JDK-8094329 : 3D rendering issues seen with Mac OS and Intel Iris chipset
  • Type: Bug
  • Component: javafx
  • Sub-Component: graphics
  • Affected Version: 8u5
  • Priority: P3
  • Status: Resolved
  • Resolution: Not an Issue
  • Submitted: 2014-07-14
  • Updated: 2015-06-12
  • Resolved: 2014-07-15
The Version table provides details related to the release that this issue/RFE will be addressed.

Unresolved : Release in which this issue/RFE will be addressed.
Resolved: Release in which this issue/RFE has been resolved.
Fixed : Release in which this issue/RFE has been fixed. The release containing this fix may be available for download as an Early Access Release or a General Availability Release.

To download the current JDK release, click here.
JDK 8
8u40Resolved
Related Reports
Relates :  
JDK-8094825 - [Mac,Retina,3D] Lighting is wrong (too dark, incorrect highlight) with pixel scaling
Description
3D samples show some sort of lighting issue with the Intel Iris chipset.
Note in the attached samples that the lower face is dark.

java -ea -DproxyHost=www-proxy.us.oracle.com -DproxyPort=80 -cp /Users/ddhill/sb/insanity/sdkmac/lib/ext/jfxrt.jar:dist/FX8-3DAPI.jar meshviewer.BoxOrientation

or 
  meshviewer.CylinderOrientation
  meshviewer.DefaultMaterial
Comments
I agree we should resolve this as "Not an issue" and file new Intel Iris related issue if we found one in the future.
15-07-2014
Playing with the color calibration on the Mac Pro (manually, I don't have a colorimeter handy). Settings->Display Color Tab Calibrate Button click expert mode I found that the gamma was set to a common "suggested" default of 2.2, and the "natural" gamma of the LCD is 2.0. Tweaking the profile, I certainly was able to bring the LCD display to be closer to the HDMI displays results. To be fair - without a colorimeter, ALL of my displays are showing various levels of inaccurate :-) ref: http://en.wikipedia.org/wiki/Colorimetry#Color_temperature_meter (There are a lot of tools out there for calibrating screens) To my reported problem - the lighting of the samples listed, in particular how soon you see the shaded face - I think this is "not an issue" because: * I can tweak the native LCD display from its default settings to get closer to the "expected" results. * a screen grab from the ARM device looks darker when displayed on my Mac laptop LCD for an approximately given angle of the rotated box. Said another way - the screen grab from the ARM which shows a brighter shadowed face, looks dark on my Mac, and looks very close to what I see when running the test app on my MAC when displaying the same approximate rotation. There may be other issues with the Iris chipset, but this is jira is not describing one of them.
15-07-2014
There are other issues related to the Iris GPU, e.g. RT-37740. Seems like either the GPU itself is "bad", or there are bugs in the driver.
15-07-2014
Looking at the screen grabs on my mac lcd monitor, I almost could not see the hidden face, just like when running the application. At Chien's suggestion, I plugged in my HDMI display, as was able to see side by side the LCD and the HDMI display result. The two were quite different. I could see a hidden/dark face a number of degrees (5 ? 10 ?) on the HDMI display sooner than on the LCD display. This suggests that it is a display color correction issue. It is not brightness because I played with that to no effect. http://www.lagom.nl/lcd-test/black.php suggests to me that is it the black level that is the issue, as I can not tell in the LCD display up to #8, and the HDMI display #2. Note: when I say LCD - I mean the built in LCD on the macbook, contrasted to the HDMI display LCD :-)
14-07-2014
Two comparision framegrabs from an i.MX6
14-07-2014