*Rant/Idealism - Why can't there be an industry standard for speakers?

[FirstReflection]

Hey all. This topic I'm posting is more for a bit of fun discussion than really trying to seriously tackle this topic. So don't get too hot under the collar, ok? :p

Basically, I take a gander at the video side of the home theater equation and notice all sorts of industry standards. Primary color points are well defined. The exact color of white is well defined. Grey scale, gamma, secondary colors, color decoders: all are well defined.

So as long as your display is calibrated and capable of meeting these well defined standards, you can be quite sure that you are seeing the images on screen more or less exactly as intended and more or less exactly as they were seen on a professional monitor right in front of the editor. There could still be slight differences in peak brightness, absolute black level and contrast. But by and large, having these industry standards means that it is possible to get extremely close to replicating identical images on (theoretically) every screen that shows those images.

But when it comes to sound, there seem to be so few industry standards. Other than Sound Pressure Level, almost nothing else is standardized.

My desire isn't to literally recreate lifelike sound. My desire isn't to have sound that is "better" than real life. My only desire is to hear what I am supposed to hear, and to have industry standards that would allow me to know whether or not I am achieving that desire.

Go from studio to studio and there can be a vast variety of different speakers and audio equipment being used by the mixer and recording engineer. Go from movie editor to movie editor and the video will look the same on their industry standardized monitors. But their audio setups could potentially be vastly different!

I ask - Why is that? Why is it that we can accept the need for industry standardization in video, but not in audio?

In order to perfectly recreate any given recording exactly as the makers of that recording heard it, I would literally have to build an identical room, have identical speakers driven by identical gear placed identically in my room. And I could do that, if there were a standard for all of that! But there isn't!

I just think it's weird and silly and I'm wondering how other people feel about it. What really irks me is that I could put together a really really good audio system. Really great speakers, driven by really great gear in a really well designed accoustic environment. I could have all of this great stuff, but I would still have zero assurance that I am actually hearing what I am supposed to hear. I could potentially be hearing a totally different sound to what the recording artists had in mind. And that just seems whack to me!

Thoughts?

[sparky77]

Simply because you cannot get perfectly flat uncolored frequency response from 5 hz all the way to 40khz. There will always will also be various forms of distortion to contend with also.

[D.R. Payne]

The only way to decide on a standard for speakers would be to standardize on A speaker, which the losers wouldn't be too kean on.

Very few manufacturers of very few products want to be "Naked on paper" next to their competition-certainly not makers of loudspeakers.

[ivseenbetter]

I believe everybody would have to be listening in a "standardized" room too...and probably in "standardized" locations.

[Joe Schmoe]

The Canadian NRC did come up with a few guidelines regarding the properties a speaker should have. These guidelines must be valid, because all the Canadian speaker manufacturers who follow them are quite successful, and are generally percieved as making good products.

What I would really like to see is a loudness standard for recordings. I hate having to crank the volume up or down by several notches every time I change CDs!

[FirstReflection]

Cool...glad to see some replies!

I've certainly considered what D.R.Payne mentioned about how manufacturers wouldn't want to suddenly be the makers of "non-standardized" speakers. But my counter to that line of thinking is that, on the video side, we have some pretty strict standards for video, but there are certainly variances from the different manufacturers. Each display manufacturer may create what is technically an inaccurate and non-standardized image, but they do this intentionally to "stand out" and to differentiate their product from the brand next to them. If you happen to prefer this variance as a matter of taste, then you will buy their product rather than someone else's.

But what I like so much about video is that there IS a standard and, if I so chose, I can buy a display that is capable of being calibrated to that standard and thus, I can know with a high degree of certainty that I am seeing what was intended by the people who made the video.

So the same sort of thing could easily be applied to speakers. Manufacturers could (and I'm sure, would) deviate from the standard (if there was one), in order to, again, "stand out" and differentiate themselves from all of the other speakers out there. And if you happen to prefer that bit of deviation, you'll buy their brand rather than another. But same thing as in video: if there were a standard, at least I'd be able to chose to purchase standardized speakers if I wanted to know, with a high degree of certainty, that I am hearing what was intended.

