Please note that this article is based on content included in Secrets of Audyssey.
Audyssey MultEQ-X has been a very powerful tool for home cinema enthusiasts, allowing for precise configuration of an Audyssey calibration. However, one of its most notable recent advancements is the introduction of Filter Smoothing, a feature that has the potential to significantly improve the listening experience. In this blog post, we’ll take a closer look at what Filter Smoothing is and how it can enhance your audio setup. For those who want to dive deeper, you can find more detailed information in Secrets of Audyssey.
The Challenge of High-Frequency Correction
One of the longstanding challenges in audio calibration is the overcorrection of high frequencies. High-frequency correction is crucial for ensuring that speakers are timbre-matched, a concept well understood by industry leaders like THX, Yamaha, Audyssey, and Dirac. The problem, however, lies in the delicate balance required to correct tonal shifts without making the sound unnatural to our ears.
As frequency increases, individual room modes become less dominant, and it’s the relative energy across frequency bands that begins to play a more significant role. Therefore, while it’s essential to correct these tonal shifts, overly detailed corrections can be detrimental, making the sound harsh or unnatural.
Enter Filter Smoothing
Recognizing the need for a more refined approach, Audyssey introduced Filter Smoothing – including in response to requests to do so from Simple Home Cinema. While not yet the progressive smoothing that we asked for, this feature is a step in the right direction.
To access this feature, you’ll need to navigate to the Filter Design Screen in Audyssey MultEQ-X. Here, you can select the Limit Smoothed option, which allows you to configure the smoothing algorithm to suit your specific needs.
Customizing Your Smoothing
The Filter Smoothing feature offers two main customizable settings:
1. Transition Frequency: This setting determines the point at which the system transitions from non-smoothed to smoothed filters. The default setting is 20,000Hz, which effectively disables smoothing. However, you can adjust this to better match your listening preferences. A good starting point might be around 1,000Hz, allowing you to hear the difference and make further adjustments based on your listening tests.
2. Smoothing Strength: This controls the intensity of the smoothing effect, with a range from 0 (off) to 24 (extreme). The default setting is 0, which again disables the feature. Starting at a middle ground of 12 is recommended, after which you can adjust up or down depending on your listening test.
A Practical Approach
One of the most useful tips is to make both the Reference and Flat curves identical in the Design Target Curve section. By leaving one curve with default settings and applying smoothing to the other, you can easily switch between them to compare the effects.
While the default settings provided by Audyssey are a good starting point, Secrets of Audyssey offers more in-depth recommendations tailored for different types of speakers and setups. These include options for Fractional Octave Smoothing and Spline Smoothing, each with suggested transition frequencies and smoothing strengths.
The Path Forward
While Filter Smoothing in Audyssey MultEQ-X represents a significant improvement, it’s still not the end of the road. Enthusiasts who require even finer control over their audio systems might explore more advanced techniques, such as psychoacoustic smoothing, which requires additional tools like REW and the Audyssey App. However, this process is intricate and best suited for those willing to invest the time and effort.
By incorporating Filter Smoothing into your Audyssey MultEQ-X calibration process, you can achieve a more natural and immersive listening experience, while still properly timber-matching speakers. Simple Home Cinema has always advocated for full-bandwidth correction and we will continue to do so. This is why we have invested so much work into research and development of our guides. We feel very strongly that this is the correct approach.
However, this feature is just one piece of the puzzle. To unlock the full potential of your home cinema system, consider diving into the detailed guidance provided in Secrets of Audyssey.
Discover more from Simple Home Cinema
Subscribe to get the latest posts sent to your email.
Hi Roland,
I have a Marantz AV10 and bought DLBC multi sub for it a few months ago for my 7.3.4 system. I always used Audyssey and followed the complex process of downloading the ady file / editing it / upload it again using the Audyssey app. The results were decent but, for me, was far from ideal.
As soon as Dirac came to Marantz/Denon I took the reduced launch price and jumped ship.
Dirac SW is much more professional and easier and although I could also invest the new SW from Audyssey I always had the (non tested) impression that I could get better results with Dirac.
I Did a few iterations where I followed the usual recommendations for Dirac and ended up with, again, decent results. I have mixed feelings about it. I tried different house curves mainly changing the bass region but when I compare Dirac on vs off I do get more focus at the front but much less punch and impact overall. It’s less exciting. It made me wonder if my sound preferences are towards a less smooth curve or if it removes or reduces the character of my speakers . This weekend I inadvertently listened to an episode of a TV series I’m following with no calibration whatsoever (distances, levels as factory and no Audyssey or Dirac) and found it quite cool!
I bought your Dirac guide (that I haven’t read yet) and am now at a crossroads: do I invest time in Dirac, follow your guide and try go achieve sound nirvana with it or do you feel I can better results with Audyssey using its new SW? Or should I go to the trouble of setting them both? The Audyssey SW is not that expensive (and I could also buy your Audyssey guide) so I wouldn’t mind having it also.
In your experience with both Dirac and Audyssey, what do you recommend?
Hey Nuno,
Well, Dirac can sound great with the right loudness compensation curves so give that a go first using the guide. Yes, I think it’s worth it.
One issue is DLBC isn’t as good as pre-calibrating subs with a MiniDSP and MSO, then overlaying Dirac Live. I’m just going to give some feedback about this to Dirac as they still have work to do!
But yes, give it a go. You can get it to sound better. But you may need to raise your sub levels if you don’t have an MSO calibration.
Thanks Roland.
On the main reasons I invested in the full DLBC license for my AV10 was to keep the system as simple possible. I have a miniDSP that previously was in charge of correcting the subs with parametric EQ. It sounded good at the MLP in terms of frequency response but, after reading about every other thing that Dirac can bring to the system, and the technical drawbacks or adverse consequences of a simple PEQ approach, I thought this was the way to ago. And also, its a pain to setup. Lots and lots of measurements with REW, setup filters and so forth… It didn’t seem the optimal solution.
I also played around with MSO but never implemented it. It is good because it also uses all pass filters to deal with nodes instead of simply cut some db and a certain frequency (that also shifts phase). From what I read, MSO is very good in conjunction with miniDSP but that is what Dirac DLBC also does.
So I went with Dirac DLBC and I loved the simplicity of it all and trusted/trust that it does much more than simply tame the peaks and dips and multiple positions. My room has just one row of 3 seats and although the MLP is the most important, I hated when previously I had a 6db drop on certain bass frequencies from one seat to the other.
After a few gripes with the SW I easily managed to setup a couple of target curves. Playing around with and without DLBC, I still feel somethings missing.
I guess I’ll read your guide and see if I can come up with better results.
Yes, I would recommend reading the guide. However, I have to be honest, DLBC is not yet on the same level as MSO in terms of extracting the very best balance between seats, and the overall balance between subs and speakers. I’m preparing some feedback for Dirac on this. So MSO + overlaying Dirac is still probably the best way to go.
So the process would be design the filters using MSO, implement them on the minidsp and then feed a single sub to the AV processor?
Or should I design the filters using MSO with only amplitude and delays, apply those filters to each sub individually and use all 3 sub outputs on the AVP?
The first option is the correct one. 🙂