• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. 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!

Which is the best DSP option: DIRAC vs Acourate vs Audiolense vs RePhase vs ?

phoenixdogfan

Major Contributor
Forum Donor
Joined
Nov 6, 2018
Messages
3,296
Likes
5,075
Location
Nashville
Thanks for the detailed comments. And there seems to be a solid consensus that it is the most novice friendly of the bunch by far. And this stuff can be daunting, I recall firsst opening my DEQX manual ( 2 of them about 150 pp in total IIRC) and thinking holy crap, what have I gotten myself into? Thankfully the users group got me to chill long enough, to slowly go through the step by step--you may not understand everything, but just do it, and I did. Now it's intuitive as a soup spoon. Well maybe not that intuitive. But we all know the drill with complex software, you dig in, you get it to work in some basic use case, and then add to ones chops piece by piece. You look back and think, that was no big deal. Photoshop was that way for me. Still no wizard by any stretch, but I can get what I need done.

One thing Mitch Barnett talks about in his e book is the sheer torture of repetition one encounters in Acourate, having to repeatedly enter the same statements over and over vs having some handles and automation that would reduce the workload by a huge chunk. That's one thing DEQX does beautifully, maybe a bit too well insofar as you sacrifice flexibility (eg target curve is flat, you have to add EQ later). Otherwise it is cursors, value fields, and check marks.
One thing I don't agree with is the superiority of using linear phase filters. Maybe I'm missing something, but based on what I've been able to research (and Amir even has a video about it) phase distortion in loudspeakers is at best an extremely subtle barely audible effect, if indeed it's audible at all. As he points out the in-room playing of a loudspeaker introduces more out of phase room reflections from every angle arriving at various times than could ever be caused by the phase shifts introduced by a speaker's crossovers or for that matter its electronics. He cites Toole who in turn cites studies by psycho acoustic researchers reporting the results of their experiments using live test subjects to corroborate that conclusion. So unless there is something very distinct and especially audible and uniquely deleterious about the phase shifts introduced by the speaker as opposed to those introduced by the room, it's very hard to see how they would be terribly significant.
 
Last edited:

gnarly

Addicted to Fun and Learning
Joined
Jun 15, 2021
Messages
990
Likes
1,390
Thanks in great part to all the folks here at ASR I feel reasonably well informed about the various options--at least in general terms. 6 months ago I wasn't even able to articulate the question properly. I had heard of REW but knew precious little else. Now I know a teeny weeny bit about microcomputers, HAT's, the myriad of flavors of players, Python, BB coding, and just how sorely needed a simple, affordable DSP/multichannel DAC that goes a step beyond the MiniDSP products in SQ and processing power, and that isn't too much more expensive. So we hobble together what we can, and talk about cats, there are a few dozen ways to skin this one.

I dig DIY for that reason--a wannabe engineer that ended up as a physician who likes to get into the nitty gritty of the hobby. I was a bystander enthusiast who was picked clean of 10k before waking up to the fact that I was on the turntable of upgrades--in fact I owned a nice VPI, Zeta arm, Koetsu MC cartridge that was just dandy, but effing $$$. I bought a DEQX showroom sample from them when they were just bringing product to market for less than the TT setup. My only knowledge of the technology was from stopping in a showroom while in Atlanta and being shown the new "digital" Meridian speaker line. It was hardly genius on my part to recognize passive XO's were a waste of money and that individually tailored EQ for drivers was a lot more saavy than buying some 32 band sound "palette" from M. Levinson. Talking 25 years ago to be exact.

It's been a lot of fun and the $$ I have saved by having been cured of upgrade-itis has been huge. There a lot of naysayers who argue at every opportunity that the savings is an illusion with DIY speakers. Which might be the case for a one off and done project--heck I have 700 dollars worth of routers and saws that were pretty much just for audio. But I have amortized these and you know once every five years or so I get the hankering to try a different approach. Those "upgrades" were more along a 500 to 1000 for drivers, and a couple hundred for lumber and whatever else per pair, versus spending 5 to 10K on a store bought product. And a built a few kits for friends. So all good. Now just needing to complete the chain by tackling the nuts and bolts of DSP. Also all good. Meanwhile I am now close to accumulating enough extra pairs of speakers to go 7.3 My assumption is that even though there will be a hodge podge mix of brands, my belief is if one uses the same target curve they will play well enough together to make for convincing envelopment. I think at that point I'm going to bite the big one and purchase a Monoprice pre/pro.

