Author Topic: ** Corona Animation ** -- relationship setting quality/time  (Read 104965 times)

2013-09-28, 17:33:19
Reply #105

rafpug

  • Active Users
  • **
  • Posts: 722
    • View Profile
Size  1920x1080

A = AA 5   Ast. 3  Ath. 0,06   | Time: 0:11:54  | Passes: 1
B = AA 5   Ast. 3  Ath. 0,05   | Time: 0:12:52  | Passes: 1
C = AA 5   Ast. 3  Ath. 0,04   | Time: 0:13:29  | Passes: 1
D = AA 6   Ast. 3  Ath. 0,06   | Time: 0:14:57  | Passes: 1
E = AA 6   Ast. 3  Ath. 0,05   | Time: 0:16:29  | Passes: 1
F = AA 6   Ast. 3  Ath. 0,04   | Time: 0:17:32  | Passes: 1


 | Time: 0:16:29  | X2 | X3 | = 97min.

E = AA 6   Ast. 3  Ath. 0,05   | Time: 1:34:16  | Passes: 1


2013-09-28, 19:15:41
Reply #106

rafpug

  • Active Users
  • **
  • Posts: 722
    • View Profile
Comparison

Method : BUCKET  PT+HD

by tests carried out, the result was as follows:


casuistry

a)   for negative values ​​of AA  greater noise
      for positive values  ​​of AA  minor   noise (uniform effect in relation to Ath.)
 
b)   AA>=Ast.     0,15 <Ath.> 0,02

c)   with values ​​of    3 <AA> 2   and   Ast.= 2      required minimum    5 Passes
      with values ​​of         AA= 4   and   Ast.= 2      required minimum 3/4 Passes

d)  with values ​​of         AA>= 4  and   Ast.>= 3   required minimum    1 Passes (Passes> 1  the higher the rendering time)


List, I report the comparison of some selected values on a base Size 1920x1080 :

A = AA 5   Ast. 3  Ath. 0,04   | Time: 1:19:22  | Passes: 1
B = AA 6   Ast. 3  Ath. 0,05   | Time: 1:34:16  | Passes: 1
C = AA 4   Ast. 4  Ath. 0,02   | Time: 1:58:57  | Passes: 1
D = AA 4   Ast. 2  Ath. 0,09   | Time: 1:37:03  | Passes: 3
E = AA 9   Ast. 3  Ath. 0,08   | Time: 2:01:09  | Passes: 1


Of course, the times are always in function of the hardware in use and the size of the rendering and setting GI secondary

2013-09-28, 19:39:34
Reply #107

rafpug

  • Active Users
  • **
  • Posts: 722
    • View Profile
TEST  with and without Bucket


Reference for comparison

with Bucket
B = AA 6   Ast. 3  Ath. 0,05   | Time: 1:34:16  | Passes: 1

2013-09-28, 23:03:30
Reply #108

rafpug

  • Active Users
  • **
  • Posts: 722
    • View Profile
Hi Ondra,

a question:
the method Bucket, is a method based on data compression?

Thanks
Raf

2013-09-29, 01:13:01
Reply #109

Ondra

  • Administrator
  • Active Users
  • *****
  • Posts: 8896
  • Turning coffee to features since 2009
    • View Profile
data compression? What? Probably a bad translation, but anyways, there is nothing even remotely related to data compression in rendering or corona
Rendering is magic.
Private scene uploader | How to get minidumps for crashed/frozen 3ds Max | Sorry for short replies, brief responses = more time to develop Corona ;)

2013-09-29, 03:01:57
Reply #110

rafpug

  • Active Users
  • **
  • Posts: 722
    • View Profile
Thank Ondra,

therefore, no compression pixels or data loss!
I asked because there is one thing that I would understand with you and Rawalanche.

Time to finish the final rendering with the Progressive method for comparison.

2013-09-29, 05:47:21
Reply #111

rafpug

  • Active Users
  • **
  • Posts: 722
    • View Profile
TEST  with and without Bucket

Reference for comparison

with method Bucket  PT+HD
A = AA 6   Ast. 3  Ath. 0,05   | Time: 1:34:16  | Passes: 1

with method Progressive  PT+HD
B = Time Limit: 1:34:16 | Passes: 98


Progressive with the method, the scene is brighter and the noise effect is more uniform.
Bucket with the method, the angle of the wall, we do not perceive

2013-09-29, 05:53:08
Reply #112

rafpug

  • Active Users
  • **
  • Posts: 722
    • View Profile
TEST  with and without Bucket

A B comparison

2013-09-29, 06:02:12
Reply #113

rafpug

  • Active Users
  • **
  • Posts: 722
    • View Profile
TEST  with and without Bucket

The image A, has left an artifact (in the middle at a right angle - this angle is repeated for all test method with Bucket)

Progressive with the method ( with Time Limit 2h or 3h , this artifact is not seen )

2013-09-29, 06:17:16
Reply #114

rafpug

  • Active Users
  • **
  • Posts: 722
    • View Profile
TEST  with and without Bucket

TEST N° 2

with method Bucket  PT+HD
A = AA 4   Ast. 4  Ath. 0,02   | Time: 1:58:57  | Passes: 1

with method Progressive  PT+HD
B = Time Limit: 2:38:56 | Passes: 168
C = Time Limit: 3:51:15 | Passes: 241  (Extrem)


The image A, is always dark,  has left an artifact ( the angle of the wall, we do not perceive)

2013-09-29, 06:25:17
Reply #115

rafpug

  • Active Users
  • **
  • Posts: 722
    • View Profile
Test result   --> Progressive - Bucket 


Progressive PT+HD 
the method is more stable and efficient


2013-09-29, 11:35:46
Reply #117

rafpug

  • Active Users
  • **
  • Posts: 722
    • View Profile
Good morning


Artifact

Test ---> Bucket - Progressive

2013-09-29, 11:39:58
Reply #118

rafpug

  • Active Users
  • **
  • Posts: 722
    • View Profile
This anomaly,
occurs on tests carried out by the method bucket

2013-09-29, 12:11:25
Reply #119

rafpug

  • Active Users
  • **
  • Posts: 722
    • View Profile
I add my observation of this test

Bucket that is really impressive!