Author Topic: Daily Builds 1.6  (Read 134020 times)

2016-12-16, 18:48:27
Reply #225

Juraj

  • Active Users
  • **
  • Posts: 4768
    • View Profile
    • studio website
Temporary workaround is that if you use "alpha channel" as pass, it will be saved correctly as separate pass.

But the alpha saved in image will still be incorrect.

In my case the alpha gets wrong with DR and refractive glass ( I am using refraction override too, didn't check it if changes something ).
Please follow my new Instagram for latest projects, tips&tricks, short video tutorials and free models
Behance  Probably best updated portfolio of my work
lysfaere.com Please check the new stuff!

2016-12-18, 23:22:35
Reply #226

snakebox

  • Active Users
  • **
  • Posts: 493
    • View Profile
    • Snakebox Media
Temporary workaround is that if you use "alpha channel" as pass, it will be saved correctly as separate pass.

But the alpha saved in image will still be incorrect.

In my case the alpha gets wrong with DR and refractive glass ( I am using refraction override too, didn't check it if changes something ).

Does all of this happen in 32bit EXR too?  or is it only in 8/16bit PS ?

2016-12-19, 10:22:43
Reply #227

romullus

  • Global Moderator
  • Active Users
  • ****
  • Posts: 8882
  • Let's move this topic, shall we?
    • View Profile
    • My Models
Sometimes when pressing save button in VFB, dump exr dialog is open instead. This happens very rarely, like one time out of hundred and i don't know how to reproduce it. Can't remember when it happened for the first time, but i think it was with introduction of 1.6 dailies.
I'm not Corona Team member. Everything i say, is my personal opinion only.
My Models | My Videos | My Pictures

2016-12-19, 10:53:40
Reply #228

Christa Noel

  • Active Users
  • **
  • Posts: 911
  • God bless us everyone
    • View Profile
    • dionch.studio
Sometimes when pressing save button in VFB, dump exr dialog is open instead. This happens very rarely, like one time out of hundred and i don't know how to reproduce it. Can't remember when it happened for the first time, but i think it was with introduction of 1.6 dailies.
imho, maybe that case just when a lag of computer interface happen then we accidentally press-hold save button and the cursor dragged to the second choice "dump EXR". :)
« Last Edit: 2016-12-19, 10:57:21 by Christa Noel »

2016-12-23, 11:44:46
Reply #229

romullus

  • Global Moderator
  • Active Users
  • ****
  • Posts: 8882
  • Let's move this topic, shall we?
    • View Profile
    • My Models
imho, maybe that case just when a lag of computer interface happen then we accidentally press-hold save button and the cursor dragged to the second choice "dump EXR". :)

If that would happen when i hold and click on save, i wouldn't even report it. Thing is that this happens when i simply click on save and when that happens it's persistent - i close dialog, click again and it still opens dump exr. So no, i don't think that laggy computer, faulty mouse or sloppy user is in fault :]
I'm not Corona Team member. Everything i say, is my personal opinion only.
My Models | My Videos | My Pictures

2016-12-23, 11:59:03
Reply #230

romullus

  • Global Moderator
  • Active Users
  • ****
  • Posts: 8882
  • Let's move this topic, shall we?
    • View Profile
    • My Models
I wonder if anyone finds current implementation of feathered region render useful? Now if i try to use it, it gives nasty heavily undersampled pixels towards region edges, which makes this feature completely useless. I think Corona should sample feathered area as much as main region and gradually lower pixels opacity towards feather edge. That would allow seamless blending with already existing image and would give expected results. Or do i misunderstand its ussage?
I'm not Corona Team member. Everything i say, is my personal opinion only.
My Models | My Videos | My Pictures

2016-12-23, 19:13:38
Reply #231

denisgo22

  • Active Users
  • **
  • Posts: 700
    • View Profile
Some bug of the Corona triplanar map in the Bump material slot///
Same instance of the Triplanar map in the Bump slot rendering different:
max 2017 daily build 13/2016

« Last Edit: 2016-12-23, 19:30:57 by denisgo22 »

2016-12-23, 23:00:27
Reply #232

Bormax

  • Active Users
  • **
  • Posts: 568
    • View Profile
I wonder if anyone finds current implementation of feathered region render useful? Now if i try to use it, it gives nasty heavily undersampled pixels towards region edges, which makes this feature completely useless. I think Corona should sample feathered area as much as main region and gradually lower pixels opacity towards feather edge. That would allow seamless blending with already existing image and would give expected results. Or do i misunderstand its ussage?
I found this feature useful. I like it. I do not have the problem you described here. Corona spreads samples evenly on zoomed region area.
I still use Max 2014

2016-12-23, 23:43:18
Reply #233

romullus

  • Global Moderator
  • Active Users
  • ****
  • Posts: 8882
  • Let's move this topic, shall we?
    • View Profile
    • My Models
That's interesting. Could you show some example?
I'm not Corona Team member. Everything i say, is my personal opinion only.
My Models | My Videos | My Pictures

2016-12-23, 23:47:26
Reply #234

Bormax

  • Active Users
  • **
  • Posts: 568
    • View Profile
Some bug of the Corona triplanar map in the Bump material slot///
Same instance of the Triplanar map in the Bump slot rendering different:
max 2017 daily build 13/2016
Max 2014 - do not have this problem. Bump and diffuse maps are in the same position

2016-12-24, 00:06:48
Reply #235

Bormax

  • Active Users
  • **
  • Posts: 568
    • View Profile
That's interesting. Could you show some example?
Sure
Here it is

2016-12-24, 00:33:30
Reply #236

romullus

  • Global Moderator
  • Active Users
  • ****
  • Posts: 8882
  • Let's move this topic, shall we?
    • View Profile
    • My Models
Hmm, that doesn't look like fuzzy / feathered region render - noise distributions looks evenly spread across all region area. Besides you should do your test on relatively clean image, otherwise undersampled pixels in feathered region is much harder to spot.
I'm not Corona Team member. Everything i say, is my personal opinion only.
My Models | My Videos | My Pictures

2016-12-24, 00:58:11
Reply #237

Bormax

  • Active Users
  • **
  • Posts: 568
    • View Profile
This is another test

2016-12-24, 01:11:08
Reply #238

Bormax

  • Active Users
  • **
  • Posts: 568
    • View Profile
And here the same zoomed region rendered longer time to get cleaner result.
Is it what you want to see?

2016-12-24, 10:31:23
Reply #239

romullus

  • Global Moderator
  • Active Users
  • ****
  • Posts: 8882
  • Let's move this topic, shall we?
    • View Profile
    • My Models
It's either you don't know what feathered region render is, or there is something wrong with my installation... :/
I'm not Corona Team member. Everything i say, is my personal opinion only.
My Models | My Videos | My Pictures