Forgot your password?   Click here   •   No account yet?   Please Register    •   Or login using  
JA ログイン
SideFX Homepage
  • 製品
    • 新機能
      • 概要
      • アニメーション
      • リギング
      • CFX
      • VFX
      • ルックデブ
      • Copernicus
      • Terrain & Modeling
    • Houdini
      • 概要
      • VFX
      • ワールド構築
      • ルックデブ
      • キャラクタ
      • モデリング
      • パイプライン と AI
    • Houdini Engine
      • 概要
      • Engine プラグイン
      • バッチ処理
    • Karma Renderer
      • 概要
      • Karma 機能比較
    • 製品比較
    • SideFX Labs
    • Partners
  • 業界
    • Film & TV
    • ゲーム開発
    • モーショングラフィクス
    • Virtual Reality
    • AI/ML 向けデータ合成
  • コミュニティ
    • フォーラム
    • ニュース
      • 概要
      • カスタマ ストーリー
      • Houdini HIVE Events
      • Contests & Jams
    • Gallery
    • イベントカレンダー
    • User Groups
    • Artist Directory
    • Houdini Merch Store
  • 学習
    • Start Here
      • 概要
      • My Learning
      • ラーニングパス
      • チュートリアル
    • コンテンツライブラリ
    • Tech Demos
    • Houdini 講演
    • 教育プログラム
      • 概要
      • 学生
      • 講師
      • 管理者
      • List of Schools
      • 学習リソース
  • サポート
    • カスタマーサポート
    • ライセンス
      • 概要
      • 商用版製品
      • Indie
      • EDUCATION 製品
    • ヘルプデスク FAQ
    • Houdini システム環境
    • ドキュメント
    • Changelog / Journal
    • Report a Bug/RFE
  • Get
    • Try
    • 購入
    • ダウンロード
    • お問い合わせ
 
Advanced Search
Forums 検索
Found 14624 posts.

Search results Show results as topic list.

Houdini Lounge » Houdini AI Assistant — Analyze, Debug, Build HDAs

User Avatar
ciusradu
2 posts
オフライン
 2025年10月23日 14:23:26


Stop Fighting with Networks. Start Creating.
Analyze. Debug. Generate. Automate.




Houdini is powerful, but endless debugging and reverse-engineering complex networks can kill creativity and waste hours.
Houdini AI Assistant is a native tool that solves this problem.





It’s not just another chatbot.
It’s your personal Assistant that reads your scene, nodes, and parameters to give you answers that are actually relevant to your project.




My goal with this tool isn't to replace that creative part that's the fun! It's to automate the boring part, like debugging a simple VEX typo at 2 AM.

The idea is to free up more time for the interesting setup work that you're talking about.

And absolutely, it's perfect for learners too! They can select a network or node they don't understand, hit 'Analyze Selection', and then simply ask: "What does this network do?" or "Explain this node in the context of the selection." The assistant, using the 'Scene Analyst' role (and maybe even the 'Explanatory' style), will provide a detailed, step-by-step explanation.

More info: https://rart.gumroad.com/l/HoudiniAIAssistant [rart.gumroad.com]


Edited by ciusradu - 2025年10月23日 14:24:21
See full post 

Technical Discussion » Shadow catcher issue in Karma

User Avatar
JasonG
5 posts
オフライン
 2025年10月23日 09:58:42
Can someone solve this for me please?

When assigning an object like a grid to be a shadow catcher somehow the casted shadow from a smaller object and casted shadows from the wall behind it don't melt into one as they should but the smaller object's shadow is darker than the shadow from the wall.
But when the shadow catcher is disabled and the grid is rendered normally, those two casted shadows melt into one, no difference as it should be... Is it a bug in shadow catcher system or I'm doing something wrong? Probably it's the latter.

Thanks

Jason
See full post 

Houdini Lounge » Why Does My Color Ramp Look Wrong in ACES?

User Avatar
amaruki
5 posts
オフライン
 2025年10月23日 06:20:53
Hi everyone,

I’m working in ACES using the default OCIO config in Houdini, and noticed something odd. In COPs when I use a Mono To RGB, with one of the default color ramp presets, the colors look too saturated or blown out.

