 
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
|
|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
174.86.28.109
One of the benefits Stereophile provides audio enthusiasts is providing real world speaker measurements. Every square inch of a printed magazine costs big bucks to print. Stereophile devotes a lot space to these measurements.
It is invaluable to know that a 94db speaker is actually 88db! It is beyond irritating when major speaker makers, who promote their speakers as being highly sensitive, are off by 4db or more. Stereophile needs to challenge the industry on this. There are at least two major makers who routinely do this and I won't even consider their product because of it. There are also reputable makers who consistently publish numbers that are accurate or very close. Triangle, Harbeth, and B&W seem to be examples of consistent accuracy.
Stereophile needs use their big stick to whack some people upside their metaphorical head. At least set out the challenge.
Follow Ups:
I would think sensitivity would be far simpler for audiophiles to generally consider simply through basic auditioning.You could probably make the following assumptions in 2020 which would hold true for 99.9% of loudspeakers
Most loudspeakers are designed with Solid State Amplifiers in mind. Most "audiophile" speakers attempting to be considered high end would probably test their speakers with at least a 100 Watt per channel SS amp of known quality (and probably 500 watts per channel) - say a Pass Labs or Krell or some such thing.
Using a high power power amplifier like this then they probably don't care a helluva lot about sensitivity. Audiophiles who buy these speakers like Magico or Wilson are "probably" not running them with 20 watt tube amps. They are "probably" buying the 500 watt Krell like amplifiers.
The sensitivity rating might matter in terms of whether the Magico or Wilson owner wants to move to a tube amplifier - like a higher powered Rogue Audio, ARC, McIntosh etc but even here - 92dB and 88dB probably aren't going to make much of a difference anyhow.
Then there are the TUBE and SET guys. But these people tend to not be too too caught up in the measured performance because hell - if they were - they would NOT be buying tube amps in the first place now would they? If they are already willing to accept that tube amps sound BETTER than SS amps in SPITE of measurements then I am sure they are the types who will be more readily willing to LISTEN to equipment that has some funky measured performance.
I remember 20 years ago when I really loved a system around a SET amplifier that people got on my case saying that it would measure badly. And my replay was always - well it HAS to measure differently to sound better. I was a measurements guru who was behind the liked of Genelec and PSB and B&W and Bryston and other SS amps. So since all these SS amps pretty much all measure the same and since no one in blind tests could tell SS amplifiers apart - then when the better SET based system came around - well yes it would HAVE TO measure considerably differently. And low and behold it does!
So the guys who buy tubes and speakers that mate to tubes tend to listen to such systems and then play their music and determine if their music can play loud enough with the given amplifier. So when I play my AC/DC at high volumes on my 8 watt per channel 2a3 amplifiers and directly A/B that amplifier to an amplifier capable of 100 Watts per channel then I can determine if the speaker needs the extra 92 watts or not. The speakers don't.
So they are easy to drive. Now granted the manufacturer's spec is not in sync with Stereophile and there are a multitude of reasons for that but it always seems a little silly to argue the manufacturer spec from the real world aspect that the speaker is usually always partnered to an 8 watt SET amplifier and can play to deafening levels with room shaking tight bass response.
Lastly sensitivity isn't the only aspect here - if manufacturer A claims a Sensitivity of 96db and only score 93db at Stereophile and manufacturer B claims 87dB and Stereophile gets 87dB. Well great for manufacturer B for being in line, but manufacturer A is STILL more sensitive and requires vastly less amplifier power to get the same perceived volume level.
Take Klipsch - They rate their corner loaded K-Horn quite a bit higher than some have measured them. But if Klipsch measures the speaker from a corner - after all they are designed for a corner - then the results will be MUCH higher than if an outfit measures the speaker free-standing. By the way that graph is per speaker. Most measurements are of one loudspeaker not two loudspeakers in room. So Klipsch may seem way off - but boundary reinforcement adds a considerably amount of Add.
