mp collision changed?

#1

I’m not sure if I’m going crazy or not but if I remember correctly in previous versions of mparticles / box2

when you had a mp collision operator in one event, then you piped your particles into another event without a mp Collision operator. There would be no collisions…

Now piping into another event with the instanced world the mp collision still registers even though its not in that event!

When we needed collisions we’d just instance the mp collision over. But now it’s always on. This limits flexibility x 100.

Johnny if you’re around. WHY DA HECK DID THEY DO THIS!!!

I’m freaking out!! Can we get it changed back!?

#2

a follow up test.

In 2015 a pflow with two events… In the first event there is no mp collision. In the second event an mp collision operator exists.

The particles in the first event still collide! Arggg :surprised

#3

Hey Nico!

I am pretty sure it was done based on performance. I seem to recall that when they were per event the World engine took orders of magnitudes longer to calculate moderately heavy sims.

I haven’t even been able to really figure out anyway to cheat around it, other than using another system *cough TP :frowning:

I agree it really could use a little more extensibility with collision registration. You can use collision exclusion groups but this is very limited, too limited. I think that those should be extended into Box#3/ADM. Input mParticles has collision registration and collided count but per particle per object/event collision inclusion/exclusion control would be epic! I can inquire about it, I have a feeling they will just say, add it to Uservoice, then we vote which is fine, I would put max votes to it. Gees, I am still pushing to get Bounding Box display to work properly for the last three versions.

You guys are doing great work on AHS Freak Show BTW :slight_smile:

#4

Johnny!

Thanks for the update. Damn, in the back of my mind i figured it was done on purpose for performance reasons.

If there’s no going back than I totally agree we need the output mparticles subop to includes object / event inclusion and exclusion on a per particle basis. Or at least a checkbox in the mp collision operator that allows for the collisions to be local to the event and not global in the world. Man.

I’m not sure what they’re cooking up for pflow in the 2016 release, but I hope it’s freaking fantastic… This was a step backwards for me!

Btw I haven’t been at Fuse for a little while, taking a big break from work. My son was just born! But I will forward you’re compliment to Charley!

#5

That’s some great news Nickolay, congratulations dude :slight_smile:

#6

Good for you man! Breaks are a necessary part of life especially for something awesome like starting a family :slight_smile: Big congrats on the new addition!

#7

Thanks Alvaro and Johnny! :slight_smile: Appreciate it!

Btw I posted this issue over at the orbaz forum and Oleg responded saying that he’ll take a look.

I’m quoting him here as saying

"Not intentional. I’ll check it out. "

http://www.orbaz.com/forum/viewtopic.php?t=4004