ttocs
Global Moderator 
I always have a wonderful time, wherever I am, whomever I'm with. (Elwood P Dowd)
Posts: 8,170
|
Post by ttocs on Jan 17, 2024 20:58:56 GMT -5
To aid in the discovery of all things ART, this thread, for the discussion of what ART is, how it works, and how it relates to Dirac Live, as well as Dirac Live Bass Control.
Dirac has a number of apps targeted for various aspects of the audio world. The three that I see as interesting to me as a home audio enthusiast are:
Dirac Live, DL
Dirac Live Bass Control, DLBC
Dirac Live Active Room Treatment, ART
Dirac Live is the main app, and DLBC and ART operate within Dirac Live.
Dirac Live is a Full Range, individual speaker, correction application with a range of 20Hz to 20,000Hz. In some processors this app is split into more segments that cover a more limited spectrum, but for Emotiva processors it's a full range product currently.
Dirac Live Bass Control is a single or multiple subwoofer correction application with a range of 20Hz to 500Hz. DLBC operates within DL.
From Dirac Support: "The Bass Control is intended to be used together with Dirac Live room correction, and it utilizes the same impulse response measurements as are used for Dirac Live."
Dirac Live Active Room Treatment is a multi-speaker treatment application with a range of 20Hz to 150Hz. ART can be configured to use speakers in the system to "address and cancel out room reflections and room modes" (Dr Ericsson, Dirac Research). Any speaker capable of 150Hz and lower can participate. ART operates within DL and only targets the range from 20Hz to 150Hz, and is tasked with combining speakers capable of helping to cancel the reflections the app deems fit to cancel.
Emotiva doesn't support DLBC or ART. The question of will Emotiva will support DLBC and/or ART at some point in the future? is uncertain. Until a Emotiva product is on the market, I'll base my understanding in the fact that DLBC and ART do not exist on an Emotiva product. So this discussion is purely an exercise to better understand how ART works. All are welcome to contribute in a positive manner. I myself am not an advocate for ART, but am interested in learning about it and sharing with you here what I might discover elsewhere.
|
|
ttocs
Global Moderator
I always have a wonderful time, wherever I am, whomever I'm with. (Elwood P Dowd)
Posts: 8,170
|
Post by ttocs on Jan 17, 2024 20:59:26 GMT -5
|
|
ttocs
Global Moderator
I always have a wonderful time, wherever I am, whomever I'm with. (Elwood P Dowd)
Posts: 8,170
|
Post by ttocs on Jan 17, 2024 20:59:42 GMT -5
Reserved for future.
|
|
|
|
Post by LuisV on Jan 17, 2024 22:46:38 GMT -5
No date, but based on this interview with Phillip Jones, Dirac ART is coming to Marantz. It would make sense that since the Cinema 40, 50 and AV10 have 4 sub outputs and support Dirac Live and Dirac Live Bass Control that they will be upgradable to ART. Fingers crossed...  Skip to 11:32 |
|
ttocs
Global Moderator
I always have a wonderful time, wherever I am, whomever I'm with. (Elwood P Dowd)
Posts: 8,170
|
Post by ttocs on Jan 22, 2024 22:59:31 GMT -5
No date, but based on this interview with Phillip Jones, Dirac ART is coming to Marantz. It would make sense that since the Cinema 40, 50 and AV10 have 4 sub outputs and support Dirac Live and Dirac Live Bass Control that they will be upgradable to ART. 4 subwoofer outputs seems to be the magic number. I heard that April is the target for M and D to get ART. I'm inclined to buy a M just for trying out ART. In order to get an ART license, the processor must first have licenses for DL and DLBC, if, there are subwoofers configured. See below. From the Dirac ART Setup Guide " Note also, that in order to use ART with one or more subwoofers, you will also need to hold a Bass Control license. You then need to
i. either, purchase and activate a Bass Control license,
ii. or, deactivate the subwoofers by removing them from the device's speaker configuration." |
|
ttocs
Global Moderator
I always have a wonderful time, wherever I am, whomever I'm with. (Elwood P Dowd)
Posts: 8,170
|
Post by ttocs on Jan 22, 2024 23:25:45 GMT -5
Reading the guides in the second post doesn't take long, and they give a pretty good sense of what to expect from anyone's current system and how helpful the Support speakers could be, or not. The videos on the other hand, take longer, but one can always speed up the video as wanted.