The next stereo project will have this sweet young thing as its heart (or maybe the Mg-Al version--1/2 the cost), running from 500 to 5kHz. Early report is vocals are like hearing angels in your living room. All waiting the Audio Compass review to see how bright this star shines. But confirmed specs are 97dB efficiency, able to hit 116dB w/o fuzz and just a smidge of compression. Haven't seen any polars yet.

It aint cheap as in open that wallet wide and prepare for 10 Benjamins to be plucked per driver. But who said endgame was supposed to be cheap. Anyhow, enough rambling.

Sounds like we have similar audio backgrounds/stories. For 30 years I was a hi-fi "high end" addict. Read all the mags, auditioned all the stores in NYC...big electrostat/ribbon/planar fan....with all the $$$ electronics, moving coil setups, yada yada. Wouldn't touch EQ with a 10 ft pole LOL.
Detoured into top drawer prosound stuff for live sound, most expensive speakers I ever bought, but they included amplification and processing.
Amazed at how close they sounded to my electrostats in term of clarity, so i studied what was going on in prosound. Made me feel like a child in elementary school, comparing what i had learned from 30 years of hi-fi, to the practical real engineering behind proaudio. A guy named Danley i had never heard of (lol) posted some plans for a killer sub (Labhorns) on Prosound forum about 20 years ago. I built some and have been DIY ever since.
I've built 30-40 speakers over the last 7 years, ranging from conventional designs, to coaxials, to all sections horn loaded, to CBTs and straight line arrays, to MEHs, and of course more types of subs. Honestly, i haven't heard any commercial design, hi-fi or prosound, that would be worth much $ to trade up from my latest MEH build. I'm thrilled really.....
I do think a big part of the DIY speaker process, or at least my process, is mastering measurements and applying appropriate DSP, first on the driver by driver level, and then at the xover summation level. I know the speaker manufacturers, both home and pro, no doubt build better boxes in terms of pure acoustic engineering with their knowledge base, driver experiences, and testing facilities. But they are also hamstrung a little in terms of what they can do with processing, given the application of there products. Few can afford the latency that comes with FIR. Many commercial installs need passives due to difficult servicing locations. Live sound can tolerate delay. Even HT has problems with video sync i hear. This is where i believe some DIYers have a real advantage, if they are music only (can tolerate latency) and can master it.

Anyway, now it's enough of my rambling. Cheers to a fellow DIYer !!
Those BlieSNa look really sweet...good luck :)
 

gnarly

Addicted to Fun and Learning
Joined
Jun 15, 2021
Messages
990
Likes
1,390
One thing I don't agree with is the superiority of using linear phase filters. Maybe I'm missing something, but based on what I've been able to research (and Amir even has a video about it) phase distortion in loudspeakers is at best an extremely subtle barely audible effect, if indeed it's audible at all. As he points out the in-room playing of a loudspeaker introduces more out of phase room reflections from every angle arriving at various times than could ever be caused by the phase shifts introduced by a speaker's crossovers or for that matter its electronics. He cites Toole who in turn cites studies by psycho acoustic researchers reporting the results of their experiments using live test subjects to corroborate that conclusion. So unless there is something very distinct and especially audible and uniquely deleterious about the phase shifts introduced by the speaker as opposed to those introduced by the room, it's very hard to see how they would be terribly significant.
I agree with your take on existing research and the majority of respected opinions....they usually diminish the importance of phase.
All that still doesn't smell right to me though...

Whenever i dig through the past studies quoted by those whose opinions we respect, i most often find the studies to really be lacking in terms of truly isolating phase audibility in real world terms.
It seems the studies have most often had to use headphones because speakers, powerful full range speakers that are necessarily multi-way to achieve realistic SPL, dynamics, and bass extension, have not had the ability to have flat phase until relatively recently.
IOW, we haven't had the ability to truly hear flat phase realistically, to even have a chance to know how rotated phase compares.
Most historical studies appear to me to have compared less-crooked to more-crooked.

