OctaneRender™ Standalone v1.0 beta3.02a [obsolete]

A forum where development builds are posted for testing by the community.
Forum rules
NOTE: The software in this forum is not %100 reliable, they are development builds and are meant for testing by experienced octane users. If you are a new octane user, we recommend to use the current stable release from the 'Commercial Product News & Releases' forum.
Post Reply
User avatar
matej
Licensed Customer
Posts: 2083
Joined: Fri Jun 25, 2010 7:54 pm
Location: Slovenia

abstrax wrote: Why don't you use just one specular material for the droplets?

Regarding the speed drop: How does it compare to 3.01? 2.59 doesn't support instances which is probably the main reason why it's faster.
It's an old project, I don't really remember why I set the material so. I probably had an "artistic" reason :D

In 301 it renders in 13:22, so 42 seconds slower than 302. Instances are not used in any of these (note at the bottom of the screenshot it says 1 meshes). But as I understand the new framework that allows instances causes some slowdown even if instances are not used?
SW: Octane 3.05 | Linux Mint 18.1 64bit | Blender 2.78 HW: EVGA GTX 1070 | i5 2500K | 16GB RAM Drivers: 375.26
cgmo.net
Top
User avatar
abstrax
OctaneRender Team
Posts: 5510
Joined: Tue May 18, 2010 11:01 am
Location: Auckland, New Zealand

matej wrote:
abstrax wrote: Why don't you use just one specular material for the droplets?

Regarding the speed drop: How does it compare to 3.01? 2.59 doesn't support instances which is probably the main reason why it's faster.
It's an old project, I don't really remember why I set the material so. I probably had an "artistic" reason :D

In 301 it renders in 13:22, so 42 seconds slower than 302. Instances are not used in any of these (note at the bottom of the screenshot it says 1 meshes). But as I understand the new framework that allows instances causes some slowdown even if instances are not used?
Thanks for the feedback. The speed numbers roughly match what you get on Windows, which is good.

Regarding instances: Yes, with beta 3.00 instances were introduced that slow down rendering speed by 10-20%, even if no instances are used. Usually it still renders faster if all geometry is in one mesh node than distributed over several instances. The reason of the slowdown even if you are not using instances is that some additional logic needed to be inserted in various places which is not for free. We will see what we can do about it (if at all), but that's how it is for now.

Cheers,
Marcus
In theory there is no difference between theory and practice. In practice there is. - Yogi Berra
Top
User avatar
pixelrush
Licensed Customer
Posts: 1618
Joined: Mon Jan 11, 2010 7:11 pm
Location: Nelson, New Zealand

I guess the performance loss is an acceptable trade off for instances capability. ;)
It would be nice to work around the additional logic if you aren't regularly using instances though.
You don't forsee any further significant losses do you? :roll: I mean hopefully Maxwell generation won't be as disappointing as Kepler. We want to keep moving forward despite the odd bit of code creep.
i7-3820 @4.3Ghz | 24gb | Win7pro-64
GTS 250 display + 2 x GTX 780 cuda| driver 331.65
Octane v1.55
Top
User avatar
marcio_max
Licensed Customer
Posts: 132
Joined: Mon Dec 05, 2011 12:17 pm
Location: Aracatuba/SP/Brazil

Benchmark GTX 670 4GB :)
Attachments
Bench_GTX-670-4GB.JPG
| Windows 8 Pro x64 | Octane for 3ds Max/Standalone | i7 860 | 12GB RAM | 1x EVGA GTX 670 4GB / 2x EVGA GTX 670 2GB | Cooler Master HAF 932 |
Top
User avatar
matej
Licensed Customer
Posts: 2083
Joined: Fri Jun 25, 2010 7:54 pm
Location: Slovenia

marcio_max wrote:Benchmark GTX 670 4GB :)
Try pathtracing and PMC ;) (please :) )
SW: Octane 3.05 | Linux Mint 18.1 64bit | Blender 2.78 HW: EVGA GTX 1070 | i5 2500K | 16GB RAM Drivers: 375.26
cgmo.net
Top
User avatar
smicha
Licensed Customer
Posts: 3151
Joined: Wed Sep 21, 2011 4:13 pm
Location: Warsaw, Poland