Should one be concerned about directionality of sound coming from Support speakers which support Main speakers, the Support speakers can be adjusted as wanted using the "F-support High" parameter to reduce the directionality of sound away from Main speakers. (See "Direction Of Arrival" in the Technical Guide: ART Channel Group and Support Settings, and see "Detailed Description of Parameters" in the Dirac Live Active Room Treatment Setup Guide)
ART seems to be accommodating in its usage, allowing great latitude in adjustment parameters, so for systems with limited "F-support Low" (see "Detailed Description of Parameters" in the Dirac Live Active Room Treatment Setup Guide in second post) type speakers, the effect can still be useful in the range the speakers have capability.
|
|
|
|
Post by marcl on May 23, 2024 8:08:33 GMT -5
|
|
|
|
Post by lrobertson on May 24, 2024 9:10:16 GMT -5
How are you guys doing? I know it's a little off subject but I still haven't heard any thoughts from anyone on the true target from magic beans. Layman's says it just deciphers the transition region or schroeder frequency of your room based on nearfield vs mlp measurements and simply creates individual speaker target curves that are easily copied into Dirac. Supposedly it supports dlbc but doesn't have a specified support for art. Does this make sense? I know some people prefer the limited dirac which only corrects for up to 500hz for a similar reason but wouldn't this seem to be more on point where not only do you get your rooms unified low frequency targets for all speakers up to the transition region but then you also get your individual speaker targets for the higher region? Does it make sense we wouldn't want to correct for reflective information at the mlp at the frequencies where the room hasn't taken over?
|
|
KeithL
Administrator
Posts: 10,275
|
Post by KeithL on May 24, 2024 11:44:52 GMT -5
The short answer is this... Everyone pretty much agrees that, at very low frequencies, sound waves act like "pressure" (which is essentially non-directional). And everyone pretty much agrees that, at mid-to-high frequencies, sound acts like a "directional wavefront", which reflects off surfaces. However there is no wide agreement about exactly where it ceases to be one and becomes the other... or how that "transition area" should be handled... (And especially how you would want to handle "the area where it's sort of one and sort of the other"... ) My rather educated guess is that "it depends"... There is no such thing as a perfect solution that works well in all situations... So there is a good chance that, in a given situation, one or the other may work better... I would assume that their system may work better with some rooms and some speakers... (But probably not all...) The arguments in favor of either are actually quite complex... For example, at 10 kHz, sound is extremely directional, and can be observed to act much like you would expect to model it with "a flashlight and some mirrors". BUT, at 10 kHz, where a wavelength is about one inch, a shift of microphone position of a half inch will COMPLETELY change the results of your measurements. And, likewise, a solution that is "perfect at a certain spot", may be "totally wrong" a mere half inch to the left or right. However, in real life, our brains seem to not notice this, and apparently manage to "average out the result over an area". Play a monaural 10 kHz tone, from two speakers, side by side, and measure the output, using a small microphone... You will detect a classic comb filter effect... where the sound gets drastically louder and quieter, every half an inch, as you move the microphone sideways. Yet, when you listen to music, with most speakers you don't notice this, because the room blurs the effect, and because your brain apparently "discounts" it to a significant degree. Incidentally, you mentioned the limited version of Dirac Live that only corrects up to 500 Hz... The full version of Dirac Live, which corrects from 20 Hz to 20 kHz, can be configured to act the same way. The "curtains" determine the range over which Dirac Live will correct... You can limit it to frequencies below 500 Hz by simply moving the upper curtain down to 500 Hz. (The "limited version of Dirac Live" is limited because you don't get the option of doing the opposite... even if you want to.) How are you guys doing? I know it's a little off subject but I still haven't heard any thoughts from anyone on the true target from magic beans. Layman's says it just deciphers the transition region or schroeder frequency of your room based on nearfield vs mlp measurements and simply creates individual speaker target curves that are easily copied into Dirac. Supposedly it supports dlbc but doesn't have a specified support for art. Does this make sense? I know some people prefer the limited dirac which only corrects for up to 500hz for a similar reason but wouldn't this seem to be more on point where not only do you get your rooms unified low frequency targets for all speakers up to the transition region but then you also get your individual speaker targets for the higher region? Does it make sense we wouldn't want to correct for reflective information at the mlp at the frequencies where the room hasn't taken over? |