A few xovers, not matter how low order, end up providing quite a bit of phase rotation. And i think there is fairly universal agreement that low order designs, in two-ways for instance, sound superior to high order designs. What is that but phase?
What is the perfect impulse and step response response everyone wants, other than both flat mag and phase?
What is impulse inversion, which is the heart of FIR, other than flattening both mag and phase?
And i guess finally, even simple minimum phase EQ's.....fix mag and you automatically fix phase.

Many, if not all, of the prosound speaker manufactures I've come to respect, stress phase. (perhaps just me looking for confirmation ??)
It's funny how the prosound world seems to be so far head of the home sound world, in terms of DSP and processing.

So anyway, there is some of my grounds for skepticism against phase doesn't matter...
and my gut says how can it not?
It's just time, how can time no matter how fine, not matter....until we can absolutely prove it doesn't?

Now, despite all the above, can i say for sure i know the flat phase i achieve in my speaker builds is the source of their great sound?
No, i cant. Like i replied earlier, it's not apples to apples exactly when i compare minimum phase processing to linear phase processing.
Minimum phase processing is much harder to dial in as smoothly as possible....simply because low order xovers are needed and drivers are called on to work wider overlapping bandwidths where rolloffs exist.
Linear phase processing is much easier to dial in smoothly.....because steep complementary xovers allow easy summation of drivers operating in their reduced bandwidth range, of better mag and phase response.

That may be the major reason my linear phase processing i use sounds better to...the simple likelihood of achieving superior frequency response.
I also have measured consistently better polars with the steep xovers.

I do wish i could figure out a better way to truly nail down phase audibility.
 
OP
J

JRS

Major Contributor
Joined
Sep 22, 2021
Messages
1,158
Likes
1,003
Location
Albuquerque, NM USA
Thanks. We will need to have a conversation some time. I'm intrigued by many of your projects, and wholeheartedly agree with the DSP approach you advocate. Start with the drivers--get those shaped up, then the XO's and do your best with the baffles with the help of a PC and thru an iteration or two finalize the driver layout, adjusting XO's as needed. Find the best spot in the room and then apply DRC--gently. It's kept me out of the audio "salons."

PS: One of these days, Danley's approach to compression loading horns begs to be explored. More of a big panel guy, but those have to be the endgame for uncolored, mind blowing dynamics.
 
OP
J

JRS

Major Contributor
Joined
Sep 22, 2021
Messages
1,158
Likes
1,003
Location
Albuquerque, NM USA
I agree with your take on existing research and the majority of respected opinions....they usually diminish the importance of phase.
All that still doesn't smell right to me though...

Whenever i dig through the past studies quoted by those whose opinions we respect, i most often find the studies to really be lacking in terms of truly isolating phase audibility in real world terms.
It seems the studies have most often had to use headphones because speakers, powerful full range speakers that are necessarily multi-way to achieve realistic SPL, dynamics, and bass extension, have not had the ability to have flat phase until relatively recently.
IOW, we haven't had the ability to truly hear flat phase realistically, to even have a chance to know how rotated phase compares.
Most historical studies appear to me to have compared less-crooked to more-crooked.

A few xovers, not matter how low order, end up providing quite a bit of phase rotation. And i think there is fairly universal agreement that low order designs, in two-ways for instance, sound superior to high order designs. What is that but phase?
What is the perfect impulse and step response response everyone wants, other than both flat mag and phase?
What is impulse inversion, which is the heart of FIR, other than flattening both mag and phase?
And i guess finally, even simple minimum phase EQ's.....fix mag and you automatically fix phase.

....
Interesting, a few posts ago I confessed to being brow beaten by Geddes over at DIY into accepting that phase issues could never be solved by DSP, and even if one could get it perfect at 1 meter, the result at the LP would bear no resemblance. So the obvious question, as I assume you can A/B filters, is whether you can identify those with perfect phase vs a more typical result using IIR's that emulate passive components. For some reason I have never really tried as it was so dirt simple to flatten phase down to modal territory. One thing I do like about Mitch's approach is to at least trim the fat from excessive phase down low. I have never been able to do that, one of the reasons AL is appealing, and am assuming Acourate can do as well.
 

Pattern

Member
Joined
Dec 6, 2021
Messages
38
Likes
53
Curious are you using the Dirac as a stand alone license or is it part of a miniDSP or AV receiver? Also are you using that with subs?
I am using it with a DDRC-24 and I am using it with subwoofers. I have a pair of passive standmounts which have each been paired with an active subwoofer crossed at 180hz. that crossover point sounds very high due to localization concerns but because I am running the subs in stereo and very close to co-located with their respective standmount (19 inches from tweeter to sub dustcap) I have effectively created a 3 way semi active floorstander via the miniDSP functions and REW measurements. the Dirac license provided with the ddrc-24 only provides two channels of correction so as far as Dirac was concerned it was correcting two speakers with deep bass extension. the results have been extremely satisfying, LS50s sing when relieved of the low registers. I run Dirac up to 500hz and use a recommended EQ based on Amirm's measurements of my speaker above that point.
 
OP
J

JRS

Major Contributor
Joined
Sep 22, 2021
Messages
1,158
Likes
1,003
Location
Albuquerque, NM USA
I am using it with a DDRC-24 and I am using it with subwoofers. I have a pair of passive standmounts which have each been paired with an active subwoofer crossed at 180hz. that crossover point sounds very high due to localization concerns but because I am running the subs in stereo and very close to co-located with their respective standmount (19 inches from tweeter to sub dustcap) I have effectively created a 3 way semi active floorstander via the miniDSP functions and REW measurements. the Dirac license provided with the ddrc-24 only provides two channels of correction so as far as Dirac was concerned it was correcting two speakers with deep bass extension. the results have been extremely satisfying, LS50s sing when relieved of the low registers. I run Dirac up to 500hz and use a recommended EQ based on Amirm's measurements of my speaker above that point.
Interesting, and yet another way to skin that poor, proverbial cat. I found what you had to say about active 3 way particularly interesting because I am in the same boat, and it was good to hear confirmed that no grievous sins have been committed. I use 15" drivers, each in a 3 cu ft cab, which provide authority and depth w/o yet another "way." (My DEQX only supported 6 channels so my hand was forced). Currently XO'ed to a 6.5" midwoofer that in its Bessel alignment could easily reach as low as 60Hz, but chose a much higher freq (375) to hand off. This decision was based on dynamic concerns and not FR.

Likewise the monitors are stand mounted and I am guessing about 20" as well from dust cap to my midbass center. And so while I don't have the freedom to move them around like subs, at least the cabinet housing the M/T is free from those vibrations w/o weighing a ton. (The woofers cabs are close to 80lbs apiece as is).

Localization seems fine on mine: even using 24dB/octave crossovers, I don't get the sense that things are smeared, and one has to remember even at 375, we are talking 3' long wavelengths, so I am just a tad over the 1/2 lambda rule of thumb at XO and you're at 1/4. Look at state of the art ww/t/ww or ww/m/t/m/ww towers and many are in the 1/4 wavelength range distance between tweeter and purely woofer driver. Sounds like a great system.

A bit of thinking outside the box (pun intended) as well, so bravo.
 

fluid

Addicted to Fun and Learning
Joined
Apr 19, 2021
Messages
691
Likes
1,196
Interesting, a few posts ago I confessed to being brow beaten by Geddes over at DIY into accepting that phase issues could never be solved by DSP, and even if one could get it perfect at 1 meter, the result at the LP would bear no resemblance. So the obvious question, as I assume you can A/B filters, is whether you can identify those with perfect phase vs a more typical result using IIR's that emulate passive components. For some reason I have never really tried as it was so dirt simple to flatten phase down to modal territory. One thing I do like about Mitch's approach is to at least trim the fat from excessive phase down low. I have never been able to do that, one of the reasons AL is appealing, and am assuming Acourate can do as well.
I have to agree with the post above from gnarly that the existing research does not test phase audibility in the way that it is manipulated by the in room DSP software being discussed. If you find yourself arguing audio fundamentals with Earl, there is either a misunderstanding or he is right :)

There is an obvious difference in result between optimizing the phase and step/impulse response anechoically vs doing the same at the listening position. It is true that DSP by itself can only fix one point in space but if it is combined with directivity, treatment, positioning etc. Then the correction can hold very well over a wider area than might be thought. Mitch demonstrates good coherency over an area covering a couch.

