LatencyMon was a Windows utility that measured DPC (Deferred Procedure Call) latency and priority interrupt latency—two metrics that indicate whether your system’s drivers are causing unexpected delays in audio processing. It was widely used by audio professionals and gamers to diagnose why their systems were dropping audio, producing glitches, or experiencing unpredictable latency spikes.
The tool was developed by Thesycon and was free to download. It provided a simple interface that showed real-time latency measurements and identified which drivers were causing spikes. For years, it was the standard go-to tool for diagnosing audio issues on Windows.
Why Use LatencyMon?
If you’re experiencing audio dropouts, crackling, or unpredictable latency spikes in your DAW or game, LatencyMon could help pinpoint the cause. It monitors kernel-level interrupt handling and identifies when driver code is preempting audio threads, causing delays.
A typical LatencyMon result looked like: “USB Audio driver causing 2,500 µs (2.5 ms) DPC latency spike every 50 ms.” That information told you exactly what to fix: update the USB audio driver, disable USB 3.0 devices, or use a different USB port.
For real-time audio work (music production, live performance, gaming), knowing your system’s DPC latency baseline and max latency was critical information.
Why LatencyMon Isn’t Maintained
LatencyMon development stopped around 2016–2017. The official website eventually went offline. No updates for Windows 10 or Windows 11 were released.
The reasons are unclear (the developer never published a public statement), but the likely culprits:
- Windows security updates made it harder for third-party tools to access the low-level kernel data LatencyMon needed
- The rise of USB 3.0 and other modern hardware reduced some of the driver latency issues LatencyMon was designed to diagnose
- Lower demand as audio drivers improved
- Time and cost: maintaining a free utility that accesses kernel internals is a burden
Regardless, LatencyMon still works on Windows 10 and 11 if you can find a copy (many sites host the old installer). But with no updates or support, many users prefer to move on.
LatencyMon Alternatives
XPERF (Windows Performance Analyzer)
XPERF, now part of Windows Performance Analyzer, is Microsoft’s official tool for measuring system performance and latency, including interrupt latency.
What it does: Captures detailed kernel traces, interrupt handling, scheduler activity, and driver behavior. Lets you analyze DPC latency, context switches, and thread activity in granular detail.
Pros: Official Microsoft tool, actively maintained, can capture everything LatencyMon could and more.
Cons: Steep learning curve. The interface is designed for performance engineers, not casual users. Generating a trace, analyzing it, and interpreting the results requires significant technical knowledge.
How to use: 1) Capture a trace using xperf -start LatencyTrace -on PROC_THREAD+DPC+INTERRUPT, run your audio software for 30–60 seconds, then xperf -stop LatencyTrace -d trace.etl. 2) Open the resulting .etl file in Windows Performance Analyzer and dig through dozens of graphs.
Verdict: If you’re comfortable with command line and data analysis, XPERF is powerful. For casual diagnosis, it’s overkill.
DPC Latency Checker (Free Alternative)
A free, lightweight tool available on GitHub and various software sites. It’s not an official Microsoft tool but provides LatencyMon-like functionality.
What it does: Monitors DPC and interrupt latency, displays real-time measurements, and logs spikes.
Pros: Free, simpler interface than XPERF, LatencyMon-like functionality.
Cons: Not officially maintained, might not work reliably on newer Windows versions, fewer features than LatencyMon.
Verdict: A decent LatencyMon replacement if you find a reliable build. Search GitHub for “DPC Latency Checker” or “Latency Monitor.”
GPU-Z
GPU-Z is primarily a GPU monitoring tool, but it includes a GPU memory bandwidth test that can indirectly measure system stability. Not ideal for DPC latency measurement, but some users use it as a quick stability check.
What it does: Displays GPU clocks, memory, temperature. Includes a test mode that can stress the GPU.
Pros: Free, widely available, easy to use.
Cons: Not designed for DPC latency measurement. Doesn’t directly tell you about driver interrupt issues.
Verdict: Use it for GPU diagnostics, not for DPC latency specifically.
Audio Interface Loopback Test
Instead of using software, use your audio interface’s hardware loopback feature (if it has one). Connect an output back to an input on the interface, and measure the actual round trip latency in your DAW or with an audio latency test tool.
What it does: Measures real-world, end-to-end latency including DPC effects.
Pros: Measures what you actually experience, not abstract kernel metrics. Bypasses the need for diagnostic software.
