- Manufacturer & Model
- MSRP
- $20,999 - $29,999
- Highlights
- Feature-rich, modular, and future-proof upgradability. DLART provides revolutionary bass optimization. Roon Ready. Dolby Atmos, Auro-Matic, and DTS:X. Outstanding support and user-friendly web UI. Endgame processor.
- Summary
- Owning the StormAudio ISP Elite MK3 Processor has been an absolute game-changer for my home theater and music listening. The MK3 and Dirac Live ART have transformed my system and room, delivering tighter bass, improved clarity, and a more immersive audio nirvana than I ever thought possible. From its deep customization to its flawless integration of speakers and subwoofers, it has exceeded my expectations at every turn. After a year and a half of extensive hands-on experience, I can confidently say this is the best sound I’ve ever had in my room—period. Highly recommended!
Index
- Introduction
- Systems
- Dedicated Room
- Features and Specifications
- Connections
- Initial Setup
- Dirac Live Calibration and ART Setup
- Child Room
- DLART Measurement Results
- Single-Point vs. Multi-Point Mic Measurements
- DLART Infra-bass Results
- Auro-Matic
- Auro-Matic YouTube Video
- StormXT
- Movies and Music
- Remote Control Options
- Modularity - Vinyl Room
- Conclusion
Introduction
Imagine telling an audiophile five years ago that AV processors would place virtual acoustic bass panels in our rooms exactly where needed. I suspect most would have chuckled. Be that as it may, progress in home audio technologies has been amazing over the last few years. Thanks to StormAudio and Dirac Live, StormAudio introduced the first processors with DLART (Dirac Live Active Room Treatment), which can digitally correct bass anomalies (frequency and decay) in our home theaters and listening rooms.
Considering what has been historically required to tame bass and room decay in frequencies below 100 Hz, I think of huge and super-thick bass traps, corner bass traps, and even Helmholtz resonators. Then, the proper placement must be determined to get the best possible results. Several years ago, I spoke with an acoustic engineer at a company that builds customized Helmholtz resonators to treat room issues below 100 Hz. The project for my room would have cost me well over the MSRP of the most expensive StormAudio processor and required substantial real estate in my dedicated room. Forget it… it is complicated, time-consuming, space-wasting, and expensive. On the contrary, StormAudio processors include DLART to do it all electronically, with precision, in only a few minutes (okay... maybe several minutes or more), and without occupying any additional space in our rooms beyond the space of a typical processor or AVR, and a few subwoofers. It might be hard to believe we have come this far, but we are here.
I had been anxious to try one of the StormAudio processors in my home system since our own @Todd Anderson reviewed the ISP 32 Analog MK2 processor in November 2020. The price tag and the pandemic were deterrents in late 2020 and most of 2021. Electronics were in low-production mode due to chip shortages, and staffing issues were debilitating. Electronics products were in short supply, and those that were available commanded a premium... if they could even be found. Discounts were few and far between. Thankfully, the dust settled, and the electronics industry normalized, although inflation has been evident. Ultimately, watching Todd's StormAudio YouTube video from CES 2023 and learning about this new DLART feature motivated me to purchase the latest MK3 model in July 2023.
I suspect most of our readers have heard of StormAudio, owned by Immersive Audio Technologies. Its manufacturing facility is in France, with distribution and service centers in France, Hong Kong, and the United States. The company prides itself on providing the finest quality audio products and has numerous years of experience leading its design and engineering team. The company is sensitive to customer needs and offers a wealth of information to assist customers and potential buyers. Support technicians are also available to help with setup by phone, should it be needed or desired. While I haven't needed a phone call for setup, I have communicated with them a few times over the last year and a half. The customer service is nothing less than superb!
This review is based on a year and a half of personal, hands-on experience since the purchase, capturing that time and perhaps justifying the investment. I hope to offer informative and genuinely helpful insights written from a personal and comprehensive user perspective.
PLEASE NOTE: Many of the images in this review are intentionally large to capture the full page of the various web interface pages during setup and REW (Room EQ Wizard) measurement graphs. However, these images must be reduced to fit our formatting requirements, which causes some to be blurry and too small to view. Click the images to enlarge them. Making sense of these large images on a mobile device may be difficult unless viewed in landscape mode.
Systems
The StormAudio ISP Elite MK3 replaced my previous Monolith HTP-1, which was used for home theater, as well as replacing my miniDSP SHD, used for two-channel music only. I have two systems set up for review with the MK3. My Main System is configured as a 9.9.6 layout: [9] base-layer speakers, [8] subwoofers, [1] bass shaker system, and [6] top-layer speakers. My Vinyl System is configured as a 2.2 layout: [2] base-layer speaker and [2] subwoofers. Below is a listing of my current equipment for each system.
Main System—Using 24 Channels
- StormAudio ISP Elite 32 Analog MK3 (Firmware Version 4.6r2)
- McIntosh MC1.25KW Monoblock Amps (Powering Front Speakers)
- RTJ Audio 410 (Front Speakers)
- StormAudio PA-16 MK3 Amp (Powering Center, Surround, and Top Speakers)
- MartinLogan Focus C-18 (Center)
- JTR Noesis 210RT (Side and Back Surrounds)
- JTR Noesis 110HT (Wide)
- JTR Noesis 110HT-SL (Top Front/Mid/Rear)
- JTR Captivator 4000ULF-TL Subwoofers x2
- JTR Captivator 2400 Subwoofers x6
- Dayton Audio BSA-200 Amp / 4 BST-1 Bass Shakers
- ROON Nucleus One
- Asustor Flashstor NAS
- SilentPower LAN iPurifier Pro (Nucleus One)
- Kaleidescape Strato C Media Player
- Kaleidescape Terra Prime 22TB Media Server
- Magnetar UDP800 Universal Media Player
- Sony 98X90L 98" BRAVIA XR Full Array LED TV
- MediaLight Mk2 v2 Flex Bias Lighting
- Cables from AudioQuest, THX PixelGen, Pangea, and Performance Audio
- miniDSP UMIK-1 Measurement Microphone
- Dirac Live Software Version 3.12.2 and 3.13.2
- Room EQ Wizard (REW) Measurement Software V5.31.3
- Lenovo 14" Thinkpad X1 Carbon Laptop
- StormAudio ISP Elite 32 Analog MK3 (Firmware Version 4.6r2)
- McIntosh MA252 2-Channel Hybrid Amp (Powering Front Speakers)
- Wharfedale Super Linton Speakers (Front Speakers)
- SVSound 3000 Micro Subwoofers x2
- Music Hall Stealth Turntable
- Ortofon 2M Blue Cartridge
- miniDSP ADept Phono Preamp
I use my X1 Carbon as a controller and remote for the processor, along with the Roon software program that streams Tidal and Qobuz and has access to our NAS, where our music collection is stored.
Some audiophiles may shun using an AV processor for critical two-channel music listening. It took me a while to accept that using my HTP-1 processor, which I used strictly for home theater, was fine for critical music listening. I went down that rabbit hole of believing the miniDSP SHD with THD+N (also known as SINAD) of 111-112 dB would be far superior to the 97-102 dB THD+N (depending on voltage output) of the HTP-1, which is why I had the SHD set up for two-channel music. The question was, could I hear THD+N at 100 dB below the signal? I highly doubt anyone can… and there is no doubt that I can't. I doubt very many can hear it at 90 dB below the signal. Chasing SINAD became unfruitful for me. I briefly brainstormed with the philosophy that I should always strive for the best specifications and measurements to ensure I wasn't missing anything, whether I could hear it or not. Again, it was brief.
While I have not found any measurements for the MK3 model, the ISP 16 MK2 that Amir measured at Audio Science Review in April 2021 was around 100 dB THD+N. I was reviewing his measurements and comments, and ironically, it was the first AV processor he recommended. He even stated, "For the first time, I thought if I needed an AV Processor, this is what I would like to have in my system!" Granted, that was back in 2021, and he has recommended several more since then, not that his recommendation is required by any means. I only mentioned it because I thought it was interesting that it was his first recommended AV processor. As it happens to be, I purchased the HTP-1 against his recommendation, which did not weigh in my decision, and I never regretted it in the least bit. It was an excellent processor, and I recommend it to anyone wanting to go lighter on the wallet. Nonetheless, I do appreciate Amir's detailed measurements. I've not found anyone who presents such excellent detail as he does in measuring audio electronics. The bottom line is that the MK3 will perform as well as anything else for critical two-channel listening, and it offers many extra features and benefits not found in other two-channel preamp-processors that cost more.
Dedicated Room
Angie (my wife) will testify that I spend several hours in our dedicated AV room daily. She would say I should refer to it as "my" room, not "our" room, as she rarely visits it. Suffice it to say, the room and equipment are used extensively. While I watch my share of TV shows, football games (Roll Tide), and movies, the most serene times I have in "my" room are listening to music. I enjoy being on my Carbon X1 laptop with Roon loaded, searching for new music that brings out the best in my system, or listening to one of my favorite albums. (Click the image to enlarge.)
The room's dimensions are 19.5 feet wide by 23.5 feet deep by 8.5 feet high. The room is separated from the house by a hallway leading to our back door entrance at that end of the house. The walls vary from about 8.5 inches to about 12 inches thick, with the ceiling being almost 2 inches thick and supported across the center by a 4-inch by 12-inch LVL beam in the attic above. The walls are staggered studs with blown newspaper insulation. The interior walls and ceiling are covered with 3/4" plywood (first layer) and 5/8" sheetrock (second layer). The ceiling has an original layer of 1/2" sheetrock underneath the same 3/4" plywood and 5/8" sheetrock. There is also blown newspaper insulation in the attic above the room. The exterior wall in the hallway on the house side of the room has two layers of 5/8" sheetrock. The only sound that has ever been noticeable in the house resembles faint thunder in the distance when I have cranked up a few movies with infrasonic bass. Cloverfield and Oppenheimer come to mind. I started watching Oppenheimer one night, and within a few minutes of starting the movie, I got a text from Angie, "Seriously?" She couldn't hear anything other than the windows rattling and the house shaking viciously. I may have had it cranked up a little too much. I had to wait until the next day to watch it while she was at work.
The front TV wall includes a couple of large custom absorption panels. They were initially built for my dipole MartinLogan speakers, but have remained. They are 91" tall x 44" wide x 7.25" deep... with heavy Roxul R30 insulation and covered with Guilford of Maine material. On the front of the side walls are large framed tapestries with no acoustic value. There are GIK 24" x 48" x 4" panels on each side about mid-wall, and GIK corner bass traps in each rear corner. On the back wall, GIK 4" acoustic panels are hung with piano hinges and serve as doors to the media shelves built into the wall. The floor is carpeted, and a thick rug covers the area in front of my recliner. A large area on the ceiling above the primary listening position has several 12" x 12" x 2" foam acoustic panels and two more GIK 24" x 48" x 2" offset acoustic panels.
- A short walk-through video of my room is available at https://avnirvana.cc.
