MultiMap and pflow

Corona Corona for 3ds Max Issues
1

Hello guys.
I’ve been having this problem since 1.6.

I’m using pflow to do some stuff and used the CoronaMultiMap to randomize some Shape Instances base color. And as you can see in the GIF they start fine but then the colors assigned get randomize every frame.

I did some of the same type of pflow animations before and did not have this problem…I’m not sure but I think I started to notice this when some of the hotfixes for 1.6 came out…but not sure at all when exactly.

giphy.gif

I’ve attached some of the settings I have.

I’m using one shape with the multimap. But this also happens while using a group of shapes…

This happens when rendering locally, in the VFB and using backburner ..

Any help would be nice, because I can not exactly pinpoint if it is a problem just of MultiMap or something from pflow..

thx!!

333.jpg
333.jpg1005×497 164 KB

2

Hi. Haven’t tested it, but a quick guess would be, that this happens when the first particles start to “die”.

So what’s your lifespan set to, and will the blinking stop if your particles live forever?

3

Hi. I forgot to say that those 2 “sets” of particles are just one, it’s part of a loop animation. So the bottom group it just the upper group but with time offset…
So the particles have no lifespan limit, they are emitted for 48 frames but no “age test” or “delete” node in pflow…

They just get emitted…all good for some time and then BAM… the ficker begins…

4

giphy.gif

This is the original sequence xD

It works using different shape for particles each with different colors…using it with multimap was very useful for quick variations

5

Works here in a quick set up (one PFlow emitter, one sphere, MultiMap applied using default colors in it rather than maps), I was able to scrub through up to frame 300 and didn’t see anything odd. Any chance you could share the scene?

6

So… I sat down with a friend and try to simplify this.

giphy.gif

If the particles get deleted by a deflector…somehow the particles “reset” and the multimap in instance mode flicks … And it happens with particle age too..

So…instead of instance mode we change to “material ID” but the multimap doesn’t seems to “read” the matID. It only renders the first ID with no randomness.

But if you use the multi/subobject and separate coronamats it works well…

It seems that the Corona MultiMap doesn’t read the random ID assign by the “material static” box in pflow…and it flicks using the instance mode…

I’ve been playing with this for not so long, so maybe I’m missing something basic and probably not entirely corona related.

Untitled-1.jpg
Untitled-1.jpg873×573 175 KB

multimap.zip (209 KB)

7

-can you check this with material static vs material dynamic?

-in pflow settings there is also something like viewport % and render % - can you try setting both to 100%

-another setting is “half-frame”, “one frame”, etc… I forgot the name, it’s like computation step. Can you also set both to the same value?

-workaround would be saving the pflow using mesher and assigning random material ids using material by element modifier

8

Hi Maru.

-same problem with both..
-all was done with 100% on both settings
-integration step the same on both on half frame
-it works, but the bad thing is that the 3dsmax file gets huuuge…and export to alembic also is huge… I’ve try this method because I had to render on cgifarm and got the flicker problem too ..

I asked a friend to do a test with other engine…works as it should

giphy.gif

with the settings attached.. Same thing as in corona: Just a material with the multimap on the diffuse..

When viewing the pflow the Particle ID itself are not changing when particles die, so Corona must be getting confuse by something while using instance mode…and not randomizing in matID mode..

beeraystuff.jpg
beeraystuff.jpg945×529 98.3 KB

9

I´m encountering the exact same bug using 3ds Max 2017 and Corona 1.6.
Has this been fixed for 1.7?
I asked for the new build (1.7) but due to my company bureaucracy, will be days before I can get the patch.
edit: I tried using mesher and material by element and still got the ids changing.