Sign in to follow this  
Followers 0
Kvalsvoll

Article: How to set up a 2-channel Sound System

40 posts in this topic

I have a bunch of uncompleted articles, and now I decided to finish this one.

 

There is content both for the expert and the novice.

 

There is a lot about sound that has no simple or absolutely correct answer. The part on Target response curves exemplifies that, when you start to look in to it, you realize there is no one answer that fits all situations.

 

There are things I would like to improve.

 

I would like to show the simple and easy solutions in text boxes throughout the other content, to make it easier to get the big lines.

 

The content could be better organized, to be easier to read and follow. And I should go through everything to verify and update the content to ensure it is in compliance with how my current understanding of sound and acoustics is.

 

 

Of special relevance here at data-bass:

 

- There is some information about deep bass content in music

- Makes an attempt to explain why full-range reproduction improves sound quality

- Covers the basics for how to add a decent bass-system to your system

- Lots of nice pictures can never be wrong

 

 

----------------------

 

Can reproduced music in your home actually sound better than live?

Based on the "How to set up a home-theater sound system" article, the basics for setting up a high quality 2-channel sound system for music is covered.

What equipment is needed, where to place speakers, and how to get the best sound.

And busting some of the audio myths that often prevents getting the best performance.

While some of the content may look very technical for some readers - with graphs and hertz and decibels - you can skip those parts, and still get a good introduction on how the experts work with their systems and learn what is important for sound.

http://www.kvalsvoll.com/Articles/Howtosetupa2chSoundSystem.htm
 

post-181-0-32931000-1485709993_thumb.jpg

Share this post


Link to post
Share on other sites

Very good article. It would make for a good guide for anyone trying to get into hifi sound. One part I disagreed with was the advice to simply leave the subs against the front wall and correct the bass response with DSP signal shaping. As you know, equalization is not a good solution to tackling cancellation dips. I would say moving the subs to placements that make for the best response overall at all listening positions is a better solution. I think the Geddes approach is the most sensible here- multiple subs asymmetrically placed.

 

This guide seems to be geared for powerful speakers and high SPL listening. Not that there is anything wrong with that, but I think SPL capability is not the problem with speakers for many folks, since a lot of people don't listen at high volume levels. I think most of the conventional bookshelf speakers and towers can get loud enough for most people, but the problem is more often frequency response and especially off-axis response. However, if one is going through the trouble to set aside a whole room for listening, might as well have speakers that are not inhibited by distortion at even very high levels. 

1 person likes this

Share this post


Link to post
Share on other sites

I disagree with your comments about SPL capability, to a point.  Obviously, many people are limited by what they can accommodate in their space, which may be an apartment with shared walls and cranky neighbors.

 

However, if the goal is maximum sound quality, then there is no replacement for SPL capability, if for realistic dynamics if nothing else.  Life-life reproduction of even mundane sound demands substantial SPL capability.  Have you ever used a microphone to measure the peak level of real-life sounds?  I remember when I clapped my hands about 12" away from my UMIK-1 and actually caused it to clip.

 

Often times, loudness and SPL capability are opposed concepts.  I've heard sound from an Apple laptop that was loud enough to hurt my ears.  But the speakers I own now sound so good that they can be totally cranked and not even sound loud.  When I upgraded my front-stage speakers from Hsu Research HC-1 (a speaker with a reasonably high 92 dB/2.83V sensitivity as far as consumer speakers go) to my DIY SEOS-15/DNA-360/TD12M, I assumed that the increase in headroom would mostly go unnoticed, at least with most music and movies.  I was wrong.  I quickly realized that there's no shortage of content with strong dynamics that fall on the mains to reproduce.  And of course, this forum has revealed that many movies make even greater demands on the sub system as well.

 

And yeah, of course people may choose to listen at lower levels.  But a good sounding speaker inspires listeners to turn it up.  I bet most people will do it without thinking about it.  Most are probably accustomed to turning things up until distortion becomes objectionable, i.e. too loud.

1 person likes this

Share this post


Link to post
Share on other sites

Thanks for the comments, I find them useful.

 

I have now read though my own article, and I think it turned out quite well. I fixed a couple spell and grammar errors. The content is sourced partially from the "How to.." article and several other non-completed writings. Some content repeats itself, and it may not be easy to get the big lines, it easily gets too complicated for people who are not deep into sound and audio.

 

I may want to re-write the first chapter, to open the door for the more religious people. The most extreme will never buy in to this message of the possibility for transparent sounding equipment, but I think there are many audio enthusiasts that would be open for new perspectives, especially when they experience how things actually work.

Share this post


Link to post
Share on other sites

Article is updated.

 

Changed some of the text in the first chapter, and removed the htpc, no one builds a htpc to play music.

 

May take a while before any "more" is coming, but I have some papers I have started working on.

 

I could elaborate on the SPL and subwoofers comments, but really, there is not much to add, what you said is right. Some shortcuts has to be made to be able to present reasonably simple solutions, even if more advanced approaches can give better results in some cases.

Share this post


Link to post
Share on other sites

I look forward to your advanced treatises, especially ones involving optimization of the sound field (SPL and PVL).  Impactful transients are so hard to do in smaller rooms, even if you have tremendous SPL capability.

 

JSS 

Share this post


Link to post
Share on other sites

I look forward to your advanced treatises, especially ones involving optimization of the sound field (SPL and PVL).  Impactful transients are so hard to do in smaller rooms, even if you have tremendous SPL capability.

 

JSS 

 

I'll just state my hunch that bass transient capability primarily has to do with whether the system is calibrated correctly.  That is, calibrated with first arrival flat except for roll-off at the very top for distance compensation and a lift at the bottom to account for first arrival boundary gain.

 

At some point, I'll start a thread about the need for bass lift for a truly correct sound.  It's entirely consistent with Josh Ricci's observation that more bass is needed outdoors.  It all goes back to the fact that the speakers used for monitoring during production are monitors that measure flat in anechoic conditions.

Share this post


Link to post
Share on other sites

However, if the goal is maximum sound quality, then there is no replacement for SPL capability, if for realistic dynamics if nothing else.  Life-life reproduction of even mundane sound demands substantial SPL capability.  Have you ever used a microphone to measure the peak level of real-life sounds?  I remember when I clapped my hands about 12" away from my UMIK-1 and actually caused it to clip.

 

+1

 

I often see people say "I'm not worried about loud SPL I want it to be low distortion and accurate and fast." or something like that. They don't realize there are a number of ways our speakers can fail to reproduce the signal well.

 

1.) Harmonic distortion

2.) Non harmonic distortions, noise, rattles, buzzing, etc.

3.) Smearing of the time signature of the signal

4.) Frequency response deviations

5.) Compression of or failure to reproduce dynamic peaks

 

Etc...It's a laundry list. The relative importance of these could be debated and be different for each person. I'd probably put #2 and #5 as most degrading to the sound for me. The point is that the harder your system is being pushed the worse it will perform in ALL of the above. A much more powerful system with greater efficiency/ power/headroom/displacement etc...Will perform much better than one capable of less SPL that is being pushed way harder.

Share this post


Link to post
Share on other sites

  It all goes back to the fact that the speakers used for monitoring during production are monitors that measure flat in anechoic conditions.

 

The monitors are designed to be flat on axis for the most part but once placed into the room they are often anything but. I'm sure there are some very high end professional places that are truly flat, but most of what I've seen of the mid range professional mixing places is very likely not.

 

If you have 1 guy who mixes on speakers with a bass response that is lifted substantially and a treble that is rolled off and the guy has a twin who mixes the same tracks on speakers with a flat or rolled off bass response and flat or slightly elevated treble the 2 mixes will end up weighted much differently. Guy 1 with the boosted bass and shelved treble speakers will try to compensate to make the mix sound good to him so his mix will end up with LESS bass and MORE treble relative to the twin who will compensate the other way. The mix will end up weighted the opposite way that their speakers respond. Think of it this way if you hear a recording that is drastically hot in the bass the mix was probably done on speakers with less bass weight than usual. If the midrange is sucked out and dark sounding the mix speakers may have been a bit forward in the mids.

Share this post


Link to post
Share on other sites

...  It all goes back to the fact that the speakers used for monitoring during production are monitors that measure flat in anechoic conditions.

 

Exactly. THIS IS IT.

Share this post


Link to post
Share on other sites

+1

 

I often see people say "I'm not worried about loud SPL I want it to be low distortion and accurate and fast." or something like that. They don't realize there are a number of ways our speakers can fail to reproduce the signal well.

 

1.) Harmonic distortion

2.) Non harmonic distortions, noise, rattles, buzzing, etc.

3.) Smearing of the time signature of the signal

4.) Frequency response deviations

5.) Compression of or failure to reproduce dynamic peaks

 

Etc...It's a laundry list. The relative importance of these could be debated and be different for each person. I'd probably put #2 and #5 as most degrading to the sound for me. The point is that the harder your system is being pushed the worse it will perform in ALL of the above. A much more powerful system with greater efficiency/ power/headroom/displacement etc...Will perform much better than one capable of less SPL that is being pushed way harder.

Maybe I will need to build those floor-ceiling arrays some day....

 

JSS

Share this post


Link to post
Share on other sites

The monitors are designed to be flat on axis for the most part but once placed into the room they are often anything but. I'm sure there are some very high end professional places that are truly flat, but most of what I've seen of the mid range professional mixing places is very likely not.

 

If you have 1 guy who mixes on speakers with a bass response that is lifted substantially and a treble that is rolled off and the guy has a twin who mixes the same tracks on speakers with a flat or rolled off bass response and flat or slightly elevated treble the 2 mixes will end up weighted much differently. Guy 1 with the boosted bass and shelved treble speakers will try to compensate to make the mix sound good to him so his mix will end up with LESS bass and MORE treble relative to the twin who will compensate the other way. The mix will end up weighted the opposite way that their speakers respond. Think of it this way if you hear a recording that is drastically hot in the bass the mix was probably done on speakers with less bass weight than usual. If the midrange is sucked out and dark sounding the mix speakers may have been a bit forward in the mids.

 

Recent productions tend to have much more bass than a few decades ago, a typical 80ies recording sound like the bass is missing if you compare it directly to a new recording.

 

Some if these differences can be due to preference, each period has its own characteristic sound, like the LOUD that now hopefully fades away.

 

But some differences are likely due to monitoring systems. Back in the earlier days, there was no dsp and room correction, the speakers were placed in the studio, some acoustic treatment, and that's it - if it didn't measure flat, there was little to do. And bass-systems, subwoofers, bass-management was not in use. I also suspect that acoustic treatment in earlier days were mostly high frequency absorption, which further tilts the tonal balance to more bass heavy and less hf.

 

Today the monitor systems are measured, and if the response is not within acceptable limits, it can be adjusted. This happens, I have been to studios, I know how they do this. Unfortunately, since there is no standard, the reference response curve may be flat, or it can be tilted, or it can be flat with a bass lift, really, who knows..

 

This is one thing that the movie producers are doing better - they have a standard, and while you can argue that the standard is "wrong", at least all monitoring is done on systems with far more similar response. Yes, that SMPTE report claims they don't follow the standard, but actually what this survey showed is that they do follow the standard, what it also showed is that eq-ing systems in different rooms with different room acoustics and speakers to the same frequency response will still not give similar sound.

Share this post


Link to post
Share on other sites

+1

 

I often see people say "I'm not worried about loud SPL I want it to be low distortion and accurate and fast." or something like that. They don't realize there are a number of ways our speakers can fail to reproduce the signal well.

 

1.) Harmonic distortion

2.) Non harmonic distortions, noise, rattles, buzzing, etc.

3.) Smearing of the time signature of the signal

4.) Frequency response deviations

5.) Compression of or failure to reproduce dynamic peaks

 

Etc...It's a laundry list. The relative importance of these could be debated and be different for each person. I'd probably put #2 and #5 as most degrading to the sound for me. The point is that the harder your system is being pushed the worse it will perform in ALL of the above. A much more powerful system with greater efficiency/ power/headroom/displacement etc...Will perform much better than one capable of less SPL that is being pushed way harder.

 

#5.

 

This is some of the things I try to communicate in my article. Having decent capacity makes a THE difference. Music at 90dB is loud, but not more than most people will find reasonable when sitting down just to listen for music, and there is no risc of disturbing others. This spl then requires capacity of at least 110dB - at the listening position - and that is more than the typical hi-fi speaker system can deliver before compressing and distorting.

 

One of the visitors in The Moderate Cinema told me that one of the reasons he wanted to hear this was to verify if it was possible to listen to movies at 0dB. His experience had been that this was too loud to be pleasant, and wanted to find out if different speakers and more controlled acoustics made a difference. According to what he says after experiencing movies at 0dB, it sure did make a difference.

1 person likes this

Share this post


Link to post
Share on other sites

+1

 

I often see people say "I'm not worried about loud SPL I want it to be low distortion and accurate and fast." or something like that. They don't realize there are a number of ways our speakers can fail to reproduce the signal well.

 

1.) Harmonic distortion

2.) Non harmonic distortions, noise, rattles, buzzing, etc.

3.) Smearing of the time signature of the signal

4.) Frequency response deviations

5.) Compression of or failure to reproduce dynamic peaks

 

Etc...It's a laundry list. The relative importance of these could be debated and be different for each person. I'd probably put #2 and #5 as most degrading to the sound for me. The point is that the harder your system is being pushed the worse it will perform in ALL of the above. A much more powerful system with greater efficiency/ power/headroom/displacement etc...Will perform much better than one capable of less SPL that is being pushed way harder.

 

I will throw in here tonal balance and argue that it trumps all of the above as far as perceived loudness vs. SPL.  Indeed, the most offensive part about harmonics distortion may be its tendency to make the tonal balance more top heavy, and even a slightly top heavy tonal balance can be excruciating when high levels get involved.

Share this post


Link to post
Share on other sites

The monitors are designed to be flat on axis for the most part but once placed into the room they are often anything but. I'm sure there are some very high end professional places that are truly flat, but most of what I've seen of the mid range professional mixing places is very likely not.

 

If you have 1 guy who mixes on speakers with a bass response that is lifted substantially and a treble that is rolled off and the guy has a twin who mixes the same tracks on speakers with a flat or rolled off bass response and flat or slightly elevated treble the 2 mixes will end up weighted much differently. Guy 1 with the boosted bass and shelved treble speakers will try to compensate to make the mix sound good to him so his mix will end up with LESS bass and MORE treble relative to the twin who will compensate the other way. The mix will end up weighted the opposite way that their speakers respond. Think of it this way if you hear a recording that is drastically hot in the bass the mix was probably done on speakers with less bass weight than usual. If the midrange is sucked out and dark sounding the mix speakers may have been a bit forward in the mids.

 

The crucial question though is: How much does the room actually alter the subjective tonal balance of the sound?  My answer is: not very much, especially for mids and highs.  Most of the difference is heard in the bass for reasons I argue in my "Time to kill the myth" thread.

 

Recent productions tend to have much more bass than a few decades ago, a typical 80ies recording sound like the bass is missing if you compare it directly to a new recording.

 

Some if these differences can be due to preference, each period has its own characteristic sound, like the LOUD that now hopefully fades away.

