Kvalsvoll

A lot has happened since the last update here. I mainly use facebook to post short messages and pictures of what is going on now, and articles are on my web site in the blog-section, products on the product page. Which means no focus on updating this thread.

A new very small speaker has been developed, it was due for launch and presentation, and then you-know-what arrived as a free gift from someone I am not very pleased about at the moment.

So launch is now postponed indefinitely, until people are able to show interest in speakers and sound again.

The new F205 has a preview page, those interested will find it on the products page, link to preview is on the F205 page.

The goal for F205 was to create a very small speaker with sound properties similar to large systems, with as few compromises as possible. The result has surprised me a bit. Since I had my F205 pair installed in Room2, I have not reconnected the larger F2, and have no urge to do so. In fact, after installing the first pair for testing, I found I really missed them after shipping them off. They are addictive. 

The F205 has the same character of clarity and transparency and dynamics as the larger F2, but it is more neutral and the highs are in a different league. Kind of like a huge horn speaker without the coloration, like a sophisticated electrostatic or magnestat with no limits in spl.

A picture from Room2, the F205 in front:

This is great, and as you mention - the challenge is how to market sound. The irony is, that it would be easier back in the 80ies - 90ies - before the internet. Today, you are easily lost between all other outstanding and remarkable offerings, and they have a larger marketing budget.

Describing sound quality in text is a hopeless exercise. You can never have the receiver understand what you mean, until they actually experience it in person.

21 hours ago, SME said:

I believe these group delay shifts are pretty modest.  IIRC, an ideal LR4 at 80 Hz has excess delay in the 5-10 ms range.  I used to think that getting more energy in the first arrival is important for the most intense slam, but I now believe that the experience of tactile slam is actually a very high level perception.

My thinking is that the brain samples information from both ears and vibro-tactile receptors and then applies sophisticated "room correction" processing to this information in order to reconstruct an accurate estimate of the actual time-frequency aspects of the original sound.  In my own room, as I tighten up the bass response accuracy, the sound and tactile sensation I perceive become remarkably more uniform throughout the room, despite what the localized measurements suggest.  If accurate enough, one can get an extremely tight "thump inside the body" effect that is more like what one experiences with a powerful system outdoors.

I'll hopefully be soon exploring this in other rooms and am particularly interested to see what I can do in tiny pathological rooms.

Yes, the crossover turns phase 90 deg for each pole, and the resulting delay is not that large compared to what room reflections can do.

21 hours ago, SME said:

Thanks for creating and posting these samples.  Can you explain a bit more about the reasoning that went into creating the particular filter response shape?  Is this supposed to be group delay related to some kind of filter (e.g. crossover), an in-room acoustic effect, or something else?  FWIW, I definitely notice a difference between the 100 ms and original with a clear reduction of impact.  Between the 20 ms and original, I think I notice a very slight reduction in impact, but I'd want to do blinded A/B/X to be confident.

Now having said this, there is a big caveat with your study because there's a big difference between GD applied electronically and GD that arises from acoustic effects.  That's why I asked my first question above.  Time and time again, engineers try to treat room acoustics as a "linear transform along a wire", when this is not the case at all.  The ears and body are capable of sampling pressure at multiple locations (the ears and tactile), and the brain is very well adapted to parsing the content of the source  (both time and frequency aspects!) from what could be a very messy sound-field with dramatic local variations in measured frequency response and group delay.  So in general, electronic changes may be far far more audible than FR and GD features of similar magnitude that appear in in-room measurements.

Another potential caveat here.  You indicate that the frequency response of your filters "is reasonably flat, considered below threshold for audibility".  I can't comment with certainty in your specific case, but in general, I would not be surprised if the frequency response changes you show were well above the audibility threshold on a system with strong accurate bass.    This alone could have substantially affected the amount of perceived impact.  Again, there is a big difference between filters applied electronically and influence of acoustics on measured sound vs. perception.  Depending on the circumstances, I believe the brain can pick out excruciatingly small changes, likely below 0.01 dB for bass.  These can be perceived most readily on transients.

Regardless of the audibility of your filters, this experiment says little about the audibility of characteristics arising from room acoustics or whether it's necessary to "correct" the group delay deviations seen in in-room measurements.  I can't emphasize enough how important it is to keep this distinguish in mind clearly when optimizing response.