See the graph above for corner loading a speaker. So Klipsh is probably not far off what they claim under ideal circumstances with solid walls concrete or brick etc
Edits: 02/05/20
"It is invaluable to know that a 94db speaker is actually 88db!"
Though circumstances may vary, I fail to see a speaker's efficieny as invaluable.
For example. Back around 2006 I owned a prized McCormack DNA2 Revision A amp rated at 300wpc@8ohm, 600wpc@4ohm, and 1200wpc@2ohm. A distributor sent me a pair of nuforce Class D amps rated at 160wpc@8ohm, 300wpc@4ohm, and 350wpc@2ohm to audition.
I was already enjoying some pretty phenomenal bass with my 300wpc amp and my full-range 3-way 84db@4ohm speakers. Conventional wisdom might lead us to believe mating the new Class D 160w the bass was going in the wrong direction.
Much to my surprise, the bass was distinctly tighter, deeper, more well-defined, more pronounced, and just more musical. And the rest of the presentation was that much more musical too. BTW, when the spirit moves me, I never hesitate to listen in the 105db area.
Even nuforce was surprised at my response as my speakers were the most inefficient they'd heard of at that time.
Of course, every situation is different and an amp's power ratings aren't always telling the whole story but just from that one experience, I fail to see where a speaker's published efficency ratings are invaluable.
To the contrary, I view a speaker's efficency ratings as barely a footnote.
The spec SHOULD, of course, represent the AVERAGE sensitivity of the speaker, over its entire frequency range, with an input of 2.83V and output measured at 1 Meter.
But if the speaker has a number of amplitude peaks and dips in its output, you can just take the level of the highest peak and call that the sensitivity.
And if the speaker has an average impedance higher than 8 Ohms, like some of the retro horn types, you can state the sensitivity in Watts per Meter. 2.83V into 16 Ohms is only half a watt, so just double the input voltage to get an impressive figure.
It's called "puffery" in marketing lingo, and nearly all manufacturers of all products engage in it. So yeah, objective testing is often required to set the record straight.
I recently bought a pair of Bache Audio Tribeca speakers. Specs are 90db , 4 ohm load. I have a pair of Audio Physic Tempo Plus speakers. Specs are 89db, 4 ohm load. Either 1db makes a huge difference or the Audio Physic only measures at one frequency. I know Bache is very thorough with measurements.
Just about no one (else) has a measurement monkey!
"It is invaluable to know that a 94db speaker is actually 88db!"
... and is recommended by Stereophile LOL!!
Well, Stereophile and Soundstage do provide good speaker measurements, though using different methods. If you don't want to use them, don't.
Personally, I see no reason to seek out a loudspeaker when I have not seen the measurements.
-----
"A fool and his money are soon parted." --- Thomas Tusser
Knowing nothing whatsoever about its technical specs or parameters, only the impressions of some subjective reviewer in his room, in his system, with his ears and sonic prejudices. Satisfaction guaranteed!
Reviews are FYI.
If you are interested in a given product a review may prod you to go hear it yourself,
or create some interest to do the same.
Everyone seems hate all Magazine or on-line Reviews.
And Reviewers.
Hmmm...
I wouldn't Not buy Kef or McIntosh because of a Good Review.
Sheesh!
"If God didn't want them sheared he wouldn't have made them sheep."
vs
At least you can eat the pig.
"Being free has nothing to do with or holds no promise of prosperity."
srdavis2000
Sorry, straw man arguments don't work with most folks.
Reviewers and magazines are LIARS.
Right?
Hello?
Of course not.
it helps when you are long term friends with the reviewer. You understand his musical taste. You understand his priorities. You've spent time together at live concerts and discussed the performance and sound quality afterwards. And spent years hearing his fabulous systems that reveal information in your recordings you never knew existed.
And while his priorities may differ with that of yours, you get a pretty clear understanding of any particular gear's gestalt by reading his observations.
I miss that today.
"Sorry, straw man arguments don't work with most folks."
Why be sorry about that happy fact? :)
Do is still have enough of my mental faculties left to recognize sarcasm?