- AV Showcase with additional photos: AV NIRVANA at CEDAR CREEK
(Click the image to enlarge.)
When we first started our speaker evaluations at my home, we tried the speaker placement recommendations from The Cardas Speaker Placement Guide. At that time, I had MartinLogan electrostatics, and we used the Cardas calculator for planar speakers, which recommended about five feet from the front wall and about five feet from the side walls. We had the primary listening position at the triangle tip but were not crazy about it. We kept moving the speakers around and brought them out to about six feet from the rear wall, and it sounded better, but it was still not quite what we were looking for. Wayne (AudiocRaver) kept leaning forward... about two feet... rocking back and forth in slow motion as he listened to his favorite tracks for speaker placement. He stated the good sound was about two feet forward. Instead of moving the speakers closer (farther from the wall), we moved my chair forward about two feet and loved it. It was the best soundstage, imaging, and depth acuity we had heard in my room. Of course, we kept fine-tuning, inch by inch, and adjusting the toe angle. The black marks are still on the floor where the feet of the MartinLogans sat (click on the image above to enlarge and look just in front of the RTJ 410 speakers).
Consequently, my speaker setup is likely not what most would consider conventional, with a perfect triangle between the primary listening position and the front speakers. The primary listening position remains inside the triangle since the first time we tried it, as it continues to bring the best of everything for my listening pleasure. All the MartinLogans in my room held that same spot for several years.
In the last evaluation we held here, the JTR 210RT speakers were in the mix and were not picky about placement, which allowed us to place them closer to the wall for a slightly better bass response without any other sacrifices. They were the closest sound to the ML Renaissance ESL 15As we had experienced, but they lacked the huge soundstage of the MLs. It was suggested that I needed to try the JTR 215s to get the larger horns, and they might be on par with the MLs. I procrastinated until Jeff Permanian released his massive RTJ 410 horn speakers. I thought that if the 210s were close, the 410s would bring that sound I was looking for, since the horns in those are larger than the 215 horns that were recommended I try. Jeff was kind enough to let me try the 410s. I promised him that if I liked them well enough to replace the 15A stats, I'd buy an entire speaker and subwoofer system from him, which I did because the 410s are marvelous speakers. They slightly bested the 15A stats, short of the low-end, which wasn't a big deal for me because I've had eight subwoofers for several years to support the low-end. Jeff didn't have a center that would work then, but he's working on one, and I hope to have it soon.
This room has hosted several evaluation events and dedicated reviews from the last 4 to 12 years. I am thankful that audio friends with ears to die for have helped me create a room that can deliver music nirvana. I am truly sad that Wayne recently passed, as he was a dear friend who visited often and was one of those who had golden ears. He always expressed that the importance of the listening position and speaker placement cannot be overemphasized. Even with all the best equipment in the world and the finest speakers, music nirvana will be challenging if listening position and speaker placement are not precise. My audio friends have been essential for experimenting and testing (ad nauseam) for the best listening position and speaker placement.
Consequently, my speaker setup is likely not what most would consider conventional, with a perfect triangle between the primary listening position and the front speakers. The primary listening position remains inside the triangle since the first time we tried it, as it continues to bring the best of everything for my listening pleasure. All the MartinLogans in my room held that same spot for several years.
In the last evaluation we held here, the JTR 210RT speakers were in the mix and were not picky about placement, which allowed us to place them closer to the wall for a slightly better bass response without any other sacrifices. They were the closest sound to the ML Renaissance ESL 15As we had experienced, but they lacked the huge soundstage of the MLs. It was suggested that I needed to try the JTR 215s to get the larger horns, and they might be on par with the MLs. I procrastinated until Jeff Permanian released his massive RTJ 410 horn speakers. I thought that if the 210s were close, the 410s would bring that sound I was looking for, since the horns in those are larger than the 215 horns that were recommended I try. Jeff was kind enough to let me try the 410s. I promised him that if I liked them well enough to replace the 15A stats, I'd buy an entire speaker and subwoofer system from him, which I did because the 410s are marvelous speakers. They slightly bested the 15A stats, short of the low-end, which wasn't a big deal for me because I've had eight subwoofers for several years to support the low-end. Jeff didn't have a center that would work then, but he's working on one, and I hope to have it soon.
This room has hosted several evaluation events and dedicated reviews from the last 4 to 12 years. I am thankful that audio friends with ears to die for have helped me create a room that can deliver music nirvana. I am truly sad that Wayne recently passed, as he was a dear friend who visited often and was one of those who had golden ears. He always expressed that the importance of the listening position and speaker placement cannot be overemphasized. Even with all the best equipment in the world and the finest speakers, music nirvana will be challenging if listening position and speaker placement are not precise. My audio friends have been essential for experimenting and testing (ad nauseam) for the best listening position and speaker placement.
Since Dirac Live has entered our homes, the room has become less of an issue, as has subwoofer placement. The seating position and front speaker placement are still critical, especially for two-channel music listening. I was excited to learn what the StormAudio ISP Elite 24 MK3 could bring to the system and the room to help achieve an even more immersive experience. I suspect most enthusiasts will agree that some impressive improvements are needed to justify the price tag.
ISP Elite MK3 Immersive Sound Processor
Features - Specifications - Pricing
StormAudio is one of the few elite companies producing high-end, high-quality audiophile processors. Designed and manufactured in France, the StormAudio ISP Elite MK3 is no exception to its portfolio of exceptional products. I originally purchased the 24-channel processor. Their ISP Elite MK3 line is modular, starting with 16 channels with the option to add eight additional channels for 24 channels total (as I did) or an option to add 16 additional channels for 32 total (as I did later). The Elite MK3 is black and has what I would consider a modern face with a large volume knob, a minimal number of smaller buttons, a well-laid-out 5" LED color display, and it weighs in at 18 lbs. The rear includes the necessary connections most would expect. It is feature-rich, with about every option imaginable on a processor.
Key features include:
- Analog Devices ADAU1966 DAC with 24-bit/192kHz PCM processing.
- Dolby Atmos, Auro-3D, DTS:X Pro, IMAX Enhanced.
- Up to 32 channels Decoding/upmixing
- StormXT ambiance expander.
- Up to 32 channels post-processing - multi-way active speakers and PEQ management.
- Bass Management - Standard or Expert with up to 6 bass zones.
- Dirac Live - Room Correction, Bass Control, and Active Room Treatment modules included.
- Up to 32 channels of analog and digital outputs - Multiple theaters and zones management.
- Roon Ready.
- 2 channels Zone2 optical output.
- 16 channels digital inputs - AES/EBU DCI compatible interface.
- 32 channels digital inputs/outputs - AES67 Ravenna/Dante compatible.
- Upgradeable HDMI - 7 inputs/2 outputs Matrix HDMI 2.0 HDCP2.2 with OSD 4K UHD with HDR10, HLG & Dolby Vision.
- Multiple control capabilities.
- Intuitive IP address web-based configuration via the web user interface.
- Remote monitoring and management.
- Room EQ Wizard integration and import.
StormAudio is also transitioning to the Hyperstream IV ESS Sabre DAC, replacing their previous DACs that have reached end-of-life. The company has plans to roll out a significant software update in Q2 2025, introducing enhanced equalization, advanced upmixing options, multi-way subwoofer management, refined output mapping, and improved latency control—all designed to simplify setup and optimize system performance. I'm always excited about these kinds of updates. The upgradability and modularity features are part of what helps me justify owning their high-end products.
- Full Specifications: https://www.stormaudio.com/wp-conte...-MK3-range-Specification-sheet-2024.12.06.pdf
- Owner's Manual: https://www.stormaudio.com/wp-content/uploads/2024/10/ISP_ISR-Owners-Manual_ALL_4.6r1_V5_A4.pdf
It may not seem like another review would offer much more than what Todd shared in his 2020 MK2 review. A few differences between the MK2 and the MK3 are the front faceplate design, which has a larger color display and a different (more stable) computing module. The major difference is that his MK2 did not have DLART. There have also been a few updates to the user interface, and we have a different room, a different system, and a new set of thoughts that should provide additional information from a different perspective. I will also focus a good portion of my review on the system setup, the DLART setup, and the results.
The MK3 is available in the following configurations and prices.
- ISP Elite MK3 16 channel analog output (XLR) $20,999
- ISP Elite MK3 24 channel analog output (XLR) $25,999
- ISP Elite MK3 32 channel analog output (XLR) $29,999
- ISP Elite MK3 32 channel AES/EBU output (RJ45) 16 channel analog output (XLR) $25,999
- ISP Elite MK3 32 channel AES67 input/output (RJ45) 16 channel analog output (XLR) $25,999
ISP Elite MK3 Connections
The first step of setup was to swap out the HTP-1 with the Elite MK3 and connect all the cables. The rear XLR outputs are labeled 1-24, so I added labels to my cable connectors to easily identify the speakers they match up with. I used my laptop sticky notes to notate each output to each speaker for setup reference. As I approached the end of the review, I decided to upgrade to the 32-channel version, which required swapping out the 8-channel module with a 16-channel module. Four of the eight extra channels will feed my vinyl system, as listed under the Systems section.
If a bass shaker is connected to one of the channels, an XLR to RCA adapter will likely be needed unless an amp with XLR inputs is used to power the shaker. Most pure bass shaker amps only have RCA inputs. As of this review, StormAudio only feeds the LFE channel to the bass shaker output, so don't expect any bass shaker action on two-channel stereo music. StormAudio advised me that a new setting will soon allow full bass output.
I needed an XLR to RCA adapter for my bass shaker amp and a couple for my SVS 3000 Micro subs. Emotiva has the Balanced XLR to Unbalanced RCA Adapter Interconnect, which works well and is inexpensive.
HDMI inputs are easy enough. Below is an image of my current inputs from the web UI (User Interface) view. The input name can be edited as desired.
With my Sony 98X90L BRAVIA TV connected to the MK3's HDMI output, my network switch connected to the Ethernet input, and my power cable and triggers connected appropriately, I was ready to storm into the setup session.
ISP Elite MK3 Setup
Todd notes in his MK2 review, "The MK2 lacks any sort of user-friendly setup wizard or on-screen GUI to guide installation." Not so with the MK3… its StormEasy setup wizard guided me in setting up the unit on my first round of setting it up. After updating the firmware, I set up the unit via Expert mode so that I could be more involved in the settings. The wizard did an excellent job of reminding me of the various settings. I recommend starting with the wizard and moving to Expert mode once familiar with the unit.
The Web UI is well laid out and easy to understand and use. After powering up the unit, the IP address will appear on its front display. The web UI will load after entering the IP address as the URL in a browser. I'll share the Web UI pages below.
The index page offers one of three choices: StormEasy, Expert Setup, and Remoet Control. As previously mentioned, StormEasy guided me through setting up my system. I now use Expert Setup, which includes all the manual settings for more detailed and customizable setup options. The online Remote Control page, which offers complete remote control, can be found on the web UI and is accessible from the Expert Setup menu. (Click on the image for a larger view.)
