Author Topic: ** Corona Animation ** -- relationship setting quality/time  (Read 103697 times)

2013-09-12, 22:39:03
Reply #45

rafpug

  • Active Users
  • **
  • Posts: 722
    • View Profile
Compare  PT+HD   PT+PT

Time limit   00:01:20  ( X2 )



PT+HD =  Time: 0:01:27 | Passes: 10

PT+PT  =  Time: 0:02:40 | Passes: 12

2013-09-12, 23:06:43
Reply #46

rafpug

  • Active Users
  • **
  • Posts: 722
    • View Profile
Compare  PT+HD   PT+PT

Time limit   00:01:20  ( X5 )


PT+HD =  Time: 0:01:27 | Passes: 10

PT+PT  =  Time: 0:06:00 | Passes: 28

2013-09-12, 23:37:13
Reply #47

rafpug

  • Active Users
  • **
  • Posts: 722
    • View Profile
Compare  PT+HD   PT+PT

Time limit   00:01:20  ( X7 )



PT+HD   =  Time: 0:01:27 | Passes: 10

PT+PT    =  Time: 0:08:40 | Passes: 39

2013-09-12, 23:57:49
Reply #48

rafpug

  • Active Users
  • **
  • Posts: 722
    • View Profile
The test proceeds in the opposite direction

PT+PT  PT+HD

Time limit   00:01:20  ( Xn )


PT+PT  = Time: 0:08:40 | Passes: 39

PT+HD = Time: 0:01:27 | Passes: 10 

2013-09-13, 22:20:55
Reply #49

rafpug

  • Active Users
  • **
  • Posts: 722
    • View Profile
Compare  PT+PT   PT+HD


Time limit   00:01:20  ( X3 )  = 00:03:60



PT+PT  = Time: 0:08:40 | Passes: 39

PT+HD = Time: 0:04:07 | Passes: 29 

2013-09-13, 22:41:02
Reply #50

rafpug

  • Active Users
  • **
  • Posts: 722
    • View Profile
Test  Renderer : PROGRESSIVE

While keeping fixed the values ​​of the secondary GI and moving from method to method  PT+PT  PT+HD  the test result was as follows :


PT+HD the method is faster and well defined !

2013-09-13, 23:41:55
Reply #51

rafpug

  • Active Users
  • **
  • Posts: 722
    • View Profile
Speed ratio PT+HD is 1,79 compared to the method PT+PT

About twice as fast

2013-09-14, 00:06:50
Reply #52

Ondra

  • Administrator
  • Active Users
  • *****
  • Posts: 8889
  • Turning coffee to features since 2009
    • View Profile
Speed ratio PT+HD is 1,79 compared to the method PT+PT

About twice as fast
that is extremely dependent on a scene. I think you overall focus too much on a specific scene and a specific implementation of HD cache, that will certainly change in the future ;)
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 ;)

2013-09-14, 10:11:36
Reply #53

rafpug

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

This test I need to figure out how to optimize time video/photo

Here, the comparison function of time
Time limit   00:01:20  ( X9 )  = 10,8  (11min.)


Method Progressive PT+HD   Aspect Ratio 16:9  Size 1920x1080


(11min.)
PT+HD = Time: 0:11:40 | Passes: 12   --> 0,8869 sec. / xPasses

(20min.)
PT+HD = Time: 0:20:40 | Passes: 23   --> 0,8869 sec. / xPasses   

(30min.)
PT+HD = Time: 0:30:39 | Passes: 35   --> 0,8682 sec. / xPasses   

(40min.)
PT+HD = Time: 0:40:41 | Passes: 43   --> 0,9397 sec. / xPasses   


The mean 0,8955 sec. / xPasses


With a size of 1920x1080
the CPU did not have temperature problems was maintained within the limits   Temperature CPU °67/°68

Beforehand:
Bucket testing with PT+PT + size 640x480 and Time limit 18min
noted repeatedly warning the seventh minute onwards   CPU °72



http://www.youtube.com/watch?v=TDQmUj8-no4

2013-09-14, 10:31:55
Reply #54

rafpug

  • Active Users
  • **
  • Posts: 722
    • View Profile
Speed ratio PT+HD is 1,79 compared to the method PT+PT

About twice as fast
that is extremely dependent on a scene. I think you overall focus too much on a specific scene and a specific implementation of HD cache, that will certainly change in the future ;)

Until now I have never used Vray, Maxwell, Indigo, ect .. to the end, because they seemed too slow!

I always simulated the scene with standard lights max and various tricks

But CoronaRender is another way ^ _ ^...

2013-09-14, 18:03:59
Reply #55

rafpug

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

I applied a light personalized .ies

Using either the default PT+PT  PT+HD light central difficulty to fade. (right / left not) you see in the image.

The lights are clones as instances


Thanks
Raf

2013-09-14, 19:54:10
Reply #56

rafpug

  • Active Users
  • **
  • Posts: 722
    • View Profile
Hello
I saw that changing the values ​​of Corona Setting,

in Material Editor slot defaul CoronaMtl changes appearance.

So everything is according to the setting parameters of the interface?


This aspect, which may affect the final scene and rendering speed?


Greats
Raf

2013-09-14, 20:27:20
Reply #57

maru

  • Corona Team
  • Active Users
  • ****
  • Posts: 9035
  • Marcin
    • View Profile

2013-09-14, 21:18:15
Reply #58

rafpug

  • Active Users
  • **
  • Posts: 722
    • View Profile
No, it's just quality of materials displayed in mat editor. :)

Hi maru  ^_^

Thank you very much
Raf

*..maru all right?*

2013-09-15, 10:14:02
Reply #59

rafpug

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

The next Test of the method Bucket,
waiting to be released version Alpha6 because I do not have access to the daily build

Bucket ---> PT+PT
Bucket ---> PT+HD

base reference Thread
http://corona-renderer.com/forum/index.php/topic,1380.msg10176.html#msg10176