 

But some differences are likely due to monitoring systems. Back in the earlier days, there was no dsp and room correction, the speakers were placed in the studio, some acoustic treatment, and that's it - if it didn't measure flat, there was little to do. And bass-systems, subwoofers, bass-management was not in use. I also suspect that acoustic treatment in earlier days were mostly high frequency absorption, which further tilts the tonal balance to more bass heavy and less hf.

 

Today the monitor systems are measured, and if the response is not within acceptable limits, it can be adjusted. This happens, I have been to studios, I know how they do this. Unfortunately, since there is no standard, the reference response curve may be flat, or it can be tilted, or it can be flat with a bass lift, really, who knows..

 

This is one thing that the movie producers are doing better - they have a standard, and while you can argue that the standard is "wrong", at least all monitoring is done on systems with far more similar response. Yes, that SMPTE report claims they don't follow the standard, but actually what this survey showed is that they do follow the standard, what it also showed is that eq-ing systems in different rooms with different room acoustics and speakers to the same frequency response will still not give similar sound.

 

I think the amount of bass in recordings also has to do with headroom on commonly available media and even pure aesthetics or preference as you say.  This is addressed in the mix though, for example by mixing the kick drum or bass instrument hotter rather than EQing up only the subwoofer frequencies.  If you just EQ the subwoofer frequencies or turn the sub level up, you eventually throw off the balance and turn a nice rhythmic punchy sound into total bass mud.  By increasing the level of dynamic range of the bass instrument, it's possible to put a lot more bass in the recording without throwing off the balance.

 

On the subject of using EQ for monitoring systems, I'm under the impression that it isn't done as much as we'd think.  One common recommendation I see is to "EQ the speaker, not the room", and if the speaker is already known to be flat, then the best course is regarded to be to just leave it alone or make only small changes.  For example, some PEQs might be used to knock down room modes, but that's it.  Any room gain that shows up is left in place, because presumably, that's what a passive consumer speaker that's anechoically flat will also do.  I wouldn't be surprised if, in those instances where the overall response shape is adjusted using EQ, that mixers do it by ear rather than trying to fit to a particular target.  I notice this approach at rock concerts.  They throw on some familiar music (Bob Marley seems real popular being that certain recordings have great punchy bass) and tweak until it sounds right.  Then presumably, the bass will be good when the band plays.

 

Even with movie content, I'm thinking that EQ is rarely used to adjust the monitoring system anymore, even when dealing with A.T. screens which may exhibit significant high frequency absorption.  Arguably, this was the main problem the X curve was trying to compensate for (along with electronics differences in the old days of analog stuff), but the X curve just didn't translate reliably enough because it's based on analysis of a continuous test signal rather than the direct sound that emerges from the screen.  The consequence was excruciating brightness in some circumstances such as drier rooms and screens with better transparency.  Instead, mixers just accept the consequences of the screen and try not to boost the highs much if at all so that they don't come out too bright in another theater.  I believe as much was confirmed by Bob Vessa from Sony in this video on mastering for the home.

Share this post


Link to post
Share on other sites

#5.

 

This is some of the things I try to communicate in my article. Having decent capacity makes a THE difference. Music at 90dB is loud, but not more than most people will find reasonable when sitting down just to listen for music, and there is no risc of disturbing others. This spl then requires capacity of at least 110dB - at the listening position - and that is more than the typical hi-fi speaker system can deliver before compressing and distorting.

 

One of the visitors in The Moderate Cinema told me that one of the reasons he wanted to hear this was to verify if it was possible to listen to movies at 0dB. His experience had been that this was too loud to be pleasant, and wanted to find out if different speakers and more controlled acoustics made a difference. According to what he says after experiencing movies at 0dB, it sure did make a difference.

 

I have yet to hear a system at Reference Level with demo-type content that was not at least a bit uncomfortable.  I guess I need to hear a very overbuilt, lots of dynamic headroom system.  

 

JSS

Share this post


Link to post
Share on other sites

I have yet to hear a system at Reference Level with demo-type content that was not at least a bit uncomfortable.  I guess I need to hear a very overbuilt, lots of dynamic headroom system.  