Check out Mr. Schneider's post below.
so meaningless as to not publish it at all.
That, and the fact that just about everybody is lying about sensitivity (and frequency range) anyway.
My speakers, as measured by JA, come in at about 83db and yet produce room-filling sound (at audio shows, at least) when driven by 50-60 Watt per Channel tube amps.
How is that possible?
Maybe there are other measurements more important than db/watt/meter?
Those three parameters, from my perspective, give a window on how easy or hard a loudspeaker is going to be as a load for an amplifier.
And that , again IMO, has implications for how well any given combination of amp and loudspeakers is going to work and more important, how it's gonna sound.:)
all the best,
mrh
Edits: 01/29/20
The speakers I own may be only 82db, by Mr. Schneider's measurements, but the key here is the 'context' part.
Relatively flat phase and 8 Ohms across most of the usable frequency range with no dips below 6 Ohms.
The measurement of sensitivity is extremely useful. Here's the thing: 2.83 volts into an 8-ohm load is 1 watt. Average sensitivity of speakers is 86-88dB. Some are higher, some are lower, but a typical front-firing dynamic designs usually hits in there.
As a result, 1W (2.83V into 8 ohms) will get you decent output at 1 meter. In fact, for most listening, amps are often only putting out several watts of power. The issue comes when you crank up the volume...
...each 3dB in speaker output requires a doubling of power. But to perceive double the volume, you need 10dB. Start doing the math and you'll realize that although you don't need much power to get most speakers singing -- you need WAY more to get them playing really, really loud.
Sensitivity is what helps to tell you that.
Doug Schneider
SoundStage!
Really, it is a voltage sensitivity, and we really don't know how many watts that would be into the varying impedance of a speaker."Sensitivity: 88.5dB (averaged 300Hz-3kHz, 2.83V/1m)"
That's from the Soundstage NRC measurements of my speaker model. It says nothing about watts, which would confuse things and would not, I think, be very useful. Sensitivity at 2.83 volts seems to be pretty standard, whether it is called 1 watt or 2.83 volts.
-----
"A fool and his money are soon parted." --- Thomas Tusser
Edits: 01/31/20
You're 100% correct -- it's voltage sensitivity.
The problem with watts is that speakers are designed for constant voltage, changing current. So you measure a speaker at 2.83V and the amp tries as hard as it can to maintain the voltage and supply the necessary current.
How that translates to power, of course, depends on the load, and as most of us know, that varies wildly on virtually every loudspeaker. So one would never know how many watts exactly across the frequency band, only at specific frequencies.
Doug Schneider
SoundStage!
And there is the question of short loudness peaks even when you are listening at 'average' levels. I've been told pianos can do over 20 dB peaks which means even fairly efficient speakers can need large amps for maximum fidelity.
You are correct! That's why it's good to know what kind of power you have and what you need.
Doug Schneider
SoundStage!
Key word: 8 Ohms
How many speakers do you measure that are even close to 8 ohms across the useful frequency range.
Joseph Audio Pulsars measured by YOU...
Sensitivity: 82.0dB (averaged 300Hz-3kHz, 2.83V/1m)
And yet...
Hi,
I was trying to make this an easy-to-follow example. You're right -- they're not 8 ohms across the board. Most these days would be considered 4 ohms. So it's 2 watts instead at 4 ohms. Again, no magic -- most manufacturers rate their amps into 8 or 4 ohms. Same rules apply.
Regardless of power, the reason 2.83 volts is used is because it equates to 1 watt at 8 ohms. The thing is -- 2.83 volts is held CONSTANT for all speakers, which means you can look across the sensitivity figures of all and make solid predictions on how much power you'll need from your amp.
Doug Schneider
SoundStage!
Not all loudspeakers are designed or use an operating principle wherein sensitivity is a useful spec.
(Keeping in mind that tubes are still a thing and don't double power as impedance is halved (yet can still operate as a voltage source if enough feedback is employed); tube power is expensive and so 1 watt/1 meter is a more useful spec.)
