@ricevs Class-D is not PWM. PWM is not "Class D type amplification". The only thing they share in common is switching devices. Purifi is a Class-D amplification product that uses the input signal and feedback to reference the output to a clean analog input. There is no equivalent in a digital PWM implementation.
I believe the first one made was the Tact Millennium back in the late 90s. The PCM digital signal is changed in software to PWM (class D type amplification). Tocatta Technologies (TACT) was headed by inspired people who now work for Purifi. It seems odd that Purifi does not have a digital amp board.....prehaps soon...they will.
Simply hooking amplifiers up to drivers and using a digital crossover only provides some, but not all the benefits of active speakers.
You give a lot of bad if not misleading advice in your article. As a business, there is a level of negligence in doing that. That the new MOFI speaker. Encouraging people to take it apart, simply replace the cross over, and then lower the tweeter is terrible advice. Very few people have the experience or tools to do this. You can make some final tunings to a speaker by ear, but 99% of the work is with tools. Getting the crossover frequency right both for evenness of frequency and disperson, the slopes, any necessary notch filters, etc. is not something you just "do" by throwing in a digital crossover. At a minimum you need a calibrated measurement microphone and knowledge of how to measure a speaker / speaker driver. Then you need to understand what those measurements mean and how to turn that into a solution. You may get lucky or more likely convince yourself it sounds good by ear, with the first music you listen to, but across a wide range of music there will be issues. I am sure Andrew put a lot of thought into the crossover frequency based on distortion of the tweeter, output at frequency, and dispersion, or more specifically matching dispersion to the woofer and using the woofer as a wave guide. Again, it is negligent to blindly tell people to drop the crossover frequency on a tweeter without knowing the impact, which is likely to be negative.
|
Actually speaker design is simple.....there are not that many factors and you can learn them all in a few hours and apply them with a digital xover and digital amps.
No, not even close, though open baffle, just one type of speaker, which may have good or terrible in room results, is easier, but even for open baffle, there are subtleties to achieve the best design.
There are tons of speakers that cross over a tweeter below 1K......yes, indeed. However, the problem is usually power handling.....and with steeper 48db per octave xover slopes it is no longer a problem.
There are few dynamic tweeters that are crossed over <1K. Horn loaded yes. AMT no. It is not a simple matter of power handling. It is a matter of dispersion, i.e. the tweeter will be too wide at those frequencies, even 27-30mm dynamic, compared to the woofer, and it is a matter of high frequency off axis energy. The larger tweeter may be flat on axis from 1K-20K, but the off axis energy drops more at higher frequencies on the larger tweeter, changing the tonal balance, and you can’t correct that with DSP without breaking on-axis response. For older people it may not matter as much. It seems that you can’t learn everything in a few hours.
Andrew probably spent a couple of hours designing it and tweaking it. Of course, he got in different proto versions of the driver and spent time with each one....but the xover design is simple and normal.....He can probably do the calculations for the parts in his head....he has done it so long. He spent no time listening to those xover parts versus more expensive more transparent one.
He probably did it using his own custom software which he has probably spent decades tuning, however, you can bet from there, he probably made several variants and listened in a few listening spaces to get a feel for how it behaved real world. He didn’t spend time listening to expensive parts for a reason, he already knew what they would sound like.
Simply hooking up a digital xover to digital amps and then to drivers not only provides ALL the benefits of normal active speakers but it gets rid of the distorting DACS, one dollar op amps on the output of the DACS and normal class D or A/B amps with all their parts and circuitry and feedback. A digital amp has NO....I repeat NO ordinary amplification stages, DACs or feedback......Usually, the shorter the signal path the more pure the result.
- It does not provide all the benefits, not even close. Please see your first paragraph, which I will point out, is incorrect as you appear to have spent more than a few hours and still have much to learn.
- A $10 DAC chip with those $1.00 op-amps implemented on a PCB has no distortion that you can hear.
- Feedback done correctly is a good thing. There just has to be enough of it. This is simple math.
- The Peachtree amplifier is a DAC. A regular DAC requires very stable power rails, low noise clocks, and effective analog filters. Stable power rails, low noise clocks, and effective filters with no frequency effects are not difficult at the power level of a DAC (10’s-100’s of mW). Now replicate that at 200W per channel with no feedback to correct issue.
- Those package DACs are based on multi-bit modulators, which provides significantly improved performance. The Peachtree will not be.
- The Peachtree filter is outside the feedback loop. It will have roll-off at high frequency.
Peachtree has not released any specifications, but I would not be surprised if their performance is lower than class leading D such as Purifi and/or Purifi plus an external DAC. Time will tell. I am sure Stereophile or Audioscience Review will publish measurements when they can. You can already buy Purifi modules inexpensively and multi-channel DACs inexpensively. This is a refinement, not a revolution.
Besides being purer sounding than a normal active system.....digital amplification allows you to tune the speaker any way you like, including get rid of bass nodes and extend the bass.
You are only making a claim of purer. Pure is in the results, and we don’t have any to compare. We already are at a state of transparent DAC+amplifier. Anything purer is academic. We can already tune a speaker any way we like, We can also already even with a passive speaker extend bass. Bass nodes has nothing to do with the speaker. That is a room issue. DSP room correction which can be external to the speaker or internal can already be done. Your streaming S/W can do it. There is no advancement being made.
Dipole speakers may give a bigger sound or they may gives you a dark sound. They may help with some room nodes, while creating all kinds of other frequency response issues. They rarely give deep bass without a sub. The sweet spot tends to be small. The off axis energy is usually low, suiting them more for near-field or quasi near-field and can be a pain to room correct for frequency response as it is harder to balance on axis response and room response. There are a reason some love Magnepan and others hate them and that professionals who do mixing and mastering do not use them. Their popularity comes and goes. If your room is not symmetric, forget about open baffle. It will be a mess.
|
@ricevs
Feel free to use a term such as subjectivist if it makes you feel better. What I am is someone who has been involved in speaker design for professional applications for approaching 2 decades now. I don't take offense when someone says you can learn everything about speaker design in 4 hours. I don't take comments like that seriously at all.