 

JSS

Demo content tends to be very lively, of course. It may be that Reference level listening is simply too loud for for listening to that type of content without it being irritating to healthy human hearing. Even without any distortion. 

Share this post


Link to post
Share on other sites

I have yet to hear a system at Reference Level with demo-type content that was not at least a bit uncomfortable.  I guess I need to hear a very overbuilt, lots of dynamic headroom system.  

 

JSS

 

Several factors come to play here, one is mentioned above - demo content.

 

If the content comes "pre-distorted" with the distortion already in place, due to excessive clipping or hitting limiters, there is no hope, no golden speakers or acoustics or ears can fix that. Demo scenes are also often chosen among the loudest action scenes, making the contrast from playing nothing to full blast very violent.

 

Calibration - if one system is 2dB hot, and has very flat response with high level at higher frequencies, it will be louder, perhaps 4-5dB, and that makes a huge difference.

 

Acoustics - loud early reflections will make it sound unpleasant at loud volumes.

 

Perception of loudness is very relative - going from quite silent to loud instantly will make it appear even louder. Also, what one person find acceptable and pleasant, can be too loud for someone else.

 

Speakers with sufficient capacity is really the easiest to fix.

Share this post


Link to post
Share on other sites

I am going to tell you a little story, because it has relevance to the 2-ch article, and simply because I want to share this with you. And I think some of you will find it entertaining.

 

I was invited to visit a dealer, and today I took upon the opportunity to get a good audiophile talk and demoed several speakers and I imagine just as many cd-players, dacs, and amplifiers as the speakers were connected to different set-ups.

 

The dealer and the manufacturers of the speakers must forgive me for what will be a very honest kind of expression of my brief listening experiences.

 

I can not reveal what speakers this was, I develop and intend to manufacture speakers myself, and as so it would be totally inappropriate to give impressions of other designer's creations.

 

But I can tell a story.

 

I had prepared a usb-drive with all music content from the DemoScenes app, remembering the last time I visited this dealer, when it was impossible to find music I knew. 

 

We started listening to the egg-speakers - a lifestyle-oriented, expensive, high-profiled product, they are very small, I have seen 16Hz extension claimed, and there was supposedly 4KW amplification. They actually cost 10 times less than I imagined - a pair for approx USD 6K is not that bad, considering all amp + playback is included, and since it is "16Hz" no need for a bass system.

 

After 12 seconds into "September In Montreal" I was stunned. Because this had to be a bad joke. Another 20 seconds and I am certain the only good thing about this so far is that Anne Bisson has some kind of holographic appearance, and there is good clarity on the vocal. Everything else is a mess. But even the vocal lacks in presence and detail compared to a good setup.

 

Flashbulb - Island on an endless plane. Reveals holography, depth of image, detail and texture in hf, impact and transient response in the bass, as well as some low extension. All of this is lost.

 

That there is bass content can be seen on the speakers as the woofers are on the sides of the round-shaped cabinet, and they move - a lot. The bass can also be heard, as a rumbling, one-note mess. Overall the bass level is experienced as quite hot. It can not be felt like it should be, with the instant impact on the transients, and the deep punch. No tactile feel whatsover. Subjectively the bass goes lower on the F2 60Hz-tuned horn.

 

So, where is the "16Hz". And where did the 4KW go. Guess it is fair to say that the old slogan "size matters" still holds. But my small S6-14 horns are not much larger that this creation, and they can make pants flap playing Valley Of Kings, and they have impact and punch on the Flashbulb - Island on an endless plane.

 

I think two things are important here - one is that marketing claims and reality often separates at some point, the other is the fact that since the S6-14 is a dedicated subwoofer unit I have them placed where the bass response is best. Placement is what makes the big difference for low frequency capability here, because even a horn-loaded 6" can only do so much. Be sure - there is still no free lunch.

 

So, moving on to something different, a small swiss-made designer-piece, I really liked their design and physical appearance. Price USD7500 - and that is for a pair, good value for something so nicely made.

 

First I claimed they play out-of-phase, cables must have been accidentally reversed on one. Wrong! Easy to verify, that's how they sound, and once that is figured out, I am fine with that. Very little bass, but what is there is also free of the rumble, sound is detailed and very bright, very spaceous, absolutely no body or presence or 3D image of any instruments, everything is a spaceous sound-cloud. Why not. If you sit close - 1.5m - and on the floor - they give a spaceous and large sound, suitable for some late-night low volume listening. They can never play loud, there is no capacity. Not my first choice - obviously, but for someone with different needs and seeking a different sound, those speakers are certainly worth checking out. They have character - both in appearance and sound.

 

Heard some other speakers in between here, but nothing interesting, more ordinary sounding, not overly-expensive. More like the standard hi-fi sound, where nothing really excels and overall impression is boring.

 

Then over to some small satellite speakers, set up with no bass system. From a small, dedicated manufacturer in Germany. Apprx USD17500/pair.

 

This was the best sound of the evening by a good margin. Now you have true holographic presentation, drums and percussion suddenly materializes, even the bass is present without being a nuisance. There is detail, presence, clarity, and instruments with body. They can not play loud, but that is not a requirement for everyone. AMT direct-radiator hf-driver, too small direct-radiator mid, too small bass driver or may be only a passive radiator on the back. A great sounding, small speaker.

 

Last was a larger speaker from the same German manufacturer, this one has a Raal 70-20 direct-radiator hf-driver. There is just something about ribbons..

 

The relevance to the 2-ch article is that all those listening experiences matches what I write about speakers - size, capacity, bass systems, sound character. A small speaker will have limitations due to laws of physics, and if you try to break those rules, the result is lesser performance overall. There is no point in reproducing 16hz if the only place it can be perceived is by looking at the data sheet.

 

When I got home, trying to recover from this rather traumatic experience, I spent the rest of the evening listening in Room2. And the speakers and the room has never sounded so good ever. Lately I have become more and more aware of some of the shortcomings of the setup, but now I could just play and enjoy the music, and if I focus on any technical aspects of the sound it is on those parts that excels, the positive, rather than the negative.

Share this post


Link to post
Share on other sites

Well, I'm not sure theatrical reference level should ever sound comfortable for theatrical content on a small room system calibrated to 85 dBC.  It goes back to the calibration standard being flawed because it relies on calibration against a continuous test signal.  The observation of this fact leads to the different official recommendations for calibration level vs. room size.  If it's the transients that are most likely to suppress our comfort threshold (usually the case), then we should expect a calibration based on first arrival to translate much more reliably.  My guess is that a better reference level may be defined to be a first arrival of 76 dB SPL over the 500-2000 Hz range using a -20 dBFS sine sweep.  I'm taking a wild guess that reflections and reverb in a typical large room theater contribute an additional 9 dB to the pink noise measurement.

 

FWIW, I'm finding these days that "-4", relative to the theatrical standard, works quite well for me for a whole lot of content.  Incidentally, this is 2-3 dB higher than is typically recommended for my room size.  I actually calibrate via first arrival as described above to 80 dB SPL instead of 76 dB SPL to give me extra room in the MV control for the occasional track with "-4" dialnorm offset, and I pretty much get 85 dBC on the dot using the usual pink noise test signal and test procedure.  It's taken me years to get to this point.  Back in the days of plain Audyssey MultEQ XT, I rarely went higher than -10, and even that was cringe worthy.  Upgrading my speakers definitely helped, but getting the tonal balance right made a world of difference.  A lot of times, "-4" doesn't really seem that loud, but it sure can sound *really really big* when the soundtrack calls for it.

Share this post


Link to post
Share on other sites

Great story about demoing speakers.

 

One point of note is that built-in DSP provides a lot more flexibility for speaker designers to get a very good, extended sound out of a very compact speaker.  Of course size still matters when it comes to max output, but it's fascinating how much quality varies between powered commercial designs despite the fact that such designs are often acoustically simpler and the fact that DSP can be used to give just about anything a very smooth on-axis response.  Clearly a lot of manufacturers either don't know or don't care how to make a good sounding speaker.  I don't doubt that with enough searching, you could find $100 speakers that are as good or better than most of the ultra-expensive models you demoed.

 

I like how, in Harman's testing, they often came across situations where a very expensive speaker was preferred to a budget model when the speakers were visible, but opinions reversed when listeners were blinded.  In one sad case, the listener owned a pair of the more expensive speaker used in the test, but preferred the cheap model once blinded.

Share this post


Link to post
Share on other sites

Well, I'm not sure theatrical reference level should ever sound comfortable for theatrical content on a small room system calibrated to 85 dBC.  It goes back to the calibration standard being flawed because it relies on calibration against a continuous test signal.  The observation of this fact leads to the different official recommendations for calibration level vs. room size.  If it's the transients that are most likely to suppress our comfort threshold (usually the case), then we should expect a calibration based on first arrival to translate much more reliably.  My guess is that a better reference level may be defined to be a first arrival of 76 dB SPL over the 500-2000 Hz range using a -20 dBFS sine sweep.  I'm taking a wild guess that reflections and reverb in a typical large room theater contribute an additional 9 dB to the pink noise measurement.

 

FWIW, I'm finding these days that "-4", relative to the theatrical standard, works quite well for me for a whole lot of content.  Incidentally, this is 2-3 dB higher than is typically recommended for my room size.  I actually calibrate via first arrival as described above to 80 dB SPL instead of 76 dB SPL to give me extra room in the MV control for the occasional track with "-4" dialnorm offset, and I pretty much get 85 dBC on the dot using the usual pink noise test signal and test procedure.  It's taken me years to get to this point.  Back in the days of plain Audyssey MultEQ XT, I rarely went higher than -10, and even that was cringe worthy.  Upgrading my speakers definitely helped, but getting the tonal balance right made a world of difference.  A lot of times, "-4" doesn't really seem that loud, but it sure can sound *really really big* when the soundtrack calls for it.

 

Your -4dB may very well be mine 0dB, due to reasons you probably already have mentioned - differences in method for calibration.

 

Perceived loudness has nothing to do with room size, it has to do with sound field properties - first arrival, decay rate, direction of sound, whether reflected decay energy is diffuse, sound field intensity of first arrival. Of other persons who also understands how this works, I remember Grimani.

 

In even larger larger rooms, with some absorption, the sound is still dominated by the direct sound from the speakers, like the measurements in the smpte report clearly shows.

Share this post


Link to post
Share on other sites

Great story about demoing speakers.

 

One point of note is that built-in DSP provides a lot more flexibility for speaker designers to get a very good, extended sound out of a very compact speaker.  Of course size still matters when it comes to max output, but it's fascinating how much quality varies between powered commercial designs despite the fact that such designs are often acoustically simpler and the fact that DSP can be used to give just about anything a very smooth on-axis response.  Clearly a lot of manufacturers either don't know or don't care how to make a good sounding speaker.  I don't doubt that with enough searching, you could find $100 speakers that are as good or better than most of the ultra-expensive models you demoed.

 

I like how, in Harman's testing, they often came across situations where a very expensive speaker was preferred to a budget model when the speakers were visible, but opinions reversed when listeners were blinded.  In one sad case, the listener owned a pair of the more expensive speaker used in the test, but preferred the cheap model once blinded.

 

Some of the smaller companies in hifi often focus perhaps a bit too much on the artistic side of the development process, and fails to do the objective quality check by doing some basic measurements.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!


Register a new account

Sign in

Already have an account? Sign in here.


Sign In Now
Sign in to follow this  
Followers 0