-
Bug
-
Resolution: Fixed
-
P2
-
fx2.0
There are lot of significant regressions in b2397 comparing to b2396.
The list of changes went in b2397 is:
1. Fix tests for API change inRT-14191 Add blendmode property to Node (detail)
2. API part ofRT-14191 Add blendmode property to Node (detail)
3. Fix apps for API change inRT-14191 Add blendmode property to Node (detail)
b2397 testing is not yet completed, so the current regressions are:
ColorfulShapes:
- Blur2Circle-Transparent: 13% (2.25 fps)
- Blur30Circle: 7% (0.75 fps)
Controls:
- Buttons-adhoc-nodes1024-skip8: 88% (50 fps)
DirtyArea:
- Test01_DDxx_xxxx_xxxx_xxxx: 92% (132 fps)
- Test06_DDDD_DDDD_DDDD_DDDD: 46% (9.2 fps) // the tests are run in full speed mode
FXTester:
- Path: 10% (2.16 fps)
I think we will have much more regressions once testing is completed.
I would suggest perhaps to rollback b2397 changes till we figure out why
new Node's blendmode property causes so huge performance regression.
Interesting, Text based benchamrks show improvements:
- FXTester.Text-shapes5000: 365% (16.72 fps)
- GUIMark2.Text: 8% (3.47 fps)
Link to Aurora results is:
http://aurora.russia.sun.com/performance/faces/ChessBoard.xhtml?reportName=FX2-graphics-scrum¶meters=%5Brefrelease%5D2.0%5Brefbuild%5D%3D+%272396%27%5Brefjdkrelease%5D1.6.0_23%5Brelease%5D%28pr.product.productRelease+IS+NOT+NULL%29%5Bbuild%5D%28pr.product.build+%3D+%272396%27%29OR%28pr.product.build+%3D+%272397%27%29%5Bjdkrelease%5D%28jdk.product.productRelease+IS+NOT+NULL%29&splitting=%5BX+axis%5DfxConf%2C+metricName%5BComplement%5Dbenchmark%2C+jdkBuild%2C+jdk%5BY+axis%5DbenchmarkName%2C+benchmarkConf%2C+fxRelease%2C+fxBuild%5BZ+axis%5Dos%2C+hwclass%2C+jdkRelease%2C+benchmarkSuite&reference=%5BOthers%5DfxRelease%2C+fxBuild%2C+jdkRelease%2C+jdkBuild%2C+jdk%2C+benchmarkSuite%2C+benchmarkName%2C+metricName%5BReference+Set%5Dbenchmark%2C+os%2C+benchmarkConf%2C+fxConf%2C+hwclass&mixReference=false&flags=&significance=empty&hideDataConfiguration=false&calculateSummary=false&showSummaryExpanded=false&showSummaryContents=true&showComplementAttributes=false&compactTables=true&viewStyle=chessboard&filter=
The list of changes went in b2397 is:
1. Fix tests for API change in
2. API part of
3. Fix apps for API change in
b2397 testing is not yet completed, so the current regressions are:
ColorfulShapes:
- Blur2Circle-Transparent: 13% (2.25 fps)
- Blur30Circle: 7% (0.75 fps)
Controls:
- Buttons-adhoc-nodes1024-skip8: 88% (50 fps)
DirtyArea:
- Test01_DDxx_xxxx_xxxx_xxxx: 92% (132 fps)
- Test06_DDDD_DDDD_DDDD_DDDD: 46% (9.2 fps) // the tests are run in full speed mode
FXTester:
- Path: 10% (2.16 fps)
I think we will have much more regressions once testing is completed.
I would suggest perhaps to rollback b2397 changes till we figure out why
new Node's blendmode property causes so huge performance regression.
Interesting, Text based benchamrks show improvements:
- FXTester.Text-shapes5000: 365% (16.72 fps)
- GUIMark2.Text: 8% (3.47 fps)
Link to Aurora results is:
http://aurora.russia.sun.com/performance/faces/ChessBoard.xhtml?reportName=FX2-graphics-scrum¶meters=%5Brefrelease%5D2.0%5Brefbuild%5D%3D+%272396%27%5Brefjdkrelease%5D1.6.0_23%5Brelease%5D%28pr.product.productRelease+IS+NOT+NULL%29%5Bbuild%5D%28pr.product.build+%3D+%272396%27%29OR%28pr.product.build+%3D+%272397%27%29%5Bjdkrelease%5D%28jdk.product.productRelease+IS+NOT+NULL%29&splitting=%5BX+axis%5DfxConf%2C+metricName%5BComplement%5Dbenchmark%2C+jdkBuild%2C+jdk%5BY+axis%5DbenchmarkName%2C+benchmarkConf%2C+fxRelease%2C+fxBuild%5BZ+axis%5Dos%2C+hwclass%2C+jdkRelease%2C+benchmarkSuite&reference=%5BOthers%5DfxRelease%2C+fxBuild%2C+jdkRelease%2C+jdkBuild%2C+jdk%2C+benchmarkSuite%2C+benchmarkName%2C+metricName%5BReference+Set%5Dbenchmark%2C+os%2C+benchmarkConf%2C+fxConf%2C+hwclass&mixReference=false&flags=&significance=empty&hideDataConfiguration=false&calculateSummary=false&showSummaryExpanded=false&showSummaryContents=true&showComplementAttributes=false&compactTables=true&viewStyle=chessboard&filter=
- duplicates
-
JDK-8113883 3D programs rendering is broken on Windows
- Closed