For example, ESLs don't behave according to the voltage rules. To start with, they aren't a driver in a box with an attendant resonance. Their impedance curve has more in common with a capacitor, varying usually about 10:1 from bass frequencies to 20KHz.
If your amplifier is a voltage source you can see that it will be too bright. Most ESLs need the same power (not voltage) to produce the same output whether 100Hz or 10KHz. Martin Logan gets around this issue (since they want their speakers to work with solid state) by making the higher frequency impedances so low (0.5ohm typically) that even serious solid state amps can't double power.
Many full-range drivers like Lowther or PHY have similar concerns and despite having free air impedance curves of typical drivers, are meant to operate with amplifiers with a higher output impedance, the latter typically operating more like a power source than a voltage source. That, and their efficiency (often higher than 97dB 1 watt/1 meter) is why you often see them used with SETs.
The over-riding issue here is gain bandwidth product, which is usually insufficient in most amplifiers employing feedback, which is to say that not enough feedback is used in 99 44/100% of all amps using feedback. This causes the application of feedback to introduce distortions of its own, typically higher ordered harmonics and IMD. This is why that application can cause an amplifier to sound brighter and harsher. If enough feedback is used though, it can then compensate for the distortion it creates. But that magic number is in excess of 35dB or so, and nearly all amps don't have enough gain bandwidth product for that. Class D is one of the few applications where this can be done.
But we have a further complication- if that amount of feedback is used, the apparent output impedance of the amp will be very low- in fact so low that nearly every loudspeaker in existence will be overdamped, resulting in that coloration known as 'tight bass'. We've been living with that problem for a while!
Sensitivity is always a useful spec. How much you weight it in your particular system scheme is a variable though.
I think Doug is simply saying....2.83 volts into 8 ohms is 1 watt. (That's not incorrect.) :)
My goodness.
Dave.
-
That's all I was saying.
Doug Schneider
www.SoundStage.com
...of anechoically measuring speakers. Dipoles, bipoles and omnis radiate a substantial amount of their energy to the rear and sides. In our listening rooms this increases the SPL, but measuring out of doors, or anechoically, this is taken out of the equation. So measurements will only take you so far.Sometimes we measure the wrong thing, and sometimes we use the wrong methodology. Whack away stick guy.
Edits: 01/28/20 01/28/20
the measurements still work but how to interpret them for planars and line sources varies from conventional speakers. The drop off of loudness for planars and line sources per unit of distance is half that of conventional speakers so for further than the measuring distance they need less power.
....of the in-room loudness levels achieved. And, since the standardized measuring distance is 1M, as you pointed out, there is a greater disconnect with measured sensitivity the further you sit from these types of speakers. The measurements being a wholly inaccurate representation of performance in real world environments.
-
Hello,
You've hit on a topic that I've long championed -- honesty in measurements.
If you want accurate sensitivity measurements, you can check our page below -- it has speakers since 2000.
Doug Schneider
www.SoundStage.com
Could you please measure Boston Acoustics A25. I bought one impressed by Stereophile review. I find the 1 inch port works like a shrill flute. The sound is so congested inside the small cabinet struggling to get out. While you are at it, please take a look at Totem Dreamcatcher too. BJR (bless his soul) found it the best speaker he had heard in 28 years of reviewing but JA One found measurements to be problematic. Great sound in spite of bad measurements.
Bill
Foe small ports the ability of the port lessens at very high levels(the port is too small a 'driver' to play loud. This means the deep bass starts rolling off faster at high levels than at lower ones since the port portion doesn't stay linear at volume.
Good Stuff ....!
Thanks Doug! I was unaware of the site.
Not a site as much as a URL to the list of measurements contained on our portal site. Measurement links accompany a lot of our speaker reviews and this more or less rounds them up.
Thanks,
Doug Schneider
SoundStage!
| FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
||||||||||||||
This post is made possible by the generous support of people like you and our sponsors:
Top of Page ]
[ Contact Us ]
[ Support/Wish List ]
[ Copyright © 2024, Audio Asylum, all rights reserved ]