The following is a brief tour of the Expert Setup pages.
Selecting Expert Setup from the previous screen (above) opens the System page (below). In addition to general information about the unit, the System page offers several options: configuration backup, restore, and firmware upgrade. (Click the image to enlarge.)
Other pages can be accessed through the menu via the three horizontal lines in the upper left corner of any page.
My Inputs page selections were shown above for connection purposes. From this page, the video and audio inputs can be set for each named input, and the preset mode for the various presets created. For example, I have set DLART-Movies for my Apple TV 4K input. As I select an input, it will use the preset I have designated here for that input. Other settings may also be useful depending on the system configuration. Footnote: The [Video in] must match the [Main audio in] for HDMI inputs. (Click on the image to enlarge.)
The Main Speakers page is where the speakers are configured. It has all kinds of goodies; I'll return to it in detail later. In summary, my AV Room configuration is shown below in the top section of the Main Speakers page. The 9.9.0.6 represents the [9] Base-Layer speakers for fronts, center, and surrounds with wides, the [9] Subwoofers, which include [1] channel for the bass shaker system, [0] Height-Layer speakers, and [6] Top-Layer speakers. This page is also where all the speakers are mapped to the channels they are connected to on the back of the unit. In the image below, note that my Left Front speaker is mapped to channel [4] output, the Center Front is mapped to channel [1], and the Right Front is mapped to channel [17]. I used the output channels closest to their respective amps. (Click on the image to enlarge.)
The Settings page offers settings that can be selected and set as shown in the menu for that page. Some of the HDMI settings are shown below. The page continues to scroll through the various sections of settings (not shown). The handheld remote has six inputs and five presets, which can be set under the IR Remote Control section (also not shown). (Click on the image to enlarge.)
The Presets page is where presets can be named, the profile for each preset selected, and other settings can be made. For example, a particular preset can be set to default to Auro-Matic, and this is where it can be set. (Click on the image to enlarge.)
The Monitoring page monitors the StormAudio gear connected to a network. The Vumeter is especially helpful in identifying each speaker's output. I recorded a short YouTube video showing how the Vumeter works to identify how DLART and Auro-Matic affect speaker output, which will be shared later in the review. (Click on the image to enlarge.)
The Remote Control page is a web-based remote control that can control the unit. Later in the review, I will also cover the mobile app for remote access. (Click on the image to enlarge.)
I initially ran through the StormEasy setup and ran Dirac Live ART with the basic default settings, and it sounded pretty good. CAUTION: PLEASE DO NOT USE THE DEFAULT SETTINGS without verifying that the speakers can produce the [F-support Low] setting that the calibration sets by default. This setting is located in the ART parameters, and I will discuss it in more detail later. In summary, it represents the lower boundary of the frequency range that other speakers or subs will cover. When I started using DLART, the default [F-support Low] was 50 Hz. The latest Dirac version (3.13.2) reads my speakers considerably lower than their capable output and must be changed to the proper setting.
Over the last year and a half, I've run through this manual setup process a few dozen times. I'll cover how I configured and set up my system. Below is a list of the steps to get me to the point of the Dirac Live ART calibration.
- Start at the Main Speakers menu page.
- Click the [+] icon to open the Configurator.
- Set up the Base Layer speaker configuration and number of subwoofer channels.
- Skip [Height Layer] and set up [Top Layer] speakers layer (for my speaker setup).
- Close the Configurator and Click the [Output Mapping] button.
- Map the speakers by matching the speaker to the channel number to which its amp is connected on the back of the processor. Save it.
- Select the edit icon for the configuration that was created and mapped.
- Measure individual speakers and subwoofers and create PEQ filters (optional).
- Explore other options in the settings.
- Setup a Baseline preset (optional, but I consider it very helpful).
- Run a Dirac calibration using DLBC to obtain levels and delay settings, which can then be saved to the Baseline preset (optional).
- Inspect delay settings and run Baseline measurements using REW.
- Make any adjustments as necessary.
- Run Dirac calibration using ART.
I selected Main Speakers from the menu.
I added a Theater to the Main Layout by clicking the [+] icon and renamed it Main System. Footnote: It is best to rename the Theater before running Dirac. If the Theater is renamed after a project is created through the Dirac calibration, the project cannot be loaded again in the Dirac software for editing. It must retain the name of the Theater for which it was created.
The Configurator page is shown below after clicking the [+] icon. The Base-Layer configuration defaults to a 2.1 setup, which I updated to match my system. The Base-Layer section includes all the base-layer speakers and subwoofers. In my system, the Base-Layer is 9.9, which includes [2] front, [1] center, [2] wide, [2] side surround, [2] back surround, plus [8] subwoofers, and [1] bass shaker system. I checked the bass shaker box for one of the subwoofer channels that feeds my bass shaker amp. The bottom right corner shows I've used 18 of the 24 channels. After clicking [Next], there is an option for the Height-Layer, where up to five speakers can be added. Height speakers are typically used for Auro 3D setups, with speakers mounted high on the walls. I have no height speakers, so I clicked [Next] again and added my [6] Top-Layer speakers (up to [7] can be added). My configuration is shown as 9.9.0.6. Clicking [Next] again shows the summary (see second image below - click on the images to enlarge).
After configuring the speakers, my next step was to map each speaker to the output on the MK3. From the Main Speakers page (image I shared previously), I clicked on the [Output Mapping] button on the right side of the page. Clicking that button will bring up the mapping page shown below. The main objective here is to match each speaker to the labeled output on the back of the processor. (Click on the image to enlarge.)
After finishing the mapping, I saved it. (Our watermark covers the Save button in the image above.)
Next, I selected the Edit icon from the Main Speakers page to explore the speaker settings area for the Theater (renamed Main System) configuration I created (see the image below—click to enlarge). This page has too much information to cover in detail, but much of it is self-explanatory.
I created a Baseline profile to start with. I always select it before running the Dirac Live calibration. It is helpful in many respects. To help set up the Baseline preset, a single measurement using a single mic point can be used with the DLBC calibration to get the initial level and delay settings, which will be entered into the Baseline preset. DLBC is used for this measurement because ART requires a minimum of three measurement points and does not support a single measurement. I copied the level and delay settings from one of my first calibrations into a new profile and saved it as my Baseline preset. I'll discuss this more later.
The MK3's built-in generator can also be used with pink noise and an SPL meter at the main listening position. The REW software SPL meter with an appropriate mic, such as the UMIK-1, can also be used. I used this method to level-match my speakers and subs before running my first DLBC calibration to create the Baseline preset. It is not as precise as DLBC, but it can narrow down any drastic level variances to help with the volume calibration when running Dirac.
After running a Dirac calibration, I investigate the delay section to ensure nothing is too far off. With all of my speakers being symmetrical, I expect most delay settings to be reasonably close for each pair of symmetrical speakers. The delay settings also confirm the center position of the mic. See the example below of my first few speakers. Had I noticed any of these paired speakers being significantly different, I would have investigated for the cause.
Using the Baseline preset with levels already set, the volume calibration levels shouldn't need to be adjusted for every Dirac calibration because the levels will already be very close to level. This saves me some time since I have 23 speakers. I have adjusted the master volume and skipped the remainder of the volume calibration for each speaker without any adverse effects thus far. The levels may change slightly after the ART calibration.
I also set up PEQ filters (see next section) to reduce drastic peaks in various speakers and subwoofers (all PEQ filters were from 44 Hz and up), then saved them to my Baseline preset. Using the Baseline preset should make the calibration process lighter for Dirac. Hidden below are a few measurements comparing the RAW output of my left front + subs vs. the BaselineEQL preset with PEQ and levels set vs. BaselineEQLD with PEQ, levels, and delay settings.
The lower frequency range will look bloated in the full-range graphs because it measures all my subwoofers as one measurement, creating a higher output than measuring only one subwoofer at a time. Comparing the left front speaker level to the level of only one subwoofer, the subwoofer level is about 3-4 dB lower. The best pre-Dirac response is the BaselineEQLD preset. That said, setting up a BaselineEQLD preset, as I have done, is not mandatory. It is simply my preference preset for starting any Dirac calibration.
It is recommended not to start a new calibration from a Dirac-calibrated preset. While there may not be any issues, this can lead to potential issues, so it is better to be cautious. With each Dirac calibration or when editing a Dirac project, I recommend using a Baseline preset or a New Profile.
The image below shows the EQ column beneath the [Channels] section. Clicking on the gear icon will bring up the PEQ filter page to create it (see next image). PEQ filters, crossovers, and shelf filters can be created here. This page is where I created my PEQ filters.
The PEQ page is shown below (click the image to enlarge).
Select the second tab on the far right to view and adjust delays and levels before running Dirac. These settings are not adjustable after running Dirac. As mentioned, I choose this Baseline profile whenever I want to run Dirac Live or edit an already-created Dirac project. My Baseline preset will always include each speaker's PEQ, level settings, and delay settings.
In most cases, only one set of Dirac Live measurements will be needed. In my case, that is helpful. Otherwise, it's 207 sweeps if I measure [9] positions for all [23] channels. No measurements are run on the Bass Shaker channel. Once I run a Dirac calibration, I will save it as a preset. I can always return to this Main Speakers page, select my Baseline preset, rerun Dirac with a new set of measurements, or open and edit a saved project containing measurements. When I edit a project where measurements have already been run, I choose the Baseline preset, click the [DIRAC] button, start the Dirac software, and open the Dirac project I want to change. I edit it, save it as another preset, and export it to the MK3. (Click the image to enlarge.)
Clicking the [DIRAC] button in the upper right corner above will open this dialog box to start the Dirac calibration process. The instructions in the image below state a new profile will be created. In other words, my Baseline profile will remain unchanged, and I will save the Dirac project I create from this profile under a new profile name that will show up as a new preset, which I will name within the Dirac calibration. My Baseline preset will not be affected. (Click the image to enlarge and see a clearer view.)
Once Dirac is run and the new project is saved as a preset, I will exit the Dirac program, and it will bring me back to another dialog box where I can save that project as a preset. The profiles I created from the initial Baseline profile are shown in the dropdown list, renamed within the Dirac program before exporting to the MK3, although I can rename them at any time. I can also select any of these to view delays and levels. Again, the PEQ filters will be retained under the new preset name. (Click the image to enlarge.)
Dirac Live Calibration and ART Setup
Calibration Overview
The StormAudio MK3 includes all the Dirac Live licenses. Dirac Live, Dirac Live Bass Management (DLBM), Dirac Live Bass Control (DLBC), and Dirac Live Active Room Treatment (DLART). While I have set up all of these on my MK3 for comparisons, I eliminated all presets other than Baseline and various ART presets because ART is superior to all the other licenses for use in my room. For the setup portion of this review, I will focus on ART.