|
|
|
Post by lrobertson on May 24, 2024 12:12:26 GMT -5
The short answer is this... Everyone pretty much agrees that, at very low frequencies, sound waves act like "pressure" (which is essentially non-directional). And everyone pretty much agrees that, at mid-to-high frequencies, sound acts like a "directional wavefront", which reflects off surfaces. However there is no wide agreement about exactly where it ceases to be one and becomes the other... or how that "transition area" should be handled... (And especially how you would want to handle "the area where it's sort of one and sort of the other"... ) My rather educated guess is that "it depends"... There is no such thing as a perfect solution that works well in all situations... So there is a good chance that, in a given situation, one or the other may work better... I would assume that their system may work better with some rooms and some speakers... (But probably not all...) The arguments in favor of either are actually quite complex... For example, at 10 kHz, sound is extremely directional, and can be observed to act much like you would expect to model it with "a flashlight and some mirrors". BUT, at 10 kHz, where a wavelength is about one inch, a shift of microphone position of a half inch will COMPLETELY change the results of your measurements. And, likewise, a solution that is "perfect at a certain spot", may be "totally wrong" a mere half inch to the left or right. However, in real life, our brains seem to not notice this, and apparently manage to "average out the result over an area". Play a monaural 10 kHz tone, from two speakers, side by side, and measure the output, using a small microphone... You will detect a classic comb filter effect... where the sound gets drastically louder and quieter, every half an inch, as you move the microphone sideways. Yet, when you listen to music, with most speakers you don't notice this, because the room blurs the effect, and because your brain apparently "discounts" it to a significant degree. Incidentally, you mentioned the limited version of Dirac Live that only corrects up to 500 Hz... The full version of Dirac Live, which corrects from 20 Hz to 20 kHz, can be configured to act the same way. The "curtains" determine the range over which Dirac Live will correct... You can limit it to frequencies below 500 Hz by simply moving the upper curtain down to 500 Hz. (The "limited version of Dirac Live" is limited because you don't get the option of doing the opposite... even if you want to.) How are you guys doing? I know it's a little off subject but I still haven't heard any thoughts from anyone on the true target from magic beans. Layman's says it just deciphers the transition region or schroeder frequency of your room based on nearfield vs mlp measurements and simply creates individual speaker target curves that are easily copied into Dirac. Supposedly it supports dlbc but doesn't have a specified support for art. Does this make sense? I know some people prefer the limited dirac which only corrects for up to 500hz for a similar reason but wouldn't this seem to be more on point where not only do you get your rooms unified low frequency targets for all speakers up to the transition region but then you also get your individual speaker targets for the higher region? Does it make sense we wouldn't want to correct for reflective information at the mlp at the frequencies where the room hasn't taken over? Thanks Keith. He did say he isn't doing anything special we couldn't do ourselves manually with rew, dirac, and a few hours of trial and error. As to whether it is worth $250 I guess I'll wait and hopefully hear some reviews. It does sound like because Dirac relies just on the mlp measurements that they will be limited on ever applying an accurate custom transition region due to the inability to know what sound is direct vs reflective. But with my limited experience it's hard to know how much juice it is I'm trying to squeeze out of the system which I'd also assume is associated more with each room's properties. I've never had success implementing REW. I would assume if Dirac ever implemented a 3d mic like trinnov it would be able to figure this out.. |
|
KeithL
Administrator
Posts: 10,275
|
Post by KeithL on May 24, 2024 14:17:22 GMT -5
