Scale

General Discussions about realflow, fluid simulations,...
markjohns
Posts: 4
Joined: Sat Jul 02, 2016 9:03 am

Scale

Postby markjohns » Sat Jul 02, 2016 10:14 am

Hi. I'm very new to Real Flow and have a question about scale.

Maya's default scale is cm.

In work, we set Maya & Max to meters.

Real Flow's default scale is meter. However, if I create a cube in Maya (using meter scale) and bring it into Real Flow, the cube is huge. It surrounds the whole grid.

If I set Maya to cm and export the cube, the cube is one unit square in size in Real Flow. Can someone explain why this is? If Real Flow works in meters and Maya is set to meters, shouldn't the cube come in as 1 unit in size?

I don't want to change the way we work in Maya and Max because of this so can someone tell me another option?

If I bring in a 1m cube from Maya (using meter scale), setting the geometry scale in scale options to 0.01 (1cm = 0.01m) brings the cube to the correct size.

Should I also bring the other scales (daemons etc) down to 0.01 ?

I set this using the scale options at the bottom right or Real Flow. In File > Preferences, you can't set them to 0.01, only 0.1.

Any help with this would be great.

Thanks


Forester
Posts: 73
Joined: Mon Mar 15, 2010 4:36 am
Contact:

Re: Scale

Postby Forester » Sun Jul 03, 2016 11:48 pm

Hi, a Maya-Realflower here....

Yes, the scales for Maya and in Realflow are NOT directly equivalent.

In general, the Autodesk advice is to use the default cm for Maya. But if you are also using MAX, you may have good reasons to set both Maya and MAX to meters, as you point out. Particularly if you are making models "to scale" and working back and forth between Max and Maya.

One thing you should know is that when in Realflow, if you change the geometry scale of an object you have imported from Maya using the Scale Option, a Realflow mesh (and presumably the *.bin file) exported from Realflow to Maya will, most of the time, end up matching the object in original in Maya that you first imported into RF. (Even though you scaled down in Realflow.) It is as if Realflow knows you scaled down, and so when you export, it is smart enough to scale up the exported product to the original. But, I need to point out, this is pretty much true only when you are working in reasonable base 10 increments, such as 1 to 0.1, and so forth.

To be explicit, let us say that you import a cube from Maya (where Maya scale is set to 1M), and then in Realflow, you use the Scale Option to change the scale of that object to 0.1M. Now you pour some fluid on that cube, build a *.bin file set or a mesh for it, and then export either of those back into your Maya scene that holds the original 1M cube and whose scene scale is set to 1m. Your *.bin file or your mesh should "fit" the cube and look the same as it did in RF. (This works when using the Scale Option to either increase or decrease the size of the objects in the RF scene.)

So, this should not be too much of an issue.

BUT you need to know that the way in which the emitters and the daemon forces interact with the objects, and the Object Interaction Settings are very much dependent upon scale within RF. Within RF, these are relative to each other. So, if you bring in an object made in Maya, where Maya is set to 1M, and do not use the Scale Option, you would have to enlarge your emitters, deamon forces and Object Interaction settings to make something realistic. And when you do that, you will have a ton of unnecessary particles, weak and poor daemon forces and very long sim times.

Conversely, if you bring in an object from Maya, where Maya scale is set to 1M, and then use the Scale Option to scale all scene objects way down to 0.1M in Realflow, it can sometimes be that you might end up with emitters that are not terribly populous in terms of particles, and your daemon forces might be a little weak. This situation would only be made worse if you also use the Scale Option to scale down the emitters and daemon forces to 0.1"scale." (And if you are using the Dyverso solver, it can become terrible because you also need to adjust the object interaction cell size down in such a way that it ends up being so tiny that Realflow chokes. (I think the technical term is "gives unpredictable results..."))

So, if your requirements are to keep the 1M scale for Maya, then you certainly do need to use the Scale Option in RF, and scale down to something reasonable such as 0.5 or 0.1 (both decent settings). But you probably will need to keep the Emitters and the Daemon Forces at their normal scale settings. And don't forget to take a look at the Object Interaction settings and possible adjust those until you get the "right" effects.

Of course, you will need to experiment because these things depend upon what you doing within your simulation. But, I think I am trying to say that, use the Scale Option to scale down in some consistant and regular fashion, don't change the Emitter and Daemon Force scales unless you really find that you must to get the effects that you are after, and then count on RF to be able to translate the results back into Maya OK. (And if it does not, and you are exporting back meshes, instead of *.BIN files, then just grab the meshes and scale those up in Maya to match your objects. You should be OK.)

Forester
Posts: 73
Joined: Mon Mar 15, 2010 4:36 am
Contact:

Re: Scale

Postby Forester » Sun Jul 03, 2016 11:51 pm

If you are using the RF Connectivity plugins to work entire sims between Maya and RF, and between MAX and RF, none of the above is true. At least, it is not at all this easy.

The connectivity plugins work quite well, even with these Scale Options for Maya and RF. But the whole process for MAX and RF is different, and is made much more complex because of MAX's native preference to work in 1M scale. (Versus Maya's 1cm native scale.) Personally, if I am using the connectivity plugins, I do everything I can in Maya and RF, and then ship what I can afterward to MAX.


Return to “General Discussion”

Who is online

Users browsing this forum: No registered users and 1 guest