I don't believe that many modern DACs, and op-amps or amplifiers are transparent. I know that used properly, by competent engineers, in competent designs, that they are, when connected to speakers and placed in a room and listened to by human beings. Computer measurement equipment may tell a difference. No human will. One of the things as a speaker vendor, mainly powered now, must know, is the dividing line between audibility of the electronics and audibility of the rest of what makes up a speaker. For a long time, the electronics, except at corner conditions, has not been an audible contributor, even for the most expensive models with very very low distortion drivers and optimized emission patterns. I don't think that. I know that from the extensive listening tests that we do. We don't just guess at these things. We do listening test after listening test, properly, to remove bias, in multiple rooms. It is not hard to make electronics color the sound, but that is not our market.
Resistor in crossovers can have a sound if poorly specified. This can come from thermal modulation, though the electronics guys tell me that technically they can have a voltage characteristic and some are high enough in inductance to show differences near the crossover points. As everything is active or DSP crossover now, that is a moot point.
I don't know what you are going on about religion or the soul. To me, you are just trying to wrap that up in some sort of attempt to discredit what I say. I don't know how else to interpret it.
How about we do keep this to a factual discussion? Class-D is not PWM. Dipoles can be great and then can be pretty awful. They are much more sensitive to the listening room and the listener position and are more difficult to fix with room correction. Subwoofer integration is also more challenging. Great results can be achieved, but it is not trivial. Built with dynamic drivers they introduce further issues with dispersion, specifically variability and lobing. Line sources normally don't run into this issue as severely, hence why line source dipoles have more market acceptance. The low distortion these line sources have also contributes.
Getting back to active speakers, simply connecting an amplifier directly to a driver provides good, but not perfect control. You eliminate the crossover, but you still have the natural resistance of the voice coil, and non-linear inductance between you and generating perfect motion.
|
@donavabdear,
For both a speaker and microphone, there are some aspects that cannot be changed with equalization and signal processing. The most predominate is dispersion. This cannot be decoupled from the room though for reproduction, the goal of ATMOS and similar methods is to not only provide a more feature rich reproduction capability, but to dampen the effects of the room. I am a firm believer that even for 2 channel playback, we have not even scratched the surface of what could be done by using more speakers during playback and signal processing.
It is probably people our age dying off or retiring that will bring about some changes in recording. Microphones are picked for their recording pattern and frequency response (tone). There is no reason for the latter any more. It can all be done in post. I remember a presentation on using a microphone array consisting of many microphones with narrower patterns, that were then combined in software. The software could simulate a single microphone of a particular pattern, or provide any number of individual outputs. Is that the future?
In my world, my customers want low distortion, flat on-axis, and smooth predictable off axis, not unlike the Harman research. For recording, mixing, mastering, they can't be second guessing what is in the recording, and what is being enhanced or suppressed due to the playback hardware. For movie/theater sound, they need to know with high confidence how it is going to sound during playback.
For all the bluster in the audiophile community for playback, I do not think we have a good handle on what drives people to prefer X over Y. Hence, while it is useful to discuss sound alteration during playback to suit personal preference, I don't think anyone has a good handle on the controls that would even be offered to the general consumer to provide that tailorable experience. I think this is one place where artificial intelligence will enter our market.
I am sure you are aware there is software to allow studio monitors to simulate other monitors / home systems / cars / etc. We do get customer interest in that area as they assume we could do a better job at it than the present offerings.
|
Audiophiles like to use the word transparency. At moderate listening levels, the speaker will always be least transparent part perhaps with the exception of a turntable. No external amplifier that is not designed to interact with and drive one specific driver, in one specific configuration, can achieve the transparency of an application specific amplification.
|
@lonemountain,
@phusis , and probably most people's concept of amplifier/driver is a simple linear voltage based amplifier perhaps with some frequency response modification. Without writing a book, that is a traditional view that is not the future of active speakers. Even active subwoofer drivers with velocity/position feedback already break that mold.
There is no set "ratio" between the LF and HF peak in any given set of music, though practically, the peaks are significantly more at the lowest frequencies. However, you can soft clip bass while preserving mids and highs unclipped and achieve a speaker that is considered more flexible, i.e. able to play louder while being perceived as still sounding good. However, that is a "club" driven too loud situation, not a professional setup where you ensure the system is not clipping with the music you are using.
|
@lonemountain ,
Was directed at me? I am not sure I understand the question.
An open question: Is this above statement about change something audiophiles agree with?
I expect little I say is something that audiophiles agree with 😁 Fortunately, I just have to make speakers that are transparent, not claim to be. There is, as noted so much lost in the translation from live music that it seems like an obtuse goal, but for what my customers mainly do, real transparency, which effectively is low distortion and controlled dispersion, is a real goal.
|
@phusis ,
How would you listen for transparency? What does it sound like? Have you heard it before? Do you know what relative distortion sounds like in a speaker over any given frequency range? at a particular sound intensity? What would you use to listen for it? Do you know the pedigree of that recording and the signal / processing chain to know the level of transparency from the time the sound was created till it comes off the recording medium?
this whole esoterically-laden secret-sauce amp-driver match baloney
When you understand the intricacies of driving a traditional dynamic driver or AMT in the most linear possible fashion, at all signal levels, with complex waveforms, perhaps your attitude will be warranted.
|
|
|
@donavabdear ,
That you for continuing to carry the torch. @lonemountain same, here and on the other thread. Some people want to learn. Some people want to move forward. Others don't want to learn. Others don't want to move forward. Some who don't want to move forward, don't even want others to move forward, and even some who do move forward don't want other to move forward faster than they are.