At the end of the previous section, I referenced the Baseline preset I established initially. This preset holds all my PEQ filters, adjusted levels, and delay settings, and I will select it before running Dirac Live. It does not matter if this is my first time running the Dirac software or if it may be subsequent runs; I always select Baseline. Next, I'll share the steps to get the Dirac Calibration started.
From the Main Speakers page... I will select the Edit icon from my Main Layout (the 9.9.0.6 configuration I created). The image below only shows the upper left corner of the page.
Pressing the Edit icon opens my AV Room management/settings page. I will select my Baseline profile, which has my PEQ filters and levels saved, and allow those to pass to the Dirac calibration. Then, I will press the [DIRAC] button. The image below only shows the upper right corner of the page.
Pressing the [DIRAC] button opens the Dirac Live calibration instructions dialog box, providing profile information (worth reading). In the image below, I show only the lower right corner of the dialog box. I then click on the [Start New Calibration] button.
Once I click the [Start New Calibration] button, I must start my Dirac Live software program (in a separate window). I prefer using the latest firmware version and regularly check the Software Downloads and Changelog from the Dirac Help Center.
I won't show images of every Dirac Live calibration process step, but I'll list them below.
- Start the Dirac software program.
- Select the StormAudio processor.
- Select the microphone, load the mic calibration file, and continue to the volume calibration.
- Calibrate the volume of each speaker and proceed to the arrangement.
- Select the appropriate arrangement and measure. (I use "Tightly focused imaging" and optimize only for my primary listening position.)
- Perform the appropriate measurements and proceed to filter design.
- This is where all the fun begins: design filters with speaker grouping, target curves, and ART parameters, and press the calculate button.
- After calculating the filter design, proceed to filter export.
- Create the preset name and export the filter.
- Save the project and exit the program, returning to the StormAudio web UI.
Measurements
Using a miniDSP UMIK-1 microphone, I only measure my primary listening position because I'm usually the only one in the room. Even if I have others in the room, whether their seat is optimized is irrelevant to them, as they could be hanging from the ceiling and enjoying it. Okay... if I have some AV enthusiasts over for an evaluation, it would matter for them, but in those cases, I let them swap out the primary listening position each night when we are watching movies. For the music evaluations, we are all focused on what we hear in the primary listening position as we rotate our listening sessions, and we want the main listening position to be as accurate as possible. I don't see anyone ever convincing me that measuring multiple seat locations won't take something away from the primary listening position, which is the seat I want to be optimized to its fullest. From this study, Stability of the Frequency Response Estimate in Listening Rooms, the conclusion was that Spatial averaging across a wide area may come with the risk of compromising the result at the main listening position. Amen!
For multi-point mic positions, the first mic location is the most crucial. I have pinpointed the height of my ears and the distance to the front wall. I verify this starting position with my Bosch laser measurer at the beginning of any new Dirac calibration. Then, I make sure the mic is centered. My front two speakers are within 1/16th of an inch from the side and front walls to each speaker with precisely matching toe-in. This precise placement allows me to measure from the mic to the side walls to get my center position in the room for the mic at my listening position. I always make sure my recliner is centered in the room.
For calibrating and tuning only the main listening position chair, there is the question of whether to leave the microphone in one location or move it around the listening position and how many measurements to take for the calibration. We've tried about every variation possible in our listening evaluations, and no one who has ever participated has been able to differentiate any differences. Granted, these were all before ART arrived. Later, in the Dirac Live: Active Room Treatment Results section, I will include the results of measurements taken with multiple microphone positions vs. leaving it in one location while using ART.
Let's elaborate on step 7. This step is where projects are built using the filter design, whether for Dirac Live, DL Bass Management, Bass Control, or ART. I will focus on ART because it is the most sophisticated of all the DL designs, and I use it exclusively for listening to music and watching movies. The options I will cover for filter design are listed below.
- Speaker Groups
- Target Curves
- [anchorlink= "slider"]Slider Shelf Target Curve[/anchorlink]
- ART Infra-bass
- [anchorlink= "point"]Control Point Target Curve[/anchorlink]
- Adjusting Level Gain
- ART Parameters
Filter Design (Speaker Groups)
The first step in the design process is grouping the speakers. While it’s possible to assign each speaker to its own group and configure nearby speakers to support it, I prefer pairing them (left and right) rather than isolating them individually. In my setup, I strategically select groups of speakers to support other speaker groups—except for the center speaker, which remains independent and does not provide support. Some enthusiasts suggest that assigning each speaker to its own group could be a viable approach to speaker grouping. While this method is time-consuming, it may be worth experimenting with. The most effective way to determine its impact is by setting up multiple presets, switching between them, and evaluating personal preference. I tried this approach but ultimately didn’t find it appealing. Below is an image of how I have structured my speaker groups.
Zoning and directional bass add another level of design, which I will not cover in this review.
My wide and side surround speakers were grouped together by default, and all eight subwoofers were grouped together by default. I separated the wides and side surrounds into their own paired group by using my computer mouse to drag the right speaker into a blank area and then the left speaker into that group. I also separated my subwoofers into pairs. The first grouping of subwoofers is my LFE subwoofers, for which I can create a target curve. The other grouped subwoofers will support some or all of my other speakers and subwoofers. All of my grouped speakers are symmetrically placed, so it's reasonable for me to group them in left and right pairs, other than the top, where I have them all grouped together.
The groups are all in one column on the right side of the filter design screen, but I cut and pasted sections of the column into the side-by-side image below. I can click anywhere inside the group box to highlight all speakers in that group and view both measurements in the measurement window, or I can click on an individual speaker to see only that speaker's measurement. In the image below, both front speakers have been selected.
Filter Design (Target Curves)
It needs to be determined what target curve is best for each system. Some enthusiasts like a Harman curve or something similar. Several sites offer custom curves that can be downloaded and installed within the Dirac software at the filter design step. Once saved to a computer, it can be loaded from the menu in Dirac >> Load target curve >> All groups (assuming the same target is used for all grouped speakers). If there are multiple target curves based on the speaker group, they can be loaded using the same method by selecting the appropriate group.
I have created several custom target curves. There are two methods of creating a target curve. The sliders on the left and right sides of the measured response graph can be used to create custom shelf target curves from 0.0 dB to +12 dB for bass frequencies and -3 dB to +4 dB for the upper frequencies... or a customized target curve can be created by using control points. The image below shows how to select each. For example, in Group [1], the slider option is selected, and in Group [2], the control point option is selected.
Slider Shelf Target Curve
The image below shows that I created a target curve using the slider option for my front speaker in Group [1] - (I have asked Dirac to let us rename our groups to match our speakers). This target curve has a +10 dB setting for bass and a -3 dB setting for the higher frequencies. I can adjust the preferred position of the slider using my mouse within its limits noted above. (Click the image to enlarge.)
While it appears I am boosting the low-end frequencies of the front speakers, with ART, it will not actually boost the front speakers because I have the [F-support Low] set to 60 Hz. The [F-support Low] setting acts similarly to a crossover in that it will not allow the front speakers to produce output below 60 Hz; instead, it will use other speakers I have set as support speakers to achieve the response, which in this case would be the subwoofer groups. The image below shows this setting at the bottom of the response graph. These are slider-controlled and can be adjusted, although they can also be adjusted in another area. Each of the sliders below represents a speaker group. If the slider is shown, that group supports the speaker group I am viewing. I'll cover the [F-support Low] setting in more detail later.
The subwoofers in the image below are being corrected, and they all support the front speakers. (Click the image to enlarge.)
The image below shows the measured results of the slider option target curve. (Click on the image to enlarge.)
Notice the brick wall that ART creates at 20 Hz. This brick wall is unique to ART and is not used with Dirac Live Bass Management or Bass Control. This leads me to the next section, which discusses how this brick wall can be removed.
ART Infra-bass
This sub-section on Infra-bass will be irrelevant for subwoofers that cannot produce output below 20 Hz.
Before we move to creating a custom target curve with control points, it is worth noting that despite using the Infra-bass options to extend the curve beyond 20 Hz and despite my subwoofers' ability to produce output below 20 Hz, the slider option with the bass boost that I prefer does not appear to attribute sufficient boost to that area. An excerpt from the Technical Guide: Infra-bass bypass in ART states the following:
Dirac has implemented an optional "infra bass bypass", enabled through a checkbox for each support channel group within each main channel group. It is intended to only become available when F-support Low (the low end of the support range) of a speaker group is lower than 21 Hz. Checking the box, which is optional, will then allow a low-passed versions of the inputs for that speaker group to pass through.
A user can make the checkbox available for a particular loudspeaker group by dragging the support range for that group down to 20 Hz. It must be stressed that this kind of manual override should only be done for speakers that have the capability to play below 20 Hz.
The target curve will be used as reference for infra bass level. All speakers that have been opted in for the infra bass bypass will receive an equal share with equal gain of the infra bass bypassed content, such that they sum up to the target curve level.
The infra bass bypass option will only be available to users with granted access to the "U4h infrabass" feature in their license.
I will share the Infra-bass measured results from my system later in the review.
Referring to the three images before the last image above, the section with the bars and boxes shows all the boxes checked (also shown in the image below). These four bars with the checked boxes represent all four pairs of my subwoofers. They include my two symmetrically placed LFE subwoofers and three more pairs of symmetrically placed support subwoofers. Each bar contains two subwoofers, as I have them grouped in pairs. These checked boxes indicate that Infra-bass is engaged. From the previous measurement graph, it does not appear that Infra-bass can boost the low-end frequencies as needed. There have been reports that Dirac is having some issues with Infra-bass, which is likely the reason for what we see.
It has been mentioned that some have had success unchecking the support subwoofers. In my case, it would be the three lower bars below. The first bar represents my two LFE subwoofers. I have not had good results with this suggested method. I have had my best overall bass sound quality and measurements by checking all the boxes. I have to check all four boxes for the front speaker group and all four boxes in the LFE group to get good measurements and bass response. It is always best to experiment since each particular system, setup, and room will differ. I have not checked these boxes for any other speaker groups because I see no need for those speakers to be set up for output below 20 Hz.
My solution, because my subwoofers can produce output below 20 Hz, is to use the control point option to create a customized target curve, allowing me to take advantage of my subwoofer's ability to produce output below 20 Hz. 10.5 Hz is the manufacturer-rated -3 dB point on my two front LFE subs, while 12 Hz is the -3 dB point on my other six supporting subs. The REW measurement graph below shows the combined Baseline response with a few parametric filters to tame a few peaks, and they measure quite well below 20 Hz. (Click the image to enlarge.)
Control Point Target Curve
Each room and system setup will determine whether the control point target curve works better than the slider target curve. The control point target curve helped me achieve my best response below 20 Hz. The image below shows that my control point target curve is approximately 10 dB lower than the slider option target curve throughout the entire frequency range. I am cutting frequencies rather than boosting them. (Click the image to enlarge.)
