Customer Cases
Pricing

Mobile Game Performance Testing in 2026: Complete Guide with PerfDog Insights from Tencent’s Founding Developer

Master mobile game optimization with insights from PerfDog’s founding developer. Learn to analyze 200+ metrics including Jank, Smooth Index, and FPower. The definitive 2026 guide for Unity & Unreal Engine developers to achieve 120FPS and reduce battery drain.
 

Last Updated: March 2026

Reading Time: 12 minutes

Author: Baojian Shen, Senior Product Manager at Tencent WeTest

Expert Interview: Awen Cao, Chairman of MTSC Forum & PerfDog Founding Developer

 


 

Table of Contents

  1. What is PerfDog and Why It Matters
  2. The PUBG Mobile Challenge: How PerfDog Was Born
  3. 200+ Performance Metrics Explained
  4. Three Revolutionary Metrics: Jank, Smooth, and FPower
  5. Industry Adoption: 200,000+ Apps Tested
  6. Expert Q&A: Future of Mobile Performance Testing
  7. PerfDog vs Traditional Tools: 2026 Comparison
  8. Getting Started: Free Trial and Best Practices

 


 

What is PerfDog and Why It Matters in 2026

Quick Answer: PerfDog is a plug-and-play mobile performance testing tool developed by Tencent WeTest that has tested over 200,000+ apps and conducted 20+ million tests worldwide without requiring ROOT/Jailbreak access.

Why Game Developers Choose PerfDog

The Problem: Traditional mobile performance testing requires:

  • ❌ ROOT access (Android) or Jailbreak (iOS)
  • ❌ Debug mode compilation (1+ day for UE engine games)
  • ❌ Multiple tools for different platforms
  • ❌ Inconsistent metrics across devices

The Solution: PerfDog provides:

  • Zero-setup testing: Plug & play, no ROOT/Jailbreak needed
  • 11 platform support: Mobile, Pad, PC, PS, Xbox, TV, VR, Switch, Simulator, Car Console
  • 200+ metrics: CPU, GPU, FPS, Memory, Battery, Temperature, Network
  • Universal compatibility: Works with all game engines (Unity, Unreal, Cocos, proprietary)

Industry Recognition (2026 Data)

140+ Major Game Companies:

  • Tencent, NetEase, miHoYo, Supercell, NCSoft
  • 37Games, Lilith, FunPlus, Riot Games
  • Garena, Netmarble

Hardware Partners:

  • Top 10 Phone Brands: Samsung, Apple, Honor, Xiaomi, Oppo, Vivo, Lenovo, Motorola
  • Chip Companies: Qualcomm, ARM, MediaTek, HiSilicon, Imagination

Global Reach:

  • ? Millions of developers worldwide
  • ? 200,000+ apps tested
  • ? 20,000,000+ test sessions

 

The PUBG Mobile Challenge: How PerfDog Was Born

The 2018 Problem That Changed Mobile Gaming QA

Background: When PUBG: Battlegrounds exploded globally in 2018 (boosting PC hardware sales by 40%), Tencent’s game development team faced an unprecedented challenge: port this next-gen quality game to mobile using Unreal Engine.

 

 

Awen Cao’s Team Faced:

iOS Testing Nightmare

  • Xcode Limitation: Only worked in Mac/debug environments
  • Compilation Hell: 1+ day to compile million-line UE code in debug mode
  • No Benchmarking: Zero competitive performance data
  • No Industry Standards: No reference metrics

Android Fragmentation Chaos

  • Thousands of Tools: Each manufacturer had custom testing tools
  • Zero Consistency: No unified analysis framework
  • Accuracy Issues: Frequent retests and misjudgments
  • Platform Variance: Customized system versions across OEMs

The Solution: PerfDog’s Birth

Development Timeline:

  1. 2018: Internal Tencent project for PUBG Mobile
  2. 2019: Public release after successful internal adoption
  3. 2020: Global launch at GDC (Game Developers Conference)
  4. 2026: Industry standard with 200,000+ apps tested

Core Philosophy: “Build a simple, highly usable tool supporting all mobile platforms with consistent, accurate metrics.”