I have to admit that I've only watched some of his video content... I have to admit that the idea of waving the microphone around, while taking continuous measurements, is interesting. By doing so you are "getting a bunch of measurements, taken over an area, sort of averaged together"... And this could be considered to be somewhat equivalent to a "smoothed and averaged measurement of the entire area over which you wave it". But I do wonder about some things, like how repeatable that is, and how the pattern you wave it around in would affect the results. The point there being that, while you are taking measurements over a broader area, the actual amount of detailed information is still limited. I'm not convinced that you're getting "an accurate profile of the near-field" simply by "painting the area manually with a microphone while a test tone is playing". (Of course, in practice, we don't need "perfect", or even "very very good", but merely "better than the other way most of the time".) I should also point out that even Trinnov's "3D microphone" is NOT collecting a detailed and complete set of data. I would consider making physical measurements using images as a reasonable analogy. A single image can be used to determine a whole set of measurements using "photogrammetry". But a pair of images, taken a few inches apart, can record a "stereo 3d image", from which you can extract a lot of information (I would equate that to their "3d microphone"). But that still won't produce as much detailed information as an actual 3d scan... And even a true 3d scan only gives you limited data... (For example a 3d scan doesn't tell you what color a surface is, or if it is shiny or dull, or if it is soft or hard, or even details about its fine texture.) The other important thing, which you alluded to, is that things like room acoustics cannot always be corrected, even if they can be measured. To pick an obvious example... Take a room, with a set of sliding glass doors on the left wall, and a large heavy tapestry on the right wall. Both walls are acoustically similar at 30 Hz... But the left wall is essentially a mirror at a few hundred Hz and above... While the right wall is going to absorb sound progressively more as the frequency goes up... And, at 10 kHz or so, that right wall will be a sort of black hole... And there is no way you can correct that using EQ, or DSP processing, or by manipulating things like phase and delay... The only way to correct that situation is to correct THE ROOM... at least to the point of making it more or less symmetrical between left and right. REW will absolutely give you a lot more information than Dirac does... Which will be a huge benefit... if you know what to do with that information. It's also a great learning tool if you just want to learn how the various measurements correlate with what you hear. ....................................... Thanks Keith. He did say he isn't doing anything special we couldn't do ourselves manually with rew, dirac, and a few hours of trial and error. As to whether it is worth $250 I guess I'll wait and hopefully hear some reviews. It does sound like because Dirac relies just on the mlp measurements that they will be limited on ever applying an accurate custom transition region due to the inability to know what sound is direct vs reflective. But with my limited experience it's hard to know how much juice it is I'm trying to squeeze out of the system which I'd also assume is associated more with each room's properties. I've never had success implementing REW. I would assume if Dirac ever implemented a 3d mic like trinnov it would be able to figure this out.. |
|
KeithL
Administrator
Posts: 10,275
|
Post by KeithL on May 24, 2024 14:36:00 GMT -5
Now... for a bit of fun... how WOULD you fix that problem room? The obvious answer is to make the walls more similar... Perhaps hang some drapes over those sliding glass doors... Or maybe replace that tapestry with something a bit less absorptive... like maybe a painting. But what if we just can't do that? How about toeing in the Front Left and Front Right speakers... They're the speakers that contribute most to left-right imaging of important stuff in the front... And by toeing them in more, or even a lot more, we minimize the effect of those dissimilar side walls... In fact, if we do a lot of stereo listening, we could toe them in a lot, and move our listening position closer to the front... This way the direct sound from those speakers would dominate what we hear... (it won't entirely fix the problem... but will make it a lot less obvious and intrusive.) ................................ To pick an obvious example... Take a room, with a set of sliding glass doors on the left wall, and a large heavy tapestry on the right wall. Both walls are acoustically similar at 30 Hz... But the left wall is essentially a mirror at a few hundred Hz and above... While the right wall is going to absorb sound progressively more as the frequency goes up... And, at 10 kHz or so, that right wall will be a sort of black hole... And there is no way you can correct that using EQ, or DSP processing, or by manipulating things like phase and delay... The only way to correct that situation is to correct THE ROOM... at least to the point of making it more or less symmetrical between left and right. .......................... ............................ |
|
|
|
Post by marcl on May 24, 2024 15:51:10 GMT -5
This ends up being an interesting discussion of some approaches to implementing Dirac ART. The approach can somewhat mimic the Trinnov wave-forming approach. Or, there's a method of using primary high output subs in combination with smaller support subs ... to compensate for surrounds with limited bass output.