I have tested correction filters for myself where one was purely minimum phase and the other phase corrected below 1000Hz. It is very difficult to determine if there is only phase changes causing a difference in sound as these are not laboratory conditions but the only difference between the filters in the programming was forcing minimum phase or allowing phase to be corrected. The difference between them is quite obvious but determining a preference is much harder and it sometimes shifts depending on the source material. In general I choose the phase corrected option as I find it to have the greatest benefit on the widest range of source material. Not everyone might choose the same. The pure minimum phase option could sound more vibrant, wilder and with more energy, sometimes this was good. I have found that when speakers are corrected to be more similar to each other in frequency and level that the sound does become somewhat more polite and that this may not initially be preferred being quite different, but that over time it does end up sounding more correct for want of better words. I found the same with correcting my headphones to the Harman Headphone target. At first it was too different but it did not take long before it clicked and everything started to sound right, going back to uncorrected then sounded obviously worse.

Approaching a speaker design as gnarly described makes it hard to determine phase audibility as the only difference because steep linear phase filters are quite different to shallower IIR filters in more than just phase.
 

hex168

Senior Member
Joined
May 29, 2020
Messages
396
Likes
338
Gnarly wrote in his excellent post, "A few xovers, not matter how low order, end up providing quite a bit of phase rotation. And i think there is fairly universal agreement that low order designs, in two-ways for instance, sound superior to high order designs. What is that but phase?"

Not picking nits,but it is also a way to have a smoother and more gradual directivity change at crossover.

 
  • Like
Reactions: JRS
OP
J

JRS

Major Contributor
Joined
Sep 22, 2021
Messages
1,158
Likes
1,003
Location
Albuquerque, NM USA
I have to agree with the post above from gnarly that the existing research does not test phase audibility in the way that it is manipulated by the in room DSP software being discussed. If you find yourself arguing audio fundamentals with Earl, there is either a misunderstanding or he is right :)

There is an obvious difference in result between optimizing the phase and step/impulse response anechoically vs doing the same at the listening position. It is true that DSP by itself can only fix one point in space but if it is combined with directivity, treatment, positioning etc. Then the correction can hold very well over a wider area than might be thought. Mitch demonstrates good coherency over an area covering a couch.

I have tested correction filters for myself where one was purely minimum phase and the other phase corrected below 1000Hz. It is very difficult to determine if there is only phase changes causing a difference in sound as these are not laboratory conditions but the only difference between the filters in the programming was forcing minimum phase or allowing phase to be corrected. The difference between them is quite obvious but determining a preference is much harder and it sometimes shifts depending on the source material. In general I choose the phase corrected option as I find it to have the greatest benefit on the widest range of source material. Not everyone might choose the same. The pure minimum phase option could sound more vibrant, wilder and with more energy, sometimes this was good. I have found that when speakers are corrected to be more similar to each other in frequency and level that the sound does become somewhat more polite and that this may not initially be preferred being quite different, but that over time it does end up sounding more correct for want of better words. I found the same with correcting my headphones to the Harman Headphone target. At first it was too different but it did not take long before it clicked and everything started to sound right, going back to uncorrected then sounded obviously worse.

Approaching a speaker design as gnarly described makes it hard to determine phase audibility as the only difference because steep linear phase filters are quite different to shallower IIR filters in more than just phase.
Well yes, re Geddes and the wrong side :eek: but your own comments speak to the intuitive notion that if one can get a bubble of perfect coherence at 1 meter from the speakers axis, your odds of getting something more coherent @ LP than not improve vs no correction at all. I just don't see that it could be any other way. It may that the odds are somewhere between nil and nothing either way. So I don't know. He was a cantankerous old sort holding court over at DIY whose input I always valued, and as I get older myself, feel that at some point, it is ones god given right to get a bit testy and cantankerous with students who haven't done their homework. But as many in the audio field he was very dogmatic on particular matters--his lasting legacy will undoubtedly be the acceptance of waveguides as almost de riguer in any system hoping to lay claim as ultra high fidelity.

