I’ve been experimenting with
Dirac ART on my Denon AVR-X4800H and wanted to share some observations and measurements for feedback from more experienced users.
Room details:
- Dimensions: 4.65 × 4.65 × 2.5 m + small alcove 1.7 × 0.7 × 2.5 m
- Topt (RT60): 0.288 s
- Estimated volume: ≈ 57 m³
- Calculated Schroeder frequency: ≈ 142 Hz
Based on this, I initially assumed the room behaves modally below ~140 Hz and diffusely above that — so I compared different Dirac ART correction ranges to see how that theory translates in practice.
Dirac ART Correction Range Comparisons
I tested several limits:
200 Hz,
800 Hz,
7 kHz, and
20 kHz.
My expectation was that restricting correction to the low end would deliver the tightest bass by concentrating filter resolution where room modes dominate.
The results, however, were more interesting than expected:
- 200 Hz correction: Good low-end control, but slightly rounder and less defined bass than ideal.
- 800 Hz correction: Noticeable improvement in bass articulation and integration; smoother transition through the lower midrange, with vocals sounding more grounded.
- 7 kHz correction: Further improvement in bass tightness and overall coherence — likely from better phase alignment across a broader spectrum.
- 20 kHz correction: Preserved the gains from previous runs and refined the tonal balance even more, giving the most consistent presentation overall.
REW Observations
Attached are
REW waterfalls , group delay and RT60 plots at:
200 Hz,
800 Hz,
7000 Hz, and
20 000 Hz correction limits.
Subjective summary:
- Bass: The 20 kHz correction maintains the improvements from earlier settings, delivering tight, well-controlled bass with excellent definition.
- Midrange: Clear and linear across all cases, with the 800 Hz and higher corrections improving coherence through the vocal range.
- Treble: The 20 kHz correction provides the most refined top end so far — smooth, natural roll-off, reduced fatigue, and cleaner high-frequency detail.
- RT60 models:
200 Hz
RT60 is
longer (~600–800 ms), especially below 100 Hz.
Can cause
boomy or muddy bass. Room modes are likely dominant here.
800 Hz
RT60 is
more controlled (~300–500 ms).
Midrange clarity is decent, but some decay may still affect speech intelligibility.
7 kHz
RT60 is
shorter (~200–400 ms), closer to ideal.
High frequencies are well-controlled, reducing harshness.
20 kHz
RT60 is
very short (~100–300 ms).
Highest frequencies decay quickly, which is typical and reduces sibilance.
Takeaway
Even though my
Schroeder frequency is around 142 Hz, limiting Dirac ART strictly below that range didn’t yield the best overall result.
Broadening the correction range to
800 Hz,
7 kHz, and finally
20 kHz consistently improved both objective and subjective performance — especially in terms of bass control, time alignment, and top-end smoothness.
It seems that in my setup, allowing ART to operate full-range enhances global phase and timing coherence, which indirectly benefits the perception of bass tightness and clarity.
I’d be very interested to hear how others have approached correction-range selection with ART on similar systems — especially regarding where you found the best balance between precision and natural tonality.