After reading your blog, I concur with @Bill_Janssen, in that you don’t seem to understand what you’re talking about. Indeed, your reluctance to read the scientific papers referred to in the links you posted is an example of confirmation bias, i.e., not considering other information.
This is a better example of confirmation bias than comparing two trivial numbers.
I also agree that when listening with high resolution headphones, distortion and noise become more apparent. I find headphone listening to be unrealistic so only use them when laid up in bed or for monitoring.
You are asking if your unique ideas of acoustics is covered not only broadly, but in a specific thread. Naturally that doesn’t exist. Your views are yours and you can write a blog and talk about that.
The topic of acoustics is extensively covered in ASR. We have a dedicated forum for it as a matter of fact:
There, you find sticky threads by me on articles I have written on the topic that have been published elsewhere. Here is an example:
While you have your ideas of why DSP is not useful, it is the most powerful tool in our arsenal to deal with impact of our room on the sound we hear. I have done reviews of automatic versions of those:
My speaker reviews both in text and video routinely include explanation of directivity, impact, etc. Here is just a quick example of KEF LS50 Meta where that is mentioned fair bit:
Net, net, you are not aware of 1% of what we discuss on ASR. And, are looking for validation of your unique point of view which naturally doesn’t exist. You are welcome to join ASR and advocate for them and see the reaction and sea of literature/research provided in return.
And no, I am not going to talk about acoustics when I am reviewing amplifiers, headphone amps, DACs. Generic, product independent talk goes in discussion threads.
These are not my unique ideas but it is common knowledge among pro audio people who are dealing with loudspeakers and rooms, may it be sound reinforcement or in a studio environment. Several pretty successful loudspeaker manufacturers took that as the main idea of their product line, and if I recall it correctly, there are even measurements of one of these on your site.
I do not expect a particular thread to cover exactly my thoughts on that topic. As far as a judgement on speaker directivity and other related parameters seems to be a part of every loudspeaker review on ASR, I expect that to be a part of every single review thread. I find it highly misleading to read about ´superb´ or ´bad´ directivity when actually both would not help anyone placing such a speaker in a living room.
See, the problem is again your verdicts like ´good directivity´ or ´perfect´ about the isobaric plot of a speaker when actually for most of people’s rooms the opposite is the case. It seems you are not understanding how speakers of different directivity sound in a given environment but rather base your verdicts on how smooth the graphs and how straight the imaginative line in the graph looks like.
That is not science nor about sound quality but rather hi-fi astrology masquerading as geometry. The straighter the line the better the verdict. Unfortunately this has nothing to do with the real world of reproduction quality. And - no - these verdicts have not base in the research Floyd Toole has done.
There is a lot of text indeed. But I have not found anything that would be helping a reader to understand what he or she should do when it comes to choice of speakers and speaker placement in his or her room. The sticky thread you have linked seems to accumulate some very old research on the topic of reflexions in a room but again nothing really helpful. Maybe I have missed a thread and you might want to give me a link.
No and No.
I did not say DSP is not useful, I strongly recommend to use one if two vitally necessary conditions are met: at first you have to have the sound quality optimized to a level that non-EQable flaws are eliminated already by choice of speakers, placement and room treatment. Secondly during setting and optimizing filters you have to know what you are doing and hearing, understanding the impact on sound of every single DSP filter you are setting. By ear and not by microphone.
And no, DSP is not the most powerful tool to achieve excellent sound quality in hi-fi environments. Most of acoustical flaws occurring in a typical living room cannot be corrected with DSP. If might be possible to reduce negative impact of the room so it does not sound annoying anymore. That is what many people sell as the miracles of automated room correction, they simply reduce the level of problematic frequency bands (such as the 3-5K region or peaks caused by room modes). But that is not a powerful way to achieve really good sound quality and it comes at a cost usually.
I have used all the three methods of automated room correction which you are suggesting in your links, at numerous occasions and in different rooms. They are all textbook examples of how room correction does not achieve excellent sound quality but rather is trading sonic flaws for different ones. You forgot to mention DIRAC in that category!