For sure, it would also be necessary to standardize room accoustics as well. But again, my argument is the same. Nothing says you have to adhere to the standard. And it would be impractical to expect everyone to do so. But it would be really really nice to have the choice to be able to do so. As it is, I don't even have the choice because there is no standard to begin with!

I guess the closest we have are things like THX recording standards of a 85dB median SPL with 20dB peaks up to 105dB in the speaker channels and 30dB peaks in the LFE. And also the "unwritten" standard of achieving flat frequency response at the seating area from all channels. But we don't really have any sort of standard at all for distortion levels or decay or combing or other aspects of the sound.

I'd really just like to be able to know, with a high degree of certainty, that I'm hearing what I'm supposed to hear...just as I can with video.

Any other thoughts?

[no. 5]

In his book, Floyd Toole laments the same thing (and I do too), as the standard for electronics has always been 'as flat as possible from DC to light', but speakers? 6dB or more variation is somehow then acceptable!?

Having some better standards would go a long way, although it would be nice, it wouldn't even need to be a standard of zero distortion or coloration from 5Hz to 40kHz to be better than what is there now, the NRC's guidelines alone would be an excellent starting point.

[AcuDefTechGuy]

Standards and Definitions are one thing, but personal preference variation is another.

Video - Everytime I calibrate my TV/Projector to "standard" using Video Essentials disc, etc., I'm never happy with the end result. Maybe it's great for some people, but not for others. I prefer my colors more vibrant and colorful. I prefer my black levels (contrast) a lot higher than standards.

Audio - I prefer a lot of tight bass. I prefer huge & heavy tower speakers. But not everyone share the same preferences.

To me, the standard for speakers is one that has a flat frequency response on-axis from 20 Hz - 20 kHz +/-3dB @ 90 w/m with or without the use of a separate dedicated subwoofer.

[Nuance AH]

My position is that it just simply can't be done. We'd all have to be millionaires or live in a communist country in order to all follow an industry standard. As you said, different rooms, different ears and different equipment = different sound. Unless we all have the same rooms, gear and ears there will never be a way to follow a standard, even if there was one.

[jliedeka]

The laws of physics make it very hard to design a perfectly linear driver much less a system like a loudspeaker or worse, a loudspeaker in a room. With drivers you can have linearity, low distortion and a wide frequency range - pick two. With cabinets there are other tradeoffs. Rooms are all over the map. Does everyone want mondo traps on their walls?

I think the only way that comes close to standardization is a standardized set of measurements. I can tell a lot about a speaker from either the NRC measurements on Soundstage or John Atkinson's measurements in Stereophile. Each set has their strengths and they won't absolutely tell you if a speaker will sound good but you can get a pretty good idea from the impedance and phase graphs or the on and off axis frequency graphs.

Jim

[FirstReflection]

I 100% agree with the notion of at least starting with standardized measurements! That seems totally do-able to me and it would at least provide some sort of "baseline" for objective comparison.

What I always find so fascinating is the common description of "hearing things I never heard before". From what I can tell, that phenomenon occurs whenever one of three things happens: when a new speaker is able to produce audible frequencies that the previous speaker simply was not able to produce; when the new speaker has so much less distortion that it “unveils” sounds that were previously obscured by distortion; or when the new speaker has a frequency response (within the room) such that certain audible frequencies are now considerably louder than with the previous speaker.

The thing is, there are actually rare times when you end up hearing things that you were never supposed to hear! I recall one anecdotal story I heard in which an older recording was played back through an extremely high fidelity system. And through this system, it was possible to make out the faint sound of voices talking. As it turned out, those voices were actually the people in the booth! There voices were unknowingly picked up by the recording mic, but during the recording, their own system at the time masked the voices with distortion and they never noticed!

[sparky77]

[QUOTE=FirstReflection;458103]I guess the closest we have are things like THX recording standards of a 85dB median SPL with 20dB peaks up to 105dB in the speaker channels and 30dB peaks in the LFE. And also the "unwritten" standard of achieving flat frequency response at the seating area from all channels. But we don't really have any sort of standard at all for distortion levels or decay or combing or other aspects of the sound.QUOTE]

THX would have more time to create those extra standards if they spent a little less time creating standards for doors, computer speakers and audio cables. What's next, speaker stands?

[dorokusai]

AMEN FR...

