How to Calculate Clarity C80 for Your Room

By James Hartley · February 16, 2026

How to Calculate Clarity C80 for Your Room

1) Introduction: what you’ll learn and why it matters

Clarity (C80) is one of the most practical room-acoustics metrics for predicting how “clear” music sounds in a space. It compares early arriving sound energy (the direct sound plus early reflections) to later arriving energy (late reflections/reverberation). If you’re tuning a rehearsal room, evaluating a small performance venue, or validating acoustic treatment in a control room used for live tracking, C80 gives you a number you can measure, compare, and improve.

This tutorial shows you how to measure an impulse response of your room and calculate C80 from it in a repeatable way, including settings you can trust, common mistakes, and what results to expect before/after acoustic changes.

2) Prerequisites / setup

Measurement mic (preferred): an omni measurement microphone with calibration file if available (e.g., miniDSP UMIK-1, Behringer ECM8000, Earthworks M30). If you only have an LDC, you can still learn the process, but your HF accuracy and directivity errors increase.
Audio interface with phantom power (if needed) and stable drivers (48 kHz or 96 kHz supported).
Measurement software: REW (Room EQ Wizard) is free and can show impulse response, ETC, and integrated energy plots. Alternatives: ARTA, Dirac, EASERA, Smaart. The steps below assume REW, but the concepts transfer.
Test signal playback: a studio monitor, PA speaker, or full-range speaker. Use one speaker at a time for meaningful results.
Quiet room: turn off HVAC if possible. Avoid measuring during traffic peaks. You don’t need an anechoic environment, but you do need a clean measurement window.
Basic geometry notes: note mic and speaker positions and distances. C80 is position-dependent.

Recommended baseline settings (good balance of precision vs workflow): sample rate 48 kHz, sweep length 256k (or 512k if the room is noisy/large), levels set so the recorded sweep peaks around -12 dBFS to -6 dBFS with no clipping.

3) Step-by-step: measuring and calculating C80

Step 1 — Define the real-world question (and choose measurement positions)

Action: Decide what listening position(s) you care about, and mark them.

Why: C80 changes across a room. A small venue may have clear front rows and muddy back rows; a rehearsal room might sound fine in the center but messy in corners. If you only measure one spot, you may “optimize” a position nobody uses.

Technique: For a music room, measure at least 3 positions: (1) typical listener/performer spot, (2) 1–2 m left/right, (3) farther back if applicable. Keep mic height consistent: for seated listening 1.2 m, for standing musicians 1.5 m.

Common pitfalls: Measuring with the mic too close to a wall (within 0.5 m) exaggerates boundary effects and can inflate early energy, making C80 look better than it is in the room’s usable space.
Step 2 — Set up a single source and control the direct sound path

Action: Place one speaker where the sound source typically is (stage position, monitor location, or the main loudspeaker position). Aim it toward the measurement position.

Why: C80 is built around the balance of early-to-late energy. If your source is moving, multiple speakers are playing, or the speaker is firing away from the mic, you’ll smear the impulse response and misrepresent early energy.

Specific placement guidance: Keep speaker at least 1 m from large boundaries if possible. If you’re evaluating a stage/PA scenario where speakers must be near a wall, document it and keep it consistent across before/after tests.

Common pitfalls: Measuring with both L/R speakers playing. Interference between channels changes the early reflections and can create false “clarity” or false “mud.” Measure one channel at a time.
Step 3 — Calibrate levels and avoid distortion (your C80 is only as good as your impulse response)

Action: Set playback level so you get a strong signal without clipping at any point: speaker, interface output, mic preamp, or ADC.

Why: Distortion adds extra energy that can appear later in time, artificially reducing C80 (making the room seem less clear). Noise reduces the accuracy of the late decay estimate and can also bias results.

Settings to use: In REW, use a sweep level that yields recorded peaks around -12 to -6 dBFS. If your room is noisy, increase sweep length to 512k rather than simply blasting volume.

Common pitfalls: Driving the speaker into limiter/compression. You’ll still get a clean-looking waveform, but the impulse response tail can be wrong. If your speaker has DSP limiting, back down until limiting is not triggered.
Step 4 — Capture the impulse response using a log sine sweep

Action: Run a measurement sweep and generate an impulse response (IR).