I have noticed just how few people from industry, even the audiophile industry post here. There is one amplifier designer / vendor. There are some who sell questionable products, but I don't count them. Given how my posts have been met, I am not surprised. I provided some links above, some heady reading, some pretty easy on where active speakers can and are heading. The only reply, derisive.
Are audiophiles confused? Perhaps. Perhaps they just don't want to move forward and don't want others to either.
|
@invalid, how do you measure purity? No driver is a perfect radiator, so they are not pure. Electronics can be, for any reasonable measure, pure. I see active speakers as the melding of electronics which can be pure, with drivers that are not, such that the combination exceeds the two combined independently.
This is only controversial at the consumer end. At the development end, for those working on active speakers, there is no controversy that an active speaker is better able to achieve measurements of purity. We define purity as reduced distortion, flatter response, better dispersion.
|
@lonemountain
It's tens of feet -maybe even a hundred feet of wire in a giant LF inductor set up for a very low crossover point.
No maybe. If someone makes a design choice to use an air core inductor on a woofer circuit in a 3 way, it does not even need to be that low of frequency for there to be a couple hundred feet. Go for a low frequency and you can be above 500 feet of wire. Then they may compound that by using Litz increasing the resistance without benefit.
|
@invalid,
I have to view purity from a throughput from the input to the amplifier to what comes out of the drivers. I don't think a case can be made, with any set of drivers, where the purity from input to output cannot be superior with an active configuration. That does not mean you are going to prefer an active speaker, or even that you will prefer an active version of a speaker over an equivalent passive as the frequency response may be different which may not be to your taste, or may not work as well in your listening space. If the speaker design does not have good matching of the dispersion between drivers at the crossover points, than crossover differences between an active and passive version of the same design could have a pronounced in room difference. Flat baffles for tweeters and even some mid-range drivers should go the way of the Dodo (IMHO).
|
@donavabdear
Takes time and acclimatization. Once you get used to that resolution, there is no going back. It's harsh at first but then you get used to it. I think the same could be said about film/digital. Disconcerting at first, then you get used to it, then you are very aware of what you are missing when you go back. You even miss the warts. Non amplified live music has none of those pleasant artifacts that are added. Even amplified music, even if the equipment is inferior, is missing the processing that often softens the music. We love it non the less. Something visceral and ultimately natural about it.
|
@donavabdear when you say bring it away from the walls do you mean speakers now somewhat up against the wall but will now be mounted some fixed distance from the wall but suspended? Basics but DSP can compensate for boundary reinforcement but it can't compensate for boundary interference (though some active speakers do ..)
|
@texbychoice
Regarding requisite personal knowledge, not a single post in this thread provides verifiable technical bona fides, only opinions or recitation of personal system usage experience.
I strongly encourage you to point out exactly where anything I said is not supported either by current products in the market or current research into active speaker design. Not one thing I have posted is an opinion. Everything I have posted is factually supported including advanced active speaker methods which I provided many of the more simple links available for people to read up on the topic not that many posts behind your own. You made 2 posts in 3 months, both negative against people with industry knowledge.
Carburettors are in constant need of adjustment, cleaning, and perform poorly across engine loading, temperature, and other environmental conditions and while being comparatively inefficient. While easy to repair, that repair rate is much higher all while experiencing a significantly inferior experience. You would be hard pressed to find a fuel injected outboard motor owner who wants to transition back to carbureted. They exist, but they are a minority.
The potential longevity of stereo equipment is at odds with the fully integrated active speaker implementation for a portion of the market. However, the potential for superior performance at a reduced system prices (speaker, amp, DAC, cables), means that reduced total potential life is balanced by lower cost of ownership and lower lost opportunity costs of tied up capital easily offsetting the lack of highly extended life. There is even the potential for entrepreneurs to offer new business models.
|
Further, @kota1 , for trying to raise the knowledge of the group I have received abusive replies like yours. Now it is like you are actively trying to stunt other people's knowledge. Why is that?
You can read them or not read them. It is your choice to educate yourself in an area you clearly are not educated in, or to not educate yourself.
|
Of course, anyone who claims to be a knowledgeable audiophile and up on active speaker technology should know some of the things I have mentioned, that go beyond simple active crossovers, are already on the market,
http://www.kiiaudio.com/acoustics.php. The Dutch and Dutch 8C also plays some DSP tricks in addition to their acoustic advances. Somewhere out there, you can also find Bruno talking about tight integration of speakers and amplifiers as simple voltage drive is not the best drive solution.
The whole premise of the B&O 90 is advanced active speaker techniques for directivity control: http://https://www.bang-olufsen.com/en/us/speakers/beolab-90
Of course, it is not like Samsung (Harman) is asleep at the switch:
https://patents.google.com/patent/US20170188150A1/en
I could go on, but I have real work to do, you know, with speakers, that people buy.
|
@lonemountain , I would "question" saying a 15" woofer costs almost the same as the 10, especially at current aluminum prices. It is not just the added material costs of the basket, but amortized tooling of what is invariably lower volume, and a much bigger tool as well. That is not even getting into cabinet size, packaging, and shipping. It may seem neither here nor there for active speakers, but a product goal of active speakers is delivering superior performance in a smaller package for those who either don't want a larger unit, or cannot support a large unit in their environment.
Reliability is a different argument for professional and consumer speakers. Professional users expect they are going to replace their speakers every 10-15 years, or sooner, and will have fully depreciated them by that time. Resale does not have a lot of meaning. Audiophiles keep their products a lot longer. To your point, the lack of connections aids reliability, and being able to control designs means being able to alleviate electrical stress. For the consumer market, at an elevated price point, the issue is not failure rate, but the ability to repair a product that may be 20+ years old. We, like other vendors (one hopes), track failure rates and adjust our spare parts and spare assemblies stock to ensure we can support a specific service life. Finance accepts that is a cost of doing business and builds it into cost. Engineering attempts to minimize BOM creep and increase reuse.
We don't have large cost gaps in our product families, but that is intentional, and is a marketing and engineering design collaboration. Know what the "best" model in the family will cost and then understand how to build out the family while maintaining product goals. You don't have to be Fortunate 500 to have a good product plan. Active speakers significantly help in regards to supporting that business model.
@texbychoice , it is not attitude it is simply a matter of fact. Mentioning frequency response is my point. It is inconsequential for a basic active speaker. Any active speaker with a DSP crossover can have an effectively perfect on axis response. Balancing perfect on-axis response with off-axis energy is where it is at. With active speakers, like the example I gave above, Kii 3, you don't have to rely purely on an inflexible acoustic design to do that. How about being able to push a driver 6db higher in output while maintaining the same distortion? How about reducing IM distortion in a small mid-woof 10db at elevated volumes. How about an electrical drive method that reduces the impact of power compression. How about an electrical drive method that can reduce breakup? If you wonder if any of those things improve the sound, they most certainly do.
|
@donavabdear the Dutch and Dutch 8C is just like the microphone with vents and all (and some DSP), the Kii (Bruno) does it with electronics. Sky is the limit.
|
I also had designs of an optical digital laser microphone that could listen from miles away. Hope no one figures out that one.
I hate to break your bubble on this one. Those have been around since the 80's probably earlier. I don't know if they were initially designed as a surveillance tool, but that was one application. They can pick up the vibrations on window glass. Now they are a common industrial tool as well.
|
I won't speak for @donavabdear, but I will speak for myself that the question is essentially irrelevant and is begging an answer. Phase corrected is essential for any working speaker design, time corrected looks much better on a marketing sheet than providing verifiable listener benefits. And yes, I have personally done the testing. Dynamic correctness, dynamic excitement seem to be implying the same thing. How long can you play, and what effects of any concerning dynamic compression. Horn loading / compression drivers is not the only way to achieve this of course. Horns provide, properly designed, constant directivity, but using a standard woofer/mif-woofer and a wave guide tweeter provides similar benefits without the side effects of vertical directivity lobing which can cause unpleasant reflections off vertical surfaces, likely one of the reasons why some people "don't like horns". I think we can agree that a real horn loaded speaker at 20Hz, even a tapped horn is rather enormous and outside the realistic realm for most people. To achieve true directivity at the frequency is just unrealistic and you are not going to avoid room modes. Velocity/position feedback eliminates power compression issues in subs, and cheap efficient amplification is plentiful. Just put in a bunch of power subs and be done with it.
@donavabdear already wrote he only has one sub for his Genelec system (maybe in a different thread, I lost track). @donavabdear , I have to expect that is contributing to some of the difference. I would consider playing around with integrating your main subs the the Gens even as an experiment. That or play with the single sub near-field. Not sure why this came to mind, but someone asked what the best sound they could get for a $1000 was. I told them $500 headphones and near field sub for the emotional impact.
|
@donavabdear didn't we go over this already :-)
MTM of WMTMW are meant to be listened to on-axis at tweeter height or whatever the tweeter height is based on the total speaker angle. At that height, there will be no phasing issues (assuming I know what you mean). The sound from the two mids-woofers at all frequencies will get to the listener at the same time. The crossover is designed as such that those frequencies all arrive at that same time. This has an advantage over a flat-front MT where the ideal response is not perpendicular to the face but typically tilted down. That can be fixed by tilting the speaker up, setting the tweeter back, or electronically. It can also be fixed with a coaxial driver. I think that is the real advantage of a coaxial driver, consistent dispersion.
The problem with MTM is the vertical directivity is narrow making the listening height more critical. I have not given a lot of though, but the wide spaced woofers in a WMTMW should provide some line source effect and reduce the floor/ceiling mode which is good as those are usually the least treated.
I personally am not a big fan of MTM, and they really are not in favor. We know enough that they do not make much sense any more. Audio Science Review probably inadvertently has given Genelec a lot of press in the consumer market. They have released a great product obviously, but that does not mean other great products not as visibile with similar design goals don't exist. As they are now going after the consumer market, it may influence that segment of the market more than anywhere. The WMTMW is an "audiophile" thing. It does not have to make sense.
|
I don’t expect AES-75 to have much or really any impact or influence on the audiophile community. It would be inaccurate to say it is targeted at the professional market only, but that will be one area where it will be used. It will also be a useful tool for professional users, i.e. engineers working on vehicle sound systems to provide a more useful and consistent measurement of how loud a system will go. It will no doubt show up in data sheets for some consumer products, and suppliers of test equipment will add it to their test suite but for the average consumer it will go unnoticed.
On measurements in general, given that many audiophiles don’t trust the science currently, including, importantly, what the limits are of the audibility of distortion, than measuring or rewarding performance below or far below scientifically validated limits is not unwarranted.
|
On come on @kota1, you should be ashamed to even provide that first link to superbestaudiofriends. That is nothing but someone upset that someone is pointing out that their high priced equipment probably does not do what they say it does. I don't know Amir, but looking at this photos and background, he does not appear to be hurting for money. To suggest he is on the take from some Chinese vendors, or to link to such an article, without any proof at all, is morally corrupt. Attacking the messenger because some companies are able to sell low cost products that perform very well? When has attacking the messenger been anything other than a deflection? Do you want me to go through that whole SBAF mess? Free gear? I saw his system. Hardly seems to need free relatively inexpensive gear. Amplifiers and DACs with inaudible distortion being rated poor? I already covered that. There was a comment about a MOTU being incorrectly compared using single-ended versus balanced. Looked at the review. That is incorrect. Balanced was used and even used at higher output than standard as the result was better.
