+- tyFlow Forum (https://forum.tyflow.com)
+-- Forum: tyFlow Discussion (https://forum.tyflow.com/forum-1.html)
+--- Forum: General Discussion (https://forum.tyflow.com/forum-2.html)
+--- Thread: Generic tyFlow methodology question. (/thread-186.html)
Generic tyFlow methodology question. - MeHoo - 04-06-2019
Is it at all possible to have tyFlow know when to re-cache previous frames or not? What I mean is.. if I am adjusting a pBomb, for instance, and it doesn't start until frame 100, is there a way for tyFlow to know that the first 99 frames don't need to be blown away and re-cached? I know this is the same with pretty much every solver I've used, I'm just curious as to why this isn't possible for it to be smart about these sorts of things (if it is, and I'm not setting it up right - lemme know please). Obviously making global changes means redoing everything, but it would seem to me that certain operations could be a bit more efficiently handled under the hood. Linear animation keys would also only need previously changed data to be re-cached, etc. Seems like there's a ton of instances where this might save precious time.
I am pretty new to particles in general though, so it's quite possible that I am just way off base here. Still, I'd like to understand why, so if anyone knows, please share the knowledge.
Thanks in advance.
RE: Generic tyFlow methodology question. - tyFlow - 04-06-2019
Nope, tyFlow caches have two parts:
There's the "cache" of the current frame, which includes a ton of different bookkeeping for each operator, and the cache of all the previous frames, which includes just particle properties needed for playback (pos, rot, etc). Since bookeeping isn't stored for each frame, the simulation cannot be restored from any point except the current frame.
If I were to save that bookkeeping data for each operator, tyFlow would run much slower and take MUCH more RAM. However, you can cascade flows as a workaround in many situations like this, where you feed one flow into another flow that has a cached sim, using Birth Flow and Flow Update....this way the first flow's cache remains intact while you make changes to your second flow.
RE: Generic tyFlow methodology question. - MeHoo - 04-06-2019
I figured there was a good reason, I just had no idea what it was, lol. Makes total sense. Thanks Tyson!