If a target curve cut is used for one group of speaker groups, there will need to be similar target curves for the other speaker groups, particularly above 100 Hz. Otherwise, the system will be out of balance. For example, I would not want a target curve for my front speakers (as shown above) and have another target curve at a 5-10 dB higher level for my surround, height, or top speakers. The only exception I made was to set a target curve for my center channel with a curved bump centered around 1000 Hz for improved dialog. I saved it to a "Movie" labeled preset for use with movies and TV shows to help improve the dialog. I do not use it for music. I also create custom target curves for my wide and top speakers, rolling off the lower end below 100 Hz. I see no reason to have the subwoofers supporting these speakers. I also roll off the bottom end of my center channel for music, using a customized target curve, but will match the level of the front speakers from 100 Hz and up for the music preset.
I created the control point target curve for my front speakers, saved it using the [Save target curve] in the Dirac software menu, and then used the [Load target curve] in the same menu to load it to [All groups]. I then replaced my center, wide, and top target curves with the previously created custom target curves.
When creating a custom target curve, the control points can be moved by dragging them with the mouse. They can also be added and deleted with a right mouse click.
One key point to remember is that if a target curve boosts the low frequencies for a speaker group with speakers that cannot produce those low frequencies, other speakers (such as the subwoofers) need to support it to accomplish the target curve. In the next couple of sections, I will show how speaker groups support other speaker groups.
We can see the subwoofers being corrected below. (Click the image to enlarge.)
Since my issue is with the sub-bass below 20 Hz, the following REW measurement will only reflect from 5-320 Hz to compare the results below 20 Hz. My control point target curve allows me to retain my subwoofers' output below 20 Hz. What is shown in the image above for the corrected frequency response below 20 Hz is not what will necessarily be measured with REW. As shown in the image below, it measures better than Dirac's corrected response in the above image. (Click the image to enlarge.)
How do the presets compare in movie action? Below is an RTA peak measurement graph of the opening minute of the movie Oppenheimer. This movie has significant output below 20 Hz right out of the gate. When I use the slider target and boost the bass +10 dB, I lose output below 20 Hz. I can maintain that same target curve shape I created with the slider target option but lower the target across the entire range by approximately 10 dB and recover that output below 20 Hz. The workaround is to bring the target level down to the level of the Infra-bass below 20 Hz. Since I reduced the overall level by 10 dB, I used that preset's +12 dB Level Gain setting to match the same output with the slider target, as shown in the Adjusting Level Gain section coming up next. (Click image to enlarge.)
I have heard of some who own a StormAudio processor but don't use ART because they cannot enjoy output below 20 Hz despite owning fully capable subwoofers. The control point target curve with a cut works for me, but it may not work for everyone since our rooms and systems can differ dramatically. I do hope that Dirac will address the Infra-bass issue very soon, as it can be a bit frustrating having to work around the issue after spending substantial money on a processor and subwoofers so we can enjoy level output down to 10 Hz.
Adjusting Level Gain
An important point to note is that if a control point target curve is used to cut the frequencies as much as 10 dB across the entire frequency range, as I have done, the Level Gain may need to be set to +9 dB or +12 dB to compensate for the cut made. The measurements seen in the graph above include my gain adjustment. This allows me to level-match my various presets, some of which may not have the same cuts as the control point target curve. I can access this setting within the Main Speakers page by editing the theater room I created and then clicking on the [Delay, level, limiter, and polarity] tab to show the Level Gain setting. Depending on the desired output level and if presets need to be level-matched, I may or may not adjust Level Gain. After each Dirac calibration, this setting will default to +6 dB. It accounts for cuts the Dirac calibration makes to the speaker levels.
Filter Design (ART Parameters)
Now that I have my target curve set, my next step for setting up ART will be to enter each speaker group's lower and upper-frequency limits depending on their output capability. The [F-support Low] and [F-support High] settings are similar to crossover points, although they differ from traditional crossovers. These settings are where I entered my lower and upper limits. I had to determine which speaker groups would support the speaker group I was setting up and to what extent they would support it. I considered each speaker's location and output capabilities to determine these settings. Dirac has labeled these settings as [Group x is supported by], [Support level], [F-support Low], and [F-support High]. I can click on the three dots within each speaker group to open these parameter settings, which will be adjusted as needed.
Clicking the three dots opens up the parameter settings box shown below. I cut and pasted these together to show the entire row in one image, as I can usually only see four groups at a time, and then I have to scroll left or right to view other groups. (Click on the image to enlarge.)
The parameters are only used for ART and will only affect frequencies between 20 and 150 Hz, depending on the speakers' capability in each group. 20 Hz would be the lower limit for [F-support Low], and 150 Hz would be the limit for [F-support High].
The Fsiso parameter box in the image above is for my front speakers Group [1]. As noted earlier, it would be super helpful if Dirac allowed us to rename the groups. Instead of Group [1], let me rename it Front, in my case here. I have ticked the boxes for each speaker group in the row labeled [Group 1 is supported by], designating which speaker groups I want to support my front speakers. Footnote: I have created several presets with different settings for each group to compare them.
Notice in this setup that I do not allow Group [2] to support the front speakers. Group [2] is my center speaker, and I don't want it to support other speakers. It seems to smear the image with music and muffles the dialog with movies, so I leave it unchecked for all speaker groups except Group [2], its own group. Each group has to have at least one other supporting speaker group. For my Group [2] center speaker, I only allow the front main speakers to support the center, and I have it set at a support level of -6 (very low support) and limit its F-support from 100-150 Hz.
The support level varies from group to group, and again, this will need to be experimented with, as each room and setup is different. I am only showing this as one example of what I have set up for my front speakers. In this scenario, I have Group [3] and Group [6] levels set to -6 (-6 is less support, and -24 is maximum support—it's backward, I know). Group [3] are my wide speakers, and Group [6] are my top Atmos speakers. Later, in the results section, I will list some of the support parameters I have used for comparison.
The [F-support Low] and [F-support High] are the frequency ranges I want each speaker group to support my front speakers. For example, Group [1] is the front speakers, and I want them to support themselves from 60-150 Hz, so I enter 60 Hz for the [F-support Low] and 150 Hz for the [F-support High]. Groups [4] and [5] are the side and rear surrounds, and they can support the front speakers from 50-150 Hz, so I enter 50 Hz for the [F-support Low] and 150 Hz for the [F-support High]. Groups 7-10 are all subwoofer groups, and they can support the front speakers from 20-150 Hz, which will also be entered.
I also have a preset where I have optimized the parameters for comparison. In this optimized preset, I have the rear surround speakers (Group 5) and rear subwoofers (Group 10) set with an [F-support High] of 80 Hz. I have read that caution should be used when speakers and subwoofers support other speakers up to 150 Hz, considering frequencies above 80 Hz can be localized. I have optimized this to compare with my preset, which has these speakers set with an [F-support High] of 150 Hz, to see if I can notice any localization from these rear speakers. While I can notice a difference in the sound, I cannot pinpoint the difference as being localization of the frequencies from 80-150 Hz, although there is no doubt the settings are causing the difference. I have yet to determine which I like the best, as they both sound good. The optimized preset pulls the bass towards the front, but it may be too much, although it may be more accurate. It will come down to preference, so it is good to experiment with and compare different settings.
The frequencies entered for [F-support Low] and [F-support High] are not brick walls. [F-support Low] has a roll-off of -12 dB per octave. I believe the [F-support High] is -24 dB per octave (I could not confirm).
Knowing each speaker's F3 helps because StormAudio recommends setting the [F-support Low] to 10 dB higher than the F3 of the speakers in each group. Each speaker can also be measured nearfield to determine its capabilities, and then 10 dB can be added to what it measured and entered for the [F-support Low] frequency.
The [F-support Low] and [F-support High] parameters can also be adjusted with the sliders at the bottom of each measurement view for each group. For example, the image below is for Group [7] in this project. Other groups can also be selected to support this group. As my mouse pointer moves over each bar, it will show the checkbox to enable it. Enabling a group will display its slider, which can be adjusted. The only setting missing with the sliders is the Support level, set at -18 dB by default. The Support level can only be set by clicking the three dots in each speaker group in the right column of the filter design screen.
There is a bit of a learning curve to determine the best support settings for all the speakers, but even the default, with only a few minor tweaks, was rather good in my room. Careful consideration needs to be given to the weight and frequency extent of how much one speaker will support another. While 80-150 Hz could be localizable depending on the distance from the speaker to the listening position, 50 Hz, being non-directional, should not be localizable, and it should be fine to support other speakers regardless of location, provided the supporting speaker can comfortably produce 50 Hz. Keep in mind that distortion can produce directional noise. If the supporting speaker is distorted while supporting another speaker, it may be noticeable, and the [F-support Low] needs to be increased to a higher frequency.
A few sections ago, I mentioned creating target curves with a boost in bass for speakers in groups that cannot produce those low frequencies. The ART parameters are where the subwoofer groups can be assigned to support those speaker groups with bass-deficient speakers.
Since we all have different rooms, some dedicated, some open, various sizes, different speaker types, sizes and numbers of speakers and subwoofers, and most of all, different preferences, our ART settings won't necessarily be the same. I suggest trying the default, but only after adjusting the [F-support Low] for each speaker group since the default may set them lower than their capability. After completing these settings and the filter design, calculating, saving, and exporting to the MK3, the preset can be saved. The same project using the same measurements can be loaded again, adjusted, and saved as a new preset. I've had twenty presets at one time. There are near-endless possibilities with presets for comparison.
The ART parameters will need to be set for each speaker group. Once this is done, that project's filter design is complete; it's time to calculate and export the filter to a preset and save it. After exporting the filter, I close out the Dirac software and return to the Dirac Live Calibration dialog box, where I can verify or change the preset name and curve type. I then click the [Save] button and return to the Main Speakers page, where I originally started the calibration. (Click the image to enlarge.)
Once back at the Main Speakers page, I click the [Delay, level, limiter, polarity] tab and adjust my [Level Gain] as needed. This setting will auto-default to +6 dB. Then I go to the Menu and select Presets, where I will see my new preset on the left labeled Preset 1. I will rename it because this name will appear in all of my preset dropdown menus, whether on the Main Speakers settings page or the Remote Control page.
Child Room
It's not what it sounds like. I have just set up this nice theater room, but I also want a strictly two-channel setup for vinyl or music that I only want to hear in stereo. I don't want any help from all these other support speakers. I only want my two front speakers playing and two subwoofers instead of all eight. This room setting is also helpful for reviewing speakers where I only want to play the front speakers, and perhaps I want to run a new Dirac calibration for only those speakers. I can create any configuration of speakers I want.
The child room is created from the Main Speakers page settings. The image below shows the upper left corner of the page. I click [Create Child].
Clicking on the [Create Child] button will open up a dialog box where I can select the speakers I'd like for the child system I am creating. I have selected my two front speakers and two subwoofers. I can also create a mono system here. Once I choose my speakers, I can tick the box to [Keep Dirac Profile] if I want to use the current Dirac settings. I can also uncheck it to listen to the raw speakers or run another Dirac calibration. I can confirm that, in all likelihood, the [Keep Dirac Profile] box needs to be unchecked, as what it retains is like a box of chocolates. It is better to rerun Dirac if a Dirac calibration is desired. I then save my selections. (Click the image to enlarge.)
