This article covers frequently asked questions regarding Dolby Atmos music room calibration. If you have questions that are not covered in this article, feel free to ask them via the forum.

Why do I have to calibrate a Dolby Atmos music studio?
What is calibration?
Is calibration the first thing I should do or are there other steps before tuning the room?
Does Dolby have recommendations on acoustics?
I like how my speakers sound. Won't calibration change that?
What target should I use for calibration?
Do I have to calibrate my room using the music curve?
I never calibrated my room for stereo. Why is Dolby Atmos different?
My calibration is already flat. Why do I want to apply equalization?
Why is the music curve the shape it is?
Why does Dolby suggest such a tight tolerance to the calibration curve, even at low frequencies?
What are the typical symptoms of poor translation?
How can I assess/ensure playback translation?

 

Why do I have to calibrate a Dolby Atmos music studio?

Room calibration is about providing consistency and confidence in mixes that are created in a Dolby Atmos studio. This includes consistency within the studio, and consistency in translation between different monitoring or playback systems.

The first aim is to ensure a consistent sound across all the speakers in the room. This is so that a sound object that is being moved around the soundfield will sound the same regardless of where it is positioned in the room, or which speaker or speakers are being used to reproduce it. While you may be using identical speakers, due to the acoustic properties of the room the speakers will sound different depending on their location. There are many factors to this such as the proximity to different reflective or absorptive surfaces, but the result is that out of the box speakers will have a different sound dependant on their placement in the room and this is something that needs to be corrected.

This is not usually an issue for stereo systems as most rooms are quite symmetrical on the x-axis so the effects of the room are similar for the left and right speakers. However this will usually have a substantial effect when comparing speakers located in the front or back of the room, and particularly overhead.

The second aim is to aid with translation of mixes between different mixing environments. By having a consistent target for the response of mixing rooms, when a mix is transferred between rooms or replayed in consumer environments the results can be more consistent and predictable and therefore closer to the original intent.

What is calibration?

Calibration is the process of applying adjustments to level, equalization and delay so the measured acoustic response of the loudspeakers meet a defined response target. Calibration is usually performed by applying these adjustments in a monitor controller or other tool at an appropriate point in the signal chain. Calibration can be performed manually, where a trained engineer applies equalization based on their assessment of the measured response, or automatically where software creates an equalization after processing the measured response.

There are various ways that calibration can be performed. For example a defined pink noise source is replayed either as a file placed in the DAW or from the Dolby Atmos Renderer application.  This is routed to a specific loudspeaker under test and the acoustic response measured with a real-time analyser (RTA). The RTA displays the measured frequency response which is influenced by many elements of the replay chain and environment including the loudspeakers, amplifiers, room acoustics and other items within the room. The measured frequency response includes both direct and reflected sound. We recommend using multiple microphones for measuring and averaging the frequency response, because the results will vary with small changes in location throughout the room, particularly at low frequency.

Is calibration the first thing I should do or are there other steps before tuning the room?

Calibration can help many of the issues that can occur in a room, but before tuning the design and acoustics of the room should be considered to look for any issues that are likely to cause problems. Due to the nature of acoustics and the way certain frequencies can build up in a room (room modes), square and rectangular rooms can suffer from large frequency response issues particularly on symmetrical lines of the room and these can also cause flutter echos if untreated at higher frequencies. Unfortunately from an acoustic perspective, rooms are usually built in a rectangular/cuboid shape as this makes the best use of space within a building and structurally is easiest to build. This is why many music mixing rooms that have been designed by an acoustician will have an internal non-rectangular shape, although are still usually built within a structural rectangular room. Adding acoustic treatment can substantially improve the response of the room so is something that should be considered before calibration. Ideally this is considered as part of the initial room design, but in practice this often has to be retrofitted to an existing space. As well as the room itself, all items within the room will impact the response of the room, particularly items such as large desks or sofas. So the location of many items in the room should also be considered, but particularly the locations of the loudspeakers as this will impact how they will interact with the room and with each other.  

If it hasn't already been done, the signal path should be verified and the phase of each speaker should be checked. Then measuring the unprocessed acoustic response of each speaker within the room can highlight if there are consistent response anomalies such as peaks and nulls that are occurring. This is a complex topic and one we can’t cover in-depth in an FAQ, but typical actions that can be taken to address identified issues in the response include repositioning particular speakers or other items in the room, moving the mix position, or adding additional acoustic treatment. Ensuring speakers aren’t directly pointing at each other, and repositioning subwoofers away from the corners of the room are examples of changes that can be made to avoid some of these issues.

Bass redirection is also important to address at this stage. Any smaller loudspeakers that do not have full range frequency response should have their lower frequencies redirected to one or more subwoofers. Care needs to be taken at this stage to ensure the crossover frequencies are appropriately set up, and this is often done in the same monitoring controller as to where calibration equalisation would be applied.

Does Dolby have recommendations on acoustics?

Dolby does not dictate the acoustics of Dolby Atmos music mixing rooms.  Well established standards, such as the international standard ITU-R BS-1116-1 cover a number of relevant topics for the ideal acoustic performance of critical listening environments . These standards and other online resources can provide guidance, insight and examples of what has worked in other rooms to assist with acoustic design. Meeting a specific acoustic performance without any training can be a very complex task so if you are not getting the results you would like, we would recommend consulting with an acoustician. They can advise on the build and acoustic treatments for a room with the particular size, design and layout to meet the acoustic standards for critical listening environments.   