If I run it through an OCIO Transform node (from Linear Rec.709 sRGB → ACEScg), the colors look correct again (to my eyes).See screenshot.

I thought Houdini handled these color conversions automatically — am I missing something? What's the correct workflow/setup here?


Update: I can leave the OCIO Transform node "To Space" on scene_linear with the same result.
Edited by amaruki - 2025年10月23日 10:37:16
See full post 

Technical Discussion » ACES 1.1 HDR View Transform

User Avatar
Tronotrond
69 posts
オフライン
 2025年10月22日 21:54:15
Going to first of all admit, I'm no expert when it comes to color spaces.
I've been using the ACES SDR transform for a while and that's working pretty well throughout my/our pipeline.

But I'm wondering what's the steps to start working with/creating more dynamic HDR content.
I figured starting in Houdini and setting viewport display space to Rec.2100-PQ and View transform to ACES 1.1 HDR would be a good start, but the viewport/IPR render gets completely washed out and de-saturated compared to the SDR view transform. There's not much information about any of this out there, so just wondering if anyone got some pointers?

I did try download the ACES OCIO v2.4 config and use that, but it didn't seem to make any difference either.
Windows and the monitor is running with HDR enabled, and the monitor is set to its HDR PQ_BT.2020 preset.
See full post 

PDG/TOPs » ML neural style transfer learning

User Avatar
bogdana
36 posts
オフライン
 2025年10月22日 16:21:12
Hi everyone!

Idea of this thread is to share our failed ML neural style transfer attempts, with some insights in what work, what doesn't. Also to share some learning resources that would help us understand better how to successfully train our own ML models.

I successfully trained ML style transfer example with noise generator given from SideFX.
After that I thought to train it on our own terrain example. Here is first result of that, posting them in the order they were trained:







First two were trained with very big range of the lowest and highest point in the terrain, while 90% of terrain was plateau. Moment I isolated only that, thing started working better. I trained with different data set size, resolution, training rate...

This kind of visualizing onnx generations alognside with graph seems to me like a good way to make a sense of the process.
Did anyone try to use this? If so, what would be best learning resources for us VFX artists to take, so we are not left behind
Edited by bogdana - 2025年10月22日 16:37:48
See full post 

Houdini Engine for Unreal » VATs Rigid body error

User Avatar
AliciaRG
1 posts
オフライン
 2025年10月22日 10:56:42
Hey guys!
Lately I have been working with VATs to import them in Unreal.
I have a problem with the rigid body export. A geometry that is a merge of multiply geometries (blades of a mill) that I imported in Houdini trought Maya, is giving me some headaches haha.
Every time that I tried to export the VAT in Rigid-body mode, it gives me an error(the image).
I found out that the rigid-body mode doesn't work well with more than one geometry, but I don't know how to change my geometry to made it undertandable for the rigid mode.
I have used a connectivity node to separate the objects but it didn't work.
I also tried to export it as a soft-body but Unreal doesn't read it good.

Help please, I'm completely lost
See full post 

Technical Discussion » TOPNET run a script locally before running ROP Fetch

User Avatar
masa90210
24 posts
オフライン
 2025年10月22日 08:32:43
Hi, I have a question about a TOP Network process (please see the attached picture).
I want to run two steps when executing Dirty and Cook:

1 - Run a Python script that updates or sets a new cache path on a File Cache node.
2 - Then run the actual simulation using a ROP Fetch node.

The problem isn’t with the Python script itself, but rather that the Python Script node seems to execute in the background Houdini session used by TOPs, instead of the current local Houdini session. I want the script to apply changes to my local session, and then trigger the ROP Fetch simulation task.

I suspect that using a Python Script node isn’t the right approach here — I probably need another method to execute the script locally before running the ROP Fetch. Or maybe I could fix the python node by changing parameters like a Generate When parm or Evaluate scripting During parm, etc, not sure at all. 

Of course, I could just run the Python script manually and then Dirty & Cook the network, but I have multiple dependent simulation steps, and I don’t want to manually update the File Cache nodes every time.