I now have another system set up within the same room.
If I click the Edit icon for my Two-Channel room, it opens my settings, where I can make changes or run a new Dirac calibration for that specific room. (Click the image to enlarge.)
Dirac Live ART: Active Room Treatment Results
Key Objectives
Dirac Live ART (Active Room Treatment) is an advanced Multiple Input Multiple Output (MIMO) impulse response correction system that optimizes speakers as a unified system. Think of it as each speaker in the room can be supported by all the other speakers in the room, all working together. By using the output of multiple speakers to actively cancel lingering bass and reduce room decay, ART enhances clarity, expands the sweet spot, and reduces the need for physical acoustic treatments. Since ART is a mathematical process, accurate calibration is crucial... bad input yields bad results. ART delivers tighter bass, a more even soundstage, and an acoustically optimized room without excessive physical treatment when properly set up.
- Below 150Hz, ART corrects direct sound and room reflections by managing phase, magnitude, and reflections using supporting speakers.
- Above 150Hz, ART applies standard Dirac Live correction for tonal balance and timing accuracy.
Unlike traditional bass management, which simply splits frequencies between speakers and subwoofers, ART dynamically redistributes bass across speakers and supporting speakers to achieve a precise target response.
In layman's terms, DLART has three main objectives.
- Equalize the frequency response (make it as smooth as possible) for single or multiple listening positions in a room, partially or fully (user choice).
- Improve bass quality by reducing room decay from 20-150 Hz.
- Leveraging speakers and subwoofers in the room to support other speakers incapable of reproducing lower frequencies below 150 Hz.
It’s remarkable how ART leverages such sophistication to achieve these results.
Measuring ART Results
I performed two different calibrations to compare the measurement results of DLART. Both were for my primary listening position, with [Tightly focused imaging] selected as the arrangement during the Dirac calibration.
Dirac Live recommends the following for microphone placement: The measurement points should have a distance of at least 30 cm (12 in) between one another. Avoid making measurements in too small a space. Even for the "Tightly focused" listening environment, it is important to spread out the microphone positions in a sphere of at least 1 meter in diameter. Too small a space will result in over-compensation, which sounds very dry and dull.
I started with the microphone centered where my head would be and the tip of the mic at ear level while sitting in my chair (the mic is always pointing up) and made the first measurement. As recommended, for the first set of measurements, I moved the microphone around the listening position for eight additional measurements, with a distance of 12 inches or more between measurements. I finished the Dirac calibration and saved the preset as MixedMic9. I then made three measurements for the second preset with the microphone stationary at the first primary head-centered location and saved the preset as FixedMic3. I used the standard control point target curves that I typically use for both presets.
ART Parameter Examples
ART Parameters for this comparison are as follows:
- Support levels remain at the -18 default (unless a change is indicated below in parenthesis).
- [F-support Low] is noted below for the speaker group (bold header) and for each supporting speaker group listed below that speaker.
- [F-support High] is set to 150 Hz for all speaker groups, which was the default for these measurements.
I also set up another preset for partial support of the front speakers for both mixed mic and fixed mic, but they did not fare quite as well in all the measurements, although in listening, I was not able to notice any glaring differences. Those settings were as follows:
Front = 60 Hz
Side = 50 Hz (-12 dB)
Back = 50 Hz (-12 dB)
LFE = 20 Hz
Support Subs = 20 Hz
Measured and Corrected Spreads
Below are the Dirac graphs showing each calibration's measured and corrected spreads. The multi-point mic measurements will show more spread because the mic is spread more around the listening position; there is more to correct. The multi-point and single-point corrections look good in the area ART corrects from 20-150 Hz and do not appear to degrade the area above 150 Hz, where Dirac performs its normal equalization process. If the spread above 150Hz appears significantly worse after correction, it could indicate too much support for the speaker, and adjustments may need to be made.
Multi-point mic measurement - left front speaker. (Click on the images to enlarge.)
Multi-point mic measurement - LFE. (Click on the images to enlarge.)
Single-point mic measurement - left front speaker. (Click on the images to enlarge.)
Single-point mic measurement - LFE. (Click on the images to enlarge.)
Frequency Response: Before and After ART
The following graphs show measurement results, including frequency response, decay, waterfall, and spectrogram. Comparing multi-point and single-point reveals very few differences, inconclusive of one being better. I shared these graphs with John Mulcahy, author of Room EQ Wizard. He confirmed very few differences besides the uneven response in the midrange area (referring to the multi-point mic measurement). He thought the decay looked well controlled for both responses.
The Baseline preset is the raw output with a few peaks removed from the subwoofers using the PEQ filters in the MK3. It also has Dirac Live delay and level settings entered from a previous Dirac calibration. It represents the output as if Dirac has been switched off but retains the delay and level settings. The MixedMic9 preset is the Dirac calibration with nine measurement positions around the primary listening position, while the FixedMic3 preset only has three single-point measurements from the primary listening position.
The REW measurement graphs below show the frequency sweeps of both front speakers from 20-20,000 Hz for the Baseline, MixedMic9, and FixedMic3 presets and one graph of the left front for all three presets combined. Later, in the Infra-bass Results section, I will show the frequency sweeps from 5-200 Hz.
All images are labeled at the top of the image. (Click on the images to enlarge.)
Decay, Waterfall, and Spectrogram Results
For the decay, waterfall, and spectrogram measurements in this section, I will only show the front left speaker (with subs) from 20-200 Hz, as the effective range for ART is 20-150 Hz, although there is some slight roll-off outside of those frequencies. (Click the images to enlarge.)
These measurements clearly show improvements, but it’s important to remember that our perception of sound can be influenced by what we view in measurements. Ideally, listening should come first before analyzing measurements, but regardless of the approach, our ears must be the ultimate judge in shaping the sound we aim to achieve.
Vumeter Monitoring: Verifying Speaker Outputs
Is ART making a difference in my system? Without a doubt, it is reducing decay as intended, and it is using support speakers appropriately. This can be verified using the monitoring VU meter, which shows the output of the support channels for the front speakers and how each speaker responds across various frequencies when ART is engaged. A demonstration of this is included in the YouTube video referenced later in the review. Additionally, I have presets for DL, DLBM, and DLBC, allowing direct comparisons with DLART. Switching between these and ART, the improvement in sound quality... especially in the bass... is undeniable. For my system, ART delivers exactly as advertised.
It might be a good idea to run a few tests to ensure each speaker group's support parameters are not causing issues outside the ART frequency range. While ART only deals with the frequencies in the 20-150 Hz range, some support parameters could worsen decay in the upper frequencies. There could be issues above 150 Hz if the spreads have increased above 150 Hz after calculating the filter design. These issues may not always be immediately noticeable in listening. It may be slightly easier to notice with two-channel music by merely listening, assuming it is reference music. Atmos and surround-encoded sources are a little more challenging because we don't always know what the sound engineer intended. The Spears & Munsil Ultra HD Benchmark disc helps check these issues. Its audio section has a panning tool that covers about every direction of panning that would be used. The following two images are screenshots of that section on the disc. The panning was spot-on for my system, so I was not concerned about my speaker groups' support parameters. (Click on the images to enlarge.)
Single-Point vs. Multi-Point Mic Measurements
The following comments apply only to optimizing for a single listening position and do not pertain to optimizing for multiple seating locations.
No disrespect to Dirac, but the single-point mic measurements and calibration are certainly not dry and dull. I'm not saying this will be the case for every room, but it has been customary for my room since first using Dirac. Several others and I, during our evaluations, have never experienced better results when using multi-point mic positions. The following study concurs that single-mic-position measurements are acceptable for "certain rooms." Perhaps I have one of those "certain rooms." Is single microphone position enough for immersive system equalization and level calibration in production monitoring?
In this study, researchers measured spatial distributions around the main listening area in 12 rooms, comparing single-point measurements to spatial averages. The study concludes that single-point equalization is reliable for measuring rooms with relatively low reverberation times and well-controlled room modes, simplifying the process without compromising accuracy. When equalizing an audio system for a limited listening area designed for a single listener, spatial averaging is unlikely to significantly alter or enhance the results. However, in rooms with pronounced modal resonances, a small spatial shift of 0.1 meter (such as measuring at each ear) can help minimize localized anomalies, leading to more consistent system equalization. While broader spatial averaging may offer benefits, it also risks compromising accuracy at the primary listening position, which has been my experience as well as those participating in our evaluations over the years. The use and selection of spatial averaging positions should be guided by the specific objectives of room equalization and the room's acoustic characteristics.
Regardless of the study, I wanted to be as fair as possible and continue to listen and compare multi-point mic measurements, even if my bias led me to believe the smoother response of the single-point mic measurements (see image below) should sound better. Admittedly, it had been a while since I'd tried the multi-point mic measurements, so it was appropriate to compare it again for this review section. I was back and forth on presets, listening to music for several hours and days, trying to discern if one was better than the other. My listening comparisons concluded that there were zero differences in what I heard for the bass, where ART works its magic, and this is indicative of the bass response for both, as seen in the response graph below. As for the issues in the midrange section (see image below), I should have been able to hear the anomalies, but they were not as noticeable as the measurements indicated. No clear favorite emerged. A few times, I heard differences, but to say which was better is difficult because it was such a minor difference, and both sounded fine. (Click on the images to enlarge.)
Below is a more recent calibration between the two, with about 2-3 dB more bass boost from 200 Hz to 20 Hz. Getting consistent results with the mixed mic placement is another challenge. The mic placements here were very similar to the above. The orange response above is compared to the red response below.
The mixed mic responses in the graphs above remind me of one of our last evaluations in my room a few years ago... before the ART days. We had one preset with the full Dirac calibration and another with partial Dirac calibration up to 600 Hz. The responses look familiar when comparing the area with no calibration from 600 Hz up during that evaluation to the area 600 Hz up on the multi-point measurements for this comparison. We all blindly and unanimously preferred the full Dirac calibration during the evaluations. Those full Dirac calibrations from the evaluation also resemble the smooth response of the fixed mic measurements above. The area from 2500 Hz and up in the multi-point mic measurements looks better than the Baseline in our earlier graphs, while the more significant variances are from 600-2500 Hz.
When I began this comparison, I initially listened exclusively to stereo music. I suspected that the multi-point midrange issue might be more noticeable in movies and TV dialogue, where the center channel is prominent, and there is less surrounding music to mask any differences.
My testing with the REW RTA shows that the most pronounced portions of the male and female voices typically fall between 100 and 1,000 Hz. Harmonics can extend farther up the frequency scale but still appear to roll off after 1,000 Hz for everything I've tested. Earlier in the review, I mentioned having a custom target curve for my movies' preset center channel. I try to avoid boosting the fundamentals of the voice because it creates a muffled sound, particularly in male voices. Instead, I boost more in the harmonics range. I tried several target curves before I landed on a target that would work best for dialog. (Click the image to enlarge.)