I like how my speakers sound. Won’t calibration change that?

Calibration is primarily about adjusting for the room and not the loudspeakers. So ultimately this isn’t about changing the sound of the loudspeakers but adapting the response of how the loudspeakers are perceived in a particular space with an aim of creating mixes that translate well. There are many aspects of a loudspeakers characteristics that are not changed by calibration and will still mean speakers have a distinct voice. The calibration provides greater consistency and this is ultimately about reducing translation issues that can occur when monitoring mixes on different systems and in different rooms. While room design and acoustic treatment should be used to address acoustic issues where possible, even the best room design will still not result in an entirely ideal measured response and this is where calibration comes in.

What target should I use for calibration?

The target curve we recommend for music mix rooms is defined within our Best Practices for Music Room Configuration. This music curve was developed in partnership with studio and label partners who required confidence in translation, such that mixes created in one room will playback consistently in studios around the world. The shape of the curve came from empirical testing with various music mixes in a number of studios utilizing the same target curve response.  In tandem the music mixes were also validated on a number of consumer playback systems to ensure predictable translation from studio to the consumer end device. 

Do I have to calibrate my room using the music curve?

In order to ensure that mixes translate well between Dolby Atmos equipped rooms, and to ensure consistency between multiple clients working on the same Dolby Atmos mixes, Dolby strongly recommends calibrating the sound systems to the music curve.  As the curve was developed by working with major label partners, creatives and studios globally, studios that meet the Best Practices for Music Room Configuration guidelines are expected to be calibrated based on the music curve in order to ensure that mixes translate to and from these reference environments. There are also a range of other room design and configuration recommendations to consider that are outlined in the best practices.

I never calibrated my room for stereo. Why is Dolby Atmos different?

Loudspeakers placed in difference positions in a room will not be perceived the same without calibration to meet a response target. While this means that the Dolby Atmos response may differ from the performance of stereo speakers without calibration in the same room, calibration of stereo loudspeakers could also assist in translation of stereo mixes between different rooms. Many stereo rooms in the past did not feature the monitor controllers or other equipment necessary to provide accurate calibration. A method of calibration is necessary in a Dolby Atmos room due to the need to balance the perception of multiple loudspeakers in different locations.

My calibration is already flat.  Why do I want to apply equalization?

The application of equalization is to take into account that even a loudspeaker with a perfectly flat response (as measured in an anechoic chamber) would not measure flat at the mixing position when it is placed a room. The measured acoustic response at the mixing position includes both direct and reverberant sound, as well as the modal response of the room. The addition of the reverberant sound will likely increase the measured low frequency response of steady state pink noise because higher frequencies are more easily absorbed by acoustic treatments in the room. The addition of the modal response of the room can cause large peaks and troughs in the measured response. If acoustic treatment measures cannot be applied to tame the modal performance, it can be highly beneficial to apply equalization predominantly on reducing the peaks to make the frequency response confirm to a target curve.

Why is the music curve the shape it is?

The curve was developed through empirical testing and designed to relate to typical speakers and responses in mix rooms. There substantial variation across stereo rooms today, so the curve doesn't necessarily align directly with any particular speaker or room combination. The LF boost was seen in a number of existing stereo rooms, and the HF roll-off relates to the balance of direct and reverberant sound in a room that is similar in size to the recommendations in the Best Practices for Music Room Configuration. If you are mixing in a room that is substantially larger or smaller than the recommended dimensions, then you may wish to apply a steeper or shallower roll-off than defined in the target curve.

Why does Dolby suggest such a tight tolerance to the calibration curve, even at low frequencies?

The closer the measured response is to the target response, the more likely that mixes will translate well between environments. This is why we recommend close alignment with the target curve. However, while good acoustic design combined with calibration can get very close to the target response, it is important to closely monitor the impact of the equalization added as part of the calibration process. An assessment should be made as to whether equalization can resolve any particular issue in the measured response. This assessment is particularly true at the lower frequencies where notches from strong room modes may mean that no amount of equalization can compensate to meet the target response due to a loudspeaker or listeners position in that room. The application of equalization to the changes observed in the measured response should be a 1:1 relationship. If you apply 3dB of equalization and you do not see a 3dB change in measured response, then equalization should not be applied for correction. For areas where large notches cannot be corrected, it may be preferable to observe the spectrum in 1/3rd octave or lower to check the main energy in this frequency region is close to the target curve.  In summary, while we recommend getting as close to the curve as possible, it is often the case that great sounding rooms have some areas that don't fully align with the target curve but respect the general trend.

What are the typical symptoms of poor translation?

The primary outcome of poor translation is that the work does not sound like it did when it was originally mixed, either in other reference rooms or on consumer devices. For example, this can be perceived as affecting the balance of the mix between different instruments or the timbre of a specific instrument. This perceived difference depends on the differences in measured acoustic responses between the mixing and playback environments.

How can I assess / ensure playback translation?

As well as looking to align closely to the target curve and other recommendations, it is key to QC the final mix on a few different systems (and ideally in other mix rooms). For example, it is often worth having a soundbar and/or home theater system, as well as a selection of headphone playback options available as part of the final mixing and QC process. If you have the option, it is also a good idea to take your mix to a different facility, particularly if you have access to a facility which has a higher specification than the one you typically mix in.