problem when using multiple VDB events
#1
Hey Tyson!
Hope you're doing well,

I still have this bug when using multiple events http://forum.tyflow.com/thread-1892.html
I use the empty birth VDB event workaround though...


But the main problem I want to point at now is the evaluation order, I don't understand how it works with VDB events.
it may be the same as standard events but It seems a bit odd with vdbs, or maybe I'm missing something...

For simple additions it's fine, but for more complex operations, with several intersections and substractions, it can be a reel pain...

For instance I'd like to patch a flow like this:    
that behave like this:    

Here's how some of my flows look like if I want a substraction to be evaluated in the right order Sad    

this file should illustrate the issue:
.max   vdb_eval_order.max (Size: 972 KB / Downloads: 11)
  Reply
#2
Quote:I still have this bug when using multiple events

Oh, weird...I thought I fixed that one.

As for your event ordering...events are evaluated breadth-first, top to bottom. So yea...things can get tricky when you need them to evaluate in a different order due to where/when you're sending data between events.

I'll add a setting in the next build so that you can offset evaluation for any event, not merely birth events. Maybe also an "evaluation order display" or something so no guesswork is involved when specifying order values.
  Reply
#3
Also, I tried loading the file in the other thread you linked and couldn't get it to crash. Do you have a set of steps to take to get the crashing to occur?
  Reply
#4
That'd be a very handy addition!
I thought it was something different than breadth-first, but that make sens.
I didn't want to ask for a new feature but maybe you could consider adding a breadth-first/depth-first switch option in the flows main settings as well...
Anyway, thank you for the quick feedback.

As for the bug, this file crashes for sure (the flow is disabled) the particle to sdf op seems to be involved...
.max   vdb_crash.max (Size: 900 KB / Downloads: 11)
  Reply
#5
Hmmm...try as I might, I can't make either file crash. I enable the flow, refresh it countless times...no crash. The VDB is initialized properly and I can see the particle mesh.

Could you send me the minidump from your crash?
  Reply
#6
Hey Tyson,
You're right, the problem occurs only on my computer at home Sad
 I've send you the log and dump files...

Thanks for your time and efforts!
  Reply
#7
Did you forget to attach? Smile
  Reply
#8
no I've send you an email on the support mailbox, you didn't receive it?

here they are

.rar   vdb_crash_log_dmp.rar (Size: 764.6 KB / Downloads: 7)
  Reply
#9
Nope...nothing in my inbox, but thanks for attaching here. I'll check it out when I get the chance.
  Reply
#10
So I think I may have actually tracked this one down. Hopefully it's fixed in the next build (v0.16108)...let me know if you run into it again.
  Reply
#11
Thumbs Up 
That's good news, you rock!
  Reply


Forum Jump: