Skip to end of metadata
Go to start of metadata

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 17 Next »

There is no significant boost in simulation speed with enabled GPU support. Why?

Dyverso is optimized for multi-core and many-core CPUs, but some GPU cards do not have enough computational power to outperform a modern 8 core or 12 core processor.

Gaming cards like the GeForce GTX 10 series are perfectly suited for Dyverso and perform very good. The typical speed boost is around 4x compared to CPU simulations.

For more information visit → "GPU-Acceleration" or "→ "Ask : GPU-Acceleration".

The mesh is not created where the particles are. Why is there is an offset between mesh and particles?

There are some cases where you will see an offset between mesh and particles:

  • The "Scene" tree object has been shifted with the the parented "Mesher" node after the particle simulation. To fix this, detach the "Mesher" or reset the "Scene" object to its original position.
  • With some MoGraph objects ("Cloner", "Array", and "Matrix"), RealFlow | Cinema 4D evaluates transformations and applies them to the "Scene" and "Mesher" nodes. This may result in an offset. 
Particles are not influenced through daemons or other fluid containers. What's happening?

In many cases, the links are established automatically by RealFlow | Cinema 4D, but there are also occasions where you have to connect interacting nodes.

Take a look the "Scene" and "Links" fields of the nodes involved:

  • If "Scene" and/or "Links" are empty the node will not be affected by other scene elements.
  • Specify the interacting nodes and "Scene" through drag and drop.
I have applied an object to the "Object" emitter, but there is no emission. What's wrong?

When the body's polygons are very small it might happen that you will not get any particle emission. To fix this, increase the "Fluid" container's "Resolution" to create more particles. You may need very high settings (> 100 or more).

Isn't there a way to specify a default cache folder?

You can define a base path in Cinema 4D's preferences under "RealFlow". For the actual cache data we recommend specifying a separate folder for every scene, because otherwise existing data might be overwritten.

I'm a RealFlow | Cinema 4D 1.0 user. Is it possible to use my simulations in 2.0 as well?

Yes, 1.0 scenes can be used with RealFlow | Cinema 4D 2.0, but there are a few things you have to consider:

  • The RPC format for particles is no longer supported and cannot be loaded or processed. RealFlow | Cinema 4D now uses the Alembic format.
  • Mesh Alembics are still valid and can be used with version 2.
  • Initial states are no longer valid, because they are also stored in the old RPC format.
  • Simulation settings from version 1 are set to "Auto".
  • Simulations with the "Liquid - PBD" particle type will be slightly different due to new fluid parameters.
Why are the cache files so big?

RealFlow | Cinema 4D stores information for hundreds of thousands or even millions of particles – and every particle requires a certain amount data for a correct simulation. This is simply the nature of simulations, esp. with huge amounts of particles. Meshes often consist of millions of polygons with vertex data and all that has to be stored somewhere as well. And the information has to be written for every frame, otherwise it's not possible to resume simulations, or create previews and renders, for example.

RF|C4D does certainly not write Terabytes, although several Gigabytes are absolutely normal for large simulations. But mass storage became very cheap over the last years. It'd be fantastic if we're able to magic away all this information, but we're not that far...

Anyway, we are working on methods to reduce the number of simulation files.

Can I use SSD drives with RealFlow | Cinema 4D?

In terms of simulation and read/write speed it definitely is a good idea to have a fast SSD drives, but these devices have a limited amount of bytes to be written –  in many cases something between 250 and 500 TB. Beyond this limit data can only be read, but no longer be written. If you consider that a large simulation can produce several gigabytes of data this lifespan is a serious thing to think about. We therefore recommend modern hard disk drives. They are slower, but much more reliable. But in the end this decision is completely up to you.

  • No labels