|
|
ttocs
Global Moderator
I always have a wonderful time, wherever I am, whomever I'm with. (Elwood P Dowd)
Posts: 8,170
|
Post by ttocs on May 24, 2024 17:01:24 GMT -5
This ends up being an interesting discussion of some approaches to implementing Dirac ART. The approach can somewhat mimic the Trinnov wave-forming approach. Or, there's a method of using primary high output subs in combination with smaller support subs ... to compensate for surrounds with limited bass output. I like Matthew. But please, stop making videos right after lunch! So many people make videos after eating and it shows. They should make the videos just before eating. But I like the design approach. And I also like his attitude of saying that one can make do with less if that's all you have, and then making the comparison of some of the next levels up. Using all subs for ART seems like it would yield a better result. |
|
|
|
Post by marcl on May 24, 2024 17:53:17 GMT -5
This ends up being an interesting discussion of some approaches to implementing Dirac ART. The approach can somewhat mimic the Trinnov wave-forming approach. Or, there's a method of using primary high output subs in combination with smaller support subs ... to compensate for surrounds with limited bass output. I like Matthew. But please, stop making videos right after lunch! So many people make videos after eating and it shows. They should make the videos just before eating. But I like the design approach. And I also like his attitude of saying that one can make do with less if that's all you have, and then making the comparison of some of the next levels up. Using all subs for ART seems like it would yield a better result. Yeah he was kind of dragging. I play him at 1.25 - 1.5 anyway. Maybe it's the other way, he only has time to do it middle of the night then does a quick edit mod-day. I think his wife helps with the edits. But yes ... supporting small subs scattered. Like a Rythmik 2x8 sealed servo. |
|
|
|
Post by lrobertson on May 24, 2024 17:59:34 GMT -5
This ends up being an interesting discussion of some approaches to implementing Dirac ART. The approach can somewhat mimic the Trinnov wave-forming approach. Or, there's a method of using primary high output subs in combination with smaller support subs ... to compensate for surrounds with limited bass output. I feel like this follows exactly what Keith is addressing. Design is always first. The fanciest lipstick you can find to put on your pig doesn't always help. You almost have to have a dedicated room designed for the purpose to really warrant the concern for using the high tech software. I went so far as to design my media room with a golden room ratio and sound isolators, hat channel, diffusers etc. I couldn't have cut all those corners and expected better results with just ART. However now I can't afford storm audio or trinnov so hopefully ART is in the RMC-1x's future. |
|
|
|
Post by bolle on May 25, 2024 4:01:52 GMT -5
I have to admit that I've only watched some of his video content... I have to admit that the idea of waving the microphone around, while taking continuous measurements, is interesting. By doing so you are "getting a bunch of measurements, taken over an area, sort of averaged together"... And this could be considered to be somewhat equivalent to a "smoothed and averaged measurement of the entire area over which you wave it". But I do wonder about some things, like how repeatable that is, and how the pattern you wave it around in would affect the results. The point there being that, while you are taking measurements over a broader area, the actual amount of detailed information is still limited. I'm not convinced that you're getting "an accurate profile of the near-field" simply by "painting the area manually with a microphone while a test tone is playing". Actually that´s how we are doing it in Car Hifi SQ installations for 20+ years. I have personally (co)calibrated some German and European champions in this way. So I guess this method is valid. |
|
|
|
Post by marcl on Jun 4, 2024 12:17:03 GMT -5
This is an excerpt from CEDIA RP22 training that was done recently. It starts with a discussion of subwoofer placement, and then goes on to describe how Trinnov Wave Forming and Dirac ART are meant to work.
|
|
ttocs
Global Moderator
I always have a wonderful time, wherever I am, whomever I'm with. (Elwood P Dowd)
Posts: 8,170
|
Post by ttocs on Jun 4, 2024 17:05:55 GMT -5
This is an excerpt from CEDIA RP22 training that was done recently. It starts with a discussion of subwoofer placement, and then goes on to describe how Trinnov Wave Forming and Dirac ART are meant to work. I haven't gotten through the entire video yet, but am about 3/5 through it so far. Enough to have gotten through the discussion about SBIR, which interests me greatly. My subs are currently lined up along the front wall as close as I could get them, specifically to combat SBIR. Curiously, the array presented a really good response without too much effort right off the bat, and when I have time to play I try different alignment scenarios - some of which prove out as positive stepping stones. Many ways to tackle many bass issues. |
|
