I read about the infra bass setting but have not looked for it yet. I am going to spend some time with the default filters before I try anything else. So far I like the way it sounds.Question, did you try to use the "infra bypass" feature to avoid the high pass filter at 20 Hz?
I tried to activate it, but didn't find the checkbox for it.How-to: Infra-bass bypass in ART | Dirac Live Active Room Treatment
Problem formulationDirac Live ART implicitly enforces a steep high-pass filter at 20 Hz, with the consequence that any content below 20 Hz present in a recording would not make its way out to any of the speakers, including infra bass-capable subwoofers. For reference, neither Dirac Live Room...helpdesk.dirac.com
-
Welcome to ASR. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!
Dirac Live standalone for PC and Mac
- Thread starter Flak
- Start date
- Joined
- Oct 14, 2025
- Messages
- 140
- Likes
- 92
Question, did you try to use the "infra bypass" feature to avoid the high pass filter at 20 Hz?
I tried to activate it, but didn't find the checkbox for it.How-to: Infra-bass bypass in ART | Dirac Live Active Room Treatment
Problem formulationDirac Live ART implicitly enforces a steep high-pass filter at 20 Hz, with the consequence that any content below 20 Hz present in a recording would not make its way out to any of the speakers, including infra bass-capable subwoofers. For reference, neither Dirac Live Room...
I got the Infra Bass setting activated by dragging the 20Hz slider past the 20Hz region into the next zone. It shows up as engaged when the Infra area is coloured in.
But I found as a couple of reviews of ART that it doesn't pass through all the Infra Bass and that from 20Hz to about 25Hz is no longer corrected. Bass Control does Infra Bass properly for comparison.
So, for hardcore Home Theatre that requires noticeable Infra Bass, ART does not seem a viable option. That's not a requirement for me so not much of a problem, but I still wouldn't mind in the future that I got all of the bass extension of my subwoofers that I paid for.
Since I had good luck with ART on my 2.2 system I tried adding 2 more subs. Unlike in the past when I tried it with DLBC, which did not work at all with 4 subs, I used subs that have very similar time domain behavior as the main subs to maximise my chance of success. Unfortunately it still does not work with 4 subs. I carefully measured twice and tried several different groupings and support levels and in all cases it gave a nonsensical delay on the left channel of ~8 ms. the right channel is correct at ~0 ms. Since I am listening mid-field at 6 ft this is very wrong and it sounds like it. The correction with the 2.2 configuration gives correct delays of ~0 ms on both right and left channels. The other delays and gains look OK but obviously the sound is broken. Not sure what to do now or how I follow up with DIRAC but I am disappointed. Maybe stereo with 4 subs never has and never will work.
alanca3
Member
I tried to google Audiolense XO, but they don’t publish a manual or any specs. I’m not sure the full extent of its capabilities. That said, none of Dirac’s offerings design user implementable filters. Everything is done “in the box.” Dirac ART does handle bass management in the sense that it sets arbitrary crossovers and filters with the purpose of damping low frequency decay.
It does not allow you to design custom crossover targets for use in eg. active speaker design. There are better tools for that anyways.
By “speakers in a box” they mean that, for the best results, you should have your loudspeakers and subwoofers roughly level matched and integrated before running ART. This is opposed to just placing random speakers in your room without touching a single setting and letting ART figure everything out.
I dont know what you googled but If you are familiar with Mitch Barnett @mitchco , you can read his review here:
In short, Audiolense XO is state of the art (as is Dirac, before the fangirls jump in)
Audiolense Digital Loudspeaker and Room Correction Software Walkthrough
In this article, I walk through the steps of using Audiolense 5.0 (AL5) to biamp my loudspeakers using a linear phase digital crossover. Followed by frequency and time domain correction, so the frequency response is tonally neutral and the direct sound is arriving at my ears all at the same time....
audiophilestyle.com
and
I also use Acourate - another state of the art.
Anyway I came here to ask about developments with Dirac - it has developed a lot which is nice but not an Audiolense replacement yet (for me).
Archimago also reviewed Audiolense XO:
Multichannel Digital Room Correction with Audiolense XO. And "What's your budget?".
A blog for audiophiles about more objective topics. Measurements of audio gear. Reasonable, realistic, no snakeoil assessment of sound, and equipment.
archimago.blogspot.com
I finally got DIRAC ART PC version to work in a 2.4 arrangement. I had to move the 2 extra subs much closer to the LP and then I got sensible delay calculations. I think ART should throw an error message if the measurements result in a nonsensical result which in my case was 8 ms difference between the right and left channels for a 6 foot listening distance. Some people might mistake this result as something "special sounding" if they are not familiar with how delays work and what a stereo image is supposed to sound like. It did sound kind of cool in some limited circumstances but having 2 distinct voices way off to the sides of the sound stage when it should be one in the center was suboptimal to say the least.
See below "before" which is a 2.2 configuration as good as I could get with REW and manual manipulation and "after" which is ART in a 2.4 configuration. Not only does the graph look better but I also prefer the sound. There is plenty of bass even though it rolls off at 20 Hz and it is smooth and tight and balanced sounding. The clarity improvement in the midrange is also noticeable. In conclusion DIRAC is going to get my money for ART
Before (homemade correction with REW and manual adjustments) 2.2 configuration
After ART 2.4 configuration
See below "before" which is a 2.2 configuration as good as I could get with REW and manual manipulation and "after" which is ART in a 2.4 configuration. Not only does the graph look better but I also prefer the sound. There is plenty of bass even though it rolls off at 20 Hz and it is smooth and tight and balanced sounding. The clarity improvement in the midrange is also noticeable. In conclusion DIRAC is going to get my money for ART
Before (homemade correction with REW and manual adjustments) 2.2 configuration
After ART 2.4 configuration
- Joined
- Jul 12, 2024
- Messages
- 3,272
- Likes
- 5,127
Good you like it. Most of the difference looks below 20hzI finally got DIRAC ART PC version to work in a 2.4 arrangement. I had to move the 2 extra subs much closer to the LP and then I got sensible delay calculations. I think ART should throw an error message if the measurements result in a nonsensical result which in my case was 8 ms difference between the right and left channels for a 6 foot listening distance. Some people might mistake this result as something "special sounding" if they are not familiar with how delays work and what a stereo image is supposed to sound like. It did sound kind of cool in some limited circumstances but having 2 distinct voices way off to the sides of the sound stage when it should be one in the center was suboptimal to say the least.
See below "before" which is a 2.2 configuration as good as I could get with REW and manual manipulation and "after" which is ART in a 2.4 configuration. Not only does the graph look better but I also prefer the sound. There is plenty of bass even though it rolls off at 20 Hz and it is smooth and tight and balanced sounding. The clarity improvement in the midrange is also noticeable. In conclusion DIRAC is going to get my money for ART
Before (homemade correction with REW and manual adjustments) 2.2 configuration
View attachment 544391
After ART 2.4 configurationView attachment 544392
- Joined
- Jan 7, 2020
- Messages
- 1,677
- Likes
- 2,393
Maybe the resonance, but there is a much smoother response upto 160Hz, the dip at 60Hz is filled in and the peak at 80-90Hz is controlled. Should be very audible and sound better
I just let it do it's default thing so full range but the kh310 's are pretty flat so it really doesn't do much of anything above 300 Hz
Yes it is audible and I prefer it. There is a lot less energy below 20 Hz that makes the graphs look different but that really doesn't come into play in most cases. The resonances are most likely mechanical (drywall hitting studs and other mechaical buzzes) which my understanding is ART can't really help with. I was hoping to see shorter decay based on other graphs I have seen but not in my case apparently. I will probably try tweaking some setting eventually but for now I am happy to listen for awhile.
It would be nice to see both spectrograms normalized to the peak SPL, so that differences in decay can be better compared.I finally got DIRAC ART PC version to work in a 2.4 arrangement. I had to move the 2 extra subs much closer to the LP and then I got sensible delay calculations. I think ART should throw an error message if the measurements result in a nonsensical result which in my case was 8 ms difference between the right and left channels for a 6 foot listening distance. Some people might mistake this result as something "special sounding" if they are not familiar with how delays work and what a stereo image is supposed to sound like. It did sound kind of cool in some limited circumstances but having 2 distinct voices way off to the sides of the sound stage when it should be one in the center was suboptimal to say the least.
See below "before" which is a 2.2 configuration as good as I could get with REW and manual manipulation and "after" which is ART in a 2.4 configuration. Not only does the graph look better but I also prefer the sound. There is plenty of bass even though it rolls off at 20 Hz and it is smooth and tight and balanced sounding. The clarity improvement in the midrange is also noticeable. In conclusion DIRAC is going to get my money for ART
Before (homemade correction with REW and manual adjustments) 2.2 configuration
View attachment 544391
After ART 2.4 configurationView attachment 544392
It seems that Dirac ART does a really great job in optimizing SPL linearity and reducing the effect of modes/ resonances below 150Hz.
