List of Reference Level, High Sensitivity & SPL Speakers - Page 5

The ONLY way we'll get better/correct data on speaker model lines or individual speakers is if people like you, who are an "expert" in that area, find anomalies, ask questions, suggest more speaker, and contribute resources.

So I really appreciate the contribution!


Good point, if many people are buying used, and that changes the price points for some speakers versus others, we should add another column.

Hmm, "Google, can I integrate a Google Spreadsheet with Google Shopper?!"

Maybe I should remove the "low distortion" from the thread title?

I was thinking that by their very nature of being either high sens or power handling, they are lower distortion than other speakers played at the same high level.

But if the distortion amounts vary wildly, and there is no distortion spec (only a few speakers list that), then I shouldn't suggest they're low distortion.

Exactly. That's exactly what's needed. It's the testing that's the tough part. I sure wish I was that capable, but I'm not even close.

And, to be clear, when I mention Ricci above, it's only in reference to his convenient listing of tested performance metrics. Which are much better than my spec-compiling!

Still, there's a big black (or at least dark gray) hole in the community's understanding and acceptance of the need for compression-free, distortion-free performance at loud-ish listening levels. So if I've filled some of that void in a small way, I'm happy.

I recall I've come across some good stuff before about this, but I'll have to dig it up. I suspect some of the better educated forum members might have such resources more at hand.

Compression is a form of distortion, in that what goes in is not what comes out. The compression/distortion effects can be seen potentially from two sources: the amplifier and the speaker.

Your speaker is sensitive enough for the volume you're attempting as long as the power delivered to the speaker doesn't result in compression/distortion effects. At some percentage of its power handling these will appear. (Still waiting to hear others confirm a generalization of where this occurs). Your amp is powerful enough as long as the watts you're being asked to output doesn't send it into clipping.

The problem is, we over-drive speakers without realizing it, and we clip amplifiers without realizing it!

Barely Good Enough
With 86db sensitivity speakers, you can listen as -10dBfs, where your amp will be at 1 watt for 75db average level. To get the 20db dynamic peaks (95db) at that volume level your amp will be tasked with 100x the watts, or 100 watts. Assuming your amp really can output 100 watts with the necessary channels driven for that program material, and the speakers' power handling are several times higher than 100 watts, then I think there won't be compression/distortion.

Low Power-Handling Speakers
With speakers whose power handling is only 150 watts, I think you'll have distortion. (but I'd like to hear from others what they think the extent would be. Is the speaker-loafing distortion of 5% going to rise to 10%? 20? Will you still get the 95db peaks?)

Too Weak an Amp
If we instead want to listen to -5dBfs, that's only 5dB higher. But this would require 340 watts! While I've heard of amps that can provide 1dB to 3dB of peak power (dynamic headroom), the length of time may well be small for repeated or sustained peaks of a few seconds. The capacitors are quickly drained. 3dB would only get you to 200 watts, anyway. You're going to be clipping, where you're producing square waves that are twice as hard on the speakers. The sound is strained, harsh, grainy, and it makes people say, "Wow, that's LOUD (uncomfortable)!" Instead of the ideal, "Wow, that's powerful, and incredible!"

I'm sure you can imagine an in-between case, where the amp was really 100 watts two channels driven, but 70 watts three channels driven. And it'll be even a little less to the front three, when the surrounds are also probably being somewhat powered.

Some great info here: http://www.rocketroberts.com/techart/powerart_a.htm

"Too Loud", or Too Distorted?
It may sound crazy to want -5db, but in a large room, or an acoustically treated room, it's not as loud as you think, and if we've heard clean dynamics before, we'd want them in our own room too! The absorbtion sucks up more of the reflected sound, and that lack of reflections lowers the overall level versus an untreated room.

When it's not distorted, people are surprised at how effortless it sounds. If you take a great singer who only does well at low volumes, and ask her to belt out operatic levels, she'll sound terrible. "Too loud!" you tell her. But really, it's that you don't like how she sounds.

When grand pianos get up to 109dB (for brief peak periods), it's not unreasonable to ask for peaks approaching that in our own rooms.


But, depending on the speaker, room, and our occasional listening desires, that amp might have to be 5,000 or 4,000 watts!

Okay, I've added the DSR 112, but this might be my biggest guess ever.

