How Funcom Increased Play Time in Lego Minifigures by 40%

Copyright © 2015, Intel Corporation. All rights reserved. *Other names and brands may be claimed as the property of others.
Antoine Cohade - Intel
Sergio de los Santos - Funcom
March 04, 2015
Power Efficient Programming:
How Funcom increased play time in Lego
Minifigures by 40%.
80%

 Copyright © 2015 Intel Corporation. All rights reserved.
 *Other names and brands may be claimed as the property of others.
 INFORMATION IN THIS DOCUMENT IS PROVIDED IN CONNECTION WITH INTEL PRODUCTS. NO LICENSE, EXPRESS OR IMPLIED,
BY ESTOPPEL OR OTHERWISE, TO ANY INTELLECTUAL PROPERTY RIGHTS IS GRANTED BY THIS DOCUMENT. EXCEPT AS PROVIDED IN INTEL'S TERMS AND CONDITIONS OF SALE FOR SUCH
PRODUCTS, INTEL ASSUMES NO LIABILITY WHATSOEVER AND INTEL DISCLAIMS ANY EXPRESS
OR IMPLIED WARRANTY, RELATING TO SALE AND/OR USE OF INTEL PRODUCTS INCLUDING LIABILITY OR WARRANTIES RELATING TO FITNESS FOR A PARTICULAR PURPOSE, MERCHANTABILITY, OR
INFRINGEMENT OF ANY PATENT, COPYRIGHT OR OTHER INTELLECTUAL PROPERTY RIGHT.
 A "Mission Critical Application" is any application in which failure of the Intel Product could result, directly or indirectly, in personal injury or death. SHOULD YOU PURCHASE OR USE INTEL'S
PRODUCTS FOR ANY SUCH MISSION CRITICAL APPLICATION, YOU SHALL INDEMNIFY AND HOLD INTEL AND ITS SUBSIDIARIES, SUBCONTRACTORS AND AFFILIATES, AND THE DIRECTORS,
OFFICERS, AND EMPLOYEES OF EACH, HARMLESS AGAINST ALL CLAIMS COSTS, DAMAGES, AND EXPENSES AND REASONABLE ATTORNEYS' FEES ARISING OUT OF, DIRECTLY OR INDIRECTLY, ANY
CLAIM OF PRODUCT LIABILITY, PERSONAL INJURY, OR DEATH ARISING IN ANY WAY OUT OF SUCH MISSION CRITICAL APPLICATION, WHETHER OR NOT INTEL OR ITS SUBCONTRACTOR WAS
NEGLIGENT IN THE DESIGN, MANUFACTURE, OR WARNING OF THE INTEL PRODUCT OR ANY OF ITS PARTS.
 Intel may make changes to specifications and product descriptions at any time, without notice.
 All products, dates, and figures specified are preliminary based on current expectations, and are subject to change without notice.
 Intel processors, chipsets, and desktop boards may contain design defects or errors known as errata, which may cause the product to deviate from published specifications. Current characterized
errata are available on request.
 Any code names featured are used internally within Intel to identify products that are in development and not yet publicly announced for release. Customers, licensees and other third parties are not
authorized by Intel to use code names in advertising, promotion or marketing of any product or services and any such use of Intel's internal code names is at the sole risk of the user.
 Intel product plans in this presentation do not constitute Intel plan of record product roadmaps. Please contact your Intel representative to obtain Intel’s current plan of record product roadmaps.
 Performance claims: Software and workloads used in performance tests may have been optimized for performance only on Intel® microprocessors. Performance tests, such as SYSmark and
MobileMark, are measured using specific computer systems, components, software, operations and functions. Any change to any of those factors may cause the results to vary. You should consult
other information and performance tests to assist you in fully evaluating your contemplated purchases, including the performance of that product when combined with other products. For more
information go to
http://www.Intel.com/performance
 Iris™ graphics is available on select systems. Consult your system manufacturer.
 Intel, Intel Inside, the Intel logo, Intel Core and Iris are trademarks of Intel Corporation in the United States and other countries.
Legal
2

 Why power matters
 General power overview
 Lego Minifigures: Power overview
 How and what to measure
 Best practices
 Lego Minifigures: Power saving mode features
Agenda
3

Why power matters
 Best practices
Agenda
4

Copyright © 2015, Intel Corporation. All rights reserved. *Other names and brands may be claimed as the property of others.Copyright © 2015, Intel Corporation. All rights reserved. *Other names and brands may be claimed as the property of others.
 User reviews: “Kills the battery.”
“Drain way too much battery life.”
“It simply DESTROYS battery life on my sgs2.”
“Kills battery and limits where one can play.”
“Can kill your battery in a blink of an eye.”
“This app is a huge battery drain.”
Why power matters
5

Why power matters
140
77 100
47 47
15
500
170
40
0
100
200
300
400
500
600
700
Extreme Gaming Desktop
( HSW-E w/ Radeon R9
295x2)
Gaming Desktop
(Core i7 3rd gen w/ Nvidia
GTX 760)
Xbox One* Gaming Laptop
(Core i7 Mobile 4th gen w/
Nvidia GTX 850M)
Gaming laptop
(Core i7 w/ Iris Pro
Graphics)
Mainstream Laptop
(Core i5 w/ Intel HD
Graphics 4400)
Power(W)
GPU + CPU TDP: From high-end to mainstream
CPU TDP GPU TDP
6
Source : See backup

 Integrated CPU and GPU share the same power envelope:
 Even if a game is 100% GPU limited, CPU workload can impact performance
 Case study:
 Synthetic workload
 GPU limited
 Running some heavy kernels
on the CPU
Why power matters
30
35
40
45
50
55
0 1 2 3 4 5 6 7 8
FPS CPU Threads
Impact of CPU utilization/power on Graphics Performance
4 cores 8 Threads system with Integrated Graphics
4 cores 8 Threads system with Discrete Graphics
7
Source : Internal test results

 Devices have varying power behaviors
Why power matters
8
0
2
4
6
8
10
12
14
16
0 3 6 9 12 15 18 21 24 27 30 33 36 39 42 45 48 51 54 57 60 63 66 69 72 75 78 81 84 87 90 93 96 99 102105108
Power(W)
Time (s)
Maximum power variation : clamshell vs. tablet mode
Power-Info-IA Core Power(Watts) Power-Info-Integrated Graphics Power(Watts)
Power-Info-Rest of Package Power(Watts) Power-Info-Package Power(Watts)
Source : Internal test using a Lenovo® ThinkPad Yoga

 The user cares
 Battery life (Lower use time)
 User experience
 Fan-less / Low TDP design
 Power optimized titles will run on lower TDP platforms
 Increases total available market
Why power matters
9

General power overview
 Best practices
Agenda
10

General power overview: ACPI in a nutshell
 SoC components (each core, graphics) have sleep states
 The deeper the component sleeps, the more power is saved
 When active, components have Performance States
 Each P-State has a given voltage and frequency
 Package C-State is determined by the higher of the cores
 Any awake component will keep the package awake
11

General power overview: C-States
C0 (active)
C0 (active)
C0 (active)
Cn
Frame n
CPU C-States
Gfx C-States
Package C-
States
C0 (active)
C0 (active)
C0 (active)
Cn
Frame n+1
C0 (active)
C0 (active)
C0 (active)
Cn
Frame n+2
C0 (active)
C0 (active)
C0 (active)
Cn
Frame n
12

C0 (active)
C0 (active) C0 (active)
RC..
C0 (act …
Cn
RC6 RC6
Frame n Frame n+1 Frame n+...
CPU C-States
Gfx C-States
Package C-
States
C0 (active) C0 (active)Cn
RC0 (active) RC6 RC0 (active)
13

CPU C-States
Gfx C-States
Package C-
States
C0
RC0
C0 (active)
RC6
Frame n Frame n+1
Cn
Cn
RC6 RC0 RC0RC6
C0 Cn C0
C0 (active) Cn C0 (active)
Package deep C-States -> Max Power Saving
Frame n+...
14

Lego Minifigures : Power overview
 Best practices
 Lego Minifigures : Power saving mode features
Agenda
15

Lego Minifigures: Power overview
16
Frames
GPU queue
CPU threads
Other metrics

How and what to measure
 Best practices
Agenda
17

Come and meet us at the Intel Hangout: 4pm today
 There are multiple tools for power and perf. analysis:
 Intel® INDE for graphics Analysis
 Intel® VTune™ Amplifier for in-depth CPU analysis
 Intel® SoC Watch (Intel® System Studio) for power states
How and Watt to measure
18

Best practices
 Lego Minifigures: Power saving mode optimizations
Agenda
19

 Numerous processors and platforms
 Differences can impact performance
 Take these factors into account and build performance bins
 Profile the platform at install time
 GPU detect / Run some short game snippet
 Select the optimal tradeoff performance/settings
Best practices: Adapt to your platform
20

Easiest way to save power, and also most efficient
– DirectX - HRESULT IDXGISwapChain::Present(SyncInterval, Flags);
– OpenGL ES - EGLBoolean eglSwapInterval(EGLDisplay display, EGLint interval);
2x less work
Cap frame rates for menus/static scenes even lower
Best practices: Cap the frame rate
Interval = 2 Interval = 1
FPS FPS Battery Life increase
Cut The Rope 2 30.2 40.3 111%
Epic Citadel 31.1 50.7 179%
Intel Fast Blur 31.0 63.0 133%
Zombie Tsunami 30.1 60.4 136%
21
Source : Internal Intel test results on a Intel® Atom™ Processor Z3745 reference design

Best practices: Find a reasonable resolution
0
10
20
30
40
50
60
70
80
800x600 1024*768 1152*864 1280*768 1360*768 1440*900 1600*900
Energy/Frame(mJ/Frame)
Resolution
mJoules / frame as a function of resolution
800*600
22

 2560x1600x60 fps => 1GB/sec
 Design points for some of the
tablets in the marketplace is 2-3W
 Changing RT format can
help save 15-20% power
Best practices: Keep bandwidth under control
[CELLRANGE]
[CELLRANGE]
[CELLRANGE]
0.40
0.45
0.50
0.55
0.60
0.65
0.70
0.75
0.80
0.85
2.87 4.01 5.12
Power(W)
Bandwidth (GB/s)
HDR Rendering :
Memory Power as a function of Bandwidth
23
Source : Internal Intel® test results on a core i7-4610Y w/ 2x2GB LPDDR3 1600

Best practices: Be careful with spin loops
Widely used:
 Reduce input latency
 Thread pools
Prevents CPU from sleeping
 Avoid using them as much as possible
 If really necessary, use the pause instruction
https://software.intel.com/en-us/articles/benefitting-power-and-performance-sleep-loops
24

Handle loss of focus
Color buffer clears
2D game optimizations using depth
Reduce CPU activity
Use of next gen APIs
Best practices: A lot more…
25

 Best practices
Lego Minifigures: Power saving mode features
Agenda
26

Funcom previous games…
Both Age of Conan and The Secret
World were designed for a mature
hardcore audience, with high end PCs.
We aimed big, at the time, with the
latest visual features… but the price
was to require high end PCs to play.
Mobile was never a target…

 Lego Minifigures is aimed for kids (PEGI
7)
 Kids play on many devices, mostly low
end PCs and tablets.
 We implemented a new render
engine, designed to run in low power
devices.
 We still like high end features, but we
need to be able to scale up and down
Lego Minifigures:

 Cap the framerate to 30Hz
 Lower quality lighting and shadows
 No anisotropic filtering
 Disable post process FXs
 Disable antialiasing
Lego Minifigures: Power saving mode optimizations
29

 Vsync vs. Sleep
 Results:
Before After Power Saved (%)
System Power (W) 16.7 12.3 26%
Soc Power (W) 10.1 6.6 35%
Memory Power (W) 1.4 1.1 25%
Lego Minifigures: Frame capping
30

 Optimization performed:
 Lower resolution shadow-maps
 Simple filtering
 Disabled AVSM
 Results:
SoC Power (W) 6.6 6.1 8%
Lego Minifigures: Low quality shadows
31

Lego Minifigures: Low quality shadows
32

 Optimizations performed:
 Still deferred – but simplified – lighting
 Render only dynamic objects to shadow map
 Disable HBAO
 Results:
Lego Minifigures: Simple lighting and shadows
SoC Power (W) 6.1 5.1 16%
33

Lego Minifigures: Simple lighting and shadows
34

 We disable all post process (DoF, GodRays, CMAA)
 Results :
SoC Power (W) 5.1 4.8 5%
Lego Minifigures: Disable post process effects
35

Lego Minifigures: Disable post process effects
36

Lego Minifigures: Final overview
Power Saving Mode Battery life Battery life Increase
Asus TP 300LD
(NVidia GT820M)
OFF 01:53:01
79%
ON 03:22:04
System with
HD Graphics 5300
OFF 01:53:02
103%
ON 03:49:04
37
Source : Internal Intel® battery rundown tests

 Users care about battery life: let them choose!
 Power optimization is performance optimization
 Focused optimizations can bring you terrific battery gain
 Come and meet us at the Intel Hangout for more info
 antoine.cohade@intel.com / sergios@funcom.com
Conclusion
38

C o p y r i g h t © 2 0 1 5 , I n t e l C o r p o r a t i o n . A l l r i g h t s r e s e r v e d . *O t h e r n a me s a n d b r a n d s ma y b e c l a i me d a s t h e p r o p e r t y o f o t h e r s .

 GPU Detect : https://software.intel.com/en-us/vcsource/samples/gpu-detect
 Loops with Pause : https://software.intel.com/en-us/articles/benefitting-power-and-
performance-sleep-loops
 Intel® System Studio 2015: https://software.intel.com/en-us/intel-system-studio
 Intel® GPA : https://software.intel.com/en-us/gpa
 Power Explorer: https://software.intel.com/en-us/blogs/2013/10/29/power-explorer
 Intel® Power Gadget: https://software.intel.com/en-us/articles/intel-power-gadget-20
 DX12 Siggraph Demo: http://blogs.msdn.com/b/directx/archive/2014/08/13/directx-12-
high-performance-and-high-power-savings.aspxt
Useful links
40

41
Slide 6:
 Extreme Gaming Desktop :
 http://ark.intel.com/products/82930/Intel-Core-i7-5960X-Processor-Extreme-Edition-20M-Cache-
up-to-3_50-GHz
 http://www.extremetech.com/computing/180043-amd-r9-295x2-review-500-watts-of-cool-
running-4k-gaming-goodness-with-a-price-tag-to-match
 Gaming Desktop :
 http://ark.intel.com/products/65719/Intel-Core-i7-3770-Processor-8M-Cache-up-to-3_90-GHz
 http://www.anandtech.com/show/7103/nvidia-geforce-gtx-760-review
 Xbox One:
 http://www.engadget.com/2013/05/21/hands-on-with-xbox-one/
 Gaming Laptop:
 http://ark.intel.com/products/78937/Intel-Core-i7-4810MQ-Processor-6M-Cache-up-to-3_80-GHz
 http://www.techpowerup.com/gpudb/2538/geforce-gtx-850m.html
 Gaming Laptop:
 http://ark.intel.com/products/83503/Intel-Core-i7-4980HQ-Processor-6M-Cache-up-to-4_00-GHz
 Mainstream Laptop:
 http://ark.intel.com/products/80343/Intel-Core-i5-4310U-Processor-3M-Cache-up-to-3_00-GHz
Sources:

How Funcom Increased Play Time in Lego Minifigures by 40%

More Related Content

What's hot

Viewers also liked

Similar to How Funcom Increased Play Time in Lego Minifigures by 40%

Recently uploaded

How Funcom Increased Play Time in Lego Minifigures by 40%