Which pre-ringing? Do you have examples?
I read that ART uses mainly minimum phase filters for bass correction, which should avoid pre-ringing - when delays are set correctly.
@Keith_W mentioned it several times, the last post I can find about it, is here: https://www.audiosciencereview.com/...oom-treatment-art-by-dirac.40743/post-2630329
But I've also noticed it with my measurements.
Thank you, I was not aware of this phenomenon.
I checked and the levels were close and when matched didn't really change the spectrograms. I think to be fair I would have to high pass the REW correction at 20 Hz but not sure about that either as the 20 Hz high pass filter ART uses is VERY steep and if I applied a steep high pass like that in REW it would really throw the time domain off. From what I can see ART / MIMO is doing things that are not straightforward to emulate using REW. I did take a look at the "clarity" tab, which is a ratio of early to late energy and it is does look like ART improves this ratio in the lower frequencies. I don't have a lot of experience correlating this measurement to what I hear but "clear" is a subjective term I would use to describe the ART results.
Clarity Baseline REW correction
Clarity ART
Keith_W
Master Contributor
Clarity is an useless measurement. It's supposedly the ratio of the first x milliseconds vs. total energy, but:
1. Think about what x milliseconds means. C50 = first 50ms. One period of 20Hz is already 50ms. And 50ms = 2.5 periods of 50Hz. That is bugger all time for the sound to decay. Because of this, "clarity" is always worse for low frequencies. Not because it is actually worse, it's because the calculation makes it look worse.
2. Clarity is location-dependent, especially for long wavelengths.
3. Nobody knows what "good" clarity is. You would think that higher clarity is better, but it isn't. Remember that an anechoic chamber will 100% clarity, and maybe a room stuffed full of treatment will have 90% clarity in the upper frequencies or so. And we all know that we don't want to overtreat our rooms and turn it into an anechoic chamber. So there is an upper limit for clarity. But what is that limit? Nobody knows.
4. Clarity is subject to the usual pitfalls of bass measurement - namely, signal to noise ratio. In a typical room, bass usually falls off in the same region room noise is increasing. REW can't tell the difference between bass energy and room noise, so it appears as if there is a lot of energy lingering around. Result: instant drop in "clarity". It's not real, it's a measurement artefact.
In short: don't bother with it. It won't give you any useful information.
1. Think about what x milliseconds means. C50 = first 50ms. One period of 20Hz is already 50ms. And 50ms = 2.5 periods of 50Hz. That is bugger all time for the sound to decay. Because of this, "clarity" is always worse for low frequencies. Not because it is actually worse, it's because the calculation makes it look worse.
2. Clarity is location-dependent, especially for long wavelengths.
3. Nobody knows what "good" clarity is. You would think that higher clarity is better, but it isn't. Remember that an anechoic chamber will 100% clarity, and maybe a room stuffed full of treatment will have 90% clarity in the upper frequencies or so. And we all know that we don't want to overtreat our rooms and turn it into an anechoic chamber. So there is an upper limit for clarity. But what is that limit? Nobody knows.
4. Clarity is subject to the usual pitfalls of bass measurement - namely, signal to noise ratio. In a typical room, bass usually falls off in the same region room noise is increasing. REW can't tell the difference between bass energy and room noise, so it appears as if there is a lot of energy lingering around. Result: instant drop in "clarity". It's not real, it's a measurement artefact.
In short: don't bother with it. It won't give you any useful information.
So based on the above it sounds like any "decay" type of measurements in REW for LF are suspect? If you wanted to see if ART really has any effect on decay (which is it's big claim to fame) what would you look at?
Keith_W
Master Contributor
The spectrogram. There are plenty of examples in the main ART thread. Make sure you "normalize for every frequency" and choose a sensible zoom. Once again, all bass decay measurements are sensitive to poor SNR. So I would choose a long sweep, maybe 1M.
By the way, when you are comparing two measurements (say before-after), it is helpful to have them on the same graph. So for your "clarity" measurement above, click on "Overlays" then click on the "Clarity" tab. If your measurements are on different graphs, it becomes very difficult to compare.
You may also notice I have a habit of making animated GIF files when comparing spectrograms. It makes the spectro very easy to read since your eyes are sensitive towards movement.
Similar threads
