Jump to content

tortoise

Members
  • Content count

    36
  • Donations

    0.00 CAD 
  • Joined

  • Last visited

  • Days Won

    2

tortoise last won the day on October 4 2019

tortoise had the most liked content!

Community Reputation

20 Excellent

About tortoise

  • Rank
    Peon

Personal Information

  • Name
    Martin Egger
  • Location
    Ludwigsburg, Baden-Württemberg
  1. Hey, in order to make work on our diploma project easier, me and a colleague developed a slightly advanced filecache simplifying the process of iteration for us and junior artists. Now to make reviewing easier, I want to simultaniously output an OpenGL-render for every frame after the geo is done. In ROPs this is super easy and my standard setup whenever I am developing some FX. This is the general layout: Since for this filecache I am relying on TOPs to make outputting quicker and more streamlined, I am now facing the issue that I can't get the ROP-Fetch to replicate what normal ROPs would do. I deliberately don't just put a ropgeo-TOP and a ropfetch with just the daily after each other because that has to start a seperate Houdini instance just for the OGL-render, which is extremely inefficient, as the render itself takes just one or two seconds - and I don't want to increase the batch size either, since that would cause a delay in the OGL-output after the sim is done. I just want it to output it right after each frame is cached. With two ROPs after each other, this works wonderfully in my experience. In either of the following two layouts the ROPnet would do what I want, which is output one frame of the GEO-ROP and one frame of the OGL-ROP one after the other: Unfortunately, now what happens with the ROP-Fetch in TOPs in the exact same layout is that apparently the very first work item already renders out the entire sequence. In the two days of debugging that I've invested so far I've tried loads of things that came to my mind, one being just using the pre-built ropgeo-TOP-node and inputting my daily-ROP there. However, there, even weirder behaviour happens. If aligned like above, it simply ignores the OGL-ROP and if I ROP-Fetch just the OGL, the very first work item outputs the full frame range and all other work items are ignored. Either I missed some hidden settings in the thousand times that I went over it or this stuff is a bit buggy. I'm hoping for the former. Cheers, Martin EDIT: Alright, after chatting with support, it turns out that the python updating the comment of the parent node was causing the crash. Well, have to go without this nice UI thing then...
  2. Hello! I'm a student in fourth year in Animation/VFX at Filmakademie Baden-Württemberg, currently starting production on our diploma project including some really awesome VFX shots. The film is a 2 minute social spot promoting renewable energies for private households, the title is "Mother Nature's Power". The story will be told with footage shot in Iceland and in a studio and then some disasters at the film's climax are VFX / Full CG parts. Here are our three key concepts for what we'll create photo-realistically later on: (Oil Rig being crashed by an iceberg) (Coal power plant falling into a sinkhole) (Volcano erupting and the lava turning into solar panels) To get to the point: We already have a big talented roster of film and animation students assigned and scheduled, but we are still looking for help in the department of lighting (Houdini/Redshift/Arnold) and assets (Anything). If you're a passionate artist looking for opportunities to collaborate then we can provide a platform for potentially awesome portfolio work. We'll be working on this for a year from now and we are very flexible with scheduling. Remote work is no issue at all as well. If you are interested, I can send you a project folder with details about our concept and approach as well as our animatic that is almost final and we can talk further. If you wanna know more about the heads of this project, here is the vita from me and the 2nd VFX-Supervisor on this project Lucas Bruchhage: https://www.linkedin.com/in/martin-egger-3d/ (Aixsponza, Mackevision, MPC) https://www.linkedin.com/in/lucas-bruchhage/ (Framestore, MPC) The main director that I share co-direction with is Nicolas Bori, student in Advertising Direction with a lot of production experience also. Check out his website here: https://www.nicolasbori.com/ We're looking forward to hearing from you! Cheers from Stuttgart, Martin Egger Contact: martin.egger@filmakademie.de
  3. RBD follow animation softly (Soft constraints)

    assemblyRnD_martin_v21_walksim_working.mp4 That's how a first version ended up looking! Just some noise to make it more alive... But it's something to start the buildup from.
  4. RBD follow animation softly (Soft constraints)

    Hahaha, emphasis was on good!!! But only with age comes wisdom, no? I'd be so lost without your wiki, man! I gotta visit you down under and buy you a couple of beers one day. I'm excited for it as well, it's a 2 min full CG short with the spirit of the forest fleeing from a wildfire and taking on all shapes and forms. My colleague Justin Braun, the director, developed a Machine Learning setup that replaces Quill-strokes with Megascan-assets - it's pretty awesome.
  5. RBD follow animation softly (Soft constraints)

    Alright, found a great setup by good old Matt Estela. http://www.tokeru.com/cgwiki/index.php?title=ConstraintNetworks2 "leaves_rbd_speedlimit_and_torque.hipnc" Depicts a pretty great workflow to set this up. Got it working with pigheads in a redo-scene. I only switched out the old spring constraint relationships to bullet soft constraint relationships. They work wonderfully.
  6. Hey guys! I'm struggling a bit with a setup that's probably quite simple but I couldn't get behind it for a couple of days now. My shot requires me to have pieces which a character consists of (Megascan-assets scattered on a character) follow the character's animation. This looks something like this: walkcycle_fox.mp4 Now what I've tried to far was to go about the following with some fancy orientation math in a SOP-Solver in the dopnet to get the euler-W-acceleration to target the goal-orientation, but somehow the pivot offsets made this way harder than expected. My next approach (which I should have rolled with from the beginning) are constraints. So I'll try to have a soft constraint for each piece to a world anchor and target the rotation. But just how exactly does rotation-constraining work with the RBD-tools? I've found very little information on this topic - only for things that are individual objects, but not 800 pieces... I'm happy for any hint towrads a solution. Cheers, Martin
  7. What So a while back here at Filmakademie, four of us got together for 4 (!) days to hack together something that would work with Dolby Atmos, a very special sound system that is supposedly "truly surround" and just has a crapton of speakers all around you. How We just built lots of different setups that were all driven by a velocity field from a smoke sim that was based on an animation matching the audio. Quite happy with what came out at the end considering the time spent on it - it was great fun to work on it too. I would gladly share all of our setups, but unfortunately, everything was wiped as we worked on a temporary network drive. *mega-facepalm* But well, now it's art. The only thing left over is this one setup for the meshed smoke, which was one that I did: meg_glowsmoke_v16.hipnc Hope you can learn something from it. A huge shout out to my colleagues, Lucas Bruchhage, Tim Markgraf and Mike Razniewski that were such a great spirit.
  8. Fill Selection