Cons: Requires hardware loopback support (not all interfaces have it).
Verdict: If your interface supports loopback, this is the most practical approach. Real latency matters more than DPC readings anyway.
DAW Latency Compensation & Monitoring
Your DAW (Ableton, Pro Tools, Logic, Reaper, etc.) has built-in latency measurement and compensation. Most DAWs test input and output latency automatically.
What it does: Measures total system latency and reports it so you can align recordings.
Pros: Integrated into your workflow, measures what you care about (actual audio latency), no external tools needed.
Cons: Doesn’t diagnose driver issues specifically, only measures the effect.
Verdict: Use your DAW’s latency test as your primary tool. If latency is unacceptably high, then use DPC diagnostic tools to find the root cause.
Custom Python/Bash Scripts
Some advanced users write custom scripts to monitor interrupt latency or system performance. This is beyond the scope for casual users, but worth mentioning.
What it does: Custom monitoring tailored to your specific use case.
Pros: Infinitely flexible, you control exactly what’s measured.
Cons: Requires programming knowledge and time investment.
Verdict: Only if you’re comfortable with coding.
Which Alternative Should You Use?
Best overall: Use your audio interface’s loopback test or your DAW’s latency measurement tool. These measure the actual latency you experience, which is what matters.
Best diagnostic: If you suspect driver issues and need to identify which driver is causing latency spikes, XPERF is the “official” option, but it’s complex. DPC Latency Checker is easier if you can find a reliable build.
Simplest: If you just want a quick LatencyMon-like tool without the learning curve, search GitHub for “DPC Latency Checker” and use a recent build. Acknowledge it’s not maintained and might not be 100% reliable on Windows 11.
How to Test DPC Latency Without LatencyMon
If you don’t want to use any diagnostic tool, here’s a practical approach:
- Record audio at low buffer sizes in your DAW (128 or 256 samples).
- Play demanding content (video, a heavy CPU-load plugin session).
- Listen for glitches, clicks, or dropouts. If the audio is clean, your DPC latency is fine. If you hear glitches, your system’s interrupt handling is causing issues.
This is the real-world test. If audio is clean, the DPC latency is acceptable, whatever the number is.
For more rigorous testing, run a round trip latency test with your audio interface to get a precise measurement of the full system latency.
When to Use These Tools
Use DPC latency diagnostics if:
- You’re getting audio glitches at low buffer sizes (128 samples or lower)
- You want to identify which driver is causing issues
- You’re building a professional audio workstation and need to baseline system performance
Don’t use DPC latency diagnostics if:
- Your audio is stable and glitch-free (a good DPC score doesn’t make audio better)
- You’re just curious about a random metric (focus on real latency instead)
- You’re troubleshooting audio problems and haven’t tried the simple fixes yet (driver updates, buffer size adjustment, CPU load reduction)
Frequently Asked Questions
Does LatencyMon work on Windows 11?
It might, but it’s not officially supported. Some users report it works fine; others report issues. Windows 11’s security and driver model have changed enough that LatencyMon might not read latency correctly. Proceed with caution and don’t rely on its results as definitive.
What’s the difference between DPC latency and round trip latency?
DPC latency is a kernel metric that measures how long the system takes to respond to interrupts. Round trip latency is the actual audio latency you measure (input to output). High DPC latency can cause high round trip latency, but they’re not the same metric. Focus on actual round trip latency; DPC is a diagnostic behind the scenes.
If my DPC latency looks bad, does that mean my audio will be bad?
Not necessarily. DPC latency is one factor. System responsiveness, buffer sizes, CPU load, and hardware design also matter. Measure your actual audio latency instead of relying on DPC numbers. Real latency is what you feel.
Can I improve DPC latency by changing settings?
Yes. Update drivers, disable USB 3.0 in BIOS (it’s a known DPC latency culprit), disable unused devices, and lower the number of background processes. These steps reduce DPC latency spikes, which reduces audio glitch risk.
Is DPC latency relevant for gaming or streaming?
Slightly. High DPC latency can cause audio glitches in your stream or game, but most gaming systems have acceptable DPC latency. If your game audio is crackling, check DPC latency. If it’s clean, DPC isn’t the problem.
Dalton is an audio testing and latency optimization writer at SoundLatencyTest. He focuses on audio latency analysis, sound delay testing, recording performance, and audio troubleshooting tools for producers, gamers, streamers, musicians, and audio engineers.