So, is there a recommended way to achieve this?
Edited by masa90210 - 2025年10月22日 08:34:32
See full post 

Houdini Indie and Apprentice » MPM Collision distance

User Avatar
N9Alex
4 posts
オフライン
 2025年10月21日 13:31:04
I've noticed that when increasing stiffness, the Collision gets worse.

And remembering the Cookie shot for H21 I'm thinking maybe I'm doing something wrong.

In the given example, I just inecreased the stiffness on the snow preset.

Appreciate any help, if this is expected for MPM it would also be great to know why its happening.


Thanks
Alex
Edited by N9Alex - 2025年10月23日 07:58:31
See full post 

Houdini Indie and Apprentice » Deconstruction - Simulation not working after unpacking

User Avatar
irissfeng
1 posts
オフライン
 2025年10月21日 13:09:05
Hi!

I've been trying to follow this youtube video that I found on object deconstruction - Vid Link [www.youtube.com] and it's been working just fine right up towards the end where the reconstruction and the unpack node. The object is shown as deconstructed and exploded but after unpacking it and bringing back a transfer attribute group the simulation is gone - its's just showing as a whole piece. And when I try to render the animtion it's just showing the object as intact all the way through. Does anyone know how I can fix this? Thank you!

Image Not Found
Image Not Found
See full post 

Technical Discussion » Set multiple Agent clip properties at once based on name

User Avatar
daniLon
6 posts
オフライン
 2025年10月20日 19:04:14
Hi,

I have some crowd agents that have a LOT of clips in their clip catalog, I need to set agent clip properties for all the nonlooping cycles to disable "Enable Looping". Is there a way that I can do this based on clip name? fFor instance all my nonlooping clips have "_nonloop" in their name. I tried using a wildcard in the clipname parameter but that doesn't seem to work. Adding each clip one by one and disabling the checkbox seems like a tedious approach, wondering if someone knows a smarter/more efficient way of doing this. Thank you!!
See full post 

Houdini Indie and Apprentice » HDA/VFL Package

User Avatar
AndrewKoehler
2 posts
オフライン
 2025年10月20日 18:26:45
Hi there!
I am pretty new to creating Houdini tools, and I currently have this project that consists of an HDA that has absolute references to a vfl utils file in the VEX code inside. Now I'm trying to figure out how to package that all together and be able to send it to others. I've found multiple tutorials on how to set up the package's .json file, but I can't figure out the process for actually putting the hda or vfl in there. If anyone could give me a link to some documentation/tutorial, or just explain it, I'd be incredibly grateful.
See full post 

Houdini Indie and Apprentice » Procedural modelling of a flower (Polianthes tuberosa)

User Avatar
vasczak
1 posts
オフライン
 2025年10月18日 06:55:59
Hello,

Coming from BLender and Cinema 4D I'm pretty new to Houdini and for an exercise I decided to recreate and rig a tuberose flower (Polianthes tuberosa). After watching the blooming flower tutorial serie I tried it on my own and nothing was pretty convincing. So I ask you Houdini users, how would you model and rig such a flower ? I've attached 2 of the reference I use.

I look forward to your answers !
See full post 

Technical Discussion » Otto Muscle and Bone Textures

User Avatar
glensandness
19 posts
オフライン
 2025年10月17日 16:27:11
Where can I find texture files for Otto's muscle and bone geometry? Both have well-defined UV maps that don't seem to align with anything in Otto's SHFS texture directory.
See full post 

How to Houdini Tutorial Contest » Contest Updates

User Avatar
rmagee
1340 posts
オフライン
 2025年10月17日 15:44:32
Two updates for the contest:

1) The deadline has been moved to January 16, 2026 so that no-one has to lose out on their new year's eve festivities.

2) There is now a Submission Form [docs.google.com] to fill out when you complete and publish your lesson. This will help us better organize the entries as they come in.
See full post 

Technical Discussion » Houdini-21 Material theme issues

User Avatar
aguiot
1 posts
オフライン
 2025年10月17日 09:51:29
Hello,

We are planning to upgrade from Houdini-20.5 to Houdini-21.
Everything seems to work great, except that the QML UIs we were using in Houdini-20.5 have some issues with Material theme in Houdini-21.
I've attached a gif showing the issue we are facing, with every hovered item having the effect shown at the top left of the window instead of on the hovered widgets.

