Kvalsvoll

Nice find, can not remember seeing this. The data in this paper shows that the relationship between port geometry and performance is very complex, and to model this for a dynamic simulation is difficult. It also shows that port compression and nonlinearities are very significant. For very low frequencies the situation gets worse, velocity increases as frequency goes down. Build a horn instead, it will more or less solve the port compression problem.

The filters are for each channel individually, LFE means LFE channel before adding any rerouted bass from other channels. So, for JRiver you should implement filters for each of the LCR+surround+LFE channels according to the description given for the film. When LFE and bass from LCR+surround is routed to the sub output, you can apply filters to the sub output if you don't have access to the individual signals. It is not the same, but should give good results in many cases. Use filters specified for LFE channel. A good bass system has a dsp able to do this filtering. The multiple filters with smaller gain vs one filter with huge gain is to avoid too much overshoot due to the q, the response is not the same.

The F2 speakers back in Room2. After temporary fixing the hf with modifications to the horn, the hf is better. And moving the speakers to the same position as the previous F1 position, improves the sound. Still, the low-mid issues remain, and of course the 6" drivers in the subwoofers limits performance, horn loading still operates within physical limits - no magic "extra octave" or "20dB extra dynamic range", but the session listening to Gary Willis at +6dB earlier today was a very good experience. Tactile in the bass is not good enough, but the overall experience is tolerable. The sound stage extends nice in the fore-aft direction, separation and tonal contrast between instruments are quite good. There is this huge soundstage extending from behind my seating towards well beyond the front wall. And the contrast between very precise and exactly located instruments up front, to the spacious effects and sounds that fill the whole room. The F1 presents a flatter soundstage with less distribution in depth. The differences has nothing to do with dynamic capacities - both have headroom to spare in this small room. It has to do with radiation characteristics. Frequency response is not very different. Radiation determines behavior in the time domain, and this is what causes the speakers to be perceived as sounding different. The differences are independent of loudness, listening at a reduced volume still reveals the same sound characters.

Looking at the pictures again, I agree about the absorption - by no means "very little absorption", still, there are large untreated surfaces (ceiling). There is more absorption in my Room2, but in the other room there is only 8x 5cm thick (2") 60x120cm suspended panels, and 4 60x120cm (2'x4') 50mm panels on the back wall. And this room still sounds better, overall. Which suggests that it is the acoustic properties of the room that matters, and not the amount of absorption - no surprise for logical thinking people.. Another interesting thing I see from the pictures is the back wall. Center section has absorption - I wolud do the same. But corners are open - they are essentially non-reflecting. Both my rooms have untreated/reflective walls in the back wall corners, because this is where the energy from the horn speakers is reflected back into the room. If this area is damped or non-reflective, you may end up with too little reflected energy and it will sound too dead. Just some thoughts, and perhaps you get some ideas for experiments with the diffusers. If you do, I look forward to learn what you experience; does it support my hypothesis, or maybe not.

A 2-by-4, I am guessing that may be similar to a 60cm x 120cm. I looked back at the pictures in the first post, nice room and system by the way. Looks like there was no ceiling absorption at all before you mounted this. It is then quite understandable that you experience a huge difference. Difference and improvement are not the same. Absorption in the ceiling usually is an improvement, but it will sound different, it will sound more dead. But envelopment and space can be restored, in a more controlled way. From the pictures I see diffusors on the back wall, and there is not much absorption in the room, so I would assume it sounds quite lively. I would mount all the ceiling absorption you have planned, and then try to add reflections/envelopment after that, if it sounds too dead/flat/soundstage-up-front. Would be interesting to see the .mdat, but I understand if you do not want to post it.

Center - have you tried blinding off visually (dark room or curtain) and listen, do you still hear sound coming from above? Like mentioned already, I think visuals have some impact here. However, it is possible that the speaker itself can be partly localized, also in the height direction. Source material can be a major factor - some sources simply sound better with the speakers, including center, disappearing much better. If we look at how hearing can detect height, we can learn what can cause the speaker to be detectable. Frequency response is one factor to height localization, but that requires a reference - a sound effect moving from center and upwards can be quite realistic, but a static voice will not have any reference, so we can not actually know where it is. If we hear a sound, and want to find out if this sound comes from above, we simply tilt our head, then we detect changes in frequency response as we move the head. Reflections and delayed sound (impulse response) can also reveal the sound source. When working on the center speaker, I found that those factors were most important. I experimented with different frequency responses around 6k-10k to try to move the sound source down, but that did not work well. What did work, was to make the off-axis response smoother and improve the impulse response. In this process I found that linear faults outside the pass-band of the lf driver had much more impact on the sound than I had assumed initially.

Internal calc in dsp is always higher resolution, and if you do processing - like bass-eq - it is more likely that you end up in situations where noise is introduced if 16 bit is used. And we all like 24-bit, overkill never hurts, at least not when there is no penalty. Working on the room today, must get rid of the low level and cancellations in the lower mid, but.. not easy. So, ended up listening to Gary Willis, to verify if more floor absorption is necessary - now flat from around 300hz and up. Sounds good, but it does not fix the cancellations, and it is not practical and it does not look good. I think it is not necessary.

