0,99)”)

any help would be appreciated, even a whole new method or approach