But that is an interesting observation of yours that the differences are indeed quite audible but not necessarily better in all cases. Personally, I have been interested in Professor Choueiri's work in BAACH. This guy is the real deal having something like a 100 patents in the aerospace field, a full professor at Princeton, and who has taken a keen interest in audio--in particular how to increase the quality of illusion of a holographic soundfield. This link is to their website, but if interested one really owes themselves to watching the YouTube videos he has done--I believe Gene at Audioholics had him on the show once. Needless to say a brilliant guy whose research has led to what I would characterize as Ambisonics on steroids and what's truly remarkable doesn't require the recording to be made or encoding in any way. Essentially any good recording using proper mic technique will render at least some of the magic. I think once I get everything up and running, I'm gonna give it a shot. My speakers are not the recommended narrow DI, but with mindful placement and reflection abatement can still achieve good results. I suspect whatever it is doing to negate cross cancellation (and it's some high powered DSP) will take precedence over whatever other DSP is doing, but am waiting to get more info regarding interaction. Stay tuned, if successful, I will definitely do a thread about it.
Cheers.
 
Last edited:

Holmz

Major Contributor
Joined
Oct 3, 2021
Messages
2,018
Likes
1,241
Location
Australia
One thing I don't agree with is the superiority of using linear phase filters. Maybe I'm missing something, but based on what I've been able to research (and Amir even has a video about it) phase distortion in loudspeakers is at best an extremely subtle barely audible effect, if indeed it's audible at all. As he points out the in-room playing of a loudspeaker introduces more out of phase room reflections from every angle arriving at various times than could ever be caused by the phase shifts introduced by a speaker's crossovers or for that matter its electronics. He cites Toole who in turn cites studies by psycho acoustic researchers reporting the results of their experiments using live test subjects to corroborate that conclusion. So unless there is something very distinct and especially audible and uniquely deleterious about the phase shifts introduced by the speaker as opposed to those introduced by the room, it's very hard to see how they would be terribly significant.

Maybe it is somewhat dependent upon the speaker’s radiation pattern width?

I know people harp on about the room reflections, but technically the direct sound’s pressure front would be closer to what the microphone picked up, when the phase is not inverted or otherwise different after coming out of the speaker.
 

fluid

Addicted to Fun and Learning
Joined
Apr 19, 2021
Messages
691
Likes
1,196
your own comments speak to the intuitive notion that if one can get a bubble of perfect coherence at 1 meter from the speakers axis, your odds of getting something more coherent @ LP than not improve vs no correction at all. I just don't see that it could be any other way. It may that the odds are somewhere between nil and nothing either way.
Sometimes these sort of discussions with the truly smart people end up hanging on single word differences. Intuition and acoustics don't always get along either. Given the variations in the modal region the likelihood of the response at 1m being the same as at 3m is quite small, phase response changes are the most position dependent too. Correcting anechoically or at the listening position seem to have the greatest chance of success. I think it is always worth to experiment though as sometimes the best subjective results don't always fit within the accepted "rules".
He was a cantankerous old sort holding court over at DIY whose input I always valued, and as I get older myself, feel that at some point, it is ones god given right to get a bit testy and cantankerous with students who haven't done their homework.
Earl does not have the smooth communication of Floyd Toole or Tom Danley, he is much more forthright and matter of fact. I don't find him cantankerous but certainly if you haven't done your homework you will get short shrift.
 

Dathzo

Active Member
Joined
Jan 20, 2021
Messages
107
Likes
68
I don't want too easy but I sure as hell don't want too hard. I just haven't heard from many people who have owned both Dirac and say Acourate, and can offer some first hand experience. Instead it's more like Dirac is amazing, but you ask compared to what? So thanks for the responses guys.

@Worth Davis mentioned the following: Been playing with Audiolense and Dirac in room. I believe audiolense sounds better, but they both improve things a good bit from untreated. Dirac is on my minidsp shd, Audiolense I use the Roon convolver.
.
.
.