It's an active speaker. Thus, I look to sensitivity of individual drivers (I've mainly based the chart on low frequency, as that's usually the limiting factor in sensitivity), the SPL output, and the watts of the amps (again, low frequency).

http://www.yamahaproaudio.com/global...ries/index.jsp

Unfortunately, Yamaha hates us. They don't qualify the max SPL 134dB 1 m:

• as anechoic, half-space, etc.
• is with one speaker or two.
• if their 1300 watts total for LF and HF is continuous, program, or peak power.

If I assume their 134db was with peak watts, and the 1300 they list is program (so peak is 2600), and that they'd use the total watts of both amps, and it's anechoic, then I can assume the sensitivity is 100. But in the spreadsheet I'm just using the LF amp (which is what I did elsewhere).

If someone has a different interpretation of their specs, or knows of a review, or can contact the company for clarification, let me know!

I'll quote Theresa again, because this is something really missing from reviews and sales & datasheet materials.

Manufacturers "say" without qualifying factors that their speaker can handle 1000 watts peak, 250 watts continuous, etc. But was this the point that the speaker failed catastrophically? Or where the distortion rose to an objectionable level?

This is akin to having an amplifier spec of "100 watts" without knowing the frequency range, whether it was RMS, and tested for how many hours, with what level of THD!

Soundstage posted THD measurements using the anechoic room at the NRC. Here's the Paradigm Monitor 5: http://www.soundstagenetwork.com/mea...digm_monitor5/

But they only test at 90dB, and if distortion is too low, they test again at 95db.

That's nice, but that's like reporting, "The manufacturer reported that the car's top speed is 140mph. We found that the car did well at 90mph."

I don't expect reviewers to take a speaker to failure-- wait, yes I do! If you buy the speaker, and want to do a great review, then test the speaker by raising input power until distortion reaches 1%, 10%, 50%, etc., and report where it failed catastrophically. If you can't break the speaker, at least go far enough that we know how much it can take at certain levels of distortion.

Secrets of Home Theater & Hi Fi test at 100dB. This is better. But it's a ways away from 105db at a typical listening position. http://www.hometheaterhifi.com/home-...r.html?start=3.

This Klipsch RF 82 II speaker, if I assume a 95dB sensitivity instead of the stated 98dB, when tested at 100dB output would require 3 watts of power (if they'd tested anechoically). 10 watts would get it to 105dB. 135 watts for 105dB @ 12 feet. How does this speaker perform differently when fed 135 watts versus 3 watts? I wanna know!

Good suggestion. I've edited the first post I think once to go back in and put more explanation, but I should revise and add to that.


You know, I don't know either. This is the most subjective of measurements, which is why I've been sorting by the "Performance" score lately, which most heavily weights dB output, sensitivity, and power handling.

Again, this doesn't mean these speakers are the best, even for a narrow class of application. We don't know distortion, aesthetics, off-axis response, and we're not scoring for low freq extension, penalizing for truncating the last 4kHz the way many cinema speakers do, etc.

But back to Review/Sentiment. 3 is high, -3 would be very low. If I left it blank it was because I didn't know. A 1 meant it was mentioned here and there, like the Mackie C200. A 2, fairly well regarded [for high-output theaters], like the CHT Sho-10 or Elemental Designs new horn speakers.

Thinking about this, I don't know if I can stay abreast of all of this. It's pretty easy to say that a JTR Triple 8 or a Seaton Catalyst is a "3." The ones and twos are more difficult. I haven't researched any of the recent speakers I've added. Maybe with help this would be easier. Or I can not use the column for now until the gaps are filled in.

One thing that drove me to do this is that you'd have a stage speaker that no one knows about (for home theater, midfield listening), that has great specs. If one scores on performance, or a little more subjectively on "value," do you think this speaker should be score as high as a JTR Triple 8? By giving the little-known speaker a "0" or "1", and the well-regarded speaker a "3", I was able to give points to the latter speaker.

This addresses situations where someone new to the scene might ask, "I'm trying to decide between the Peavey speaker and the JTR. Are they just as good for my theater?"

What do you think? Would you find it difficult to do some searching on AVS, and look for some pro reviews, to categorize these into a few simple "bins"?

I think you bought the best center speaker for good value in their currently-produced lineup. I've tried to steer people to the Center 3 instead of the Center 1. It looks like you'd be fine at -10dBfs (if your receiver is calibrated such that "0" is reference level, then you should turn the volume no higher than -10 for films...although TV channels and Netflix might use different levels).

Paradigm drastically reduced this line's sensitivity. The CC-390, for instance, was 94db, instead of the Center 3's 90.