 

200+ Performance Metrics Explained 

 

 

8 Major Metric Categories

1. System Performance Metrics

  • CPU Usage: Normalized for Android frequency changes
  • Memory Usage: RAM, Texture Memory, Mono Memory (Unity)
  • Disk I/O: Read/write performance

Why It Matters: CPU normalization solves Android’s frequent CPU frequency throttling, giving accurate performance readings even under thermal constraints.

2. Graphics & Rendering Metrics

  • Frame Rate (FPS): Average, Min, Max, Variance
  • GPU Usage: Utilization percentage, GPU time
  • Draw Calls: Number of rendering commands per frame
  • Texture Memory: VRAM usage

Real-World Impact: Helps identify whether lag is CPU-bound, GPU-bound, or memory-bound.

3. Power & Temperature Metrics

  • Battery Power: Current draw (mA), power consumption (W)
  • Temperature: Battery, CPU, GPU thermal readings
  • FPower: Revolutionary metric (explained below)

Industry Benchmark: Top mobile games target <2°C temperature rise per hour.

4. User Experience Metrics

  • App Launch Time: Cold start, warm start
  • Responsiveness: Touch input latency
  • UI Smoothness: Jank and Stutter detection

Player Impact: 53% of mobile gamers uninstall apps that lag (2025 industry survey).

5. Network Testing Metrics

  • Latency: Ping, jitter
  • Packet Loss: Connection stability
  • Network Speed: Upload/download bandwidth

Mobile Gaming Critical: 70% of mobile games are online multiplayer.

6. Application-Specific Metrics

  • Crash/Exception Monitoring: Integrated with CrashSight
  • Resource Usage: App-specific counters
  • Custom Data Extension: Developer-defined metrics

Developer Flexibility: Add up to 50 custom metrics via API.

7. Deep Analysis Metrics

  • CPU/GPU Profiling: Thread-level analysis
  • Memory Tracking: Allocation tracking, leak detection
  • Bottleneck Identification: Automated performance issue detection

Advanced Use Case: Unity Mono GC events, Unreal stat commands integration.

8. Custom Data Extension

  • API Integration: C++, C#, Java, Python, Lua support
  • Real-time Sync: 1-2% CPU overhead for 20,000 calls/sec
  • Data Types: Float, Integer, String

Example: Track in-game economy metrics (gold earned/hour) alongside technical metrics.

 

Three Revolutionary Metrics: Jank, Smooth, and FPower 

 

 

1. PerfDog Jank: Why FPS Alone is Misleading

The Apple WWDC 18 Revelation

Discovery: At WWDC 18, Apple demonstrated that high FPS ≠ smooth experience.

Example:

  • Left: 40 FPS game with obvious stutter (one frame >100ms)
  • Right: 30 FPS game with perfect smoothness (consistent 33ms/frame)

Conclusion: Frame time consistency matters more than raw FPS.

Google’s Project Butter (Android 4.4)

Google’s Jank Definition: “If display doesn’t refresh for a subsequent v-sync, it’s a jank.”

Problem: Google’s method didn’t match user perception in real-world gaming.

PerfDog’s Enhanced Jank Algorithm

Evolution:

  1. PerfDog Jank (2019): Improved detection based on visual perception + frame time
  2. PerfDog SmallJank (2020): Detects micro-stutters on 120Hz+ displays

Calculation Logic:

 

 

Jank occurs when:
- FrameTime > 2× average of last 3 frames
- AND FrameTime > 84ms (Single Jank) or 125ms (Big Jank)

 

Industry Adoption: Now the de facto standard for mobile game smoothness testing.

Real-World Benchmark (10-minute gameplay):

Device

Graphics

Jank Count

Big Jank

Stutter %

iPhone 7 Plus

Medium

1.8

0.2

0.03%

OPPO R11

High

2.1

1.2

0.04%

Redmi Note 4X

Medium

4.2

2.9

0.07%

(Source: PerfDog industry benchmarks for leading mobile games)

 

 

2. Smooth Index: Beyond Jank

 

Why Smooth Index Was Created

Problem: Two games with same FPS and Jank count can feel different.

Example Scenario:

  • Game A: 3 janks, each 100ms
  • Game B: 3 janks, two 90ms + one 200ms

User Perception: Game B feels worse despite same jank count.

Smooth Index Solution: Weighs jank duration and severity.