I like the things that a loudspeaker does and doesn't do well. This is probably more of a vintage thing in regards to the overall picture. I love my DQ10's but its not a perfect speaker. I love my Magnepan but it's not a perfect speaker either. Who cares? For me, that's where the magic is....let them do what they do, based on their own merits.

There is way too much thought in this hobby and less about media and enviroment. A crossover upgrade isn't the solution, just a tweak, and at that point....it isn't the same speaker anymore.

Mark

[jamie2112]

They should sound Good......:D

[annunaki]

I could see a standard that showed the linearity of frequency response as long as it included an output level anechoic.

For instance:

Level One - Low resolution device +/- 10db from 40hz-20khz at 90db at 1m anechoic

Level Two - Moderate resolution device +/- 6db from 40hz-20khz at 90db at 1m anechoic

Level Three - High resolution, limited range device +/- 3db from 40hz-20khz at 90db at 1m anechoic

Level Four - High resolution full range device +/- 3db from 20hz-20khz at 90db at 1m anechoic

Level Five - Studio Quality device limited range +/- 1.5db from 40hz-20khz at 90db at 1m anechoic

Level Six - Studio Quality device full range +/- 1.5db from 20hz-20khz at 90db at 1m anechoic

*The wattage required to meet the spec would be stated on all speakers as well.

If a speaker did not fit at minimum Level One, it would receive an Unratable device ranking and be considered a very low resolution item.

These rankings can also apply to full speaker systems if they meet such criteria.

[jliedeka]

That's not a bad idea but it will never sell. Marketers prefer to make up the reasons you should buy their speakers. :p

Jim

[no. 5]

I could see a standard that showed the linearity of frequency response as long as it included an output level anechoic.

For instance:

Level One - Low resolution device +/- 10db from 40hz-20khz at 90db at 1m anechoic

Level Two - Moderate resolution device +/- 6db from 40hz-20khz at 90db at 1m anechoic

Level Three - High resolution, limited range device +/- 3db from 40hz-20khz at 90db at 1m anechoic

Level Four - High resolution full range device +/- 3db from 20hz-20khz at 90db at 1m anechoic

Level Five - Studio Quality device limited range +/- 1.5db from 40hz-20khz at 90db at 1m anechoic

Level Six - Studio Quality device full range +/- 1.5db from 20hz-20khz at 90db at 1m anechoic

*The wattage required to meet the spec would be stated on all speakers as well.

If a speaker did not fit at minimum Level One, it would receive an Unratable device ranking and be considered a very low resolution item.

These rankings can also apply to full speaker systems if they meet such criteria.

I like it! :)

Because a standard need not be, "all loudspeakers must have zero resonance, have no frequency deviation greater than 0.1dB between 120Hz and 15kHz, have a fully active DSP crossover system and the ability to reach 105dB".

It has been brought up a number of times that there would be a difficulty in setting a standard because of the room, but just because the room is in control below ~200Hz (the speaker is mostly in control above that), does not mean that the consumer, or recording engineer can't (or shouldn't) start with a speaker that is at least mostly accurate.

The video standard is there to allow the ability to see at home (or local theater) very nearly what was seen on the editing screen, an audio standard would be the same thing, it need not require "perfect" reproduction, just the guarantee that what you're hearing is very close to what the artist heard. It wouldn't be that hard to do, speakers with reasonably tight tolerances for on and off axis frequency response, and subwoofers or receivers with a DSP system that would allow the low frequencies to be smoothed out. Even if speaker manufacturers were required to provide meaningful third party measurements, that would be a huge step forward.

[Sheep]

Where is WmAx?

Lets look into what you're saying. Real life has nothing do with it. You're not loading real life into your DVD player then hitting play, it's a CD, DVD, Blu Ray, whatever. This is the source of sound. Since sound is subjective, and art(CREATING, not reproducing), it will vary from person to person. The job of the stereo is to reproduce the sounds on the CD, exactly as the CD reads. Having a CD player that can do this is very simple, and already achieved. Next you need an amplifier to take the signal and put it out to the loudspeakers without adding distortion or skewing the FR. Also achieved.

Wires... Achieved.

Loudspeakers, the weakest link. The loudspeaker must produce the sound form 20Hz to 20KHz without deviating more then 1.5dB above or below the source. Hard? Yes. Impossible? No. Also, you don't need 5Hz to 40KHz. You're wife may call you a dog, but you still don't have their hearing. B&W have made speakers that priduce 20H to 20KHz, WmAx modded some Infinity Primus speakers to do this. This isn't the hard part. The hard part is removing the cabinet coloration and resonance from the speaker. It's not impossible, but it requires a lot of labor and would be extremely expensive to mass produce. So now we have a clear sound, from the speaker. No coloration. But what about the room? It's going to change that perfectly flat signal into a Californian beach. You need to Treat the room with bass traps and acoustics diffusers and the like, but how much? And where? Well, how the speaker produces the sound will tell you. This is the key point to having surreal playback. Sound emanates from life like sounds, at more then 1 axis. If you want the same realistic sound in your stereo, your speaker has to also have a flat off axis response. This again is no small task, but it is doable, and has been done. The normal train of though is to treat the first reflection points, but if you have clear, clean sound waves going there, treating that spot is going to suck up that reflection that gives you a sense of depth and realism. I once knew the proper way to treat a room for Omnipolar speakers but I can't remember it. Hopefully WmAx will see this thread and post.

Cost is one of the things keeping audio standards out of the loudspeaker arena, that and complexity. Most people don't want to go through all this to hear music, and are comfortable with what they have from their TV. Without a market, no product will soar, regardless of it's excellence.

SheepStar

[speakerman39]

Where is WmAx?

Lets look into what you're saying. Real life has nothing do with it. You're not loading real life into your DVD player then hitting play, it's a CD, DVD, Blu Ray, whatever. This is the source of sound. Since sound is subjective, and art(CREATING, not reproducing), it will vary from person to person. The job of the stereo is to reproduce the sounds on the CD, exactly as the CD reads. Having a CD player that can do this is very simple, and already achieved. Next you need an amplifier to take the signal and put it out to the loudspeakers without adding distortion or skewing the FR. Also achieved.

Wires... Achieved.

Loudspeakers, the weakest link. The loudspeaker must produce the sound form 20Hz to 20KHz without deviating more then 1.5dB above or below the source. Hard? Yes. Impossible? No. Also, you don't need 5Hz to 40KHz. You're wife may call you a dog, but you still don't have their hearing. B&W have made speakers that priduce 20H to 20KHz, WmAx modded some Infinity Primus speakers to do this. This isn't the hard part. The hard part is removing the cabinet coloration and resonance from the speaker. It's not impossible, but it requires a lot of labor and would be extremely expensive to mass produce. So now we have a clear sound, from the speaker. No coloration. But what about the room? It's going to change that perfectly flat signal into a Californian beach. You're need to Treat the room with bass traps and acoustics diffusers and the like, but how much? And where? Well, how the speaker produces the sound will tell you. This is the key point to having surreal playback. Sound emanates from life like sounds, at more then 1 axis. If you want the same realistic sound in your stereo, your speaker has to also have a flat off axis response. This again is no small task, but it is doable, and has been done. The normal train of though is to treat the first reflection points, but if you have clear, clean sound waves going there, treating that spot is going to suck up that reflection that gives you a sense of depth and realism. I once knew the proper way to treat a room for Omnipolar speakers but I can't remember it. Hopefully WmAx will see this thread and post.

Cost is one of the things keeping audio standards out of the loudspeaker arena, that and complexity. Most people don't want to go through all this to hear music, and are comfortable with what they have from their TV. Without a market, no product will soar, regardless of it's excellence.

SheepStar
Wow Sheep very well spoken. I mean, you really hit the nail on the head so to speak. I agree with what you are saying and as such could not have said it any better. Way to go Brian!

Cheers,

Phil

[Sheep]

Wow Sheep very well spoken. I mean, you really hit the nail on the head so to speak. I agree with what you are saying and as such could not have said it any better. Way to go Brian!

Cheers,

Phil

I guess after 8000 posts you learn a thing or 2 :D

SheepStar

[speakerman39]

I guess after 8000 posts you learn a thing or 2 :D

SheepStar
No doubt. But, it is also how you worded it. I can write well when I want to, but you just came right out with it that truly shows skill. And, as you said your 8K posts surely hasn't hurt anything. It is good that you have learned and grown in this hobby. I have as well. The best part is quite simple-I often learn something about this hobby everyday which keeps things interesting. Otherwise, burnout would set in quite quickly. Anyways, good job on the write-up Brian.