I have strange cracks after importing obj into Octane (2.58e and 3.01, 3.02a tested - same problem.) I attach ocs (and .skp model). Could you please test it? Why I see these holes?
Attachments
3.jpg
home.rar
(160.71 KiB) Downloaded 114 times
3090, Titan, Quadro, Xeon Scalable Supermicro, 768GB RAM; Sketchup Pro, Classical Architecture.
Custom alloy powder coated laser cut cases, Autodesk metal-sheet 3D modelling.
build-log http://render.otoy.com/forum/viewtopic.php?f=9&t=42540
Top
User avatar
abstrax
OctaneRender Team
Posts: 5510
Joined: Tue May 18, 2010 11:01 am
Location: Auckland, New Zealand

smicha wrote:I have strange cracks after importing obj into Octane (2.58e and 3.01, 3.02a tested - same problem.) I attach ocs (and .skp model). Could you please test it? Why I see these holes?
You see these holes, because your geometry in the OBJ file uses n-gons (i.e. polygons with more than 3 or 4 vertices). Octane tries to triangulate them, but it uses a fairly simple algorithm which doesn't always work with concave n-gons.

Solution: Triangulate the geometry during or before the export to OBJ.

Cheers,
Marcus
In theory there is no difference between theory and practice. In practice there is. - Yogi Berra
Top
User avatar
smicha
Licensed Customer
Posts: 3151
Joined: Wed Sep 21, 2011 4:13 pm
Location: Warsaw, Poland

Marcus,

I'd used triangulation before and I tried it again in two different ways - once in SU and once during exporting. Both did't solve my problem. I still see cracks in Octane.
3090, Titan, Quadro, Xeon Scalable Supermicro, 768GB RAM; Sketchup Pro, Classical Architecture.
Custom alloy powder coated laser cut cases, Autodesk metal-sheet 3D modelling.
build-log http://render.otoy.com/forum/viewtopic.php?f=9&t=42540
Top
User avatar
abstrax
OctaneRender Team
Posts: 5510
Joined: Tue May 18, 2010 11:01 am
Location: Auckland, New Zealand

smicha wrote:Marcus,

I'd used triangulation before and I tried it again in two different ways - once in SU and once during exporting. Both did't solve my problem. I still see cracks in Octane.
Then it's probably best, if you could send me the scene (see my PM).

Cheers,
Marcus
In theory there is no difference between theory and practice. In practice there is. - Yogi Berra
Top
User avatar
abstrax
OctaneRender Team
Posts: 5510
Joined: Tue May 18, 2010 11:01 am
Location: Auckland, New Zealand

Rickky wrote:any chance to show us a test scene with the index/fresnel effect.
Are you sure it works ?
Ooops, I get it now :)
Yes, of course we are sure it works ;) In the following examples the six balls have roughness 0, 0.2, 0.4, 0.6, 0.8, 1.0 (left to right) and I modified only the specular value and index of refraction:

specular = 1.0
index = 2.0 (strong Fresnel reflection)
spec_1.0_ior_2.0.png
specular = 0.3
index = 3.0 (very strong Fresnel reflection)
spec_0.3_ior_3.0.png
specular = 1.0
index = 1.5 (normal Fresnel reflection)
spec_1.0_ior_1.5.png
specular = 0.3
index = 0.0 (no Fresnel reflection)
spec_0.3_ior_0.0.png
specular = 1.0
index = 0.0 (no Fresnel reflection)
spec_1.0_ior_0.0.png
I hope that clears the questions.

Cheers,
Marcus
In theory there is no difference between theory and practice. In practice there is. - Yogi Berra
Top
Post Reply

Return to “Development Build Releases”