Formula (simplified):

 

 

Smooth Index = 100 - (Weighted Jank Severity Score)

 

Target: Smooth Index >95 for AAA mobile games.

 

3. Frame Power (FPower): The Game-Changer for Power

Optimization

The Developer Request

Problem: “We need quantitative metrics to assess power consumption performance without sacrificing user experience.”

Traditional Approach Failure:

  • Reduce frame rate → Lower power → Worse experience ❌
  • Reduce graphics quality → Lower power → Player complaints ❌

FPower: The Scientific Solution

Definition: Energy consumption per frame

Formula:

 

 

FPower = Total Power Consumption (W) / Frame Rate (FPS)

 

Key Insight: “Frame rate depends on waiting + computation. Power consumption depends on computation + activation. Only by optimizing computation can power be truly optimized.”

Optimization Goal: Reduce power consumption without reducing FPS.

Real-World FPower Optimization Case Study

Before Optimization:

  • FPS: 60
  • Power: 3.6W
  • FPower: 60mW/frame

After Optimization:

  • FPS: 60 (unchanged)
  • Power: 2.8W (-22%)
  • FPower: 46.7mW/frame (-22%)

Result: Same user experience, 22% longer battery life.

Industry Impact: FPower is now a KPI (Key Performance Indicator) for power optimization at major studios.

 

Industry Adoption: 200,000+ Apps Tested

 

How PerfDog Built a Robust User Ecosystem

Phase 1: Game Developers (2018-2019)

Initial Use Case: PUBG Mobile development team

  • Cross-platform testing (iOS + Android)
  • Competitive benchmarking
  • No ROOT/Jailbreak requirement

Result: Internal Tencent adoption across 100+ projects.

Phase 2: OEM Partners (2019-2020)

Partnership Model: Hardware manufacturers (Samsung, Xiaomi, Oppo, Vivo) sent flagship devices to game teams for testing.

Mutual Benefit:

  • Game Developers: Test on real devices
  • OEMs: Identify hardware weaknesses before launch
  • Players: Better optimized games

Phase 3: Tech Influencers & Players (2020-2021)

Social Media Impact: YouTubers and tech reviewers started using PerfDog for:

  • Device performance reviews
  • Game optimization analysis
  • Purchase recommendations

Why Players Prefer PerfDog:

  • Traditional Benchmarks: Show hardware capability (AnTuTu, Geekbench)
  • PerfDog: Shows real gaming experience

Influence: Device performance rankings based on PerfDog tests impact millions of purchase decisions.

Phase 4: Chip Manufacturers (2021-2022)

Strategic Partnerships:

  • Qualcomm: GPU counter integration for Adreno GPUs
  • ARM: Mali GPU utilization metrics
  • Imagination: PowerVR GPU analysis

Result: Hardware-level metrics now accessible via PerfDog.

Phase 5: Beyond Gaming (2022-Present)

Expanded Applications:

  • Video streaming apps (YouTube, Netflix)
  • Browser performance (Chrome, Safari)
  • IoT devices
  • Smart TVs
  • VR/AR applications

Authoritative Certification: Collaborated with National Institute of Metrology for metric accuracy validation.

 

Expert Q&A: Future of Mobile Performance Testing 

Interview with Awen Cao, PerfDog Founding Developer

Chairman of MTSC (Make Tester Super Cool) Forum

Senior Testing Technology Director at Tencent

 

Q1: What deeper-level metrics can we expect from PerfDog?

Awen’s Answer:

"Absolutely, we aim to provide users with more comprehensive and in-depth low-level metric information. We are actively collaborating with hardware manufacturers like Qualcomm and Imagination.

In the near future, we hope PerfDog will be able to offer users detailed information at the hardware and even driver levels, facilitating better and faster performance issue identification.

Example: Currently, PerfDog shows ‘GPU 80% usage.’ Future versions will show:

  • Vertex shader utilization
  • Fragment shader utilization
  • Memory bandwidth bottlenecks
  • Specific GPU block bottlenecks (TMU, ROP, etc.)

 

Q2: For new game projects, which performance metrics should developers focus on first?

Awen’s Answer:

"Games are incredibly complex from an app perspective, second only to operating systems. During game testing, looking at just one or a few metrics won’t reveal the true performance standards.

We often advise examining a comprehensive range of metrics.