Why: C80 is calculated from the energy over time in the IR. A log sine sweep deconvolves into a high-SNR impulse response and separates harmonic distortion better than an impulse/noise burst.

REW technique (practical defaults):
- Sample rate: 48 kHz (96 kHz is fine if your interface is stable; it increases time resolution but isn’t mandatory for C80).
- Sweep range: 20 Hz–20 kHz for full-range loudspeakers; if your system can’t reproduce deep lows, use 40 Hz–20 kHz.
- Sweep length: 256k (increase if the room is noisy or large).
- Use one input channel and confirm the correct mic is selected.
Common pitfalls: HVAC noise or people moving during the sweep. Footsteps and chair squeaks show up as late energy and can lower C80. If the tail of the IR looks “bumpy” or the noise floor is high, redo the sweep in a quieter moment.
Step 5 — Time-align the impulse response (find time zero correctly)

Action: Identify the direct sound arrival and set it as t = 0 for analysis.

Why: C80 splits the IR at 80 ms after the direct sound. If time zero is wrong (even by a few milliseconds), you’ll misclassify energy between “early” and “late,” which can swing your C80 by noticeable amounts—especially in small rooms where reflections are dense and early arrivals are close together.

Technique: In REW, view the IR/ETC and locate the first strong peak (direct sound). Confirm it makes sense based on distance: speed of sound ≈ 343 m/s. Example: speaker-to-mic distance of 3.4 m implies direct arrival around 10 ms. If your direct arrival is wildly different, something is misconfigured (wrong routing, loopback error, or the “direct” peak is actually a reflection).

Common pitfalls: Choosing a reflection as the first arrival because the direct sound is weak (speaker aimed away, mic blocked). Fix the physical setup rather than “forcing” time zero to a convenient peak.
Step 6 — Apply sensible windowing and frequency treatment (don’t over-process the IR)

Action: Use minimal processing so the IR remains physically meaningful while still readable.

Why: Over-aggressive gating can remove legitimate late energy and inflate C80; heavy smoothing can hide frequency-dependent problems. C80 is often examined in octave or 1/3-octave bands because clarity is strongly frequency-dependent (mud often lives in low-mids; harshness relates to higher bands).

Recommended approach:
- For full-band C80: keep the full decay (don’t gate at 200 ms unless you’re explicitly isolating early reflections).
- For banded analysis: use 1/1 octave or 1/3 octave band filtering if your software supports it. Common reporting bands for music: 500 Hz, 1 kHz, 2 kHz.
Common pitfalls: Using a short right-window because “it looks cleaner.” It will look cleaner because you threw away the late field—the very thing C80 compares against.
Step 7 — Calculate C80 from the impulse response (the actual math)

Action: Compute the ratio of early energy (0–80 ms) to late energy (80 ms to end) and convert to dB.

Why: C80 expresses a perceptual idea—definition/clarity for music—as an energy ratio. More early energy relative to late energy generally increases articulation and separation (within reason; extremes can feel dry or aggressive).

Definition:

C80 (dB) = 10 · log10( E_0–80ms / E_80ms–∞ )
where E is the integrated squared impulse response: E = ∫ h(t)² dt.

How to do it in practice:
- In software: Many acoustics tools show C80 directly under “Clarity” or “Room Acoustics” metrics once an IR is loaded. In REW, you can access room acoustics parameters after measuring; if your version doesn’t display C80 directly, export the IR and compute externally.
- Manual calculation (spreadsheet / Python / MATLAB): Export the IR as WAV. Square the samples to get instantaneous energy. Convert 80 ms to samples: at 48 kHz, 80 ms = 0.08 s × 48000 = 3840 samples. Sum energy from sample 0 to 3839 (early). Sum energy from 3840 to the end of the decay (late). Then compute 10·log10(ratio).
Technique detail: “End of the decay” should be where the IR tail is essentially noise. If you integrate far into the noise floor, you’ll add late energy that isn’t room decay and C80 will drop. A practical method is to truncate when the squared IR stays near the noise floor for a sustained period (for example, when the energy is within 3 dB of the noise floor for 200 ms).

Common pitfalls: Using amplitude instead of energy (you must square the IR). Using 20·log10 instead of 10·log10 (20·log10 is for amplitude ratios, not energy ratios).
Step 8 — Repeat and average results (build confidence, not just a single number)

Action: Take at least 3 measurements per position and average C80, then repeat at the other positions you marked.