Here is the minimal example used in the gif:

# main.py
import sys

try:
    from PySide6.QtGui import QGuiApplication
    from PySide6.QtCore import QSize, QUrl
    from PySide6.QtQuickControls2 import QQuickStyle
    from PySide6.QtQuick import QQuickView
    PYSIDE_VERSION = 6
except ImportError:
    from PySide2.QtGui import QGuiApplication
    from PySide2.QtCore import QSize, QUrl
    from PySide2.QtQuick import QQuickView
    from PySide2.QtQuickControls2 import QQuickStyle
    PYSIDE_VERSION = 2

QQuickStyle.setStyle("Material")


class BugReporterUI(QQuickView):
    def __init__(self):
        super(BugReporterUI, self).__init__()
        self.setTitle("Bug Reporter")
        self.setSource(QUrl.fromLocalFile("main.qml"))
        self.setMinimumSize(QSize(350, 400))


def launch():
    global bugReporter
    bugReporter = BugReporterUI()
    bugReporter.show()


if __name__ == "__main__":
    import sys
    app = QGuiApplication(sys.argv)
    launch()
    if PYSIDE_VERSION == 6:
        app.exec()
    else:
        app.exec_()

# main.qml
import QtQuick 2.15
import QtQuick.Controls 2.15
import QtQuick.Controls.Material 2.15
import QtQuick.Layouts 1.3

Rectangle {
    visible: true
    anchors.fill: parent

    Material.accent: "#DC8D02"
    Material.background: "#191919"
    Material.foreground: "#EEEEEE"
    Material.primary: "#607A87"
    Material.theme: Material.Dark

    color: Material.background


    ColumnLayout {
        id: root

        anchors.fill: parent

        Label {
            Layout.fillWidth: true
            Layout.leftMargin: 15
            Layout.rightMargin: 10
            Layout.topMargin: 5
            text: qsTr("Assigned To")
        }

        TextField {
            Layout.bottomMargin: 5
            Layout.fillWidth: true
            Layout.leftMargin: 10
            Layout.rightMargin: 10
            selectByMouse: true
        }

        Label {
            Layout.fillWidth: true
            Layout.leftMargin: 15
            Layout.rightMargin: 10
            text: qsTr("Impact")
        }

        ComboBox {
            id: presetsCombo

            Layout.bottomMargin: 5
            Layout.fillWidth: true
            Layout.leftMargin: 10
            Layout.rightMargin: 10
            currentIndex: 3
            implicitHeight: 35
            model: ["This bug prevents my whole department (or more) from working", "This bug concerns a few people", "This bug concerns me and me alone", "This bug doesn't prevent me from working", "I don't know, bring me a lawyer"]
        }

        Label {
            Layout.fillWidth: true
            Layout.leftMargin: 15
            Layout.rightMargin: 10
            text: qsTr("Bug Description")
        }

        TabBar {
            id: bar

            Layout.fillWidth: true
            Layout.leftMargin: 10
            Layout.rightMargin: 10

            TabButton {
                text: qsTr("Bug Report")
            }

            TabButton {
                text: qsTr("Context")
            }
        }

        StackLayout {
            Layout.fillHeight: true
            Layout.fillWidth: true
            Layout.leftMargin: 10
            Layout.rightMargin: 10
            currentIndex: bar.currentIndex

            ScrollView {
                id: report

                clip: true

                TextArea {
                    clip: true
                    selectByMouse: true
                    text: "a"
                    wrapMode: TextArea.Wrap
                }
            }

            ScrollView {
                id: context

                clip: true

                TextArea {
                    clip: true
                    readOnly: true
                    selectByMouse: true
                    text: "b"
                    wrapMode: TextArea.Wrap
                }
            }
        }

        Button {
            height: 30
            Layout.fillWidth: true
            text: "submit"
        }
    }
}

# in houdini
import main
main.launch()

# in standalone
hython main.py
See full post 

Technical Discussion » CUDA slower than Direct ML

User Avatar
jemabaris
8 posts
オフライン
 2025年10月16日 10:14:25
I am doing some tests with the new ML volume upres node in H21 and weirdly choosing Direct ML as the execution provider is a lot faster than CUDA. I verified that Houdini actually uses CUDA Toolkit 12.8 using:

import os
print(os.environ.get("CUDA_PATH"))
print(os.environ.get("PATH"))


My cuDNN version is 9.13 and it's also on my PATH environment variable. My understanding is that CUDA should be the fastest option but it's in fact 3-5x slower than Direct ML.