Yes, about that need for a 144dB dynamic range.. The answer to this may not be obvious to all, and certainly difficult for anyone to read my mind to see how I am thinking. First, the ? is "need for 144dB" because that is what we have with 24-bit, if we use all the dynamic range available. Lucky for us, we don't need to, neither do we want to, nor is it possible technically. 1. The "high-end" speakers will compress and distort way before reaching 144dB spl, regardless of how powerful the amplifier is. Now, if we consider a noise floor in the 20dB to 40dB range in a very quiet room, we actually need 164-184dB.. 2. This spl in the midrange/higher frequencies will lead to instant hearing loss. We don't want that. we don't want sound louder than we perceive as reasonably pleasant, and certainly not so loud that it will cause permanent hearing damage. 3. No recording has a dynamic range anything close to 144dB, in fact it is physically impossible to achieve if parts of the production is made from recording voices or acoustic instruments using a microphone. 4. No electronics have more than around 120dB dynamic range, this is partly a physical limit, the noise level can not be lowered more. So, has 24 bit audio any advantage, is the performance increase audible compared to say 16 bit? Yes, it is audible, but it has no practical advantage in sound reproduction. It is audible, if you play silent audio in 16 bit the noise floor is audible around -3dB MV. For 24 bit, it depends on the DAC and preamplifier stages, typically noise will be audible around +8 to +12dB MV. In more practical terms, 24 bit audio lets you hear silence a little louder. When playing music, the music itself will mask the noise floor, and even in the totally silent parts the noise from the recording will totally mask the noise floor of 16 bit audio.

Didn't think about that solution. Anyway, I solved it by converting all my music to 8-bit, now I can still use my vintage class-A amplifier.

Part from the dips in the lower midrange, I think the chair is the most limiting factor of this room right now, it simply is not very comfortable.. Saw some of the avshowroom youtube-casts yesterday, if for nothing else it is always a good laugh. Ignorance, religion and incompetence rules in this business, making it hard to get in the market for any serious - or, is it really an opportunity, depends how you look at it. Like this statement from a manufacturer rep: "Oh, if you do the math on 24-bit audio you see why we need that powerful amplifier.." I can see at least 3 flaws in this statement. Can you spot any of them?

I did not calculate - on the 100hz, only a very rough estimate. Only a simulation/measurement graph can give a true picture of the situation, but a rough estimate is all we need - we understand it is not happening around 2k, it is closer to 100hz, or 200hz. Interesting to see you are thinking about what happens to the sound field when there are objects around. Just thinking out loud here, but can it be that a directional sound field is more robust for such disturbances. This is a problem when doing measurements in-room. Do you leave the listening chair, should the chair/seating be damped. This has huge impact on the measured response. In the Room 2 I just remove the chair, because it is very easy to do. In the Moderate Cinema I usually place some damping in the sofa, and adjusting the size, material and shape of this damping makes it possible to achieve a very wide variety of measured responses. Just mentioning, in case someone really believes a +-1dB unsmoothed response is a reasonable target..

It is a great idea. But it does not need reflective floor-ceiling to act like a that, it only needs to be large compared to the wavelength. For bass frequencies it is also a good idea, and now the floor-ceiling will have an effect, and any absorption will not destroy this behavior because the absorption does not work that low. But since the floor-ceiling problem is eliminated, it may not be necessary to do anything with floor or ceiling. In a practical implementation things are a little more complex, but it works well. I built some magnestat panel speakers late in the 80's with true ribbon panels, the HF section was a very long, narrow ribbon, delivering a dipole cylindrical radiation pattern. Those speakers presented a very realistic soundstage, and only now do I have speakers that starts to bring back some of this sound character.

I would still consider absorption in the ceiling, the line array uses reflection from ceiling only at very low frequencies, as the array itself is directional at higher frequencies, and the absorption does not work at lower frequencies. With normal room height and a large line array the ceiling will start to have effect at around below 100hz. So, why bother? Opportunity to control and tune later reflections coming from the ceiling.

Just to share some thoughts around speakers and acoustics; Look at what happens to the phase, when early reflections are sufficiently removed - it starts to look very much like the anechoic response: The IR also shows that early reflection level is low: But the IR also reveals there are lots of later arriving reflections, and they are quite high in level. The interesting thing to observe here is that the later reflections has little effect on the frequency response and phase. Both steady-state and transient response is largely determined by early reflections. This means it is possible to achieve a smooth frequency response and good transient response even if the room/speaker combination is quite lively, with lots of room contribution. That is exactly what a large horn speaker does - the directivity reduces early reflection level, and the sound reflected from the back of the room creates lots of later reflections. This results in a big sound with great clarity. The small F1 speakers need acoustic absorption or huuuge distance to side walls to achieve this. As the measurements show, it certainly can be done, with the small F1 speakers in this small room, with sufficient absorption. So, does this sound very dead - the IR drops to below -30dB before 0.5ms. No, the soundstage is larger, and perception of room information from the recording is better.