I will take on one specific topping in that attempted hatchet job because it applies to this topic and that is specifically the comments made about his review of the SVS Ultra Bookshelf versus the JBL305 (at which time he compares the JBL control 1). Now the point the author seemed to be making is special treatment of JBL. As a first point, he effectively said the JBL Control 1 is really awful. Hardly special treatment. He did say he liked the JBL 305 a lot, but not the SVS Ultra. As noted, his listening area appears to be untreated. He listens single speaker, not sure if near field or not. Let's dissect the dissection that the person incensed with audio science. He makes a claim about the calculated in room response being pretty good, and said Amir's comments didn't match the graph. Amir's comments were warmth or brightness depending on how your draw the line. That is exaclty correctly. The calculated is just that, calculated based on a "typical". Depending on your room, it may be bright or warm. The incensed also took issue that the JBL room response was not as smooth as the SVS, and felt this could be corrected.
Perhaps the biggest issue with Audio Science is that it presents highly technical measurements that are then interpreted incorrectly by people who don't understand them. You can't make conclusions about a speaker by looking at one graph. The SVS has several glaring issues. It has some dips in the on-axis response. These will be audible. Worse, it has a serious off-axis response at 3KHz and at 7Khz with varying issue in between. Right there, you have broken 2 generally fundamental aspects of designing a good speaker by modern standards. What that means if you attempt to correct the room response as the SBAF person suggested, you would wreck the on axis response even more causing worse issues, not fixing them. There is also some additional directivity issues from 2-3KHz you don't want in a modern design, and the cabinet resonances seem high. The JBL305 is not perfect, but the on axis response has not broad irregularities. The ones that are there are narrow, so far less audible, but not so narrow they look like resonances except around 1.7Khz but that could be crossover overlap (Amir notes are resonance). Off-axis is very smooth and simple a sloped shift from on-axis. That is very important as it means you can correct the room response fairly well without breaking the on-axis response. The 305 does not appear to have any directivity discontinuities. When you look at the totality of the measurements, and you understand what they mean and how they interrelate, then it is not at all surprising that Amir did not like the sound and that he could not EQ it to fix the issues. He did comment it played loud without distortion by the way.
|
I can only assume @kota1 , that by whack job you mean Amir at audio science? Do you feel you are in a position of knowledge or experience to make that conclusion.
The BS that AES75 recognizes and attempts to address is that you need verifiable standards, not some whack job with a volt meter measuring stuff in his thread bare living room:
It did not take long to figure out that audio science is using a Klippel for speaker measurements. This is hardly a volt meter, and represents the state of the art in audio measurement. I think he runs it in his garage, but it almost does not matter as long as you have enough space. The point of the Klippel is that it does not need an anechoic space or treated room to measure accurately. It is a great tool, though a bit slow as a development tool. The Klippel will export a CEA2034 compliant test report. That is a far more extensive test standard than AES75. The reports that audio science publishes are from within the CEA2034 measurement set and appear to cover most of it. CEA2034 would also be considered "independently verifiable", as it sets out the full test standard, methods of test, equipment requirements, reporting, etc.
CEA2034: Standard Method of Measurement for In-Home Loudspeakers
This standard describes how to determine the frequency response, directivity and maximum output capability of a residential loudspeaker. It is intended to determine the audio performance of a loudspeaker, not the loudspeaker’s ability to survive a given input signal. This standard applies only to loudspeaker systems, and not to raw transducers.
This contrasts with the AES75 standard, which has one, and only one function,
Abstract: This standard details a procedure for measuring maximum linear sound levels of a loudspeaker system or driver using a test signal called M-Noise.
I don’t consider his apparently very high end electrical test gear "a volt meter" either and fail to see how his thread bare living room will make any difference on the measurements. From my colleagues, apparently the standards around electrical performance tests are not extensive and all over the place. They also say it does not matter much as long as the fundamentals of the test is communicated. The speaker testing is well beyond anything anyone else has done previously in online reviews. I don’t know all the ins and out of the electrical testing, but even if there are flaws, it still appears far more detailed than what has been done previously.
I will comment that listening is not done as per AES20, but no audio reviewer comes even close. AES20 requires a stereo pair, but it also places requirements on the room and placement, as well as the requirement for blind testing. Without an optimized conforming listening room, single speaker listening will provide the most repeatable results which appears to be done.
|
@kota1 , I am not sure what that link to the other thread here on audio science review is supposed to prove? A quick skim from the end and working back and I could not find any good examples of where anyone provided a solid founded argument to support what you claim. Looking from the outside in, I saw a lot of emotion driven writing, but little in the way of fact driven, logical arguments with supporting documentation. Some of the people responding should be giving their "adult" cards back.
Giving myself a reality check, I am not sure what any of this has to do with powered speakers. I feel like the new AES75 standard was just used as an excuse.
|
@kota1,
If you posting your system's pictures is supposed to lend any credence to what you say, I am sorry to say that for me, it does not sway me one way or the other. Like musicians often having crappy sound systems, I know some excellent people involved in speaker design who have relatively modest systems. Personal and professional passions don't always align. However, if you know speakers, it would take about 2 minutes of talking to them to know they know their stuff. Hence I place more value in what people say and know.
I will give you an example. You post pictures of near flat room corrected responses in your system pages using Audyssey apparently. Critical is your front left and right. I assume, based on some things you wrote that you think this is a really good thing. If you know the limitations of the Audyssey correction system (and Arc) you know this is not a good thing. That flat of a room response given the pre-corrected response means that other aspects of the response that are also critical were compromised. That is why advanced systems like Lynggdorf and Trinnov have both better measurement (out of necessity to work properly), and more flexibility on correction. Dirac is somewhere in the middle, though I am looking forward to how Dirac Active performs in the wild, not to mention the expected patent battles.
|
@kota1 the topic of the thread is "Powered speakers show audiophiles are confused", not does @thespeakerdude have a nice system. I have talked extensively about the technical details and underpinnings of active speakers. I can and do that all day off the top of my head. Have you posted anything relevant in regards to active speakers, how they work, why they can do what they do, why they will only get better, what technology underpins those advancement? No you have not. So how about you stick to the topic, and stop trying to make yourself look better at my expense.
|
@phusis,
Have you been following what I said I do? Speakers for professional applications? Do you think that just means studio speakers? @donavabdear mixes for movies where you are trying to "recreate" real dynamic events.