Furthermore, differences between game types mean performance metrics can vary widely, so it’s not advisable to arbitrarily set standards for projects.

My Recommendation: Use industry benchmarks from similar competitors as a reference for optimization. PerfDog already offers this functionality, and I encourage users to explore it.

Example Workflow:

  1. Identify 3-5 top-performing competitor games in your genre
  2. Run PerfDog benchmarks on those games
  3. Compare your game’s metrics against competitor averages
  4. Prioritize optimization for metrics where you lag significantly

PerfDog Feature: Competitive Analysis Dashboard (Premium Feature)"

 

Q3: Can PerfDog test multiple devices simultaneously with one account?

Awen’s Answer:

"Testing multiple phones with a single account on one computer is already a feature of PerfDog.

By launching PerfDog software multiple times on a PC, you can test up to three phones simultaneously.

Why the 3-device Limit?

Users might wonder why it’s limited to three devices. Given the extensive performance metrics and UI displayed during testing, testing too many phones simultaneously could lead to:

  • ❌ Incomplete test results
  • ❌ System overloads
  • ❌ Data synchronization issues

We recommend testing on a maximum of three devices.

Pro Tip: For testing more devices, use PerfDog Service (automation API) which supports parallel testing on 10+ devices via CI/CD pipeline integration."

 

PerfDog vs Traditional Tools: 2026 Comparison

Comprehensive Feature Comparison

Feature

PerfDog

Xcode Instruments

Android Studio Profiler

GameBench

Setup Required

None (Plug & Play)

Mac + Debug mode

Debug mode

App install

ROOT/Jailbreak

❌ Not required

❌ Not required

❌ Not required

❌ Not required

iOS Support

✅ Full

✅ Full (Mac only)

❌ No

✅ Limited

Android Support

✅ Full

❌ No

✅ Full

✅ Full

PC Games

✅ Yes

❌ No

❌ No

✅ Yes

Console (PS/Xbox)

✅ Yes

❌ No

❌ No

❌ No

Metric Count

200+

50+

80+

30+

FPower Metric

✅ Yes

❌ No

❌ No

❌ No

Jank Analysis

✅ Advanced

✅ Basic

✅ Basic

✅ Basic

GPU Counters

✅ Qualcomm, Mali, PowerVR

✅ Metal only

✅ Limited

❌ No

Real-time Testing

✅ Yes

✅ Yes

✅ Yes

✅ Yes

Custom Metrics

✅ API support

❌ No

❌ No

❌ No

Multi-device Testing

✅ 3 simultaneous

❌ 1 at a time

❌ 1 at a time

✅ Multiple

Automation API

✅ PerfDog Service

❌ No

❌ No

✅ Yes

Competitive Benchmarking

✅ Built-in

❌ No

❌ No

✅ Limited

Price

Free + Premium

Free (Mac required)

Free

Subscription

 

Key Differentiators

Why PerfDog Leads in 2026

1. Platform Universality

  • ✅ Only tool supporting iOS + Android + PC + Console + VR
  • ✅ No platform lock-in (works everywhere)

2. Zero Setup Friction

  • ✅ No compilation required (vs 1+ day for Xcode debug)
  • ✅ No ROOT/Jailbreak (major security benefit)

3. Unique Metrics

  • ✅ FPower: Energy per frame (industry-first)
  • ✅ Advanced Jank: SmallJank for 120Hz+ displays
  • ✅ Smooth Index: Weighted smoothness scoring

4. Hardware-Level Integration

  • ✅ Qualcomm Adreno GPU counters
  • ✅ ARM Mali GPU utilization
  • ✅ PowerVR analysis

5. Industry Adoption

  • ✅ 140+ major game companies
  • ✅ Top 10 phone brands + chip manufacturers

 

Getting Started: Free Trial and Best Practices

Quick Start Guide (5 Minutes)

Step 1: Download PerfDog

Official Website: https://perfdog.wetest.net

Supported OS:

  • Windows 7/8/10/11
  • macOS 10.13+
  • Ubuntu 18.04+

File Size: ~50MB

 

Step 2: Connect Device

Mobile (iOS/Android):

  1. Enable USB Debugging (Android) or Trust Computer (iOS)
  2. Connect via USB cable
  3. PerfDog auto-detects device

PC Games:

  1. Launch PerfDog
  2. Select target process from dropdown

Console:

  1. Connect dev kit via network
  2. Enter IP address in PerfDog

 

Step 3: Start Testing

Basic Testing:

  1. Click “Start Test” button
  2. Launch your game/app
  3. Play for 5-10 minutes
  4. Stop test
  5. Review metrics dashboard

Advanced Testing:

  1. Enable specific metrics (GPU counters, custom data)
  2. Set performance targets (FPS threshold, temperature limit)
  3. Configure alerts (notify when metric exceeds threshold)
  4. Export data (CSV, Excel, JSON)

 

Best Practices for Accurate Testing

1. Device Preparation

Before Testing:

  • ✅ Fully charge device (>80%)
  • ✅ Close background apps
  • ✅ Disable auto-brightness
  • ✅ Connect to stable Wi-Fi (for online games)
  • ✅ Wait 5 minutes for device to stabilize

Why: Background processes and low battery affect CPU throttling.

 

2. Test Duration

Minimum Duration by Game Type:

  • Hyper-casual: 3 minutes
  • Mid-core: 10 minutes
  • AAA/MMO: 20-30 minutes

Why: Thermal throttling typically occurs after 10-15 minutes.

 

3. Multiple Test Runs

Best Practice: Run 3-5 test sessions and average results.

Why: Device performance varies based on:

  • Thermal state
  • Background processes
  • Network conditions
  • OS updates

Statistical Validity: 3+ runs reduce variance by 40%.

 

4. Competitive Benchmarking

Workflow:

  1. Identify 3-5 top competitors in your genre
  2. Download and test competitor games
  3. Run identical test scenarios (same level, same duration)
  4. Compare metrics side-by-side

PerfDog Feature: Benchmarking Dashboard saves competitor data for easy comparison.

 

5. Automation for Regression Testing

Use Case: Test every build to catch performance regressions early.

Solution: PerfDog Service API

Example Workflow (Python):

 

 

from perfdog import PerfDogService

# Initialize
pd = PerfDogService(api_key="your_key")

# Start test
session = pd.start_test(
    device="device_id",
    app="com.yourcompany.game",
    metrics=["FPS", "CPU", "GPU", "Memory", "FPower"],
    duration=600  # 10 minutes
)

# Wait for completion
result = pd.wait_for_result(session.id)

# Assert performance targets
assert result.avg_fps >= 58, "FPS regression detected"
assert result.avg_fpower <= 50, "Power efficiency regression"

# Generate report
pd.export_report(session.id, format="PDF")

 

Supported Languages: C#, C++, Java, Python, PHP, Go, JavaScript

 

PerfDog Service: Industrial Performance Management

What is PerfDog Service?

An enterprise-grade automation API for integrating PerfDog into CI/CD pipelines.

Key Features:

1. Multi-Device Parallel Testing

  • Test 10+ devices simultaneously
  • Distributed test execution
  • Cloud device farm integration

2. CI/CD Integration

  • Jenkins plugin
  • GitHub Actions support
  • GitLab CI/CD integration

3. Automated Reporting

  • PDF/Excel/HTML reports
  • Email/Slack notifications
  • Dashboard visualization

4. Historical Trending

  • Track performance across builds
  • Regression detection
  • Performance baseline management

Pricing: Contact WeTest for enterprise quote.

 

Integration with WeTest Ecosystem

Complete Quality Assurance Solution

PerfDog is part of WeTest’s comprehensive game QA platform:

1. PerfDog (Performance Testing)

  • Use Case: Development & testing phase
  • Method: Plug & play tool
  • Data Source: Specific test devices

 

2. PerfSight (Performance Monitoring)

  • Use Case: Live operations phase
  • Method: SDK integration
  • Data Source: All player devices (real-world data)

Workflow: Test with PerfDog → Launch → Monitor with PerfSight

 

3. CrashSight (Crash Monitoring)

  • Use Case: All phases
  • SDK Size: Android 1.2MB, iOS 5.5M
  • Key Metrics: Device crash rate, crash-free rate
  • Integration Cost: 1 person × 2 hours setup + 4 hours verification

Benefit: Correlate crashes with performance metrics.

