Reference level

[MDS]

If you calibrate your system to reference level the max peak level can vary considerably (ie M&C...WOTW etc) from DD to DTS soundtracks. Are you also going to dispute that as well ? I know first hand how much louder many of these soundtracks are......even after calibrating to the correct levels. It's not rocket science, anyone can verify this for themselves.

The SPL you hear can be greater than 115 dB because the average level of the soundtrack is greater than the average level of the test tones you used to calibrate the system.

The test tones are bandwidth limited pink noise - not quite the same as real music or other audio. You calibrate to approximate reference level. It is never going to be exactly 115 dB at all times and even your room can emphasize certain frequencies, increasing the value your SPL meter displays. But it is not because combining bass from the other channels with the bass from the LFE channel increases the SPL.

 

[WaynePflughaupt]

The math tells me I should get a 6 db gain when adding two equal signals (coherent) You might get only 3 db with random frequency random phase pink noise, but I have seen a real world, measured at my seat 6 db gain when I measure sine sweeps on 2 channels verses 1.

The test tones are bandwidth limited pink noise - not quite the same as real music or other audio. You calibrate to approximate reference level.

Sounds like a co-located subs experiment. The extra 3 dB would come from boundary gain. Take the subs outside and you’d only be getting 3 dB with the second one. Pink noise is the better test signal for this, not sine waves. Pink noise, being a random signal, is actually closer to music or movie sound tracks than sine waves.

Regards,
Wayne A. Pflughaupt

 

[mtrycrafts]

No, his math is correct: If you combine three acoustical sources generating 105 dB, the total SPL is 109.7 dB. Check it yourself and see: Generate a pink noise signal from one of your speakers and take an SPL measurement. Then add a second and you’ll see a ~3 dB increase in the reading.

....Regards,
Wayne A. Pflughaupt

But you would get the same 3 dB spl increase if you just double the amp power into that one driver, right? What happens if you also increase the radiating surface with a 2nd driver and an amp? Why would that not also increase by a 3dB regardless of boundary stuff?

 

[mtrycrafts]

...and the only time I've been able to achieve those kind of numbers are when the family is out of the house.

And, maybe even in another county Wives do have super hearing

 

[skers_54]

That assumes RANDOM PHASE ! I'm not talking about random phase. If phase and frequency are coherent it is 6 dB's -- not 3 dB's.



If you calibrate your system to reference level the max peak level can vary considerably (ie M&C...WOTW etc) from DD to DTS soundtracks. Are you also going to dispute that as well ? I know first hand how much louder many of these soundtracks are......even after calibrating to the correct levels. It's not rocket science, anyone can verify this for themselves.



But each of those main channels have considerable low bass recorded as well as the LFE. The requirements if redirected would tax the subwoofer more than simply the LFE on its own. Unless in your world LFE+redirected bass is the same as just reproducing LFE.

If the acoustic sources are not co-located in space with identical dispersion patterns, phase at the listening position is random. Every reflection off every surface in the room from every speaker contributes and a significant portion are not in phase. That's why you don't get an ideal 6 dB increase with non-colocated acoustic sources. Sure the direct sound is in phase but the reflections that contribute the majority of the sound are not. Again, just because the electrical signal is in phase doesn't mean that's true of the acoustic signal. Different distances, different dispersion patterns, different proximity to boundaries and inherent room asymmetries all factor in reducing the ideal gain.

DD and DTS are different standards using potentially different mixes and different encoding schemes. You have no idea if the tracks were mixed to the same reference or the exact amount of headroom each consumed. Might as well be comparing Bambi to Terminator.

The issue isn't whether bass management makes it harder for the sub (it does). It's whether bass management allows real program material to exceed the headroom available in the Dolby Digital format. To confirm that, you must have the exact amplitude values present on the disc. For example, Pulse could have been recorded with 110/100 dB peaks on the subs and mains but have been designed to be played back 8 dB above reference. This would yield output above 115 dB when played back as intended but not at true reference.