20CuFt is not really enough for a proper horn loading at 20Hz.
Tell me @phusis, what is your personal definition of "dynamics"?
What is a sufficient peak db level?
What is a peak db level listening to an orchestra say 10 rows from the stage?
How often are you trying to recreate a Saturn V launch?
|
Yes you have talked extensively @kota1 , that is not in question.
|
|
@donavabdear ,
For you Genelec, the mid is the waveguide for the tweeter and the front is the waveguide for the mid, so we better hope things have gotten better.
Reflections, higher order modes, throat and mouth and edge diffraction, resonances, it is not as easy as it seems.
The answer is the problem still exists, but we have gotten much better at making them, not the least because we have software now (even in the DIY community) that really helps to automate and optimize aspects of the designs so instead of making a 100 revisions or more CNCing plastic, you are now making 5-10 and printing them. How many you need usually comes down to how good your model is of the underlying driver.
Our software like our competitors is a mix of proprietary software and off the shelf generic tools and plug-ins. Obviously can't post that, but I can post stuff out in the "community" that is really good. This reminded me of a interview with Earl Geddes, always as crotchety as ever, but also always interesting. I like the guy, don't get me wrong, but he is like that friend you have that is always cranky :-)
The video is rather apropos, as he mentions active and passive speakers too, though I consider some of this comments outdated. Some however, are not. The most important one is that active speaker design does not magically solve all issues. You need to start first by being a good "passive" or in essence fundamental speaker designer.
https://www.youtube.com/watch?v=nhe8VfuTg08
More relevant to the discussion is his mention of what he considers, and many agree, of the importance of constant directivity to ensure a speaker sounds good in most environments. This is a implementation embodiment of the Toole\Harman research showing smooth off-axis response is important. That does not solve the waveguide issue, it just give credence to why it is important to solve it. Very relevant to the discussion is the use of software for wave guide design. Geddes says that even the DIY S/W, or at least semi open source, does a better job than he was able to do. The software and a comprehensive discussion happened here:
https://www.diyaudio.com/community/threads/acoustic-horn-design-the-easy-way-ath4.338806/
|
You wrote ".. a real horn loaded speaker at 20Hz," and I’m telling you a 20cf. quarter wave tapped horn with a tune a ~22Hz will do honest and proper 20Hz - period. What isn’t proper is asking a smaller size, lower eff. direct radiator doing the same, even with a surplus of power. And horn sub iterations can be in multiples as well.
The realities of path length and flare angle means that for a practical sub-woofer, a vented box will always have the advantage at the lowest octaves. The horn is significantly more efficiency at higher bass, but at the deepest, it is not. If you are recreating the Saturn V, you need the energy at the lowest frequencies, 20Hz and below. You are better off vented. For pure music, you don't have a lot below 40Hz and dropping quick, horn is great for a sub. I realized the Saturn V was both hyperbole, but also representational. You want to recreate life's audible events.
A pair of corner placed (i.e.: with boundary gain) high eff. tapped horn subs and high eff. pro cinema main speakers with a combined 2.3kW actively per channel can shell out +125dB’s at the LP (~11ft. listening distance), full-range, so backwards math gives an easy 105dB’s with +20dB’s of headroom - within my actual required range.
That's fair. Essentially rock concert level, somewhat close to the stage.
I am glad you don't listen to it regular. As much as I love live music, I don't say yes to the frequent invites I get any more for amplified events and even for the last long while, I have generally enjoyed with ear plugs. I take enough of a "hit" professionally. Have to respect your ears.
The clean (and full-range) dynamic bandwidth not least comes in handy with Blu-ray/4K UHD playback of movies.
@donavabdear can comment on this better, but the target playback levels, we could call it the intended levels, are far below what your system is capable of.
10th row with a large symphony orchestra during tutties? I’d say 105-110dB’s.
10 seats out, 105 would be the max, and usually lower. If you were up on the stage, it could hit 120 with some pieces having extended 110db+ sections. This is starting to become a big issue, starting initially in Europe. Due to the amount of practicing, the musicians total exposure can be at ear damaging levels, especially in the brass section, even worse than percussion though percussion can have higher peaks. Lots of talk w.r.t. regulation, creating practice spaces with more distance between performers, positional changes to reduce total exposure, etc.
|
And @kingharold , I noticed that the THT measurements are all corner loaded, hence 1/8 space and a 9db gain. Yet the website compares to Genelec readily available measurements which are 1/2 space, i.e. only 3db gain. Not exactly an apples to apples comparison.
My comment, so we are both being clear, was a 20Hz flat horn would be enormous. Your what appear to be 24" wide THT, as you said 18cu ft is not exactly "flat" to 20Hz (and of note the 1/2 the size vented catches up at 20Hz).
Even the large 27cu ft version is not flat to 20Hz. This is corner placed so maximum boundary reinforcement. It is not going to get better than this. If you had a vented 18cu sub, it would have a significant advantage in the deepest bass. This is an inherent issue with practical bass horns at least for home theater / effects. For a given size, they can’t reach as low as a vented box can. Better efficiency once you get past deep bass, but not as good for deepest bass. There have been designs that try to address that, but everything is a compromise. Fix one issue, create another.
Not that I would expect it to be by the way, just calling out hyperbole.