The mixed mic measurements caused the center to have a higher output of up to 3 dB throughout the boosted range. Undoubtedly, it made the dialog slightly louder when switching back and forth. Occasionally, I noticed that this boost caused a slight heaviness in male voices, particularly in the fundamental area I preferred not to boost. However, it was not noticeable with most programs. It didn't bother me enough for it to be a major concern. Regardless, I will use the fixed mic preset since I have the option, and it works perfectly with all the programming I watch.
Below is my center channel target curve for music, such as concerts in various surround formats. I did not hear any differences in these two target curves for music with center channel output. (Click the image to enlarge.)
Measuring and comparing music voices is more challenging, considering the background instruments and extended voice harmonics. To make it easier, I used two a cappella songs for listening comparison and to measure differences. I compared the mixed and fixed mic music presets using the REW RTA. The first graph is a 15-second clip from Ed Sheeran's Shape of You, a cappella version. The second graph is from the first minute of Whitney Houston's I Will Always Love You. I would have thought the 700 Hz anomaly in the mixed mic preset would have been more pronounced and noticeable, but I'm not sure even the best ears could pick between these in a blind AB test. I certainly couldn't hear any differences. (Click the images to enlarge.)
Ultimately, despite the few hiccups in the mixed mic measurements, the differences had no major consequence for me. I recommend trying fixed and mixed mic positions, measuring, listening, and determining if one is better than the other.
Pondering on the study I referenced concerning single mic measurements, many years ago, during our evaluations, we measured three points, the center of the head and each ear (a very small area), as the study suggested might be beneficial in some rooms. After comparing this method with leaving the mic in the center position only, we did not hear any differences; thus, we started using the single-point mic position, and I still use it. Yet, over the years, I have also tested various multi-point mic position measurements, randomly and symmetrically positioning the mic over smaller and larger areas. I have not measured or heard any differences that would convince me that multi-point mic positions benefit me in my room.
I was curious about the average response for the Fixed Mic vs. Mixed Mic presets, which had three measurements, one at the center of the head and one at each ear. Below is the REW graph of those averages. Averages cannot generate decay, waterfall, and spectrogram plots. Averaging significantly diminishes the differences. (Click the image to enlarge.)
For those hesitant about my measurement method... I'm not suggesting anyone completely ignore the multi-position mic measuring that Dirac recommends. I would recommend it, along with a single-point mic measurement. Once compared, if differences cannot be measured or heard, I suggest future multi-point mic measurements may be fruitless unless something changes to warrant another comparison.
ART Infra-bass Results
Using Infra-bass in my setup significantly increased decay below 20 Hz. However, in my experience, switching between presets with and without Infra-bass enabled had no noticeable impact on music listening. Neither did it introduce issues in the few movies I’ve watched with infrasonic bass output. Some enthusiasts claim that Infra-bass can lead to bloated bass or a muddied midrange, but I haven’t observed these effects in my system. That said, evaluating such claims critically is important, as not all widely shared opinions are backed by evidence.
Once again, I asked my trusted friend, John Mulcahy, what he thought about decay below 30 Hz. He said that because these are individual modal resonances and we are insensitive to decay at such low frequencies, it wouldn't matter much, but it would make sense to equalize the peaks. I did not challenge him, but I did quiz him about his thoughts, and he pointed me to this study: Perception of Temporal Decay of Low-frequency Room Modes.
While not definitive, this study suggests that our ability to perceive temporal details in sound deteriorates significantly below 100 Hz, with a slight decline noticeable from mid-frequencies to 100 Hz. The temporal aspects of modal decay are particularly critical down to this threshold, where magnitude equalization is essential, and excessively long decay times should be addressed through acoustic treatment or modal equalization. However, below 100 Hz—especially at or below 50 Hz—extended decay times (even over 2 seconds) may go unnoticed if the magnitude response is well equalized. This study raises doubts about the necessity of modal equalization at very low frequencies, except in extreme cases.
This study may explain why I do not notice any issues with significant decay at 20 Hz and below when my Infra-bass preset is used. I would still like to see Dirac fine-tune Infra-bass and at least allow us to shape the response below 20 Hz with the target curve. It doesn't appear that attempting to fix decay in that area makes much sense.
Below are a few graphs of my Infra-bass presets. I will only show the left front speaker graphs (with subs) from 5-200 Hz because this focuses on the area below 20 Hz, which Infra-bass allows to pass. These are labeled accordingly. (Click the images to enlarge.)
For my curiosity, I moved my target curve up 2 dB across the entire target range, as the output below 20 Hz will not change regardless of the target curve. This adjustment improved the lingering decay, which, again, is probably a moot point, considering I didn't notice it to begin with. The waterfall best shows this change.
Auro-Matic
Auro-Matic is Auro-3D's upmixer. Auro-Matic in the StormAudio processor upmixes two-channel stereo to an Auro-3D effect. It can also upmix some of the other Dolby and DTS native formats. It does not have any effect on Dolby Atmos encoded formats. Auro-Matic can be selected from the menu's Remote Control page under the PREFERRED UPMIX dropdown box, as shown below. (Click on the image to enlarge.)
Auro-Matic has two settings in its menu... AURO PRESET and AURO STRENGTH. The preset options are Small, Medium, Large, and Speech. I initially favored the Large setting, but I complained about losing the precise imaging I enjoyed with my two-channel system with music. While Large delivers a slightly heightened surround effect, it also expands sonic cues in a way that can obscure the pinpoint localization of instruments, resulting in broader but less focused imaging. @JStewart suggested I try the Small setting, which I did. I now use Small, which tightens the imaging. While it is still not as precise as my native two-channel, it is good, and considering how much I like the overall effect of Auro-Matic for two-channel music, the trade-off is worth it (for me). The strength setting is flexible, but I've found 5 to be sufficient in most cases. I started out liking 10 but then realized it was too much, and somewhere from 3-7 works for most of my concerts and two-channel music. These are pure preference settings for each individual. Listen and set accordingly.
Once settings are made and the user does not desire to change them, the handheld remote can be used to engage Auro-Matic, but it does not allow the user to change the settings.
I used the monitoring Vumeter mentioned in the ISP Elite MK3 Setup section to confirm how Auro-Matic redirects and affects the input signal while engaging Auro-Matic and changing the Auro-Matic strength setting. The Vumeter shows the signal input and the signal output to all speakers. With a stereo input, Auro-Matic uses the top speakers to add effects. The two native DTS-HD 7.1 formats I tested, Eagles: Live From The Forum MMXVIII and Pink Floyd: Delicate Sound of Thunder, both on Kaleidescape, used the back surround speakers and the top speakers. I also tested the same with David Gilmour: Live In Concert on DVD, recorded in Dolby 5.1, where it added the rear surround speakers and top speakers.
I perceive a slight boost in the bass with Auro-Matic, although I could not verify this with measurements or monitoring. I used REW to measure the bass sweeps with Auro-Matic off and engaged at various presets and strength settings, and there were no measurable differences. It may be a perceived boost based on the speakers being added. It does not seem too dramatic, and I like what it does. I am curious about what is happening with the bass. I opened a ticket with Auro and StormAudio, and both acknowledged the tickets. StormAudio passed on their response from Auro: What they mentioned is that there is no intentional part of the algorithm that could emphasize the bass. Their current assumption is that this is related to Auro-Matic generating additional channels that include uncorrelated signals at lower frequencies that, when added to the subwoofer output create an increased perceived bass level. If no subwoofer is involved, then this is likely simply a consequence of having more channels playing bass at similar levels… They will need to test further. If I get any further information, I will update it here. The Chief Technology Officer at Auro-3D requested more information concerning measurements and my perceptions, and I obliged.
The YouTube video below shows how Auro-Matic affects the speaker outputs and also how ART affects the support speakers for my front speakers. This video does not include all the testing I performed, but it provides an idea of what Auro-Matic is doing based on the recorded testing.
Auro-Matic is relatively new (to me); I discovered it in the last couple of months. I had been brushing it off as another Atmos music gimmick I wouldn't like. I have not found any Atmos music I cared for on Amazon Prime, iTunes Music, or Tidal. None of it sounds right to me. The only Atmos music I have enjoyed is the Hans Zimmer: Live in Prague concert on Kaleidescape. After giving Auro-Matic a try one evening, I found myself using Auro-Matic more often than not with music. It has found a home in my system with two-channel music, especially with the ability to control the effect. I'm not losing my stereo effect with Auro-Matic in the way I lose it with Atmos. While I am immersed in the music when listening in stereo, I am immersed deeper with Auro-Matic, despite the slight loss in pinpoint imaging. As for movies, I have not watched any that were not Dolby Atmos, so I haven't been able to try it. On one stereo TV show I've watched, I preferred Dolby Surround. The three concerts I referenced above, from the Eagles, Pink Floyd, and David Gilmour, sounded excellent with Auro-Matic.
StormXT
StormXT is StormAudio's ambience expander. It adds output to the wide speakers, which I tested and monitored with the Vumeter. I do not particularly care for it with music, as it affects the imaging. For movies, it is too difficult for me to pinpoint the effect, so unfortunately, I can't elaborate any further about its effectiveness. I leave it off.
Movies and Music
I have watched a few dozen movies over the last year and a half, along with numerous TV shows, and thoroughly enjoyed them all with this processor. It has never disappointed me. That is about the best I can say because comparing a movie viewed through one processor to the same movie viewed through another processor is not feasible for me. In every case, it has been at least a year, if not several years, between watching the same movie. It's impossible for me to even think I could remember details for comparison's sake. Not only have my processors changed over the years, but so have my speakers and subwoofers. I feel like everything is better with movies and TV because of several improvements in my system, including the MK3 processor and ART. I can confirm that every movie I have watched using the MK3 processor in my system has been from extremely satisfactory to extraordinary. There is no doubt it is getting the job done in the manner I would expect it to. It checks every box on my list of must-have features for watching movies and then adds more.
Music, on the other hand, provides a more reliable reference for comparison because I’ve played certain songs countless times... perhaps over a hundred times... right here in this same room. Music has never sounded better, thanks to the StormAudio MK3 processor and DLART technology. The bass, in particular, has improved dramatically; it’s clearer, more pronounced, and better defined. Some might describe it as "tighter" due to reduced lingering bass, though I wouldn't attribute the improvement solely to that. A key factor in this transformation is ART's ability to use additional speakers, especially subwoofers, to support other speakers, creating a more cohesive and impactful low-frequency experience. When I set up the MK3 a year and a half ago, my immediate reaction was “Wow!”—or maybe it was “WOW!”. It was difficult to articulate then, but I instantly felt it sounded significantly better than my previous processor. The imaging, soundstage, and depth acuity were all present as before, but they felt noticeably enhanced. Without question, the bass improvement played a major role in elevating my overall listening experience.