 

[skers_54]

But you would get the same 3 dB spl increase if you just double the amp power into that one driver, right? What happens if you also increase the radiating surface with a 2nd driver and an amp? Why would that not also increase by a 3dB regardless of boundary stuff?

You get a 3 dB increase either way because not all the acoustic energy reaches the listening position in phase and coincidently. If you stack 2 subs, everything is doubled and the energy propogates in the same way, yielding a 6 dB increase.

 

[Vaughan Odendaa]

I never said any of this would be practical. But that it's possible. Not that I would listen at reference level :

If the acoustic sources are not co-located in space with identical dispersion patterns, phase at the listening position is random.

The electrical signals will net a 6 dB increase (if in phase). Pans Law. The acoustical signals can net a 6 dB increase too. It doesn't necessarily matter where the speakers are in this case because we are talking about signals well below the x-over point. If all the information is summed together with the LFE there is no reason to assume it wouldn't be in phase.

Every reflection off every surface in the room from every speaker contributes and a significant portion are not in phase. That's why you don't get an ideal 6 dB increase with non-colocated acoustic sources.

Let's assume the summed redirected bass is in phase. Distributing that signal over multiple subs, you should have less overall cancellation. But the signal requirements....which is the crux of the matter for me is different from sub handling LFE vs handling LFE and redirected bass. There can be no free lunch.

FYI, I believe one can still achieve a full co-located 6 dB output at low frequencies if the distance of both sources are within 1/4 wavelength of the frequency you are trying to reproduce but that's another subject for debate.

Sure the direct sound is in phase but the reflections that contribute the majority of the sound are not.

You may even have higher overall sound if those reflections combine constructively. But let's assume the room acoustics are close to ideal in a home theater set up. I mean, reference level isn't practical 98% of the time because 98% of systems simply can't handle reference level bass output at low distortion, but reference level is reference level...

Again, just because the electrical signal is in phase doesn't mean that's true of the acoustic signal. Different distances, different dispersion patterns, different proximity to boundaries and inherent room asymmetries all factor in reducing the ideal gain.

If summed together and sent to a capable enough sub, there shouldn't be cancellation (assuming the signals are in phase at 0 dB's) especially if distributing over multiple subs, which would be recommended. Again, let us assume room acoustics are close to ideal. Let's assume close to ideal circumstances since it's a given that most won't reproduce reference level cleanly....it's simply not practical in most cases.

 

[Vaughan Odendaa]

I think it's safe to say that my instructor was incorrect on both counts. First, a sub that handles LFE only vs one that handles LFE + redirected bass will inevitable place a much higher burden on that sub.

Secondly, he wasn't aware that deep bass existed in the main channels. A big one. When asked, he didn't know. So......

....if I am designing a theater and I want to achieve maximum dynamic impact given what I know, I wouldn't want a sub that can hit 115 dB's. I would target 125 dB's and cater for at least 2-3 dB's of clean headroom. Response would need to go down to 10 Hz to cover most of the infrasonic content. This would require 4 potent subwoofers (Submersive HP) all positioned in each wall corner. Main speakers would use BM. Acoustics would be close to ideal in a home environment, so close to flat response.

 

[mtrycrafts]

.,... If you stack 2 subs, everything is doubled and the energy propogates in the same way, yielding a 6 dB increase.

And, that is what I was driving at

 

[digicidal]

Might as well be comparing Bambi to Terminator.

I would assume that Bambi would put harder demands on my subwoofer simply because it was mixed by Disney sound engineers and Thumper's feet are really big and flat.

I guess the bottom line is that in order to feel totally 'secure' in being able to deliver every signal to the absolute maximums possible in a HT environment, and not being able to philosophically or mathematically determine what boundry enforcement or cancellation might come into play until you have a REAL room to measure and work with... then YES - get a sub capable of hitting 125db @ 15-20Hz or better yet 4-6 of them.

To my thinking if you're looking at this more academically than practically (which seems to be the case) then I would simply state this in cases where you are redirecting bass signals from the individual channels: If you want to assure adequate levels and presentation - simply make sure that you have n+1 subwoofers capable of ~115db peaks (where n is the number of channels being played). That would get ridiculous pretty fast - but should provide an easy solution since you'd stop having to worry about summing bass information from the speakers - because each one would have a sub dedicated to it's own bass info - and there'd be an extra for the pure LFE information. Of course you now have a 1000sqft theater that can only seat 6 people max because there are literally speakers all over the freakin' place... but that seems to be what some people like.



On the other hand... you could simply save yourself a huge headache and just put in true full-range speakers for each of the channels and run everything LARGE instead - which is essentially the same thing as getting each channel a dedicated sub - with the exception of the amp required.

 

[Vaughan Odendaa]

To my thinking if you're looking at this more academically than practically (which seems to be the case) then I would simply state this in cases where you are redirecting bass signals from the individual channels: If you want to assure adequate levels and presentation - simply make sure that you have n+1 subwoofers capable of ~115db peaks (where n is the number of channels being played).

Yes, that sounds logical. Makes sense. I just wanted to clarify what the sub would ultimately be required to handle with LFE+redirected bass. The solution isn't really relevant to this thread, just the understanding of it.

The total peak output using BM or not should remain the same. But my lecturer said that whether the sub handled LFE or LFE + redirected bass it shouldn't make a difference. What he NEGLECTED to mention was the added strain of redirected bass from every other channel together with the LFE requirement and he didn't even realize that deep bass existed in the main channels ! His course revolves around BM. It's basically a requirement. Well let me tell you something, it sure would make a difference to the sub. In that case he would be dead wrong.

On the other hand... you could simply save yourself a huge headache and just put in true full-range speakers for each of the channels and run everything LARGE instead - which is essentially the same thing as getting each channel a dedicated sub - with the exception of the amp required

Yeah you could do that. Although I don't think your bass response would be great. Usually positioning for smoothest bass response isn't the same for imaging.

 

[digicidal]

Yeah you could do that. Although I don't think your bass response would be great. Usually positioning for smoothest bass response isn't the same for imaging.

True, however if ALL of your discreet channels were full range speakers, you probably wouldn't have any problem with dispersion (however, you are correct in that you probably wouldn't have smooth bass response - considering that you might have more cancellation problems, etc..).

I would agree with you (again academically) that you would need to be able to achieve levels of >120-125db at the sub if you redirected all bass from 5-7 discreet channels to the sub. However, I would add this to further the argument from the practical side of things:

Although redirecting the bass could potentially create that large a demand on the sub... I would warrant that in 99.9% of all cases it would never require that. Simply because of the way recordings are mixed (especially in movies) you would never see a full peak intended to be directed to all of the speakers as well as the LFE channel simultaneously. Even with an explosion, or thunderstorm, etc... the whole point of mixing the surround track is to create the spacial illusion in the viewing space... which requires delays on the surround channels from the mains, etc.

Add to that the fact that because of the way the Dolby/DTS/etc.. specs and documentation read - it's reasonable to assume that no sound engineer is creating a recording that even commercial theatres will not be equipped for, and certainly no home theaters will be.

 

[Vaughan Odendaa]

Then we are in agreement. I agree that most material will never require 123-125 dB output from the sub+redirected bass (assuming a flat calibration ). But simply that it CAN happen. Some movies have bass mixed in that is coherent, others incoherent and/or a mix of coherent and incoherent.

So while my nightmare scenario (0 dB's in all channels including LFE) is certainly improbable, it is not impossible. I think in many cases, films with lots of deep bass content (I posted in another thread here by linking a thread to AVS where members posted charts of deep bass content recorded in the LFE + other individual channels) would still require at least 2-4 dB's over the 115 dB spec if one is redirecting bass from all other channels. So personally I would bank 118-119 dB's max in many of the bass intensive films out there. Dialnorm will affect the total SPL too....not all films are mixed the same, some louder than others, even when calibrated correctly. Watch Star Wars AOTC. The bass is ridiculous.

For films with lots of deep bass content (under 30 Hz) look at the thread I linked to (in one of the bass management threads here in the 'general AV' area). The Hulk has ridiculous low bass in the main channels. It's really startling how deep this bass is. I feel sorry for those who bump up the bass more than usual. Imagine those who increase their sub levels by 4-6 dB's.....and then asking their subs to handle the peaks in these movies at reference (redirected bass + LFE).

I myself don't listen at reference because neither my system nor my room acoustics make it easy (wayyyy too fatiguing not to mention my system is dynamically incapable so it's a mute point for me). Some of my clients can achieve this, many can't and many don't require it. This Cedia course made the assumption that we had to deliver what the director intended and one of those requirements is achieving the *reference level*.

 

[Vaughan Odendaa]

it's reasonable to assume that no sound engineer is creating a recording that even commercial theatres will not be equipped for, and certainly no home theaters will be.

Commercial cinemas aren't capable of producing very deep bass (25 Hz and below) generally speaking. The size of the space makes it simply too difficult to pressurize at those frequencies. The output required in such a space would be enormous.

Black Hawk Down has bass that goes right down to 7 Hz if I'm not mistaken AND it is audible (it can be perceived with enough raw displacement at LF's). I CANNOT believe that a cheap PC with the right software can pick up this content but a professional mixing engineer with his sophisticated hardware can't. Don't buy it AT ALL. I'm 100% sure that they can monitor this content with little trouble. If the average Joe can see it.....then...

Apparently they used something like 10-12 Bagend subwoofers to handle the low end for that soundtrack. One thing is for sure, I certainly won't be reproducing 7 Hz anytime soon.

 

[jostenmeat]

Apparently they used something like 10-12 Bagend subwoofers to handle the low end for that soundtrack. One thing is for sure, I certainly won't be reproducing 7 Hz anytime soon.

About 2.5 years ago, FilmMixer was talking about where he worked, and showed some photos. At the time, they were using 18 Bagend subs, all with 18" drivers, if I am not mistaken. See all of those white marks below the screen? They are all subs. Did Todd AO do the sound on BHD?

 

[InTheIndustry]

About 2.5 years ago, FilmMixer was talking about where he worked, and showed some photos. At the time, they were using 18 Bagend subs, all with 18" drivers, if I am not mistaken. See all of those white marks below the screen? They are all subs. Did Todd AO do the sound on BHD?

Yes, I believe that's the film studio that did the movie.

 

[Vaughan Odendaa]

Nice studio.

A single Bagend sub on it's own is nothing special. The linear excursion on those 18 inch woofers is not that great, as I understand it. But in large numbers, you have a lot of displacement, especially 18 of them. I think Keith Yates did a sub comparison using the Bagend and measured something like 80 dB's at 7 Hz or something to that effect (would have to check the article to be certain)....which is useless...for a single sub.

 

[digicidal]

I guess my question from all of this (which does in many ways apply directly to the OP although not entirely) is: would any sound engineer ever create a need for anything over the listed specs for a given audio track? Perhaps someone with direct or 2nd hand experience could elaborate.

What I mean by this is say there is an explosion for which the desired effect involves an absolute peak output of 112db at 40Hz (not saying this is realistic or accurate... but for simplicity sake). Would it not be always the case for that engineer to direct all of the 40Hz output to the LFE channel? Or are there cases where instead he would wish for a 106db output from the LFE channel and 40Hz at say 97db from each of 5 channels (or whatever level would result in an audible level of 112db at the listening position).

Since bass is non-positional - wouldn't it never actually matter as far as the actual impact in the listening position (apart from perhaps some better distribution of the sound)? Now as far as sound higher than 80Hz I would understand completely, but isn't bass just bass for all intents and purposes, and therefore the listed specs from Dolby and DTS simply be summarized by saying "a maximum bass peak output of 115db" PERIOD.