Any ideas what could be going wrong here?
See full post 

PDG/TOPs » Image Magick / cache help

User Avatar
oliverprimus
1 posts
オフライン
 2025年10月13日 13:10:17
Hey,

My current TOPs setup:

1. Wedge with 2 attributes and a wedge count of 4.
2. Rop Geometry: Grabbing a pyro sim. I'm writing this out to a bgeo ($HIPNAME.simcache4.`@wedgeindex`.$F.bgeo.sc), cooking frames as a single work item. It's set to 'in-process'.
3. Rop fetch: This fetches the Octane ROP. Cooking frames as a single work item, and in process too.

The reason I opted for 'cook frames as a single work item' is that I've found (likely to my misuse of the setup) is that the cache will be ignored for the rendering side, so it's constantly caching the pyro sim before rendering the image.

My current setup technically works, but from here what I'd like to be able to do is run this into imagemagick to create a montage of each frame (so frame 0 of each sim, frame 1 of each sim etc). With the current setup - where I have 4 work items (a work item per frame sequence) - I'm not entirely sure how to tell it to look within those sequences, find frame 0 for each sequence, and render an image of that. In previous setups where I've had access to every work item for the sequence, it's been possible, but not sure how to approach it here.

Any help would be appreciated!
See full post 

Technical Discussion » Tearable Cloth / Paper Effect in Houdini 20

User Avatar
RaFiH007
2 posts
オフライン
 2025年10月12日 11:23:55
Hi, I am trying to achieve this kind of tearing effect in Houdini. but not getting the desired result. Can anyone tell me how I can get this kind of effect where I can get tears in my masked areas? And also what would be the best method for masking? like point, edges or primitive selection for masking
https://imgur.com/a/mR1BFzM [imgur.com]
Edited by RaFiH007 - 2025年10月12日 11:26:35
See full post 

Technical Discussion » Metahuman Houdini Plugin for MacOS?

User Avatar
gmarc
39 posts
オンライン
 2025年10月11日 00:45:42
It seems that the Metahuman Plugin for Houdini 21 is only available for Windows. Does anybody know if it's possible to install it on MacOS? Aren't plugins for Houdini generally OS independent because of Python?
See full post 

Houdini Indie and Apprentice » How can I emit fluid in a changing direction

User Avatar
Desper
1 posts
オフライン
 2025年10月10日 18:50:18
I have a static FLIP DOP source and I need to give it velocity in a direction and have it change, not using the initial velocity values in the FLIP DOP source or the FLIP object nodes (unless there is a way to do it with that, I just don't want to manually control the direction using those). My idea was to have a grid face the direction I want the fluid to emit and somehow copying it's surface normal to the FLIP DOP source as the velocity but that feels like too "hacky" of a solution for a situation that sounds way too common. I also tried using an attrib wrangle but couldn't figure out how to get vex to read any of the other inputs I gave to the wrangle.

See full post 
  • 最初
  • 2
  • 3
  • 4
  • 5
  • 最後
  • Quick Links
Search links
Show recent posts
Show unanswered posts
製品
  • Houdini
  • Houdini Engine
  • Karma Renderer
学習
  • Houdini 講演
  • 教育プログラム
サポート
  • カスタマーサポート
  • ヘルプデスク FAQ
  • Houdini システム環境
  • ドキュメント
  • Report a Bug/RFE
LEGAL
  • Terms of Use (英語)
  • Privacy Policy (英語)
  • License Agreement (英語)
  • Accessibility (英語)
  • Responsible Disclosure Program
COMPANY
  • SideFX社について
  • Careers
  • Press
  • Houdini Merch Store
  • Internships
  • お問い合わせ
Copyright © SideFX 2025. All Rights Reserved.

使用言語