Also a comment about the sample material.  The kick seemed a bit soft, diffuse, and fluttery to begin with, and it didn't really sound consistent between beats.  Audacity spec analysis suggests that the kick has some high Q ringing at various frequencies, which also does not appear to be consistent between different beats.  Differences in group delay might be a lot more apparent on tighter transients that don't ring so much.

I chose processing that resembles some of the problems I see in measurements from systems around. I have some measurements from customers now, and some of those follow the calibration process from start until the customer reaches the crossing between bored and good enough.

Some subwoofers introduce so high gd it is impossible to correct for in ordinary av-processors/receivers. Then there is the room.

Yes, I could do a better job with frequency response accuracy.

Sample material was quickly chosen, drums with more dry character may be easier to hear - or, may be not. Must try it, to confirm. But my reasoning was, that if it shuold be possible to hear on normal music with transient drums,

I have updated the article 2 times, with comments, and new sample files.

Thank you for taking time to listen to this, and it is interesting to see your impression is close to my own experience. I suspect it is required to have a very good bass-system with decent response - both frequency and timing - to hear this.

I just updated the article with spectrogram charts and subjective impressions from listening in Room2, now at more realistic volume.

10 minutes ago, maxmercy said:

How is this done?  In AVRs, the subwoofer crossover is not adjustable for slope in most cases.

JSS

Delay is adjustable, and the rest is fixed in the dsp on the bass-system. For a text-book filter there is a delay, and if you adjust delay on mains to remove this, it wqill not sum correctly. But if the slope on the bass-system is different from the text-book, it is possible to end up with something that sums correctly and has zero delay. you end up with flat phase and flat GD.

@maxmercy, yes, this is true. But it is actually possible to remove all gd/phase from the crossover itself, by adjusting slopes and delay. This is also possible to do higher up in frequency, in even in passive crossovers, but then you don't have the option of adjustable delay, so you depend more on simulations to get it right.

Audibility of group delay at low frequencies:

https://www.kvalsvoll.com/…/audibility-of-group-delay-at-l…/

It is then confirmed and proven that group delay at low frequencies is audible. 

This means timing - GD, phase - matter for sound in the bass range, it is not sufficient to tune towards a flat frequency response alone.

(Yes, we knew, told you so..  But, now it is actually proven with evidence in a replicable,  described, controlled experiement.)

16 hours ago, SME said:

Wow!  What complete and utter rubbish.  I'm not going to question the author's reputation as a mixer, which may or may not be well deserved, but most of what's written there is embarrassingly wrong.  It's hardly unique to the author.  He's just perpetuating myths that are widespread throughout that community.  These myths come about because people are trying to relate their subjective experiences with sound to objective principles, without sufficiently understanding the latter.

...

The disaster here is of course the writer of that article will never read you comment, not because he does not want to or does not want to learn how this really works, but because he will never see it.

The main problem with audio today seems to be more about communication - getting through with the message. We have the technical solutions, we know the theory, we know how to make this work in a practical audio system.

7 hours ago, 3ll3d00d said:

Re the holes on the side, translate says they are "acoustically filtered", does it mean some damping material behind the holes? I looked into doing this once but the amount of trial and error (and/or lack of modelling tools) put me off. One thing I do remember though is that any such speaker tended to be pretty deep in order to get enough delay on the side output to affect directivity in a desirable way (iirc). These are v shallow speakers though by the looks of it. Are you using them for this purpose or something else?

Yes, something inside the cabinet behind the ports to control acoustic resistance. This is modeled and simulated, at least to some extent. Dimensions and placement of ports, cabinet volume, is nice to get reasonably correct. The damping will always need some experimenting, because it is not straight forward to get good models of the acoustic properties of the damping material. This is not a problem, it does not take much time to get it right.

The cabinet does not need to be very deep, but the location of the ports, baffle width, speaker driver size all matter for resulting response.

I plan to develop more of this type speakers, with damped ports to control radiation. The F105 is small, with low-cost drivers, I designed them to test the concept. And it works.

New articles in the blog-section, published quite recently:

The F105 loudspeaker - review-style article on this small loudspeaker with some quite interesting technical solutions:

https://www.kvalsvoll.com/blog/2019/04/23/the-f105-loudspeaker/

(Use google translate for the text that did not make it through the author's translation service.)

Bass and sound quality - 4 real world examples:

https://www.kvalsvoll.com/blog/2019/03/10/bass-og-lydkvalitet-4-eksempel-fra-den-virkelige-verden/

(Use google translate, it continues to improve, and has now reached a performance level sufficient to make most of my articles understandable in English.)

Looking at what is happening around on the net and otherwise, it is quite apparent that audio has died and become sort "Reign of the nonsense". There is no interest for technology, little innovation in new products, what is left of audio press are rendered totally irrelevant, lots of nonsense products gets whatever remains of attention. And there is simply no advancement in knowledge and technology - you find yourself just repeating and trying to explain things that by now should be known ("it is known"), which of course hinders further development up to the next level because you are stuck with this debunking of myths that should not be alive. Most of the really serious enthusiasts now build their own custom systems, which is good, but also not so good for audio as a business, because there simply is no market.

Read the articles. Is this interesting? Is it relevant? Is it entertaining? Does it make you inspired to learn more about audio, put on some music and listen, find out what you can do with your own system?

There was some discussion about some subwoofers that were linked to my name, in a different thread. Please feel free to ask or comment, and you can use this thread, no need to start a new one.

If you go to my web site, you will see product presentations of the subwoofers I sell, and articles related to those subwoofers and bass in general. All presentations and product info are available in English language, most of the articles as well.

What I am going to say now, may seem a little strange:

I am not on data-bass to promote my products and services. Actually, I am not that keen to have a discussion about either products or technology, here. There are reasons for that. I want this to be my "free-space", where I can post and talk about things, and not worry about whether what I am saying is good for my own products. But feel free to ask, and I will try to answer. There is a contact-page on the web site, where you can find information on how to get in touch with my company. The company is also on facebook, where you can post public questions, or send private messages if you prefer that.

The obscenely large "horn" pictured in that other thread has a story behind it. The driver is a 24", so that thing is quite a bit larger that what it seems like from a quick look at the picture. The origins of this design was that the builder was curious about whether a compact-horn using some 24" drivers he already had, was possible. And it is possible, but even with the chosen tuning it gets very large, and the performance per size-unit is not particularly good. The driver is simply too large. It is a tuning with very large rear chamber and short horn channel, closer to a ported pox with huge port, than a real horn. This design performs a little better than 2x V110 - but those V110 would be half the size of one of these overly huge cabinets. The sound quality, however, should be quite good, as there are no resonances and very smooth response in the intended pass-band, it also has a very low cut-off. He has built several of those now, and it would surprise most of you what they have replaced, how sound quality improved, and I have never seen any complaints or concerns about capacity. Which does not surprise me at all, and I have not even heard them.

As for the question - why? They are too large, similar performance can be had in half the size, on paper it may seem like even much smaller than that using regular "mickey-mouse" subwoofers could do the job. So, why on earth choose something like this, even if the price-per-size-unit is extremely good due to cabinet made in construction-chipboard. My take is, that the 'I have a 24"' is the major factor, and combine that with the knowledge that when it comes to bass, the real thing requires something very different from what you find in the typical shop, price is low, and suddenly you find yourself looking at ways to make room for a couple of those small houses inside you listening room.

1 hour ago, Ricci said:

Cool...I didn't translate and read the thread. I did have a brief look at your website. It appears you are selling subwoofers or will be soon? I don't care if your stuff is discussed. By all means do so, but I don't think the Skram thread is the best place for their discussion. It appears that member Alexel may have questions.

No place for this in this thread, regardless of what is for sale or not. Go to the Room2-thread, I will make a comment there.

On 4/29/2019 at 6:07 PM, Ricci said:

Kvalsvoll is a member here...Discussing his products for sale is kinda off topic for this thread. What he has going are not "technically" horns but that's just semantics that doesn't really matter in the scheme of things. The tuned resonators being employed are the most interesting aspect of them to me, because they aren't employed very often,  but they do increase the size and complexity of the system. Not sure I would call them compact relative to the driver compliment in the cabinets, but I would like to see things like measured impedance, voltage sensitivity, distortion, compression and group delay on one of his designs.

If you have questions about my designs, you can ask in the Room2-thread, I may even answer.

As for this specific construction pictured above here, that is not something I sell or build, this is a diy based on technical input provided by me.

Guide to this season of GOT:

1. Go to Braavos to lend money in the bank for a new screen with night-vision.

2. Valyrian steel is definitely not required for the magnet system in your subwoofers.

3. Most important step: Go to data-bass and learn how to apply bass-eq.  Oh, it's this thread.

4. Turn it up and enjoy. Part from the non-existing low bass the sound is good enough.

New article about bass-system calibration:

https://www.kvalsvoll.com/blog/2019/02/10/calibrating-a-bass-system/

Small will still not be the same, because directivity will be different. Unless directivity is designed to be similar, of course.

Active does not make any difference, and eq can not change decay profile, which is a result from room and speaker radiation. The solution lies in the physical acoustic design of the speaker.

Active is necessary to solve some issues, such as crossover at lower frequencies, and delay between drivers where this is necessary. Proper crossover between bass system and main speaker must be active, because it is too difficult and costly to make a passive network that works, and the quite large delay needed on mains can not be solved passive. Subwoofer usually requires more power, too. But then all this is not necessarily true either, because when I designed the C2 (1992) I made a passive crossover at 150hz, so it is possible, using computer simulation. But timing will be off, this ends up as a text-book 4. order crossover, with option for individual delay it is possible to design crossovers with no timing problems.

The new F105 is a small speaker with 5" lf driver and dome hf with moderate horn loading. The cabinet is a damped dipole with resistive acoustic ports. Directivity is controlled all the way down to 100hz, where it already has started to roll of. Design f range is 120hz and up. With a small subwoofer - will be half size of a V6 - that can be placed near boundaries, with dsp and eq, this gives a small system that will work well in normal rooms. The response is much smoother and more similar in different locations and rooms due to the directivity, even when this directivity control is quite small. But it doen't sound like a big speaker, regardless of sound volume. This small speaker would not benefit from active configuration, but the system as a whole is active where the crossover to the bass system needs to be done in a processor, and the bass system need dsp with eq. The next speaker will be a little larger, with 2x 5" lf drivers of a very different type. The goal for this one is to achieve good transient response - as long as you keep the volume down.

English version of the small speaker article:

https://www.kvalsvoll.com/blog/2018/10/20/can-a-small-speaker-perform-like-a-big/

To discuss or comment, you can reply here in this thread.

Small speakers without the sound compromise have always been wanted, and there are several new speaker offerings claiming the problem is solved - small sounds as good as huge. But it doesn't. This article tries to describe some of the most important reasons for this.

New article about power requirements:

https://www.kvalsvoll.com/blog/2018/12/02/how-much-power-do-i-need/

Info on power requirements for active or bass-managed systems is interesting - reducing the bandwidth does not reduce peak power requirements.

13 hours ago, SME said:

I agree with you in part, but I think the differences can be greater than you'd think.  A lot will depend on the particular mix and also the particular playback system and possibly some subjective preference.

In TLJ, the failure to recover ULF from the surrounds is a sin of omission, which is relatively minor.  Yes, it does mean that a spaceship might lose its weightiness as it pans from the front, overhead and to the rear, but at least the sound is not worse than what you started with.   I picked the surrounds in my example because the difference is quite dramatic on paper and is one that we could all agree would be very audible with those discrete surround effects.

However the front LRC channels are another story.  Even though they roll-off at a similar point, their shapes are still quite different from LFE.  So an EQ solution that is optimized to the mono sum average (which is dominated by LFE), could introduce new humps or bumps into the front LRC that weren't there before.  Here's where we *can disagree* about what's audible and what's not.  Though arguing from personal experience, even quite small bumps can be audibly degrading.  Much depends on shape and bandwidth of the feature, in addition to the level, and also ...

Audibility of differences will depend on the playback system.  Systems with substantial bass problems may not reveal degrading resonances as readily.  (That's not a virtue as such systems also fail to reveal a lot of content.)  For example, a BEQ filter applied to front LRC that increases ULF while adding a slight bump around 55 Hz may have a pronounced boom around that area in general, but on a system with a severe boomy room mode at 45 Hz, the problem at 55 Hz may be hardly noticed.  The BEQ might be an unqualified improvement on this flawed system, but on a system with very clean bass response, the 55 Hz bump may be much more obvious and degrading.

If you had to choose between full ULF extension and balanced response between the deep bass, mid-bass, and upper bass, which would you choose?  Personally, I'll take the balanced response over the ULF extension any day.  IMO, the ULF is the least important frequency range.