    No need for a script - just hold down H. Like holding "A" to make a "shortest path" selection with left click or middle click to do a loop. This page is an awesome read: https://www.sidefx.com/docs/houdini/basics/select.html
  9. Flat Fliptank Wave Issue (Moving Narrowband)

    This image illustrates quite well what is happening on the first sim frame. At the bounding regions, the sourced fluid is at the surface level when the initial source (By shelf tool default) Lies at Y=-0.15. When I try to raise this initial source however, the extension is by that amount higher as well. It's a bit mind-boggling.
  10. Flat Fliptank Wave Issue (Moving Narrowband)

    Alright, after spending almost two workdays on trying to fix this I have decided to ask SideFX for support - but I still would like to post my latest findings. Here is a file where I tried to recreate it in the simplest way possible: A sphere (in boat size/shape) travelling in +Z with the vanilla "flat tank" shelf tool applied to it. This image clearly shows the issue: Here is the scene: Skipping_Basesetup_Flattankissues_v01.hip With speeds of 1m/Frame it is completely fine, but as soon as I go below 0.1m/Frame this wave starts occuring really badly. And no amount of tweaking that I could think of would reduce it. Apparently, the particles created at the boundary layer have a weird offset and velocity upwards. In certain testing cases, this offset upwards was the partice seperation. Eg.: When water level is Y=0, the initial particles rest at Y=0-particleseperation, however, sourced at the boundary layer the particles had Y=0 leading to an uneven surface and thus a wave. I'm thankful for any advice at all, Martin
  11. Hello! So at the moment I'm doing a sim for penguins jumping in a pool, and since the camera pans along with them, I decided to roll with a moving narrow-band container, to reduce the cost of the sim. In an earlier shot the narrowband worked nicely for jumping in, however, in this shot, I'm having problems with the container producing a sort of wave. It's going in -Z (To right of image). So for testing purposes, the simbox is super tiny, but no amount of extending the narrowband towards the direction of movement had any impact. Does anyone know what's causing this? I'm quite stuck at the moment. Greetings from 127 Wardour Street London, Martin
  12. Because today was another day where I went nuts trying to figure out retiming to be 'right', I now took a bit of time to build this retimer, which works a bit more intuitively. It works via a solver - so now you can animate the speed of the input, like in Premiere, and the output duration essentially gets stretched and squashed automatically. Because the solver is simply a recursive calculation, the solving is instant - when you re-animate the input time, just hit "Reset Simulation" and you're good. Maybe it's useful to somebody! Here it is for download: retime_solver.hdanc (H17.5) (While it's just a non-commercial asset, it can easily be rebuilt in a couple of minutes. The groundwork is done. All the best, Martin PS: I also added a fun-toggle, where you can tweak your retime with spinning test-geo in case you have heavy input geo! (... which we always do, right?)
  13. Simple Noiser

    I appreciate it. Alright, in the latest update of the tool on Orbolt, I made the long needed change of making the noise non-time-dependent when temporal frequency is set to 0 - so the noise is static. Until now, it was necessary to have a timeshift freeze after it, which was... not optimal. Also now it's working with the latest Houdini releases. :)) Here's the link to it again: https://www.orbolt.com/asset/Tortoise::Noiser (I also updated the original post with the new version.) Cheers, Martin
  14. Martin's Procedural Setups

    Hey! With the desire to contribute to this community, I will use this thread to collectively post some of my setups here, in an attempt to help others learn Houdini. My experience was that going through established setups was a great way to learn - and that breaking working setups and learning what breaks them, teaches things that tutorials never really touch. So whenever I found a page sharing advanced setups like that, my heart lit up, and this thread will be an attempt to recreate this feeling in others. Being a current student at Filmakademie BW, this thread is also an hommage to the work of my (just graduated) great colleague Juraj Tomori - and his "Juraj's Playgound". (Here is the link) Enough babbling about, let me start off by showing my reel of the works that accumulated over the past two years in which I worked with Houdini. The first post will also be sharing the first setup shown: The arena shot of the plasma cannon destroying a concrete pillar. The entire first sequence was my bachelor thesis about "Procedural 3D for Film Using Houdini" where I tried explaining a procedural approach in written form and detailing my setups. If anyone is interested, I will also be sharing it here. The arena environment was also procedurally modelled in Houdini, which I will also be sharing a setup on in the next couple of days. I hope you will like it, Cheers, Martin
  15. So after posting this, my project changed directions quite a bit and we no longer pursued generative sound design - somewhat unfortunately. However, I did get around at a later time to recreate a setup of a sequencer in Houdini in a very basic solver and a CHOP-Net. Here's a quick video with sound: hou-sequencer_double_1 (1).mp4 And the HIP is here: meg_hou-sequencer_v07.hipnc And this is the basic Copy-CHOP-Setup:
×