Now, you say the midranges are 16 feet from the bass? So I have to assume a much bigger listening room than most would be using? This is also implying that the cutoff for the woofers is above 80Hz? Your definition of midrange may not match what most people’s are. I will reserve judgement till you clarify, but if you are running the bass bins with any output past 120Hz, .... well better not to say anything.
@donavabdear , this brings up back to what you were saying about reflections and phase issues. Folded horns, if you have the space, can get some high efficiency and reasonably low distortion. Think of bass bins essentially. Keep the output below 100-120Hz, and phase issues aren’t a problem as we don’t detect even large group delay at those frequencies. Allow them to leak past 120Hz and other than the obvious locatable sub issue, the oscillating phase makes for audible group delay issues.
|
@thespeakerdude what is your view on the damage the internal vibration inflicts on the amp fidelity in active speaker designs? Have you ever tested this.
Yes to test. For a DSP based crossover, virtually non existent. There is nothing anywhere in the system that a signal passes through a capacitor which can be susceptible. Weak potential analog signal level crossover, but that is mainly a component selection issue. We use a soft silicone encapsulation in some areas in some speakers and our electronics are not open to the interior, but that is as much for reliability. If the speaker is cranking 100db, the distortion is probably already climbing, and the simply large levels mask anything low level, if it was happening. I can't go into details of testing, but as you would expect, multi-tone stimulus comes into play, a tone that could perhaps resonate something on the boards, and then a second reference tone(s), with both of them monitored.
|
Surround recording should be recorded with a fake head in a binaural format because that is how we hear all other recordings are improper. They may sound wonderful but they aren’t accurate by definition. I have done very few surround recordings but adding the channels by force for the XYZ axis information to be interpreted by your brain somehow to me seems like looking down the wrong
A recording with a head and torso simulation would only work for playback if listened to on headphone corrected for equivalent flat response. Even then, it will only be approximate for any person as it is not their head, torso, or pinna.
For the Spatial Mic, they don’t seem to be tied to ADAT.
No need for expensive multi-microphone & preamp setups – just plug-in and record with either Dante audio networking or USB/ADAT models.
I had this bookmarked if anyone is interested:
https://www.dpamicrophones.com/mic-university/immersive-sound-object-based-audio-and-microphones
I vaguely remember a paper that discussed using 4 omni microphones in a pyramid shape with a lot of signal processing to extract objects for surround. I did a search but could not find it quickly.
|
@kota1 ,
For an 85" TV, the recommended THX viewing distance is 9.5 feet. For an 8K set, you need to be about 3 feet away to get the full advantage of the resolution. For a 4K set, you need to be about 5 feet. HD, about 11 feet. A 4K set makes sense to get the best resolution at realistic viewing distances. 8K for home really does not make sense.
It looks like 8K LED\LCD in Europe may be dead in the starting blocks at least in European sized sets. They won't pass efficiency requirements due to the high light loss from the reduced aperture. Don't remember, but expect at 85" you should still have a big aperture if everything scales, but maybe there are some limitations. The 8K, 85" SONY uses 50% more power than the 4K.
|
You can do it with speakers you just need your head in a clamp. Fails quite miserably at the human factors aspect of product design. Small movements off the centerline of your head destroy timing. No way around that except head tracking which there have been a few systems demonstrated.
Lipshitz? proposed something with two additional speakers near your head on either side.
Want to try something fun? Do you have open air headphones? Put headphones on, turn the main speakers off and run your subs up to 100-120hz.
|
I spent the first 10 years buried in engineering and didn’t get to hob knob with celebrities like some have on this topic :-) ... just a bit envious, though I have been star struck a few times since but back to the topic at hand.
If you don’t mind your amplifier and speaker adding something to the music that was not there in the recording, then by all means go with the Pass amp and passive speakers. You have to like what you are hearing. If you want a improved level of accuracy, then you need to get rid of the passive crossover. If you want the absolute lowest distortion possible in a given form factor, then you will need to ditch the Pass amplifier as that requires a tight integration of amplifier, driver, enclosure and more than a "hint" of advanced signal processing. If you want a single speaker that can adapt to your environment, your mood, or your music, more than just simple equalization and time alignment, than that is going to require an active speaker and that is what you will start to see more of. Active speakers are in their infancy.
Where it is going to get complicated is "deciding" where an active speaker ends, and where a room correction system begins.
|
@lonemountain ,
It is one area where we think the ATMOS standard / specification is insufficient and lacks the framework for future extensibility, tuning, and features.
ATMOS provides a rough definition for the dispersion pattern of speakers depending on where used, but it is fixed, without consideration of actual pattern, cutoff, dispersion versus frequency, etc. For an extensible object oriented speaker system, there should be much better ability to define the speakers and eventually the appropriate algorithms to maximize the "experience". Though ATMOS is object based, those objects are assumed to be fixed in terms of what they are with a fluffy definition.
|
Most of the DTS professional documentation is business-business only (and NDA) so I cannot post it here. The Dolby Cinema specification is here:
https://professional.dolby.com/siteassets/cinema-products---documents/dolby-atmos-specifications.pdf
I won’t say that Dolby rushed the specification, but like all of us, they have time to market pressures and first to market pressures. While ATMOS may be object oriented, the rather fluffy speaker specifications including fairly wide dispersion for some and very limited in terms of detail cutoff specification places limits on how many useful objects you can have.
There is also another significant limitation at least in conceptual implementation. The basic premise of ATMOS is that everyone will hear the same thing. That is fine for music,sufficient for mainstream theater, but lacking for where the largest entertainment industry wants to go, namely video games. Another expected growth area is immersive experience spaces.
There are obviously ways around it (headphones and custom software of course), but industry likes standards, because standards bring down costs, reduce training time, and improve time to market, and reduce risk.
|
@donavabdear ,
The microphones will be flat. They are calibrated when used for room correction.
The Lyngdorf should allow you to adjust the target curve. If you don’t find it warm enough, up the bass a bit, lower the mid-upper highs, and then adjust the mid-range to suit. You may get closer to what you like.