Why: Small changes—mic moved 10 cm, someone shifting, a door opening—can alter early reflections. Averaging reduces the risk you’ll treat the room based on an outlier measurement.

Technique: Keep the mic within a small “cluster” (e.g., a 30 cm radius) around the listening point to represent head movement. Average the results to get a more realistic clarity estimate.

Common pitfalls: Comparing a “before” measurement at one spot to an “after” measurement taken 1 m away. If you changed position, you changed the experiment.

4) Expected results: before/after comparison

C80 is context-dependent (music style, room size, distance from source), but these are practical expectations you can use to sanity-check your numbers:

Very low C80 (e.g., -2 to +1 dB): Often perceived as muddy or washed out for rhythm-heavy music. Common in reflective rehearsal rooms with hard parallel surfaces and little absorption/diffusion, especially at mid frequencies.
Moderate C80 (e.g., +2 to +6 dB): Typically reads as clear yet still “roomy.” Many small venues and treated rehearsal spaces land here in the mid bands (500 Hz–2 kHz).
High C80 (e.g., +7 dB and up): Very defined and forward. Can be great for articulation, but if it’s achieved by over-absorbing high frequencies while leaving low-mids uncontrolled, the room can feel both “dead” and “boomy” depending on band.

Real-world before/after scenario: A band rehearsal room (6 m × 4 m × 2.6 m) with painted drywall and tile floor measures around +0 to +2 dB at 1 kHz in the center. After adding a thick rug plus 100 mm broadband absorption panels at first reflection points and back wall (with a 100 mm air gap), it’s common to see +3 to +6 dB at 1 kHz, with a more noticeable improvement in vocal intelligibility and snare definition.

5) Pro tips to take it further

Measure C80 by frequency band, not just broadband. If your broadband C80 improves but the room still feels muddy, check 250 Hz and 500 Hz. Many rooms “look clear” at 2 kHz while low-mid decay masks definition.
Control early reflections strategically. C80 depends heavily on what arrives in the first 80 ms. Treating the first reflection points (side walls, ceiling, sometimes floor via rug) often increases clarity more efficiently than random panels.
Separate “clarity” from “loudness.” If your “after” measurement has higher playback level, you didn’t improve C80—you changed gain. Keep output gain fixed, or calibrate SPL (e.g., 75 dB SPL C-weighted slow at the mic) before each measurement.
Use distance as a variable. Measure at 1 m, 3 m, 6 m from the source. Many rooms are clear up close and fall apart at the back. This guides decisions like adding absorption/diffusion on rear walls or changing speaker directivity/aim.
Document everything. Speaker position, mic height, room conditions, doors open/closed, seating, curtains. C80 is repeatable only if your procedure is repeatable.

Troubleshooting: when the numbers don’t make sense

C80 is extremely high (e.g., +15 dB) in a live room: Check if you accidentally gated the IR too short, truncating late energy. Also verify time zero; if the direct sound is misidentified, the 80 ms split can land in the wrong place.
C80 is extremely low (e.g., -10 dB) even in a treated room: Look for background noise included in the tail (HVAC, street noise). Increase sweep length to 512k, measure when quieter, and truncate integration at the noise floor as described.
Results vary wildly between repeated measurements: Ensure the mic and stand are stable, the speaker isn’t moving, and no one is walking around. Also check for automatic processing in the playback chain (EQ, room correction, dynamics).
Direct sound peak is weak or not first: Speaker may be aimed away, mic may be obstructed, or you may be measuring a reflected path due to placement. Fix geometry first; don’t “edit” your way out of a physical problem.

6) Wrap-up: build skill through repetition

C80 is a disciplined way to connect what you hear (“this room is smeary”) to what you can measure (“late energy dominates after 80 ms at 1 kHz”). Once you can reliably capture an impulse response, set time zero correctly, and compute the early/late energy ratio, you can validate acoustic changes with numbers instead of guessing.

Practice by measuring your room in three positions, then make one controlled change—hang a thick curtain, add two broadband panels, lay down a rug—and re-measure. The goal isn’t chasing a magical C80 value; it’s learning how placement, reflections, and treatment move clarity in predictable directions.