Author Topic: Water caustics - fireflies  (Read 29758 times)

2017-03-14, 22:26:20

synolog

  • Active Users
  • **
  • Posts: 112
    • View Profile
Hi,
i'm testing caustics using:

- corona sun+sky
- water with simple plane and material with reflection+refraction(caustic)+opacity+displacement+volume, cubic projection
- solids with a diffuse only material

- render engine bidir/vcm

I've noticed some fireflies that are connected with the kind of lightsource and material parameters. Are they phys correct?

Thanks

2017-03-14, 23:02:45
Reply #1

cecofuli

  • Active Users
  • **
  • Posts: 1577
    • View Profile
    • www.francescolegrenzi.com
Try to add some glossy reflection to the water. I think that from 0.9 to 0.98 it's enough + a little bit of Bloom and Glare.

2017-03-15, 20:14:02
Reply #2

synolog

  • Active Users
  • **
  • Posts: 112
    • View Profile
Thanks but problem still present. Best using PPM engine

2017-03-15, 20:23:49
Reply #3

FrostKiwi

  • Active Users
  • **
  • Posts: 686
    • View Profile
    • YouTube
In Bidir, there are certain high energy rays, that will never (in human years ;]  ) average out against the middle bright values of the scene.

Use the firefly denoise option, it will clear the fireflies and leave the scene intact.
Also, if water is a simple plane, then refraction is not correct. Water needs to be a volume.
I'm 🐥 not 🥝, pls don't eat me ( ;  ;   )

2017-03-15, 21:27:13
Reply #4

synolog

  • Active Users
  • **
  • Posts: 112
    • View Profile
Hi, thanks, but:
Bidir: denoise function doesnt have any effect (in firefly and full mode).
Volume vs plane: plane caustics seems to be more natural; with volume caustics are still the same in deep water and surface.

2017-03-15, 23:34:58
Reply #5

FrostKiwi

  • Active Users
  • **
  • Posts: 686
    • View Profile
    • YouTube
Bidir: denoise function doesnt have any effect (in firefly and full mode).
Ohhh, I just noticed, that we are in C4D >.>
Pretty sure it does in max.

No problem though, there is a workaround in post. You can even use a free tool. I posted about the exact same situation in another thread.
I'm 🐥 not 🥝, pls don't eat me ( ;  ;   )

2017-03-16, 07:23:21
Reply #6

synolog

  • Active Users
  • **
  • Posts: 112
    • View Profile
Thank you, i'll follow your instructions.

edit:solved thanks
« Last Edit: 2017-03-16, 09:03:04 by synolog »

2017-03-29, 20:43:05
Reply #7

Lars Vader

  • Active Users
  • **
  • Posts: 15
    • View Profile
    • Lars Lubkowitz Visualization
I never even once managed to see any kind of caustics in my renders. Would anyone of you be so kind and give me some tips on how to achieve them? I followed your path but no caustics what so ever. I tried water as a simple plane and as a box, used displacement, checked "caustics" in the refraction channel and checked volume as well, although I'm not quite sure what values I'm supposed to use.

2017-03-29, 22:17:27
Reply #8

synolog

  • Active Users
  • **
  • Posts: 112
    • View Profile
I never even once managed to see any kind of caustics in my renders. Would anyone of you be so kind and give me some tips on how to achieve them? I followed your path but no caustics what so ever. I tried water as a simple plane and as a box, used displacement, checked "caustics" in the refraction channel and checked volume as well, although I'm not quite sure what values I'm supposed to use.

Hi, there are many ways to achieve caustics. Here an example:

1st-render settings: in performance settings enable devel/debug mode; in develpment/experimental stuff under basic configuration in render engine select Bidir/VCM.

2nd-model a swimming pool and assign a corona material with diffuse white 80%.

3rd-insert a plane for water surface: for it use a material with no diffuse; activate reflection; activate refraction; activate opacity with brightness 50%; activate displacement with c4d_water shader in texture; activate volume with color 5% cyan and absorption distance 5cm (or more depending of the plane water surface size), volume scattering deep blue color, brightness 25%, mix mode add, mix strength 50%.