The imaging is likely better from a more consistent channel-channel frequency response and less distortion.
Also there is electronic DSP that creates optimized curves based on information from the way the speakers interact with the acoustics of the room, these curves will most likely not correspond exactly to all the other curves our modern systems have in them
The fundamental challenge in room correction is balancing these two conflicting requirements:
- As close to a flat direct frequency response as possible, at least from about 1-5KHz, ideally from 200Hz - 6 or 7 KHz for each speaker
- Consistent left and right channel room response
- A room response that is somewhat close to the researched Harman room curve that seems to suit most people.
DSP can only do so much. You need to start with an acoustically tolerant space.
There are other aspects of the audio signal that change the curve like phase knobs, separate EQ on subs, different pass filter fall off rates, built in EQ on preamps, A to D converters, D to A converts, etc.
I wouldn’t get too hung up on this. For one, if you are using passive speakers that are not time aligned, the speaker is likely the biggest contributor. Two, many room correction systems will do time alignment. Three, reflections make a mess. Last, four, we are not that sensitive to phase changes. Well researched area. In design, our critical measure is always no phase discontinuities. You never want the phase to change quickly.
No only are audiophiles confused but manufactures are confused because playback systems are not made like your ears, in electronics it is very difficult to not introduce feedback into the circuit for efficiency sake, so even at the base of a simple amplifier circuit we are already swimming upstream.
Are you implying electronic feedback is bad? Poorly implemented amplifiers are bad. Feedback is not bad.
I have went out and recorded choirs with 1 very good stereo microphone (multi microphones always have phase issues by definition) and plugged that direct recording into an amp and listened on 2 speakers (also a phase issue). The result is like eating a fish that you just caught out of a mountain stream, it is an entirely different test and experience than when you buy fish at the market.
Is it real or is it Memorex? When I was young, and would go to clubs, I hated when they had a band and preferred DJ. Now I would prefer a band as long as they are decent. Does age hone our skills to detect authenticity? I have always preferred a well recorded live album, but that is not a simple stereo pair microphone. Untapped market for artists that are authentically talented?
|
@donavabdear ,
I am not up on all the latest in home audio equipment to even begin to spend $200K, and it would definitely take a good review of the space, understand what potential acoustic issues there are, etc.
However, before throwing out the baby with the bath water, maybe we need to change the water first? One comment on the tubes, I am not an expert on them, but I thought the tubes in pre-amps or pre-amp sections could last 10 years? That does not seem like much of a hassle if you like the sound.
Just a quick perusal. You only have a single left and right balanced output, which I assume is going to the amplifier? What are you running to the sub? One of the unbalanced outputs? Those are not the quietest outputs, but not the worst.
- The amplifier gain (Stereophile) is high and you can't control it
- The noise on unbalanced outputs of the preamp is high, but looks like about 90db on the balanced. Not solid state, but not awful
- The output level of the preamp is high (and looks like that may be where the SNR is done)
- You probably have a ground loop to your sub. Isolation transformer to power your sub, audio isolation transformer to isolate the single ended connection if that is what you have
- With the amp gain so high, you need to gain up your sub to match, making the buzz/noise worse.
Your amp is MOSFET output, so it won't behave like a regular tube amplifier with output transformers and tube amp output resistance. I think that is what that Sunfire was trying to emulate. Maybe its the distortion of the tube preamp you like, maybe the tube section of the amp, maybe the two together, or maybe something completely different. I think I would start with a new SS pre-amp with lots of balanced outputs including one for the subs/mono and preferably one that lets you set the relative outputs. With your budget, I am sure you could get a loner. I would start with something with really low distortion and noise floor. Maybe you will like it, but at least it will give you a baseline.
|
@kota1, @donavabdear amplifier, the BHK300, is not a tube amplifier. It has a tube input stage. The output is a high bias MOSFET AB. The damping factor is 100. That is not very high for a solid state amplifier, but would be very high for a tube amplifier. The Sunfire 7401 is 150 in voltage source mode and 8 in current source mode. The Persona 9H drops to 2.4 ohms. It has low impedance dips at a few spots. That is not a speaker I would want with a damping factor 8 amplifier.
|
I admit my limitations as opposed to doing no research and recommending an out of production amplifier not suited to the system. Run that amp in the constant current mode and you would add new +/- 3db peaks/valleys to the frequency response that are not currently there
@donavabdear , creating a separate topic makes sense. Are you planning to add room correction to this system, or just keep it the way it is?
@thespeakerdude
I am not up on all the latest in home audio equipment to even begin to spend $200K,
+1 (+10!)🤣
|
|
@donavabdear , the Lyngdorf Class-D is either off the shelf Purifi or customized. The brains is Bruno and Lara who started Purifi with Peter. Those are Peter's words by the way. Class D is almost a technical necessity for the bass due to high power in active, but for mid / tweet, not essential though a switching power supply becomes a packaging requirement. In professional products no one asks you what's the power supply. They only care about the sound. In the home market there are a lot of preconceived notions. There are also a lot more single ended electrical connections more susceptible to noise.
That Meridien system was nice. Not a fan of the room and lack of treatment but they have to work with what the customer allows. They dropped the ceiling though, so could have created a 6" fake acoustic wall. All digital interconnect. Just standard Cat6.
|
Funny, @kota1 , they Sonos has tried to recruit me twice. How about you?
However, at least to the latest information I have, their emphasis, at least from a pure audio standpoint, is best audio in small size. There are a lot of parallels from a technology standpoint with achieving maximum performance in a large size, but parallels and goals are not the same. I did think their purchase of Mayht was wise though for their goals and it will be interesting to see if they push that technology up to larger sizes.
Let me know when you purchases your Meridien system, though at that point, like now, you will just be someone who purchased active speakers.
|