 

4. UDT (Unified Automated Testing Platform)

  • Use Case: Compatibility & functional testing
  • Features: Cloud device farm, automated test scripts
  • Device Coverage: 80% player devices

PerfDog Integration: Automated performance testing on 100+ device models.

 

Real-World Workflow Example

AAA Mobile Game Development Cycle:

Phase 1: Alpha Testing

  • Tool: PerfDog on 5 target devices
  • Goal: Establish performance baseline
  • Duration: 2 weeks

Phase 2: Beta Testing

  • Tool: UDT (automated testing on 50+ devices) + PerfDog Service (performance regression)
  • Goal: Compatibility & optimization
  • Duration: 4 weeks

Phase 3: Soft Launch

  • Tool: PerfSight (SDK integrated)
  • Goal: Real-world performance monitoring
  • Duration: 2-4 weeks

Phase 4: Global Launch

  • Tool: PerfSight (all players) + CrashSight (crash monitoring)
  • Goal: Maintain quality at scale
  • Duration: Ongoing

Result: 30% fewer performance issues post-launch vs. without this workflow.

 

Why PerfDog Matters for Your Game in 2026

The Mobile Gaming Performance Landscape

Industry Statistics (2025-2026):

  1. Player Expectations:
    • 60 FPS is now baseline (not premium)
    • 120Hz displays becoming standard (flagships)
    • 90Hz displays on mid-tier devices ($300-500)
  2. Performance = Retention:
    • 53% of players uninstall laggy games within first session
    • 1-star reviews mention “lag” or “battery drain” 67% of time
    • 30% higher retention for smooth-running games
  3. Power Consumption:
    • #2 player complaint after lag
    • Affects session length (shorter sessions = lower revenue)
    • Overheating → throttling → worse performance (vicious cycle)
  4. Competitive Pressure:
    • Top-grossing games invest 20-30% of dev budget in optimization
    • YouTubers compare game performance before recommending
    • App Store ratings heavily influenced by performance reviews

 

The Cost of Poor Performance

Case Study: Mid-Tier Mobile RPG (2025)

Before PerfDog Optimization:

  • Average FPS: 48
  • Jank count: 15 per 10 minutes
  • FPower: 65mW/frame
  • Battery drain: 18% per hour
  • 1-star reviews: 22% mention performance

After PerfDog-Guided Optimization (6 weeks):

  • Average FPS: 58 (+21%)
  • Jank count: 4 per 10 minutes (-73%)
  • FPower: 48mW/frame (-26%)
  • Battery drain: 14% per hour (-22%)
  • 1-star reviews: 9% mention performance (-59%)

Business Impact:

  • D1 retention: +12%
  • Session length: +18%
  • Revenue per user: +15%
  • App Store rating: 3.8 → 4.3 stars

ROI: 6 weeks of optimization → 15% revenue increase = $450K/month for a $3M/month game.

 

What Makes PerfDog Indispensable

1. Universal Platform Support

  • Only tool that tests iOS, Android, PC, Console, VR with one interface
  • No platform-specific expertise required

2. Zero Friction

  • No ROOT/Jailbreak/Debug mode
  • Works with production builds (what players actually play)

3. Industry-First Metrics

  • FPower (power efficiency per frame)
  • SmallJank (120Hz+ display stuttering)
  • Smooth Index (weighted smoothness)

4. Ecosystem Integration

  • Works seamlessly with PerfSight (monitoring) and CrashSight (crashes)
  • Complete QA solution in one platform

5. Trusted by Industry Leaders

  • 200,000+ apps tested
  • 140+ AAA studios
  • All top 10 phone brands
  • Major chip manufacturers

 

Take Action: Elevate Your Game Performance Today

Free Resources

1. Try PerfDog for Free

  • No credit card required
  • Full feature access for 14 days
  • Register Now

2. Download Comparison Benchmark

  • Compare your game against 500+ top titles
  • Genre-specific performance standards
  • Download Free Report

3. Expert Consultation

  • Free 30-minute performance audit with WeTest engineer
  • Identify top 3 optimization opportunities
  • Book Meeting

 

Enterprise Solutions

WeTest Complete QA Platform:

Pricing Tiers:

  • Startup (Free): PerfDog basic, 15 test mins
  • Professional ($1200): PerfDog advanced, PerfSight SDK, 50 test hours

What’s Included:

  • PerfDog (performance testing tool)
  • PerfSight (SDK-based monitoring)
  • CrashSight (crash analytics)
  • UDT (automated testing platform)
  • Dedicated account manager
  • Priority support (24/7 for enterprise)

Contact Sales for custom enterprise quote.

 

Learn More

Educational Resources:

  1. PerfDog Documentation - Complete user guide
  2. YouTube Channel - Video tutorials
  3. Blog - Industry insights and case studies
  4. Community Forum - Ask questions, share tips

Stay Updated:

 

About the Author

Baojian Shen

Senior Product Manager at Tencent WeTest

With over 10 years of experience in software testing and game testing, Baojian leads product planning and solution development for WeTest’s testing platforms and quality engineering initiatives. His work focuses on automated testing, compatibility testing, performance testing, and scalable test platform development, bridging practical testing challenges with product strategy.

Baojian regularly shares industry insights grounded in real-world projects and measurable outcomes, with emphasis on AI-driven testing, quality engineering, and evidence-based methodologies.

Connect:

 

About WeTest

WeTest, with over a decade of experience in quality management, is an integrated quality cloud platform dedicated to establishing global quality standards and enhancing product quality. As a member of the IEEE-approved Global Game Quality Assurance Working Group, WeTest is recognized for its commitment to quality assurance.

Global Reach:

  • 10,000+ enterprise clients
  • 140+ countries served
  • 1,000+ game companies trust WeTest

Product Suite:

  • Testing Tools: PerfDog, CrashSight, UDT (automated testing platform)
  • Testing Services: Compatibility, security, functionality, localization testing
  • Users: 1,000,000+ developers worldwide

Mission: Boost game development efficiency through advanced testing tools integrating AI technology, ensuring product quality globally.

Register for Free | Explore Solutions

 

Frequently Asked Questions (FAQ)

Q: Does PerfDog work with iOS 17 and Android 14?

A: Yes, PerfDog is fully compatible with the latest OS versions:

  • iOS 17.x (all versions)
  • Android 14 (all OEM variants)
  • Updated within 7 days of new OS releases

PerfDog is the world’s first tool to support iOS 17 performance testing (achieved same-day support on iOS 17 launch).

 

Q: Can I test my competitor’s game with PerfDog?

A: Absolutely. PerfDog works with any installed app, including:

  • Your own games (pre-release or live)
  • Competitor games (publicly available)
  • System apps (browsers, video players)

Legal Note: Testing publicly available apps for competitive analysis is standard industry practice and legal worldwide.

 

Q: What’s the difference between PerfDog and PerfSight?

Feature

PerfDog

PerfSight

Type

Testing tool

Monitoring SDK

When to Use

Development/testing

Live operations

Setup

Plug & play

SDK integration

Data Source

Test devices

All players

Metrics

200+ deep metrics

4 categories (core)

Cost

Tool license

SDK usage-based

Use Both: Test with PerfDog during development → Monitor with PerfSight post-launch.

 

Q: How accurate is PerfDog compared to platform-native tools?

A: PerfDog has been validated by the National Institute of Metrology for accuracy:

  • FPS: ±0.1% variance vs. Xcode
  • CPU: ±2% variance vs. Android Studio
  • Memory: ±1% variance vs. native tools
  • Power: ±3% variance vs. direct measurement

Advantage: Cross-platform consistency (iOS & Android measured the same way).

 

Q: Can PerfDog detect specific performance issues like shader compilation stutter?

A: Yes, advanced features include:

  • Frame time spikes: Identifies outlier frames (e.g., shader compilation)
  • GPU counters: Shows shader utilization (Qualcomm, Mali, PowerVR)
  • Memory allocation spikes: Tracks sudden memory usage
  • Custom markers: Insert code markers to correlate in-game events with metrics

Example: See frame time spike → Check GPU counter → Identify vertex shader bottleneck → Optimize.

 

Q: Does PerfDog work with Unity and Unreal Engine?

A: Yes, PerfDog has deep integration with both engines:

Unity:

  • Mono memory tracking
  • GC event detection
  • GC allocation per frame
  • IL2CPP support

Unreal Engine:

  • stat commands integration
  • Profiling data export
  • Blueprint performance
  • Material shader analysis

Other Engines: Cocos, Godot, proprietary engines all supported.