1. Both Dirac and Audiolense (or Acourate or Focus Fidelity) improve the sound compared to no DRC
2. There is a tendency to mention that both Audiolense and Acourate are better than Dirac.
3. The factual difference between Dirac and Audiolense (or Acourate, or FF) is that Dirac cannot correct excess phase at low frequencies (only possible with FIR). Still, does that result in an audible improvement?
4. Target curve (tonal control) can be tweaked with all the systems to user preference
5. Key point: what about sound quality differences between Dirac and Audiolense (or Acourate/FF)?

I am yet to find a proper comparison (maybe ABX?) between filters generated by Dirac and one of the other FIR software technologies to answer point 5.
 
OP
J

JRS

Major Contributor
Joined
Sep 22, 2021
Messages
1,158
Likes
1,003
Location
Albuquerque, NM USA
Thanks for the concise summary. I agree about the excess phase correction. On the one hand there have been a bunch of studies indicating phase is not a significant issue, but on the other hand, if there is any place that it does matter, it is likely the bass where it counts. I do like the notion in theory--that is excess phase correlates with resonances in the system, and cleaning up both the amplitude and phase is helpful in maintai ing impact and articulation while the amplitude correction helps with the FR. I just don't know. Likewise with your comments about many advocates of the others who claim dirac is inferior and not by a small margin. And I've yet to hear from someone who has optimized both on the same system. That would indeed be an interesting ABX test.
 

f1shb0n3

Senior Member
Forum Donor
Joined
Aug 29, 2020
Messages
322
Likes
498
Location
Seattle Area
Likewise with your comments about many advocates of the others who claim dirac is inferior and not by a small margin. And I've yet to hear from someone who has optimized both on the same system. That would indeed be an interesting ABX test.
I'm also interested to see a comprehensive comparison between room correction systems - from my experience they provide the biggest audible and tangible improvement in any of my systems. I'll be doing my own comparison of Audyssey XT32 vs Dirac Live with stereo and 5.1 surround once my Onkyo TX-RZ50 arrives. I like what Audyssey does to my HT and I also like Dirac Live on my near-field stereo, so the comparison will be interesting to hear.
 

fluid

Addicted to Fun and Learning
Joined
Apr 19, 2021
Messages
691
Likes
1,196
Dirac is quite different to the others mentioned in a number of ways. Audiolense and Acourate both have the ability to integrate crossovers of different types into the overall correction. This allows directivity to be influenced by how the drivers blend together and either a complete active speaker design to be realised or an existing one converted.

In pure room correction terms Dirac does not allow as much freedom for the user to adjust the parameters, the core algorithm is set.

Correcting a speaker from in room measurements is not as clear cut as it is based on anechoic measurements. Each speaker has different directivity and is in a different location with different reflection patterns etc. There is no universal way to correct this to a predefined target and be guaranteed a good result. This is the prime reason why room correction gets a bad name. It is easy to make a graph look good and have the sound be bad.

This is where having more control over the parameters of the algorithm can help an experienced user to tweak the processing to get a subjectively preferred result. The difference between good and great can often be very small.

Changes in phase correction at low frequencies are absolutely audible, I don't think this invalidates any research but that what is being done in these programs was not tested by that research.

One major benefit of correction is making both speakers more similar in level and tonal balance, this can dismissed as simple but it is one of major factors in making an improvement over stock and all of them do this bit.
 

Holmz

Major Contributor
Joined
Oct 3, 2021
Messages
2,018
Likes
1,241
Location
Australia

In pure room correction terms Dirac does not allow as much freedom for the user to adjust the parameters, the core algorithm is set.

And in fact some of the systems do not allow the retrieval, nor loading, of the FIR taps.
 

dominikz

Addicted to Fun and Learning
Forum Donor
Joined
Oct 10, 2020
Messages
803
Likes
2,626
I am yet to find a proper comparison (maybe ABX?) between filters generated by Dirac and one of the other FIR software technologies to answer point 5.
Perhaps this post in my comparison thread on DRCs might be interesting; foobar2000 ABX comparator results are in the second half of the post.
Links to posts with various other DRC tests I did can be found at the end of the first post of the thread.
 
