How Fast Is NVIDIA’s Stretchy Soft Body Simulation?

TL;DR

NVIDIA’s stretchy simulation processes a soft body made of 80 thousand elements at 8 milliseconds per frame, enabling real-time performance on a modern graphics card. The method withstands extreme deformation, largely preserves volume, and has much simpler pseudocode than an earlier material-synthesis technique. Its combination of speed, resilience, and quick implementation makes the detailed comparisons worth reading.

Transcript

Dear Fellow Scholars, this is Two Minute  Papers with Dr. Károly Zsolnai-Fehér. Today we are going to have a look at beautiful  simulations from a quick paper. But wait,   how can a research paper be quick?  Well, it is quick for two key reasons. Reason number one. Look at this complex soft body  simulation. This is not a jumpsuit, this showcases  ... Read More

Key Insights

• 👻 The simulation algorithm in the paper allows for stunning soft body simulations with realistic deformation and movement.
• 🚄 It runs at a high speed, allowing real-time rendering on modern graphics cards.
• 🔇 The algorithm preserves volume, setting it apart from previous methods that suffer from volume dissipation.
• 😄 Its ease of implementation makes it a valuable tool for computer graphics researchers.
• 👨‍🔬 The authors of the paper, Miles Macklin and Matthias Müller, are renowned research scientists at NVIDIA.
• ❓ The advancements in this algorithm suggest potential future progress in simulation techniques.
• 💻 The simulation algorithm showcases the possibilities of material synthesis in computer graphics.

Questions & Answers

Q: How fast is NVIDIA’s stretchy simulation?

The simulation runs at 8 milliseconds per frame. According to the presentation, that is fast enough to run easily in real time on a modern graphics card.

Q: How complex is the elephant soft body simulation?

The elephant simulation is made of 80 thousand elements. These elements represent the geometry of the elephant’s outer tissue while it stretches, moves, and deforms.

Q: Can the algorithm recover from extreme deformation?

Yes, the algorithm withstands severe deformation in the demonstrated dragon test. After the dragon is compressed and scrambled into a tiny plane, releasing the forces allows it to regain its original shape.

Q: Does the new simulation method preserve volume?

The method retains the entirety of the volume in one demonstrated comparison, unlike traditional techniques that lose material over time. In a more extreme cube test, the volume changes a little, but most of it returns after the forces are released.

Q: How does the method compare with previous simulation techniques?

Previous methods shown in the presentation suffer from volume dissipation during stretching and deformation. The new method better preserves volume while remaining quick and handling challenging transformations.

Q: Why is the simulation algorithm described as quick?

It is quick for two reasons: it runs quickly and is quick to implement. Its performance reaches 8 milliseconds per frame, while its pseudocode is much simpler than that of the earlier material-synthesis technique shown.

Q: How long could the simulation algorithm take to implement?

The presenter estimates that a competent computer graphics research scientist could implement it in a day if everything goes well. He describes that as rare for a modern simulation algorithm.

Q: Who created NVIDIA’s stretchy simulation algorithm?

The work was written by Miles Macklin and Matthias Müller. The presentation identifies both authors as research scientists at NVIDIA.

Summary & Key Takeaways

• Definition: The method simulates soft bodies that stretch, move, and deform while responding to applied forces.

• Number: The elephant’s outer-tissue geometry is made of 80 thousand elements.

• Number: The simulation runs at 8 milliseconds per frame.

• Tool: A modern graphics card can run the demonstrated algorithm easily in real time.

• Step 1: Forces compress and scramble the dragon into a tiny plane.

• Step 2: Releasing the forces lets the dragon regain its original shape.

• Compare: Traditional techniques can lose volume over time, while the new method retains volume more effectively.

• Compare: Extreme cube transformations change the volume a little, but most of it returns after release.

• Compare: The algorithm’s pseudocode is much simpler than the earlier material-synthesis technique shown.

• When: A competent computer graphics research scientist could implement the algorithm in a day if everything goes well.

• Who: Miles Macklin and Matthias Müller, research scientists at NVIDIA, wrote the work.