Author Topic: when will new caustics solver be done?  (Read 2261 times)

2020-08-06, 19:09:29
Reply #15

johnmarc

  • Active Users
  • **
  • Posts: 23
    • View Profile
Also caustics should work natively without much tweaking to any random interior and outdoor scene on turbo squid without producing fireflies all over the place that never go away and take 500 passes to even begin to remotely clean up especially when there is shiny curved objects such as metal and such

2020-08-06, 19:57:11
Reply #16

PROH

  • Active Users
  • **
  • Posts: 1143
    • View Profile
According to your definition ALL renderer are mediocre? It's not only Corona that is "exposed" (your wording) as such - but ALL renderer.

We all would like better caustic, but your wording isn't getting things moving forward.

2020-08-06, 21:37:36
Reply #17

lolec

  • Active Users
  • **
  • Posts: 155
    • View Profile
Not reading comments from other users, ignoring basic questions, repeating the same chant over and over. Not sure we need more evidence of a troll among us.

2020-08-07, 01:03:31
Reply #18

johnmarc

  • Active Users
  • **
  • Posts: 23
    • View Profile
Pointing out terrible solvers for sure I am troll

2020-08-07, 01:23:53
Reply #19

lolec

  • Active Users
  • **
  • Posts: 155
    • View Profile
Answer 3 basic questions then:

1. Which render engine has a great caustic solver, that lives to your standard?

2. What kind of render scenes NEED caustics, otherwise they are not usable (for example, a pool without caustics is not usable. Jewelry render without caustics or dispersion is not usable)?

The best artists in the world are out there doing amazing renderings that I bet you can't match. And they are not complaining like a baby screaming and demanding features no one wants.

Caustics are not going to improve your renders.

2020-08-07, 15:45:54
Reply #20

johnmarc

  • Active Users
  • **
  • Posts: 23
    • View Profile
johnmarc, if you will continue insulting Corona team and other forum users, i will ask admins to suspend your account. There will be no more warnings.
« Last Edit: 2020-08-07, 16:36:14 by romullus »

2020-08-07, 16:35:05
Reply #21

lolec

  • Active Users
  • **
  • Posts: 155
    • View Profile
3 times longer than what?

I really hope your animation skills are better than your ability to answer simple questions.

Here they are again, made them shorter to help you out.

1. Which render engine has caustics that you like?

2. What kind of commercial render scenes would benefit the most from having caustics?


2020-08-07, 16:39:52
Reply #22

romullus

  • Global Moderator
  • Active Users
  • ****
  • Posts: 6676
  • Let's move this topic, shall we?
    • View Profile
    • My Models
lolec, i consider you to be very reasonable person, but please stop feeding this troll. I think it's pretty obvious that his intentions are not to seek for Corona improvement, but something else entirely.
I'm not Corona Team member. Everything i say, is my personal opinion only.
My Models | My Videos | My Pictures

2020-08-07, 16:41:42
Reply #23

lolec

  • Active Users
  • **
  • Posts: 155
    • View Profile
Romullus, you are right, it got to me. Done.

2020-08-08, 01:36:58
Reply #24

johnmarc

  • Active Users
  • **
  • Posts: 23
    • View Profile
I never insulted any corona team members only lolec the person who thinks the caustics solver is good which it is clearly not

2020-08-08, 02:08:45
Reply #25

TomG

  • Corona Team
  • Active Users
  • ****
  • Posts: 3338
    • View Profile
I beg to differ. I will speak personally here - calling Corona "mediocre" is something I find insulting, due to how hard we all work on the engine. Calling the caustics solver "terrible" I find insulting, due to how hard everyone worked on that and just how complex a topic that is.

I have yet to see you point to another engine that does it better - rather you point to some ideal situation that is likely not possible for any engine or any developers, where caustics are magically possible for every scenario while only adding a modicum of extra render time, and say that since it doesn't live up to that it's "terrible".

So, speaking personally, I am insulted by both of those comments, and the way they are delivered - not in a way hopeful for what improvements the future may bring and saying what you would love to have in there and giving pointers all while understanding the timescales and difficulties involved, but rather by simply denigrating what is there already. We're always open to suggestions for improvement, but calling something mediocre and something terrible is not a suggestion for an improvement, it is just that - name calling.

2020-08-08, 02:20:23
Reply #26

johnmarc

  • Active Users
  • **
  • Posts: 23
    • View Profile



just rendered this in rc1 and caustics just nuke entire scene with fireflies that never go away even after insane numbers of passes 500 plus frames take way too long even with 3970x

Tom G just admit the caustics solver needs major work 

2020-08-08, 02:22:06
Reply #27

TomG

  • Corona Team
  • Active Users
  • ****
  • Posts: 3338
    • View Profile
I am not here to "admit" anything to you. Of course there is always room for improvement. But that is not what is being discussed here, it's the way you put down what is there already.

2020-08-08, 15:16:45
Reply #28

maru

  • Corona Team
  • Active Users
  • ****
  • Posts: 10194
  • Marcin
    • View Profile
We are aware that the caustics solver is not bullet-proof. We will keep improving it.
It is absolutely fine to report bugs and request features, but we will not tolerate trolling and offending other users and our team via forum posts and private messages.

Moving this thread to feature requests, and locking it.