Maya nParticles VS RealFlow in liquid simulations

These are 4 tests that I threw together to test Maya 2009's new nParticles liquid simulation option. I compared it to RealFlow, a fluid and particle simulating standalone program to see how the results held up. Overall Maya's nParticles did pretty well, better than I expected and could probably cover most simple liquid simulation needs.

-Additional Test Information-
I tried to keep the simulations pretty equal for things like particle count and the radius of each particle in the end result to make it about as even of a challenge as I could. From what I saw the mesh output option in Maya seems to be about as fully featured as RealFlows, with support for motion streaking, vertex motion baking, smoothing/filtering, so to save time I opted to do test renders with simple Maya Blobby particles in the software renderer. I did not see an option for wet-maps in Maya, though.

Test 1 - Ramp Splashing
In this test I wasn't too sure what I was going for, just first starting to play with nParticle liquids, so there's two pointless extrusions, I guess they break up the flow a bit more for some interesting effect. I had the emitters fill a bit of the ramp, blocked by a plane that jumps out of the way after a bit. After playing with this I was pleased enough to do a few more tests.

Test 2 - Objects Splashing
The objects were simulated before any water, with their animation baked beforehand, so they are not in any way affected by the particles. On that note; I couldn't find any way yet to make objects get pushed about/float with nParticles... Since the plank fell flat against the ground plane it crushed the particles into an infinitely small space, hence the splashes underneath.

Test 3 - Waves
Just a box oscilating to push some waves about. The nParticles kept falling too thin, so I increased the radius a bit beyond what I felt they should be, and when I went for the final higher quality simulation they ended up occupying a bit more volume than the RealFlow test, but the results didn't seem to be affected too much. It's interesting to note how RealFlow has waves of pressure and speed run through the particles, whereas Maya just has a messy gradient of force imparted straight on. It showcases a bit of the additional technical accuracy in RealFlow, though standing by itself the Maya result seems to look just fine. I also found it odd how the nParticles would actually leave the bounds of the box and then pop back into it. Also; I have no idea what's with the one random particle sitting in mid-air on the RealFlow side. It's a glitch that occurred only at render once the particles were brought into Maya.

Test 4 - Pressurized Spray
Here I made a bottleneck and a bit of a pipe on top of a cylinder, and forced the particles into way too small of a volume. Again RealFlow exhibits better accuracy, as well as it seems to strive to maintain volume more aggresively, best seen when the cylinder first starts pushing. I found it interesting that the Maya simulation didn't really blow up untill a wave of speed bounced down off the top of the container, then back up off the pushing cylinder.

Conclusion - nParticles do a pretty good job. Overall, I expected less, and though RealFlow does do a better job in terms of fine quality as well as works harder being multi-threaded, nParticles manage to get a pretty convincing result in overall less time. Personally I'll keep using RealFlow when it comes to liquids, but nParticles certainly hold their own.

--Extra notes--
It was told after these tests that Maya's nParticles default to a substep calculation of 3, whereas RealFlow by default uses an adaptive sampling of up to 333, which could greatly account for why Maya was marginally faster, but lower of lower accuracy.

Tests ran on a Q6600 Quad Core at 2.7GHZ with 4GB of ram under 64-bit XP Pro, a few re-tested off-record on an HP with a Quad Core Xeon at 2.6GHZ, 32-bit XP Pro.

Category:

Film & Animation

Tags:

• Maya
• 2009
• nparticle
• particle
• RealFlow
• liquid
• simulation
• dynamics
• test
• comparison
• wave
• fluid
• splash
• spray
• benchmark
• performance
• waves
• water
• particles

[D4RK809]

You didn't mention If the times you posted where in hours or minutes or whatever, especially the first comparison is that 35 hours or minutes?

[ICPJuggalo1988]

@D4RK809 Sorry, yeah all the times are in hours:minutes:seconds, with hours only showing provided it took that long. All of these scenes were run over the course of one evening, so the longest test would be the second one with objects crushing particles, where Realflow took 3 and a half hours.

[joemoet2009]

EMACHINES! = VERY BAD!my cpu goes 100% when simulating

[ICPJuggalo1988]

Having a CPU crank at 100% just means it's doing the hardest work it can, and is a good thing when you're asking for it. Having less than 1GB of ram could certainly be a bottleneck, these scenes were run with 4GB, but soon after I had to replace two sticks, and the drop even to 2GB was severe for simulations and rendering especially.

[aghaanantyab]

realflow = good but maximum ram used

maya = does not require much memory but its bad

[ribenawrath]

this i great and all, but I have to say that from these demos realflow looks the most natural.

[michaelwbell]

Maya 2012 nParticle simulations are meant to be multi-threaded now, i believe? Should be interesting!

[bharatkumargupta]

while real flow is a seperate product and nparticles (maya) is built in, u can easily see the quality that maya tools offer in competition to stuff around , maya is amazing app

[4b10xs]

thanks for the insight.

[zappenduster66]

awesome comparison. could you do this for realflow 5 and maya 2011/2 too? =)

[wasitrealy]

@realflow100 That's not RAM, that's Virtual RAM

[realflow100]

@supermix1337 lol :)

[supermix1337]

@realflow100 Oh, 6 months ago I made that comment:D

[realflow100]

@supermix1337 oh and i type too much now my hand hurts lol.