Cheers,

Phil

[no. 5]

Cost is one of the things keeping audio standards out of the loudspeaker arena, that and complexity. Most people don't want to go through all this to hear music, and are comfortable with what they have from their TV. Without a market, no product will soar, regardless of it's excellence.

Absolutely, and I agree, but an effective standard need not be as refined and detailed as what you outlined (although it would be preferable) to bring us an improvement. Right now there are zero standards for the acoustical performance of a loudspeaker, just look at what $3,600 (http://www.soundstagemagazine.com/measurements/zucable_druid/) and $274 (http://www.soundstagenetwork.com/measurements/speakers/psb_alpha_b1/) can buy you. I do not think the issue in setting a standard lies in a manufacture being able to build a marketable speaker that meets reasonable tolerances of acoustic performance, but in getting people to agree on what the standard should be; many here would likely gladly accept a standard along the lines of what current perceptual research has outlined as "good sound", but what of speaker designers? Will they view off axis response as more important than phase response? What about resonance? Will every designer feel that they should go as far as possible to eliminate it?

[avaserfi]

In a way a standard has been created through perceptual research and in fact, many loudspeaker design companies seem to follow parts of it, but none follow it fully in all their commercial lines [not feasible] and very few use this research as an exemplar for their top tier speaker.

The first thing to note is that depending on application the standard would need vary. For example, if a loudspeaker was to be used for studio mixing in a near field environment the ideal loudspeaker and room requirements are vastly different from a far field studio environment and even father from the requirements of a loudspeaker whose goal is ideal perceived reproduction.

With regard to the example of a rating system based on speaker linearity: Such a system would be completely useless. Firstly, there are other variables that are just as important as a speakers on axis frequency response [as shown by credible perceptual research] such as distortion, polar response, cabinet and driver resonance etc... Also, the Q of each deviation is of utmost importance as well as the deviations relation to the axis [a dip or peak]. This is so because higher Q peaks/dips are less audible than low Q dips/peaks in many situations. Further, dips are less audible than peaks in many circumstances. So such a rating system would be virtually useless as loudspeakers that fall in class 1 or below could outperform loudspeakers in class 6 and even within class large deviations in quality would be had.

If a proper standard based on credible research was to be set into a list it would require any speaker that adheres to this list to have a variety of qualities which are not only unreasonable by most commercial design standards, but also vastly more expensive due to current production techniques. Also, as previously mentioned there would need to be various subsections of said standard accounting for near field listening [not as demanding] and far field listening as well as room interaction which is of utmost importance in typical home listening environments.

Now the JAES has created a standardized room for which much of the perceptual research is done. This room is designed in such a way to be representative of the average listeners living room furnished in various ways. So this would be a step in such a direction, but a typical living room will never be able to allow for optimal sound reproduction due to the necessity of specific [and fairly ugly] treatments for the ideal speaker.

So in an ideal world what would such a standard be for music reproduction?

For optimal sound reproduction [please note this is not necessarily a completely conclusive list, but just an example based on perceptual research and its findings on loudspeaker preference]:

-On axis frequency response with no deviation greater than 0.5dB 20Hz-20Hz [some could argue greater extension is required in both directions for such instruments as pipe organs and those who have hearing above the norm]

-Polar response matching on axis response in a full 360 degree pattern horizontally as well as +/- 90 degrees vertically

-No audible cabinet or driver resonances

-No audible distortion during normal playback or transient peaks [while this is rarely an issue many DIY and commercial designs seem to focus on distortion behavior so much that it is detrimental to other aspects of design]. Related: Dynamic capability such that during reasonable SPL there is no created distortion during transient peaks.

-Customizable tonality such that desired response can be had while maintaining all other previously explained aspects.

Now these are just requirements of the loudspeaker. There would need to be strict requirements for the room and all equipment being used. Not only this, but the recording methods used today. Since dynamic compression is so prevalent most modern recordings would sound no better on such an ideal system when compared to a system of far lower capability.

With regard to loudspeaker designers disagreeing on various aspects of design: Each aspect can be correlated to a specific part of loudspeaker performance. If a designer chooses to sacrifice one area it can be done while knowing what actual performance aspect is being sacrificed if the research is properly understood which from my experience seems rare. This is one reason I suspect that phase response is commonly dwelled upon when actual research shows its subjective effect to be minimal to nonexistent just like group delay or why distortion is commonly leaps and bounds below audibility yet resonance is almost always audible.

In the end, as with most consumer goods the final burden lay with the consumer. Doing proper research and understanding the issue at hand is paramount. Even with such standards in place some entity would need to be created to enforce them etc...There are still many areas of gray.

[FirstReflection]

Where is WmAx?

Lets look into what you're saying. Real life has nothing do with it. You're not loading real life into your DVD player then hitting play, it's a CD, DVD, Blu Ray, whatever. This is the source of sound. Since sound is subjective, and art(CREATING, not reproducing), it will vary from person to person. The job of the stereo is to reproduce the sounds on the CD, exactly as the CD reads. Having a CD player that can do this is very simple, and already achieved. Next you need an amplifier to take the signal and put it out to the loudspeakers without adding distortion or skewing the FR. Also achieved.

Wires... Achieved.

Loudspeakers, the weakest link. The loudspeaker must produce the sound form 20Hz to 20KHz without deviating more then 1.5dB above or below the source. Hard? Yes. Impossible? No. Also, you don't need 5Hz to 40KHz. You're wife may call you a dog, but you still don't have their hearing. B&W have made speakers that priduce 20H to 20KHz, WmAx modded some Infinity Primus speakers to do this. This isn't the hard part. The hard part is removing the cabinet coloration and resonance from the speaker. It's not impossible, but it requires a lot of labor and would be extremely expensive to mass produce. So now we have a clear sound, from the speaker. No coloration. But what about the room? It's going to change that perfectly flat signal into a Californian beach. You need to Treat the room with bass traps and acoustics diffusers and the like, but how much? And where? Well, how the speaker produces the sound will tell you. This is the key point to having surreal playback. Sound emanates from life like sounds, at more then 1 axis. If you want the same realistic sound in your stereo, your speaker has to also have a flat off axis response. This again is no small task, but it is doable, and has been done. The normal train of though is to treat the first reflection points, but if you have clear, clean sound waves going there, treating that spot is going to suck up that reflection that gives you a sense of depth and realism. I once knew the proper way to treat a room for Omnipolar speakers but I can't remember it. Hopefully WmAx will see this thread and post.

Cost is one of the things keeping audio standards out of the loudspeaker arena, that and complexity. Most people don't want to go through all this to hear music, and are comfortable with what they have from their TV. Without a market, no product will soar, regardless of it's excellence.

SheepStar

Awesome post. Very nicely done. And this is exactly the sort of industry standard I would LOVE to see put in place!

I think some people may have misinterpreted what I've said.

I never said I believed that ALL speakers must meet this hypothetical industry standard. In fact, I stated that I figured manufacturers would still purposely deviate from any such standard - just as they do with video.

My wish was simply that there BE a standard. So that just like video, when you go from one recording studio to another, nearly the exact same sound could be replicated. Then, if you were so willing and able, you could then have the option to recreate this same audio environment (room and speakers) in your home.

In video, the option to buy a display that can be calibrated to industry standards exists. Such a display typically costs more though. And the number of people who actually go to this trouble is very small! It's certainly not as though every TV out there MUST meet the industry standard - far from it! But what I'm saying is that the video standard does exist. You can go from studio to studio and see very nearly the exact same image on all of their monitors. And if I so chose, I can do this at home as well!

So I LOVE the idea of such tight tolerances as outlined by Sheep for audio!

Recording studios lay out A LOT of money for their gear and speakers. I truly do not believe it would be at all unreasonable to them to meet standards as strict as what Sheep has described.

The real point would be to have this standard and to have as many recording studios as possible follow it. On the consumer side, it's still a totally open ball game. Manufacturers can make whatever they like and consumers can buy whatever they like. And I'm positive they would!

But the OPTION would exist for me to build a room, treat it and populate it with speakers that meet or very nearly meet the industry standards. That would be my call and my option. But if it existed and I went to this trouble, I could then be reasonably sure that I was hearing, in my home, exactly what the recording artists heard in their recording studio! THAT is what I want :D