4th-standard corona sky and corona sun, insert corona camera with exposure -3(EV).

I wish this will help you.

2017-03-29, 22:33:58
Reply #9

RolandB

  • Active Users
  • **
  • Posts: 335
  • Corona fan
    • View Profile
    • Béhance portfolio
1st-render settings: in performance settings enable devel/debug mode; in develpment/experimental stuff under basic configuration in render engine select Bidir/VCM.
Hi synolog
Thanks for your answer but (I'm on Max and with the 1.5), I've no devel/debug mode in performance settings. So without Bidir/VCM, is it possible to produce caustics (i don't think so, I've tried...)
Thanks
Roland
Portfolio on Béhance
http://www.behance.net/GCStudio

2017-03-29, 22:50:58
Reply #10

synolog

  • Active Users
  • **
  • Posts: 112
    • View Profile
In Cinema 4d you need Bidir/VCM or PPM. I dont know if the progressive mode in Corona for max 1.6 supports caustics and how to activate Bidir/VCM in Corona for Max 1.5. Maybe you could find more info in the Corona/Max forum.

2017-03-30, 07:25:13
Reply #11

Lars Vader

  • Active Users
  • **
  • Posts: 15
    • View Profile
    • Lars Lubkowitz Visualization
I never even once managed to see any kind of caustics in my renders. Would anyone of you be so kind and give me some tips on how to achieve them? I followed your path but no caustics what so ever. I tried water as a simple plane and as a box, used displacement, checked "caustics" in the refraction channel and checked volume as well, although I'm not quite sure what values I'm supposed to use.

Hi, there are many ways to achieve caustics. Here an example:

1st-render settings: in performance settings enable devel/debug mode; in develpment/experimental stuff under basic configuration in render engine select Bidir/VCM.

2nd-model a swimming pool and assign a corona material with diffuse white 80%.

3rd-insert a plane for water surface: for it use a material with no diffuse; activate reflection; activate refraction; activate opacity with brightness 50%; activate displacement with c4d_water shader in texture; activate volume with color 5% cyan and absorption distance 5cm (or more depending of the plane water surface size), volume scattering deep blue color, brightness 25%, mix mode add, mix strength 50%.

4th-standard corona sky and corona sun, insert corona camera with exposure -3(EV).

I wish this will help you.

Thank you, synolog!
I did everything you mentioned but still no result. It must be jinxed, I don't know. Seems so easy and yet it doesn't work.

Btw: Could you also tell us how you solved the firefly problem? I'd really like to know that once I figure out the damn caustics. :-D   

2017-03-30, 09:19:31
Reply #12

synolog

  • Active Users
  • **
  • Posts: 112
    • View Profile
I forgot the most important setting: in the refraction settings of the water material turn on (under refraction behavior) Caustic (slow). If it doent work send me your c4d file.
For fireflies look this https://forum.corona-renderer.com/index.php/topic,12284.msg89798.html#msg89798
« Last Edit: 2017-03-30, 09:24:35 by synolog »

2017-03-30, 20:48:53
Reply #13

Lars Vader

  • Active Users
  • **
  • Posts: 15
    • View Profile
    • Lars Lubkowitz Visualization
I forgot the most important setting: in the refraction settings of the water material turn on (under refraction behavior) Caustic (slow). If it doent work send me your c4d file.
For fireflies look this https://forum.corona-renderer.com/index.php/topic,12284.msg89798.html#msg89798
Sent you a PM. Still nothing. Curious to find out what my mistake was.

2017-03-31, 06:36:15
Reply #14

synolog

  • Active Users
  • **
  • Posts: 112
    • View Profile
Hi,

the problem in your file is the volume parameters of the water material:

In absorption color HSV parameters put H200 S2% V100%; Absorption brightness 100%; Absorption distance 8cm (if you increase this water is more transparent).

In scattering color HSV put H230, S100%, V15%.

Your file is from c4d R15 and if i save it with my R18 you cant open it.

Let me know if you solve