Manta Interactive Ray Tracer Development Mailing List

Text archives Help


Re: [Manta] r1869 - in trunk: . Core Core/Color Core/Containers Core/Exceptions Core/Geometry Core/Math Core/Persistent Core/Thread Core/Util Core/XUtils DynLT Engine Engine/Control Engine/Display Engine/Factory Engine/IdleModes Engine/ImageTraversers Engine/LoadBalancers Engine/PixelSamplers Engine/Renderers Engine/SampleGenerators Engine/Shadows Image Interface Model/AmbientLights Model/Backgrounds Model/Cameras Model/Groups Model/Instances Model/Materials Model/MiscObjects Model/Primitives Model/Readers Model/TexCoordMappers Model/Textures SCIRun StandAlone SwigInterface UserInterface scenes scenes/galileo tests


Chronological Thread 
  • From: James Bigler <bigler@cs.utah.edu>
  • Cc: manta@sci.utah.edu
  • Subject: Re: [Manta] r1869 - in trunk: . Core Core/Color Core/Containers Core/Exceptions Core/Geometry Core/Math Core/Persistent Core/Thread Core/Util Core/XUtils DynLT Engine Engine/Control Engine/Display Engine/Factory Engine/IdleModes Engine/ImageTraversers Engine/LoadBalancers Engine/PixelSamplers Engine/Renderers Engine/SampleGenerators Engine/Shadows Image Interface Model/AmbientLights Model/Backgrounds Model/Cameras Model/Groups Model/Instances Model/Materials Model/MiscObjects Model/Primitives Model/Readers Model/TexCoordMappers Model/Textures SCIRun StandAlone SwigInterface UserInterface scenes scenes/galileo tests
  • Date: Thu, 06 Dec 2007 12:37:56 -0700
If you look at this again, the numbers have settled down as well:

http://dashboard.sci.utah.edu:8081/Manta/Dashboard/Test?testname=.Test.StandAlone.DefaultSceneNoDisplayBench_NP1&submissionid=201&measurement=total_time
http://dashboard.sci.utah.edu:8081/Manta/Dashboard/Test?testname=.Test.StandAlone.DefaultSceneNoDisplayBench_NP4&submissionid=201&measurement=total_time

James

Steven G. Parker wrote:

On Nov 30, 2007, at 2:33 PM, Solomon Boulos wrote:

Looking at the dashboard, this commit somehow reduced performance "noticeably":

http://dashboard.sci.utah.edu:8081/Manta/Dashboard/Test?testname=.Test.StandAlone.DefaultSceneNoDisplayBench_NP1&submissionid=183&measurement=frames_per_second

Any thoughts?

I have spent some time looking at this, and it looks like it was not this commit. It coincides with the day that I updated to 10.4.11 and rebooted. We should get some more machines on the test list so that we can isolate these a bit better.


[taz:~/Develop/Manta/macopt] sparker% bin/manta -bench (HEAD)
Benchmark completed in 3.71269 seconds (100 frames, 26.9346 frames per second)
[taz:~/Develop/Manta/macopt] sparker% cd ../macopt2  (-r1868)
[taz:~/Develop/Manta/macopt2] sparker% bin/manta -bench
Benchmark completed in 3.72277 seconds (100 frames, 26.8617 frames per second)

Steve



Archive powered by MHonArc 2.6.16.

Top of page