Most reviewers provide highly detailed descriptions of what they hear with various songs. After years of reading reviews, I’ve realized that what I hear often differs from what they describe. In many cases, I’m not even a fan of the music they use for evaluation, though I’ll admit that I have discovered a few great tracks from reviews that have made it onto my reference list. Beyond musical preferences, countless variables are at play: our rooms, our systems, and, most importantly, our personal preferences all shape our listening experiences. Then, we each have our own way of describing sound, making subjective comparisons even more challenging. That said, I firmly believe that when paired with the right system, the StormAudio MK3 delivers outstanding performance. It will never be the weak link in a setup, especially given its extensive features and near-endless customization options. To suggest that the MK3 doesn’t sound as good as another processor? Well, in my humble opinion, that would be a pretty silly claim.
I will at least provide some of my favorite songs with good imaging, soundstage, depth of acuity, and what I am listening for.
- Ode To A Butterfly - Nickel Creek
- Mandolin is left of center stage – front edge of stage.
- Fiddle is middle of the stage – front edge of stage.
- Guitar is right of center – front edge of stage.
- Deep Bass is behind and between the fiddle and guitar
- Strange Fruit – Cassandra Wilson
- Voice is almost center (leans left), forward and high.
- Double Bass is slightly right of center and slightly behind her voice and the dobro guitar.
- Dobro guitar is farther right of center stage and lower than her voice.
- Cornet horn is farther left of center and rear of stage.
- Til Tomorrow – Yello / Til Bronner – 50 Hz
- Alive – oblyx – 40 Hz
- Turner's Ship – Malia / Boris Blank – 20 Hz
- Stay – Yello – low 20 Hz range up to 60 Hz
- Get On – Yello – mid 20 Hz range up to 60 Hz
- Bass I Love You Too – Bassotronics – 15 Hz and 30 Hz
- Colors (Pt. 2) – Halsey – 20 Hz
- House of the Rising Sun - Hanne Boel - 30 Hz up to 60 Hz
- Uninvited - Alanis Morisette - 20 Hz and up
- Queen Mary - Francine Thirteen - 30 Hz
A few weeks later, my new replacement MK3 arrived. I sat down and listened to several of my reference songs one last time with the HTP-1, paying particular attention to the bass songs I listen to most often. I then reconnected my MK3 and began the setup process, including ART. About an hour later, I had it ready for listening. After a very short listening session, a sigh of relief turned into several hours of listening. That somewhat magical bass was back, and I was seriously diggin' it, and I've been diggin' it since.
Remote Control Options
Controlling the MK3 is effortless, thanks to multiple options ranging from a physical remote to mobile and web-based interfaces. I've already shared an image of the web user interface remote control, which gives the most detailed control of most functions on one screen. The other two options are the physical remote and the mobile app StormRemote (Android or iOS). I use the web UI remote more because I usually listen to music while using Roon on my laptop. It is also convenient for measuring and testing since I'm using REW on the same laptop.
The physical remote is small and handy for making a quick change while watching a movie or TV. It covers all the basics. One minor nitpick for me is having to scroll through my presets when comparing, but I'm usually on my laptop when comparing presets, and it probably wouldn't be an issue for most. Eventually, I'll narrow my presets down to only a few.
The StormRemote app is a no-nonsense remote. While I haven't used it often on my Android phone, it has come in handy a few times to switch between presets and engage Auro-Matic while using my Roon Android app. I consider it a nice convenience. It may be just the right remote for someone constantly on their mobile phone (who isn't these days). I can certainly see how it might be convenient for phone surfers while listening to music. Surely, phones are put aside during movies, right?
I won't show all the possible screens of the StormRemote app, but I'll share a few. (Click on the image to enlarge.)
Modularity: Vinyl Room
It might seem like this section should have appeared earlier in the review, but it was close to the last piece of the puzzle—so to speak. One of my vinyl setups shares space in my dedicated room with my main home theater and listening system. This vinyl setup has a Music Hall Stealth turntable feeding into a miniDSP ADept phono stage. Previously, it fed a miniDSP SHD processor. The SHD outputs were connected to a pair of SVSound 3000 Micro subwoofers and a McIntosh MA252 amplifier powering my Wharfedale Super Linton speakers. The miniDSP SHD only has basic Dirac Live and lacks the robustness of DLBC or DLART, and I haven't been able to get it to sound like I wanted. So... I upgraded my 24-channel MK3 to 32 channels and integrated my vinyl system into the MK3 to gain access to DLART for my vinyl system in hopes of getting the sound I wanted. While winding down this epic novel, I figured, why not add one last chapter?
The upgrade must be performed by an authorized StormAudio dealer. Fortunately, I have a close friend who is a dealer, and the process took less than an hour. The upgrade involved replacing the 8-channel output panel with a 16-channel output board. Since I already had the 8-channel panel, an additional DAC board was not required. However, for a 16-channel unit, the DAC board and the output panel must be replaced. The upgrade was plug-and-play, requiring no firmware updates or additional configuration. Once powered back on, the eight extra channels appeared automatically. I used four of these additional channels for my vinyl system. This upgrade eliminated the need for the SHD. I can now switch presets in the MK3 for vinyl playback, assuming my McIntosh amp is powered on. Additionally, I can still route any source through the vinyl system. Like the Child Room, this setup makes swapping out amplifiers and evaluating speakers for two-channel listening incredibly easy. Additionally, if needed or desired, I can create a new layout with a new theater and incorporate any number of my larger subwoofers from my main system instead of relying solely on the SVS 3000 Micro subwoofers.
Creating the Vinyl System is easy enough. I opened the Main Speaker page from the menu and selected the [+] icon to add a new Theater.
Clicking this [+] icon opens the Configurator, where I add a subwoofer to change the default configuration from 2.1 to 2.2.
After saving the configuration, I am returned to the Main Speakers page, which will show my added Theater, which I renamed Vinyl System. Next, I will click on the [Output mapping] button and map the speakers appropriately to the channels they are each connected to on the rear of the unit. I have already saved the mapping in the image below. (Click on the image to enlarge.)
Now, I am ready to create a Baseline profile and preset. I click the [Edit] icon to open the speaker settings page. (Click on the image to enlarge.)
I changed the default New Profile 1 to BaselineV on the speaker settings page, representing my Baseline vinyl preset. As a precaution, I set the subsonic filter to 20 Hz since the SVS 3000 Micro subwoofers are only rated to 23 Hz. (Click the image to enlarge.)
I ran the Dirac calibration as a first run to obtain my levels and delay (not required) as a convenience for future calibrations. These values were entered into the BaselineV preset and saved. Afterward, I ran my normal Dirac calibration for ART. Since this is a fairly simple ART setup with only a 2.2-speaker configuration. I left the [Support Levels] at the default of -18 dB, set the [F-support Low] to 80 dB for the speakers and 25 dB for the subwoofers, and did not change the default [F-support High] of 150 Hz. I allowed them to support each other. After running ART, I saved my preset and named it Vinyl. When I select the Vinyl preset, the MK3 automatically switches the system from Main to Vinyl and vice versa. As mentioned earlier, my McIntosh amp will need to be powered on for the vinyl system to play, as well as the turntable and phono preamp, if I desire to listen to vinyl, which is the reason for the vinyl system.
Below are the REW measurements of the left front and subwoofers before and after ART. Since system decay was minimal for this system, and ART produced no meaningful improvement, I did not see a need to post any of those graphs. (Click the image to enlarge.)
I had been seriously struggling to make my vinyl system sound right in my dedicated room with the SHD processor leading the system. I had second thoughts about leaving it in the room, not that I had anywhere else to set it up. After integrating it with the MK3, I discovered that one issue may have been improper time alignment between the speakers and subwoofers. I manually measured the distance and entered it into the SHD settings, but it was significantly off, according to StormAudio’s Dirac calibration. Another challenge was crossover integration. No matter how many different crossover points I tried for the speakers and subwoofers, I couldn't achieve a smooth transition in the measured response. Placement may have also been a limitation. I had previously used the SHD in this room with my main speakers and multiple subwoofers, and it performed excellently—but I had greater flexibility in speaker and subwoofer placement then. Bottom line... I was disappointed with what was happening with my vinyl system and my inability to get it tuned right. Upgrading to the extra channels in the StormAudio MK3 and integrating the vinyl system into the MK3 was my last resort to see if it would help.
Once again, the StormAudio MK3 and ART proved to be nothing short of impressive. After setting the system up with ART, I was astounded by how good it sounded. The SVS 3000 Micro subwoofers blended perfectly with the Wharfedale Super Lintons. Switching the presets back and forth and comparing the main system with the vinyl system while listening to Roon was a treat. Naturally, the vinyl system couldn't keep up with the main system at louder volumes or with extended bass response, nor was the soundstage as large. However, at lower volumes, they were not miles apart. The imaging was spot-on in both systems, and the overall tonal qualities were similar. Each system has its own character, but it is not as drastic as some might think it would be. The vinyl system is perfect for vinyl listening because I listen at lower volumes, mostly for background music while writing, researching, or chatting online. Trying new speakers in the room will also be fun as opportunities arise. Thanks to the StormAudio MK3 and DLART, this vinyl system has found a permanent home in my dedicated room. Touchdown, MK3... keep piling on the points and running up the score!
Wrapping It Up
If I were to cover every aspect of this processor, this review could possibly require a sequel, but my main goal was to cover as much of my ART experience as possible. I hope to eventually find time to explore more configurations and options, but I’m satisfied with my current setup and would rather enjoy it than keep tinkering.
In summary, the StormAudio ISP Elite MK3 consistently delivers reference-level performance for both music and movies. Dirac Live ART is a standout feature, tightening bass and enhancing clarity beyond what traditional room correction can achieve. This processor provides exceptional customization and truly next-level sound.
Similar to pairing the right quarterback (the brains of the offense) with the right offensive line and receivers, it's a team effort with the StormAudio MK3 and DLART, as with any high-end processor (the brains of an AV system)... it needs to be paired with good speakers and subwoofers. However, with the StormAudio MK3, the room may not have to be perfect, and subwoofer placement doesn't necessarily have to be perfect, especially if there are multiple subwoofers. I have moved my eight subwoofers around in my room to nearly every possible combination of placement, and they all sounded good with the MK3 and ART. Hence, I ultimately placed them where I preferred them. The MK3 will encourage trading for all the right players to create a winning team, even if the playing field (room) is not the best. A good coach will develop a good team.
The possibilities of what can be accomplished with the MK3 are astonishing. Regardless of what is being listened to or watched, it's hard to imagine anyone not finding pure enjoyment and happiness with this processor. I might be negligent if I didn't say it's an endgame processor. It is my Super Bowl winning quarterback and the most valuable player in my system. With its modularity and upgradeability, the MK3 isn’t just a processor... it’s a long-term MVP. Whether for music or movies, it has delivered the best sound I’ve ever experienced in my room… and that’s a game-changer.
The End! Almost... Highly Recommended!