Funny sidenote: The in-room measurement of your Odyssey and Lyngdorf experiment threads are clearly showing that something is fundamentally wrong with the modes, the speaker placement and the way speakers are interacting with the room (which will inevitably lead to non-EQable flaws and bad sound quality). Yes, I mean the dips around 50-65Hz and the peaks around 110Hz with the Revels as well as 100 and 180Hz regions with the Genelecs.
Any room correction method which I would consider to be useful MUST stop at this point and give advice to alter the speaker placement or treat the room in the affected frequency bands as the first step. In contrary all three are proceeding, most probably making things worse from the perspective of sound quality (with Audyssy and Genelec showing the most ridiculous algorithms, behind Lyngdorf´s method there is at least a strategy).
But again, the question is if you want the best sound quality in a given environment, or if you prefer hi-fi astrology and hunt for nice-looking curves instead?
Thanks for the LS50 meta video link as it is exemplarily showing the contradictions when it comes to judging a speaker’s directivity. In the video you mention direct and indirect sound should ideally have the same or similar tonality - that is exactly my point, and I agree to that statement.
But a minute later you explain that tweeters are usually narrowing down the radiation pattern resulting in an increasing DI towards higher frequencies and just comment ´that is fine, you want to see that´, ´very smooth line´ and ´very good directivity´. NO, IT IS NOT! It is the utmost contradiction to the previous statement on direct and indirect sound both supposed to show similar tonality.
Do you understand how an increasing DI (or decreasing RT60) towards higher frequencies sounds in a living room environment compared to a constant directivity or decreasing DI? Or are you again hunting for nice graphs and straight lines instead of optimizing sound quality?
Pretty fair, and I did not expect that. But I expect you to apply your own standards of scientific evidence on what is audible and what not when you formulate verdicts solely based on amplifier and DAC measurements. It seems that you this time are not hunting for straight lines but for specs well below audibility (like -110dB of noisefloor) and again clean-looking graphs like in jitter plots. According to your own standards this is clearly inaudible but you dare to either praise or condemn products which should sound the same if your measurements tell the truth.
This is what I call hi-fi astrology and highly misleading. Could laugh about it if there weren’t people out there believing that stuff and buying speakers which surely will create bad sound in their living rooms, listening to badly EQed systems in an unsuitable room if only the response looks flat and disparage electronic components of other listeners which clearly sound excellent.
I absolutely support the idea of basing hi-fi verdicts on science, solid experience and measurements while fighting esoteric beliefs. But it looks like ASR has moved to rather the opposite, i.e. creating its own esoteric beliefs, just the name of the gods is different.
This was always one of my concerns about the purported science at achieving the best sound. When subjective preference is so fundamental and we’re all different, how would you know that someone’s eqed preference towards what they perceive as a flat or “original” sound or “best” produces better results for you for a given headphone or speaker conditions? You could do the double blind test, which should help. It’s surprising you don’t see such data on reviews in ASR and instead lots of discussing about likely inaudible measurement data points considering the overall quality level in general of gear today.
We are not remotely “all different.” We are far more alike than not in our preferences for neutral sound. Tons of research shows this, a lot of which I have posted about here and elsewhere. This is the foundation of four decades of research into what makes good sound in a room or in a headphone. Just like the fact that majority of people like ice cream and chicken, majority of people like sound that is not colored.
There are some people who are different but your starting point should be that you are the same unless proven differently. If so, then you can use DSP to adjust to taste. This is all written and talked about in ASR in numerous threads, articles and discussions.
That couldn’t be more wrong. We not only discuss such topics, but we do so with the very people who have performed such research, such as Dr. Sean Olive, and Dr. Floyd Toole. Here is a short article I have written on this very topic:
I appreciate the response, but you are continuing to misunderstand me and my preferences, and perhaps the human subjective experience in general. This may be part of the confusion that you continue to have in dialogue with others. There truly does seem to be a world-view problem here that is being outlined for you if you could listen.
I have many different headphones that produce sound differently across the FR range. There is more bass or treble, or some do sound more “neutral” vs. more bass heavy vs. more clear on the high end. They sound different with different genres and depending on the quality of the sound recording. Some have switches or filters, so you can modify the FR. I even have a custom one that someone tuned in a specific way that would not be considered neutral at all, but I love listening to it especially for certain genres or types of music. DSP is not a solution that is necessarily helpful for these experiences I have with music.
But isn’t DSP a way to achieve custom switches, filters and tunings? Isn’t that what Amir is saying? I’m not sure I get your point.
Nope. Your statement was clear. Research shows just as clearly that your assumptions are wrong that we are all different. That in a controlled tests, listeners mostly prefer similar headphones and speakers with neutral frequency response. There is room for some variation in bass and treble for which, DSP can easily provide, obviating the need to spend money on multiple headphones/speakers to get there.
Sounds like you haven’t used one to know. I have EQed more that 400 speakers and headphones. The power it provides to both correct response errors and shape it to what you want is beyond measure, pun intended. And the way it is integrated into Roon is superb as you can create as many profiles as you like and switch on demand. I have a massive list for all the speakers and headphones I have tested.
Please take the time to read and understand the research and spend quality time with equalization. You will not get a better education in audio fidelity than this. It is all there for free if you are a Roon user.
@Amir_Majidimehr Take a look at the last post in this thread (Are Our Preferences Different in Audio? | Page 4 | Audio Science Review (ASR) Forum) that you shared. There are so many factors including eartips and pads that can matter. DSP tweaking in Roon has not resulted in improvements in my listening experience.
To be honest, the fact we all have a different perception, is not as relevant for aspects like tonal balance, at least when it comes to judging a loudspeaker setup. Reason being we all have a similar calibration mechanism for natural sounds and those reproduced by loudspeakers. What is perceived as natural in an acoustic concert by two completely different listeners is basically the same sound field to which the individual senses are calibrated. So both listeners would perceive the same reproduction system as natural being similarly balanced in direct sound, early reflexions and diffuse sound field compared to the live event (although two-channel stereo and live sound field vary in the angles from which different fractions of the sound field are hitting the listener but that is not really relevant in terms of perceived tonal balance).
Please note this is not the case with headphones and earphones as these are bypassing a significant part of the individual hearing process such as the HRTF (Head related transfer function) or even parts of the auricle and ear canal transfer function in case earbuds are used. What is perceived as well-balanced with headphones or earphones is a matter of individual perception.
Would should also be emphasized is the fact a sound field reproduced by loudspeakers will only be recognized as tonally balanced if direct sound, early reflexions and diffuse components are all equally balanced, roughly coming from the expected angles and keeping the level ratio between each other.
That is the reason why referring to in-room frequency responses is utterly pointless as they usually show only a sum of all of these or a fraction thereof plus they discard all information about angles. A huge fraction of flaws created by automated or frequency-response-based room correction methods originates from the same phenomena as these cannot affect the aforementioned ratios and cannot keep the different parts of the sound field into account. And, no, the early reflections are not being perceived as a part of direct sound or in a similar way. That is one of the fundamental errors of early psychoacoustical research.
A double-blind listening test is possible but it is quite an effort as you have to compare two pairs of speakers of different directivity being calibrated to either identical on-axis response or in-room response. I have done such experiments in the past and it is really eye-opening.
Could you please give an example of a controlled test in which the majority of listeners have preferred similar headphones in terms of frequency response? What was the definition of ´similar´ in dB?
This ist not very credible, as headphones are bypassing the HRTF which could be called some angle-dependened, interchannel EQ created by our head and auricles. Similar headphones do not remotely result in a similar frequency response at the eardrum if shape of the head and auricles differ.
And they differ from person to person. Otherwise all HRTF-related technologies would work flawlessly with all listeners equally, such as binaural recordings, virtual Dolby Atmos and alike. And no, they do not.
Every time I discuss sounds of headphones, particular in-ear monitors, with other experts, I am surprised about the variety of what people personally find tonally well-balanced.
With loudspeakers, there is a certain truth to people having similar preferences, but it is pointless to narrow this down solely to a linear frequency response in an anechoic chamber while other factors defining sound quality of a speaker are ignored. In controlled environment, consumers tend to prefer a little more (lower) bass than linear response would suggest. And without taking direct and indirect sound (and other factors defining perception of tonal balance) into consideration separately, there is not much point in the whole experiment as you can only test which particular flaws people find less annoying.
And even if many people prefer a more or less balanced loudspeaker reproduction, there is still significant room for personal taste or at least judging with your ears as most of aspects of sound quality are not covered.
You meant a different poster but I had to grin. I can tell you I understand all research I have read so far on the topic but I find it difficult to spend quality time with equalization. I use EQ on a professional level. Do you also tell your gardener to spend quality time with his shovel? I thought he knows how to handle it…
You claim to have used EQs extensively, but it seems not much of understanding has evolved regarding what they do and what they cannot. Ask a recording engineer or mastering engineer or PA technician! They will tell you the same I tell you: EQs are tools to encounter certain problems or manipulate certain aspects of a mix or a reproduction system. They are good for that and used on a daily base. But they help little to nothing when it comes to achieving superb sound quality, solve room-induced problems or altering the character of a mix. Most aspects of perceived music reproduction have little to nothing to do with what can be manipulated using an EQ. Same is true to sound quality of loudspeakers in a room.
Funny sidenote: the only thing EQs are really almighty in is manipulating the frequency response on loudspeaker axis. So as you advocate the extensive use of EQ, what point is there in judging loudspeakers by their frequency response? Even if the latter looks like flawed - no problem, we EQ it to 0.1dB per 1/3 of an octave band! So why judge speakers by frequency response?
I would not recommend to use roon´s DSP for the purpose you are claiming to use is. Professional solutions work much better. Solely for the reason roon creates a little gap when altering parametric EQ parameters and there is no possibility for seamless and stepless adjustments. So it is very difficult to make a direct comparison or judge the temporary bypass of one filter. But that is exactly what you need when making room correction adjustments or manipulating headphone response to your taste.
Of course not if you are not using your ears to compare but instead of that stare on frequency response graphs and make videos about why a little dip or peak should be corrected or not.
@Arindal I get some of the points you are trying to get across, but as stated before your methods/solutions are not practical for the average home listener.
Have a look at the Roon showcase thread to see how far from reality your ideas are. See what your average Roon user is doing.
Performant speakers and Roon DSP is going to be the limit for most.
To make it clear, I absolutely understand there is a necessity to have a compromise in a living room environment. May the budget limit or the room´s interior or the size of gear or the experience of the owner be the limit. Everything understood.
I have started the discussion from the point how ideally things should work not to tell people they should hire studio installation professionals, spend 100 grand and aim for the ideal. It is rather to understand which factors of loudspeaker choice, placement thereof, basic room treatment and DSP are a promising way to get the maximum sound quality in a given environment and which ones are most probably a way to dissatisfaction.
What makes me sad is people like Amir telling music listeners in a very self-convinced manner from a pseudo-scientific point of view to focus on completely inaudible criteria, non-relevant parameters and choose methods which for sure do not work (like equalizing when having major, non-EQable room-induced flaws) in order to achieve a good compromise. Moreover they seem to misinterpret important parameters like speaker´s directivity for the sake of having smooth, neat-looking diagrams but leading to a non-compatible combination of room, placement and speaker. It is a shortcut to dissatisfaction and I have experienced that numerous times.
To highlight that: I am absolutely not saying anything against Roon DSP from a quality point of view. The DSP functions are well-coded, the filters sound superb, they do what they are supposed to do and simulation graph is awesome. I understand why they produce a gap after altering parameters and that is not roon´s fault as it is not a professional manipulation software.
The only reason why they are not helpful for a proper optimization process is the gaps after altering parameters. It makes it difficult even for trained listeners to compare the effect of a particular filter and the reproduction without. Recording and master engineers as well as sound reinforcement experts like do ´dial in´ narrow parametric EQ filters which is only possible if you have a seamless adjustment method especially of frequency and level plus a gapless bypass option.
While for an experienced expert it is already difficult to dial in such filters with a gap and without bypass option, for a laymen it is impossible without A/B comparison. I do not believe many people playing around with parametric EQs in roon, Amir included, are really capable of fully understanding what they are doing and achieving an optimization in sound quality which would stand the test of long-term listening. The fact that many rely on measurements trying to fight some most probably inaudible peaks speaks louder than a listening test.
If anyone is really interested in proper EQing their own system I recommend to rent a professional parametric standalone EQ which you can get from a PA rental company or alike for little money per day. Once you have dialed in your filters you can transfer the parameters to roon and the result will be as desired.
The question is: What are performant speakers for home use? How to choose the right one for a given environment?
I can tell you from vast experience with different rooms that all these things being taught on ASR about which criteria are important and which speakers are good or bad are in many cases leading to disappointment.
The sad thing is that measurements published there seem to be mostly correct and could be a helpful base for an informed decision. Especially the spinorama graphs, calculated indirect sound response and DI. One of the major problems is not telling people about qualities which are NOT in the measurements but require a listening test. The other big elephant in the room is the way verdicts on directivity are evolved as it seems to be merely judging how neat the graph looks like and how straight the line is you can draw through it.
That has nothing to do with choosing the right speaker for a given environment, and will bring a lot of people to dissatisfying choice of gear. And, yes, I am aware there has to be a compromise and there are affordable products out there which deliver a fair compromise.
They have no useful knowledge to share. I have interacted with many and while they are good at what they do, their understanding of proper sound reproduction in the room, psychoacoustics of the same, preference factors with respect to speakers and rooms is no better than average audiophile. Ask them to quote a single piece of research in their choice of room treatment and you will not get any. Just like your posts which have been devoid of any references.
Ask me the same question and I will quote you extensively on every aspect of the advice I give. So please, don’t keep appealing to this (vague) authority. They are following old school ideas about acoustics which has long been shown to be wrong.
This is very different than research into consumer sound reproduction where we have decades of research to draw upon as far as room treatment, directivity, factors impacting preference, etc.
Nope. First, we are not talking about creating music so let’s not bring that in. Second, room modes are minimum phase phenomenon which means parametric EQ can most definitely correct them both in time and frequency domains. Since room modes are predicted 100% by physics, you must deploy DSP for proper response. There is no way you can remotely get there with any acoustical treatment as wavelengths are massive. Even anechoic chambers tend to have room modes below 80 to 100 Hz!
What? 1/3 octave? That has no place in the discussion we are having. Room modes are very narrow and require analysis at 1/12 octave or even finer. You can use more filtering (up to 1/6th octave) if you just want to look at overall tonality, setting target curves, etc.
CEA-2034 standard requires 1/20th octave analysis by the way and that I what I use for my speaker measurements.
It is clear you have never used Roon EQ, or any parametric EQ system in general. If you had, you would know that each filter can independently be turned on and off. Indeed, I develop my speaker/headphone filters one at a time using this technique. I make sure each one provides an improvement by itself, and then collectively all together. The technique absolutely works and is used by countless audiophiles already.
Research shows efficacy of such equalization to improve listener preference. Here is one example (AES paper):
The Subjective and Objective Evaluation of Room Correction Products
Sean E. Olive, John Jackson, Allan Devantier, David Hunt and Sean M. Hess
A listening experiment was conducted where eight
trained listeners evaluated the sound quality of five
different room corrections applied to a high quality
loudspeaker/subwoofer in an acoustically typical
domestic listening room. The same loudspeakersubwoofer
without room correction was included in the
test as a hidden anchor. Listeners gave comparative
ratings of the room corrections based on spectral
balance and overall preference. Acoustical
measurements of the room corrections applied to the
loudspeaker/subwoofer were made at the six listening
seats, and the primary listening seat, to study
correlations between the objective and subjective
measurements.
Room correction, when done properly, can
provide significant improvements in the sound
quality of loudspeakers in rooms. Three of the
five room corrections produced significantly
higher preference ratings than the uncorrected
loudspeaker/subwoofer.
Here is the preference score graphically: Target Room Response and Cinema X-curve | Audio Science Review (ASR) Forum
Automated systems like this are quite useful if someone doesn’t want to learn how to do things manually, or are dealing with multichannel systems where the work can get very tedious if not impossible to do by hand.
Here is another article i published a few years back on power of DSP when combined with multiple subwoofers:
Bottom line, there is no good sounding system without DSP. Period.
I leave that sentence uncommented as it insults all practically educated experts in that field and speaks for itself.
That is kind of funny and showing that you have not really talked to anyone with proficiency in that field. If you have ever been witnessing and understanding how huge PA systems are planned, particularly when it comes to directivity and room simulation, you would stop talking about old school ideas. These people have advanced technology at hand and they know how to apply it.
Why should I ask people who are setting up studio listening rooms, studio-grade cinema sound systems and planning sound reinforcement systems for scientific research? This is not what they are doing, their job is to create a certain level of sound quality in a given environment. And I can with my own ears and similar yearslong experience hear the results and they match with what many of these experts are explaining about their concepts and theoretical base and why they do things the way they do.
With what you are claiming, it is rather the opposite. I read your verdicts on speaker directivity like ´perfect´ knowing this or that particular speaker and the flaws resulting from incompatible DI being massive in most of home environments. I read your statements about decreasing target curves and narrowing-down DI towards higher frequencies being ´what you want´ knowing exactly how that will sound like and how dissatisfying it is to most of listeners. And finally, the close-to-religious belief in the almighty DSP and room correction systems - once you have heard how these systems are trading acoustic flaws for others, you are done with such claims.
I asked you already which research was exactly confirming your claims people in mass tests prefer very similar sounding headphones and why a significantly decreasing in-room frequency response towards higher frequencies is proven to be preferred by many listeners as better/more natural compared to a linear one with both balanced direct and indirect sound field (which would be the natural conclusion of your ideas that everything should be linear and neutral).
Maybe you do not see these contradictions in your claims. Other people see them.
You obviously failed to understand what a minimum phase system with a resonator is when it comes to room modes. That is particularly funny as the idea of a resonator and the way it is affecting movement or sound waves in the time domain can be understood by an infant after dropping a ball and letting it bounce or sitting on a swing.
Room modes are caused by resonation and they usually resonate for an extended period of time. If this time is long enough, the sustain is audible in the form of booming and slower bass decay regardless the actual frequency response. The parametric EQ for obvious reasons can only affect or correct the latter but not make the sustain disappear. On top of that, the way room modes are stimulated is neither linear in terms of stimulation time nor sound pressure level. So, on top of resonation and sustain, they are nonlinear by dynamic means. Comparing a constant-frequency sine measurement with a DIRAC based one or one at 60dB SPL with another at 100dB should give you enough of proof your ideas about minimum phase modes are fundamentally wrong.
And this you want to correct with a parametric EQ? Really funny.
Obviously, a lot of people planning listening rooms and cinemas are getting to that point. It is just a matter of the right tools. Yes, wavelengths are massive, you might need a significant amount of absorption area but it is absolutely doable.
And no-one said it is necessary to eliminate room modes passively in their entirety. It is enough to get the resulting audible flaws in the time and dynamic domain under the threshold of audibility as well as controlling cancellation effects caused by them. The rest you can indeed correct with PEQ, as mentioned. I am not against DSP but you should be aware of what its limits are.
As mentioned, I work with PEQs for professional purpose and I have tried the implemented DSP options in roon extensively.
Obviously you try to twist my words but that looks increasingly desparate. I did not refer to the Enabled/Disabled option for every single parametric filter but to the fact every change or alteration in parametric filters including this defeat mechanism is causing a little gap. That is making it close to impossible to dial in single filters seamlessly, especially high-Q ones, for frequency and level by direct comparison.
It is obvious that you believe it is working in terms of an improvement, but that is just a typical audiophile’s belief based on overconfidence and misjudgment of the tools you have at hand. The way you do things and the tools you have speak a different language.
I have no doubts equalization can improve listener acceptance in case of mode-dominated, flawed reproduction in a room. In many cases it is enough to reduce the level or cut out the boomy band with a notch and people find it less annoying.
But that usually comes at a cost and it does not make a good sound quality nor is it any proof that room correction solely based on parametric equalizing can achieve same or better results than a method of avoiding such flaws from the ground up.
I strongly disagree with that statement. I would rather say, achieving a good sound quality is impossible if people rely solely on DSP and ignore basic rules of loudspeaker design, room acoustics, room treatment, speaker choice (particularly DI, bass quality and imaging) and placement. The risk of having massive flaws in a room+system combination which are not EQable is pretty close to 100% in an untreated living room environment if there is no particular concept of preventing such in place (such as dedicated near-field listening or full-range high-DI, CD speakers, line sources, dipoles, cardiods or alike).
The most flawless, most naturally balanced system offering best imaging installed in a control room I have ever heard was fully analogue, completely without DSP. Unfortunately the room was pretty expensive so not particularly an example for home use.
I have no problem with DSP and recommend to use one once the basic reproduction quality is met by traditional methods of room treatment, speaker choice and optimizing placement.
This is how we all do it. Although room treatment is our furnishings, rugs, curtains etc.
Spinorama helps us identify performant speakers. Yes there are guides we can follow to poisition speakers, but not always practical. Then use DSP to suit.
Not sure what you are getting so worked up about?
By the way you can use Dirac ART or HAF Room Shaper for long resonances (both are DSP as well…)
Every professional community has lots of word-of-mouth lore that circulates. Some of it is obsolete knowledge, some of it is rank superstition, and some of it is accurate. (The least supportable bits of lore seem to be the ones most passionately put forth by their supporters.) The trick is distinguishing between those categories.
This is brilliant. Thanks so much for sharing, as it’s is the exact description of the problem i feel I’m having and especially with headphones, which you already pointed out are a completely different beast from speakers.
I fully agree, and when it comes to sound quality and technology to achieve it, you need lots of knowledge, listening practice and a solid knowledge base on how to match theory and result. And although many people in that business have their own methods and workflow, a lot of them get really convincing results and seem not to be prone to superstition and misleading theories.
I wished it would be the case. Looking at the way many hi-fi people - including those claiming to follow a ´no-frills´approach, relying on measurements, ´scientific´ knowledge and technology like DSP - ignoring basic laws of electroacoustics and psychoacoustics, setting up their system improperly, combining loudspeaker and room which do not match while not getting results meeting a minimum level of sound quality, I have reason to doubt that.
Not just because I meet a lot of dissatisfied listeners and encounter shocking sonic results in many living rooms. Looking at pictures of home setups, seeing measurements telling a story of extreme acoustic problems (such as Amir’s example with Revel Ultimas and automated room correction routines) and reading how people give each other ridiculous advicse on how to choose or set up equipment - it seems a huge fraction of this scene is stuck in absurdity, the esoteric guys to the same extend as those dancing around straight graphs and DSPs.
On the other hand, if someone does not want to achieve studio-grade sound quality for the sake of not compromising, that’s fine. On the other hand i don’t quite get the contradiction between spending a lot of time and money on hi-fi if you live with massive flaws for the sake of not moving or exchanging your speakers or treating the room.
With all due respect, this in most of cases does not mean room treatment but making things worse, as these objects usually tend to absorb treble and let midrange reflexions pass more or less unfiltered resulting in a typical decreasing RT60 towards higher frequencies.
This would not be such a big problem if one is opting for a nearfield-like listening setup, or choosing speakers with a matching DI such as constant directivity on a higher level or even better slightly decreasing DI towards higher frequencies.
And this is where my dissatisfaction with Amir’s advices is becoming obvious: he seems to regularly recommend speakers showing a continuously increasing DI towards higher frequencies at times even showing steps towards higher DI in critical frequency bands. NO, PLEASE NOT! I can already imagine the result in a room like the one you have described.
In theory, yes. Practically I have not seen a single recommendation based on spinorama data suitable for a room like the one you have described (which is a very common situation in living rooms). To really identify which speaker is performant in a given environment you have to know the RT60 of the room and the approximate listening distance. Following ASR´s recommendations for ´perfect directivity´ with increasing DI or stepping up DI will lead to a disaster in most of cases.
That’s a contradiction according to my experience. What optimizing speaker positioning can achieve is mainly reducing mode-dependent or wall reflexion induced booming, improving bad localization or torn imaging caused by early reflexions, reducing the ratio between indirect sound and direct sound towards the latter. These are the textbook flaws you cannot EQ to satisfaction under any condition. If you are not flexible with speaker positioning, exchange the speakers for proper ones or treat the room. DSP might reduce the level of annoyance of these phenomena but would not mean a cure. It is as easy as that.
In theory this sounds like a way to fight mode-induced booming but from first practical experience with a demo I was attending I tend to be skeptical. Have to admit that I did not really work with this ART system myself so cannot give a final verdict.
I have had countless discussions with people in the Pro industry and one of my friends is a mastering engineer. You can look for my arguments in GS forum. As I said, and you prove yet again, you never, ever see a reference to research paper for anything they claim. The main argument is “I am a Pro therefore I know what I am doing and you don’t.”
As to “huge PA” systems, they have nothing to do with sound reproduction in home listening spaces, or studios for that matter. Large acoustic spaces don’t have the modal region that we have in our much smaller rooms. As such, measurements like RT60 have more meaning there than they do in our rooms. On the other hand, they have major issues like dealing with long distances, covering high SPL focused on listeners, etc. Again, issues that have no cross section with home listening spaces. So please don’t keep mentioning them. What they do – whether right or wrong – is unrelated to problems of home acoustic spaces.
I already linked you to an article I wrote on wrong practices in that space such as X-Curve. And totally messed up and variable frequency responses even in “dolby certified” rooms. Here is the first part of that article:
“The reverberation times of a number of the Dolby-accredited rooms above 1 kHz are commensurate with those recommended in the 1994 Dolby Standard and yet these rooms do not show the assumed frequency-response characteristic during reverberant build up. We therefore conclude that it is likely that the X-curve has not been valid at least since 1994, and quite possibly earlier.”
“The authors conclude that the use of the X-curve is detrimental to the enjoyment of cinema.”
And from the previous follow up article I linked to last night:
Let’s remember that the X curve like all such responses is a target curve. The reality is different as I menionted. The Audio Engineering Society paper, A Survey Study Of In-Situ Stereo And Multi-Channel Monitoring Conditions, shows how far out of compliance production systems can be. The paper has measurements of 250 Genelec speakers used in professional control rooms. Their composite statistics are shown in Figure 3. The poor X curve shown in solid black line is lost in a sea of wide variations no matter which statistical measure you use.
Practices such as using 1/3 octave measurements is at the core of this problem as is using antiquated concepts such as X-Curve which have had no research behind it.
As I said, room modes are minimum phase so correction in frequency domain absolutely corrects time domain. Conversely, frequency domain variations are absolutely there in case of room modes. This is one of those myths that the PROs keep repeating due to lack of fundamental understanding of signal processing. I have discussed this at length so much that I wrote an article on it:
See how effective these filters are in reducing both variations in frequency domain and time domain. There is no way you can make this kind of impact with acoustic products.
Even if you tried, you would wind up with a dead and lifeless room due to how much of the surface of your room you would have to cover.
What you ask is simply read by spending just a bit of time on ASR or reading my posts here. But here we go.
Simply put, we want speakers with flat anechoic frequency response. And smooth directivity. The directivity of most speakers narrows at higher frequencies. This means that if you measure the room without gating, you would see a sloping down response. This, is considered “neutral” by listeners.
This sloping down target response is desired also because we are used to some amount of room gain in bass response. As such a flat response would sound too bright.
The slope of the bass to treble however, is subjective to some extent. This is why you must have DSP so that you can create the target to match your taste and music you listen to. This is mostly needed because audio lacks any production standards. Every professional space is different per above meaning every music we listen to has been mixed and mastered to a different response. So we have no choice but to have variability in our playback system.
As to headphones, there is a decade of research into preference there. I have written about it extensively on ASR and any casual read of the forum would give you an idea of that. Read that and if you have questions, ask.