I believe the notion that "[global] BEQ that gives 80-90% of the improvement" is overly optimistic, but I am also inclined to judge the soundtrack for what it will sound like on a revealing system vs. an "average" one.  So practically speaking, a global BEQ may be an improvement for most people who choose to use it, even if it does degrade other aspects of the bass somewhat.  And I do understand that most people who have EQ capability at all can only use it on the sub output.  I agree many filtered tracks can be improved to an extent with a global BEQ and that it's worth doing even if an independent channels BEQ would sound better.

But I'm skeptical that a global BEQ will always be better than nothing at all.  Focusing only on ULF, a BEQed track will always seem to be an improvement, but if one considers the sound as a whole, BEQ that introduces new bass resonances in some of the channels could end up sounding worse than nothing at all.  Again, a lot is going to depend on the playback system.  When doing these BEQs (whether global or channel-independent), it's very important to listen to the results on a system that is as accurate and revealing as possible.  (This is probably my biggest gripe with the AVSForum thread where it appears BEQs are being developed using all eyes and no ears.)

Doing a proper bass-eq takes time and effort, and as you do more of them you start to notice that experience is nice to have. Yes, you need to listen, and since a movie is quite long, with lots of scenes and sound effects, this will be a time-consuming process if you want to be sure you get the best possible result within constraints given by you skills and the soundtrack you have. 

But we only watch the movie once - usually. Kind of like how GOT (a person living in Norway) put it about skiing - the conditions really does not matter when climbing and skiing a mountain, because you only ski it once. It is what it is, that one time. If it was perfect, well, good, but if it was icy and crusty, or the bass in the movie was less than perfect in some scenes, that is what we had. But we do not have to re-live it over and over again.

This also means there is a limit to how much effort you want to put into fixing someone else's mistakes on a movie. So I usually end up picking a couple scenes, and do beq on lcr+lfe.

I see your point where individual bass-eq for all channels including surround is the ultimate solution, and can in some cases be a significant improvement, as when surround effects that has wide frequency range are not mixed in to the lfe channel, and those surround channels are filtered somewhere in the process.

But still, I claim most of the performance gain with bass-eq can be had with beq on the bass-managed signal, which is the only practical solution for most people. Then you miss out the correct correction for surround - as well as lcr - because the adjustments you make are weighted +10dB hot on the lfe channel, giving too little boost on lcr+sr.

In a world where even very, VERY few dedicated enthusiasts knows what bass-eq means, there is a huge potential for improvement that will be possible on most serious systems, because they all have dsp on the bass-system, which can be used to implement BEQ that gives 80-90% of the improvement. Compared to no BEQ, there is a huge improvement still, and the dedicated channels BEQ will only be a small step above that.

40hz or 50hz for surrounds is more like the same, when talking about the real thing, which means full frequency range - full capacity.

It is not strange that something with +60dB ulf boost on surround channels looks odd, because this is very likely mostly noise.

Keep in mind that no studio or cinema can reproduce low frequencies on the surround channels, whatever is there below around 50hz was never part of the sound design.

The most important to fix is lfe and lcr, that is where the big difference lies. 

@3ll3d00d, I do follow and appreciate the excellent work you have done on the beqdesigner. Especially since it is obvious now that the best we can hope for is movie sound that responds well to bass-eq, there will never be a situation where you can assume the sound is perfect from the provider.

I also follow, or at least make an attempt to see what is going on in the soundbar-forum thread.

Apply gain -10dB before filtering, that should leave enough headroom. After applying all filters, you can fix the gain (+10dB), and perhaps give a warning if clipping occurs.

But remember that for dsp processing during playback there may be more headroom available, if the dsp is located after master volume in the signal chain.

4 hours ago, 3ll3d00d said:

@Kvalsvoll I built the latest build on my (debian testing) box, I don't have any other distros here to try it on but it works on this machine so give it a try and see if it works.

https://github.com/3ll3d00d/beqdesigner/releases/download/0.0.3/beqdesigner

I may be the only one trying to run beqdesigner on linux, so it's like you did this for me, and I feel obliged to test it. I will see if I get the motivation to give it another try, requires some effort to upgrade to the right version of everything, including the right python version.

I will report back when I have tried it, and thanks.