Last edited:
OP
J

JRS

Major Contributor
Joined
Sep 22, 2021
Messages
1,158
Likes
1,003
Location
Albuquerque, NM USA
Dirac is quite different to the others mentioned in a number of ways. Audiolense and Acourate both have the ability to integrate crossovers of different types into the overall correction. This allows directivity to be influenced by how the drivers blend together and either a complete active speaker design to be realised or an existing one converted.

In pure room correction terms Dirac does not allow as much freedom for the user to adjust the parameters, the core algorithm is set.

Correcting a speaker from in room measurements is not as clear cut as it is based on anechoic measurements. Each speaker has different directivity and is in a different location with different reflection patterns etc. There is no universal way to correct this to a predefined target and be guaranteed a good result. This is the prime reason why room correction gets a bad name. It is easy to make a graph look good and have the sound be bad.

This is where having more control over the parameters of the algorithm can help an experienced user to tweak the processing to get a subjectively preferred result. The difference between good and great can often be very small.

Changes in phase correction at low frequencies are absolutely audible, I don't think this invalidates any research but that what is being done in these programs was not tested by that research.

One major benefit of correction is making both speakers more similar in level and tonal balance, this can dismissed as simple but it is one of major factors in making an improvement over stock and all of them do this bit.
Thanks for the input. I'm working with three way actives +/- SWs depending. I may be mistaken but I believe the fully featured version of Dirac does multichannel with specified XO's, but there us a chance I'm thinking when used in conjunction with one of the media center apps like JRiver. So I'm wondering what improvements you are talking about when cutting excess phase. You mentioned better speaker matching, anything else?

One of the things i dont like about Dirac is it must be solving a large matrix to dig oot and repair the IR. I may gave mentioned above that my preference at least intellectially is to make quasi-anechoic measurements, fix the driver response insofar as possible, then do msmts at LP (averaged over thelistening area), and then do the DRC.
 

fluid

Addicted to Fun and Learning
Joined
Apr 19, 2021
Messages
691
Likes
1,196
Thanks for the input. I'm working with three way actives +/- SWs depending. I may be mistaken but I believe the fully featured version of Dirac does multichannel with specified XO's, but there us a chance I'm thinking when used in conjunction with one of the media center apps like JRiver. So I'm wondering what improvements you are talking about when cutting excess phase. You mentioned better speaker matching, anything else?

One of the things i dont like about Dirac is it must be solving a large matrix to dig oot and repair the IR. I may gave mentioned above that my preference at least intellectially is to make quasi-anechoic measurements, fix the driver response insofar as possible, then do msmts at LP (averaged over thelistening area), and then do the DRC.
My understanding from reading the website is that Dirac can either be stereo or multichannel but there is no mention of crossovers in either, there is the option to add Live Bass Control for one or multiple subwoofers.

It is neater to have the crossover and overall correction for each driver created as a multichannel convolution, but it can be done in stages as you suggest. For your intended use (DEQX replacement) I don't believe that Dirac offers what you need without adding something else.

Getting the speaker design as good as it can be through quasi anechoic measurements and listening is the first step. I'm not sure what you mean by solving a large matrix. All the room correction software works in a very similar way overall. The impulse response is windowed based on some settings, the top and bottom are truncated to avoid numerical instability, there is dip limiting, maximum gain settings and target set then the response is inverted to get the filter. There can be other functions built into that process and it may be minimum, mixed or linear phase.

Speaker matching doesn't have much to do with excess phase, but more to do with the stereo format in that it only works as well as it can if the two sources are as close to each other in frequency response and level.

Excess phase correction can work to undo crossover group delay but I don't find that to be very audible, I find it to be most audible when used from 1000Hz down. It changes the perception of the bass and for me tends to make the presentation tighter and cleaner with less boom. But it can also remove some vibrancy and energy in the same way that very accurate speakers can sound a bit boring. You might like it or not but I am sure that you will hear a difference.

Before investing money in any of the paid options, I would work on the speaker design using whatever EQ and crossover solution you prefer in the computer and then try free DRC_FIR over the top. It is not point and shoot but every setting can be changed and allows a full exploration of what the options do if that is something you like.
 
Top Bottom