The Studio CC-490's sensitivity is 87, with no more power handling.

The Signature C1 is 85!

I dunno. As I posted in the Paradigm Owners Thread: Paradigm must hate dynamic theaters.

Good point. I see why its size wouldn't work for many people.

Still, let's look at the Mini Monitor, where the size didn't change but the sensitivity did:
Version 6 Mini Monitor
Weight: 17lbs
Sensitivity: 89 anechoic
Freq Response -2dB: 72 Hz -20 kHz

Version 7 Mini Monitor
Weight: 17lbs
Sensitivity: 87 anechoic
Freq Response -2dB: 75 Hz -22 kHz

And, they even lost some extension going to v7.

But, like the center speakers, looking at Monitor 7 and 11, I see their weight and size decreased as well to v7.

This snippet from a review of version 6 is interesting:

Audioholics says that the center was reduced to fit modern furniture better.
So it sounds like they went backwards in size for centers to fit furniture. Smaller/lighter in the towers for some reason, and even when they didn't drop in size or weight (as in the Mini's), the drivers are less sensitive. They gained in some areas, but lost in the ability to output peak dynamics at higher volume settings / well-treated rooms.

Heh "Squawker".

I added a couple speakers to the spreadsheet:
The JBL MRX512M is a pro sound stage / monitor speaker that's come up in a few AVS discussions.

Seaton Sound's (the famous Mark Seaton) Spark LCR active speaker.

This would be a good question to ask of manufacturers.
"Where's the speaker that will give me the 20dB peaks from my listening position without power compression at X dBFS?"

Compression starts at some small fraction of the speaker's power handling (lately I suspect this is actually a fraction of continuous or program watts, not peak watts as I've been using). This may well be 20 watts for many speakers!

So, if you want to listen at -10 dBFS, and don't want to go above 20 watts, you need a 94 db sensitive speaker (anechoic, assuming a treated room). I just don't see that many speakers that sensitive. Why not? And you'll need more for greater distances.

One theory is that extension sells better than sensitivity. It looks better to say, "Look, our -3dB point is 45 Hz!" I feel it is silly to only produce such speakers, if the majority of the theaters will have subwoofers crossed over higher than that! Is this only a consumer education issue?

We need more >= 8" woofer designs that are geared for sensitivity, and either domes that can take the power and have good off-axis response (and probably use in treated rooms), or waveguide compression drivers.

Or maybe the 6.5" woofers are sufficient when there are two per cabinet?

What other reasons do y'all think these speakers are hard to find?

I've referred to showing the fraction of watts where power compression starts, and I've mentioned this was something I was unsure about.

I finally found the statement by MJG100 here: http://www.avsforum.com/avs-vb/showt...&postcount=107


So, I've included a column showing dB at 1/8 peak watts (where power compression starts).
I'm not sure if MJG100 was referring to Continuous Watts or Program Watts. Because he used a fairly low number of "100" in his example, I suspect he meant Continuous Watts. Uh oh, this doesn't bode well!

Using even 1/8th of Continuous Watts handling, which is 1/4 Peak Watts, means that the dB reached will be 6dB even lower!

Now, another question I have is: do we see that same power compression when talking about peaks? I suspect not. One source of compression is the voice coil heating up and gaining resistance, requiring more power from the amp, which heats it up more, etc. in vicious cycle. In a transient peak, you don't have so much time to heat up the voice coil. So, what are the compression effects for transient peaks?

So, I'm led back in a circle to where I started. Maybe power compression starts at 1/10 to 1/8 power, for whatever your program is and what you're measuring.

• Thus, if you're sending a continuous signal with a low crest factor to the speaker, compression will typically begin 1/8 of the speaker's continuous watts handling.
• If you're sending quick peaks to the speaker, which it can handle better, then the fraction of watts where compression begins is based on the speaker's peak watt handling (usually given as 4x continuous).

In which case, my speculative column is fine, since this whole list and thread is about handling dynamic peaks, not the average levels.

I hope some people well versed in speaker measurements and power handling can chime in!

Incidentally, here's a speaker that actually gives a graph of compression at 1 watt, 30 watts, and 100 watts! JBL studio monitor LSR6332: PDF of graphs.

Wow, an example of a manufacturer who, at least for their studio monitor line of speakers, provide much more information about the speaker's directivity, power handling, and distortion.

If only consumer model lines / manufacturers would provide such info!

Link to JBL LSR6336 PDF.


Plus 96dB and 102dB graphs in the PDF.