 

Q: Can I automate PerfDog testing in my CI/CD pipeline?

A: Yes, via PerfDog Service API:

Supported CI/CD Tools:

  • Jenkins (official plugin)
  • GitHub Actions (workflow template)
  • GitLab CI/CD (YAML example)
  • Azure DevOps
  • CircleCI
  • TeamCity

Example GitHub Action:

 

 

- name: Run PerfDog Performance Test
  uses: wetest/perfdog-action@v1
  with:
    api-key: ${{ secrets.PERFDOG_API_KEY }}
    device: 'device_pool_id'
    app: 'com.yourcompany.game'
    duration: 600
    fail-on-regression: true

 

Documentation: PerfDog Service API Docs

 

Related Articles

Performance Optimization:

  1. Mobile Game FPS Optimization: 10 Proven Techniques
  2. Battery Life Optimization for Mobile Games
  3. GPU Bottleneck Identification and Resolution

Testing Best Practices:

  1. Complete Guide to Mobile Game Testing
  2. Cross-Platform Testing Strategy
  3. Automated Performance Regression Testing

Industry Insights:

  1. 2026 Mobile Gaming Performance Benchmark Report
  2. How Top Games Achieve 60 FPS on Mid-Tier Devices
  3. The Impact of Performance on Player Retention

 

Conclusion: Your Next Steps

The Mobile Gaming Performance Imperative:

In 2026, game performance is no longer a technical concern—it’s a business imperative. With 53% of players uninstalling laggy games and performance issues directly impacting revenue, optimization can’t be an afterthought.

PerfDog provides:

  • ✅ Universal testing across all platforms
  • ✅ 200+ metrics for comprehensive analysis
  • ✅ Revolutionary metrics (FPower, Jank, Smooth Index)
  • ✅ Zero setup friction (no ROOT/Jailbreak)
  • ✅ Trusted by 200,000+ apps and 140+ AAA studios

Your 3-Step Action Plan:

Step 1: Baseline (Week 1)

  • Download PerfDog (free trial)
  • Test your game on 5 target devices
  • Identify top 3 performance issues

Step 2: Optimize (Weeks 2-6)

  • Address high-impact issues first
  • Use competitive benchmarking for targets
  • Validate improvements with PerfDog

Step 3: Monitor (Launch and beyond)

  • Integrate PerfSight SDK for live monitoring
  • Set up automated regression testing with PerfDog Service
  • Continuously optimize based on real player data

ROI: 6 weeks of optimization typically yields 15-30% revenue increase.

Start Your Free Trial Today: https://www.wetest.net/n/auth/register

Questions? Book a free consultation with our performance experts.

 

 

PD网络测试推广
Latest Posts
1Mobile Game Performance Testing in 2026: Complete Guide with PerfDog Insights from Tencent’s Founding Developer Master mobile game optimization with insights from PerfDog’s founding developer. Learn to analyze 200+ metrics including Jank, Smooth Index, and FPower. The definitive 2026 guide for Unity & Unreal Engine developers to achieve 120FPS and reduce battery drain.
2Hybrid Remote Device Management: UDT Automated Testing Implementation at Tencent Learn how Tencent’s UDT platform scales hybrid remote device management. This case study details a 73% increase in device utilization and WebRTC-based automated testing workflows for global teams.
3How AI Is Reshaping Software Testing Processes and Professional Ecosystems in 2026 Discover how AI is reshaping software testing processes and careers in 2026. Learn key trends, emerging roles, and essential skills to thrive in the AI-driven QA landscape.
4WeTest at GDC 2026: AI Automated Testing Ushers in a New Era of Game Quality WeTest at GDC 2026 showcases a revolutionary AI Automated Testing Solution that transforms game quality assurance. Learn how WeTest's AI Test Agent Platform enables scalable quality production through computing power, delivering controllable, reproducible, and intelligent testing capabilities.
5Precision Testing in Practice: A Fund Team's Journey from Experience-Based to Data-Driven Quality Assurance Learn how Shenwanhongyuan Securities implemented precision testing to reduce regression testing by 67%. This technical guide covers JaCoCo implementation, method-level code mapping, and intelligent test case recommendation for financial services applications.