The difference is in the radiation pattern of the speakers, the F1 horn has better pattern and 2 bass drivers makes it possible to achieve a better pattern in the midrange. I will fix the remaining issues, but I am not sure how to proceed right now. Next up is a new small bass system, to be tested in this room.

@SME, thank you for the 'nice', was going to comment on that, when I realize I don't know if you mean the visuals or the technical performance.. 3D is quite good, with much better placement and separation in depth compared to The Moderate Cinema. Vocal is better, But of course the resulting sound depends on the speakers. I have never heard a traditional hifi-speaker that can give a realistic presentation like when instruments appear like physical objects, regardless of room acoustics, I believe it is necessary to have horns or larger planar panels to get that. Those F1 speakers are too small to have good directivity control at lower frequencies, but still they present sound in a very physical and realistic way. In this small room it works quite well. The tonal balance is too bright, and it does not sound as relaxed and pleasant as the other room. The cancellations and reflections in the low midrange is a problem that needs to be solved, you can see it on the frequency response and the decay - there is too little energy in the 200hz - 600hz range.

That is the left sidewall absorber. It is much larger than the first version, now covering the lower part of the window and 20cm thick. Experiments with two 120cm x 60cm absorbers on the left wall showed significant improvements, so I had to change it. The room is quite narrow in width, and that is a problem.

This is how it looks now: Measures like this (20ms interval, 20ms pre-window): Velocity measurements reveals the directivity is very good, better than the larger F2 speakers from previous measurements. The sound can be described as addictive.

Here http://data-bass.ipbhost.com/index.php?/topic/538-bulding-the-room2-listening-room/page-4#entry13919 is a picture of the back, there is a large opening and not practical to do much more. The opening obviously does not reflect any sound. There are 4 ceiling absorbers, you can see the 2 back units in the picture. The floor reflection is far from completely absorbed, the small boxes only changes things a little to the better in the midrange, the F1 benefits from having something on the floor in front of them.

Testing new F1-p speakers in Room2: The room got a new left sidewall absorber, the small one below the window is not good enough. I am now using the room to work on new speakers, and it is easier and better to work in compared to untreated. Both measurements and listening is more consistent and easier, and it also sounds much better. The F1-p presents a sound different from the unfinished F2. I have some measurements I may want to share later, initially it looks like they measure about the same, but there are differences, but they are not very obvious.

Exactly. But a simulation - of the linear system model - is still useful, as it will give you the velocity. Then you can use the information on the page @SME linked to, to estimate performance.

Of course it can be modeled, but it is difficult to make a model accurate enough to give useful results. The behavior is nonlinear, mainly 2 parts are important - viscous losses along the walls, and turbulent losses where flow area changes, as it does at the exit. The viscous part is not very difficult, the turbulent part is very difficult. Then you need to find some software that can simulate the nonlinear model, and if this model also includes the rest of the speaker system, this will be quite complex. You may write your own simulator software to do it. Akabak can not simulate nonlinear systems. But it can give you very useful information about the flow - speed, pressure. Then you can estimate what happens, you know the size and shape of the port, and with some experience you will learn when flow speed vs port shape and size becomes a problem. Flared exits behave much better, larger port area tolerates higher velocities. Port compression is a significant problem at very low frequencies, and one of the reasons why even a small horn outperforms a traditional ported box.

But some hifi-nuts do have capacity. They are the ones building their own systems, like most readers here do. The only difference is that there are only 2 channels, and often no need for extreme sub-20hz capacity, because the music they play never contain ulf, part from unintended noise. I want full frequency range for music as well, it makes a huge difference on lots of electronic music. The audio world is becoming divided - those who still believe the reviewers and high-end suppliers, and those who take on the more scientific approach and end up building things themselves, because the performance they want is not available in any ordinary shop. The believers think that 120dB+ bass is never needed, in fact not desired at all. They have never experienced a good full-capacity, full-range system. A good system also sounds much better at very low volume, because the response is smooth without huge dips or peaks. But that is not what we were discussing here, this was about what capacity is actually needed when you turn up, all the way up. And for that, the limits are when it gets so loud it is not pleasant anymore. I have experienced that some people can get problems with the pressure at low frequencies, at spl below 130dB, guessing peak values around 125-128dB.

I have just now learned that all posters in this thread are wrong, 140dB is not needed, 130dB is not needed, even 123dB is far off the scale. Around 100dB seems to be right, and movies are to be played at -15 or occasionally -10dB if you want it very loud. I learned this from reading on the local hifi-forum. <insert double-facepalm picture here>

You are right about the hf transients - peak level can be high enough to be comparable in level with the full frequency range signal. I analyzed this once, see the attached files. Comparing full range to 6K high-pass filtered. While rms shows 18.6dB difference, the peak difference is only around 6dB. This means it is necessary to dimension amplifiers for the hf section so that peak levels comparable to the lower freq sections match up. Using the power spectrum to dimension amplifiers is wrong. Full fr spectrum: 6K filtered: rms -18.6dB: But as we can see from waveforms, peak level is only around -6dB: