Closed Bug 984396 Opened 11 years ago Closed 3 years ago

High power consumption on animation from old Vimeo homepage (no video)

Categories

(Core :: Graphics: Canvas2D, defect)

x86
Windows 8.1
defect

Tracking

()

RESOLVED INCOMPLETE

People

(Reporter: rvitillo, Unassigned)

References

()

Details

(Keywords: power, Whiteboard: [Power:P2][platform-rel-Vimeo])

Attachments

(1 file)

Idling on the vimeo homepage seems to consume significantly more energy than on IE.
Attached file Testcase 1
The power profile on http://people.mozilla.org/~rvitillo/dashboard/#vimeo shows that the GPU power consumption of FF is about 7 times greater than IE. The culprit seems to be the canvas based particle generator shown in the testcase.
Component: General → Canvas: 2D
Whiteboard: [Power]
Summary: High power consumption on Vimeo → High power consumption on Vimeo homepage (no video playing)
I tested the animation (which doesn't appear to no longer be on the vimeo.com homepage) on Mac. Nightly: > total W = _pkg_ (cores + _gpu_ + other) + _ram_ W > #01 15.75 W = 12.38 ( 1.36 + 3.89 + 7.13) + 3.37 W > > 1 sample taken over a period of 30.000 seconds > > Name ID CPU ms/s User% Deadlines (<2 ms, 2-5 ms) Wakeups (Intr, Pkg idle) GPU ms/s > com.apple.Terminal 224 461.32 84.93 48.31 164.04 > firefox 69583 149.46 73.64 0.07 0.13 71.74 39.25 150.91 > plugin-container 69585 310.66 86.62 3.63 3.07 12.56 8.76 13.14 > Terminal 359 5.19 88.96 0.00 0.00 0.57 0.27 0.00 > com.apple.WindowServer 68 97.08 67.41 38.55 253.53 > WindowServer 142 98.93 72.23 7.16 0.03 67.41 38.55 253.54 > kernel_coalition 1 22.79 240.44 131.88 0.00 > kernel_task 0 19.87 0.00 6.90 0.00 240.31 131.82 0.00 Chrome: > total W = _pkg_ (cores + _gpu_ + other) + _ram_ W > #01 9.89 W = 7.84 ( 1.44 + 0.57 + 5.83) + 2.05 W > > 1 sample taken over a period of 30.000 seconds > > Name ID CPU ms/s User% Deadlines (<2 ms, 2-5 ms) Wakeups (Intr, Pkg idle) GPU ms/s > com.google.Chrome 309 580.30 453.17 178.12 33.65 > Google Chrome Helper 69651 359.29 86.93 119.97 0.00 180.52 25.69 0.00 > Google Chrome Helper 69616 124.32 65.45 0.00 0.00 118.87 106.04 33.65 > Google Chrome 69612 96.45 67.94 92.54 0.17 153.79 46.39 0.00 > com.apple.WindowServer 68 81.45 64.01 31.06 25.02 > WindowServer 142 81.57 61.53 3.90 0.07 64.02 31.06 25.02 > kernel_coalition 1 40.63 339.57 159.02 0.00 > kernel_task 0 38.07 0.00 4.13 0.03 339.27 158.79 0.00 Safari: > total W = _pkg_ (cores + _gpu_ + other) + _ram_ W > #01 6.98 W = 5.14 ( 0.39 + 0.31 + 4.45) + 1.84 W > > 1 sample taken over a period of 30.000 seconds > > Name ID CPU ms/s User% Deadlines (<2 ms, 2-5 ms) Wakeups (Intr, Pkg idle) GPU ms/s > com.apple.Safari 310 151.16 64.98 56.58 14.52 > com.apple.WebKit.WebContent 69691 150.64 80.90 0.00 0.00 63.28 55.05 14.52 > Safari 69684 0.50 53.51 0.00 0.00 1.07 1.03 0.00 > com.apple.WebKit.Networking 69687 0.15 43.56 0.00 0.00 0.43 0.37 0.00 > com.apple.WebKit.WebContent 69689 0.04 37.20 0.00 0.00 0.20 0.13 0.00 > com.apple.WindowServer 68 66.67 61.88 53.51 17.64 > WindowServer 142 66.71 69.64 1.67 0.13 61.91 53.52 17.65 > kernel_coalition 1 13.48 151.71 75.87 0.00 > kernel_task 0 10.83 0.00 3.60 0.00 151.48 75.68 0.00 We do quite badly. In particular we hit the GPU a lot harder than Chrome and Safari.
Summary: High power consumption on Vimeo homepage (no video playing) → High power consumption on animation from old Vimeo homepage (no video)
Whiteboard: [Power] → [Power:P2]
platform-rel: --- → ?
Whiteboard: [Power:P2] → [Power:P2][platform-rel-Vimeo]
platform-rel: ? → ---
Severity: normal → S3

All the testcases don't work anymore so it's hard to say if that is still a problem with recent versions of Firefox. As such closing the bug as incomplete.

Status: NEW → RESOLVED
Closed: 3 years ago
Resolution: --- → INCOMPLETE
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: