DECAF is a tool that systematically analyze and dissect the performance of cloud gaming platforms across different game genres and game platforms. DECAF is highly automated and requires minimum manual intervention. By applying DECAF, we measure the performance of three commercial cloud gaming platforms including Google Stadia, Amazon Luna, and Nvidia GeForceNow, and uncover a number of important findings.
Our work is published ACM SIGMETRICS 2022. For details on DECAF capabilities and the interesting findings on the performance of three cloud gaming providers, read the paper available here.
Citation:
@article{10.1145/3491043,
author = {Iqbal, Hassan and Khalid, Ayesha and Shahzad, Muhammad},
title = {Dissecting Cloud Gaming Performance with DECAF},
year = {2021},
issue_date = {December 2021},
publisher = {Association for Computing Machinery},
address = {New York, NY, USA},
volume = {5},
number = {3},
url = {https://doi.org/10.1145/3491043},
doi = {10.1145/3491043},
journal = {Proc. ACM Meas. Anal. Comput. Syst.},
month = {dec},
articleno = {31},
numpages = {27},
keywords = {streaming bitrate, measurement, deep learning, network utilization, cloud gaming, game bot, performance evaluation, latency}
}
For data collection, DECAF is compatible with Windows operating system.
First step is to download or git clone decaf from github in your C directory. It should look like C:\decaf
Please install the following:
- Python3: Install using MS store or PowerShell
- Scapy: Run the command in PowerShell
python3 -m pip install scapy - Win32api: Run the command in PowerShell
python3 -m pip install pywin32 - Install FFmpeg for windows by following instructions here
- Install WireShark from here
Make sure that the WireShark is installed in direcotory C:\Program Files\Wireshark. Otherwise, find the directory where WireShark is installed and update it in the launcher\paths.json by replacing the value of key "tshark".
Go the C:\decaf directory and run the script create_directories.py. This will create all the required directories.
Download instrumented chromium from here.
Inside the C:\chromium\src\out\ that is already created, extract the downloaded chromium. The extracted files will be in folder titled Default.
To run chromium, open the Windows PowerShell in administrator mode and change directory to C:\chromium\src. Then run the command .\out\Default\chrome.exe --no-sandbox. This will launch chromium. Browse gaming platform such as Google Stadia and start the game say Far Cry 5.
Launcher runs various components of Decaf such as game bot, FFmpeg, wireshark to collect the data. Once the game is running in chromium, e.g., Far Cry 5 in Stadia, run the launcher.py to run to start the game bot and start collecting data. The usage is as follows:
Usage: python3 launcher.py platform game total_runtime_min bot_runtime_min folder_name
For example, if command python3 launcher.py stadia fc5 10 8 testing is run, it implies that fc5 will be played by game bot on stadia for 8 minutes and the it will launcher.py stop running after 10 minutes. The data collected will be stored in folder named test which will be present at C:\decaf\dataset\stadia\test. The bot_runtime_min can be less than or equal to total_runtime_min.
Currently, DECAF can perform data collection for three game: Far Cry 5, Assassin's Creed Valhalla, and Crew/Crew2. It is tested for three platforms: Google Stadia, Amazon Luna, and Nvidia Geforce Now.
These game setting are required for gamebot.
Set forward movement of the character from w to mouse middle button
Set forward movement of the character from w to mouse middle button
Enable on-screen keyboard and place it on top left of the screen
To process chromium logs and game recording, refer to readme.md in directory data_processing.