List of Reference Level, High Sensitivity & SPL Speakers - Page 3

Umm, actually, I am okay with testing the speakers myself. Just substitute my basement for the N.R.C. and let me keep each speaker set for a month.

Oh, and I'll start at the top of the list.

I'm glad you answered that and linked to a great thread. I hate to get drawn so far off track on the "what's" and "why's" when there are already so many threads dealing with those topics.

There are already a lot of issues and questions here just with a list, calculations, and these speakers!

Added three Genelec to the google spreadsheet. Had to extrapolate sensitivity because they're active.

Added one Triad but will add more.

And I for sure need to get more Seatons on there like Sparks. And Danley's.

What other manufacturers would you expect to see? Any favorite go-to speakers at the $1,000 $2,000 $4,000 (or small/medium/large room) price points / requirements?

New Speakers:
JBL Pro Monitors and Cinema (just a start, got many more to add, some discontinued and on the used market)!

New Fields:
Active or Passive; Original Purpose (Cinema, Home, Live, Monitor)

 

Reference-Capable-Speakers-1_5.pdf 454.3125k . file

I'm glad you asked, because this is one of many issues with which I need help.

In Flux
Like any project that is rushed out the door, the color coding has changed. It started as quick conditional formatting, and then as more data was added, I saw I was being inconsistent in colors, and I've tried to address that over the weekend. And I want feedback on the data presentation, so I expect more changes are in order.

Ha, and the question just prompted me to look at the ranges and see if Green = 105db, and I saw that the "100 Watts dB" column had 105 as yellow. This is due to how Google treats the number '105' with respect to 'less than 105db' and 'more than 105db.' It apparently was placing 105db in the yellow bin. I've updated the spreadsheet to use 104.99 as yellow.

Make Visual Sense of Data
One purpose of the color coding is to convey more rapidly the relative differences between the numbers. It's more difficult to discern patterns in a sea of black text little numbers.
To this effect, some kind of gradient needed to be applied using conditional formatting.

If Google had some sort of auto-coloring: "pick range granularity, toggle low=good or low=bad, pick low color, mid color, high color," it could do its own "color-normalization" of the data. Alas, it doesn't so I'm left to pick the ranges.

I chose reds for "badder" and blue for "gooder," Oranges are "not as bad," yellows "almost good enough," greens "reached acceptable," light blues "more than good enough," and super blue "superlative."

While this is dividing the available specs, and calculated fields, up into ranges, there's still a judgement call being made by what's considered "acceptable." Many publications avoid this by not passing any judgement and just presenting the "sea of numbers." But I think part of the purpose of this list is educational.

I also did a manual normalization by noticing, for instance, a lot of one color green, and very few speakers using, say, yellow. So I adjusted the ranges for that column so that a more-equal number of speakers fell into each color range.

Judgement
Another purpose was to provide some sort of editorial judgement call.

By default, I'm aiming for 105db from 12 feet. But the other sheet named "Calcs" in the document has the values that are used in the calculations: 105db, 12ft, etc. Changing these would change the "watts to reach 105db" calc and all the SPLs that are calculated from 12ft. For instance, if your listening position was 15 feet, you could change that value, and the SPLs would drop. Unfortunately, changing the 105db target wouldn't render the colors as meaningful, because the conditional formatting ranges are static and not formulaic (eg referencing a field of data).

I inwardly wince at each area that I'm passing judgement. But the pragmatic reality is that there are already subjective, loose definitions involves, such as "High SPL" and "High Dynamics." I thought it too narrow to say, "These speakers that will do 105db at some distance 'Pass' and these that don't 'Fail'. So the color gradient is a way to more gently make these distinctions.

Anomaly Detection
While this is more related to the first point, that it more rapidly conveys info, I found it interesting that some speakers stuck out like sore thumbs. In a sea of blue, for instance, they'd have some spec that was orange.

In some cases, to my chagrin, it was a data-entry problem, either copying from the published spec, or my previous manual process of using an SPL calculator (this is now automatically calculated from distance/watts/sensitivity).

But in other cases, it serves to highlight speakers that are extraordinary in some fashion. For instance, the only way that a low-power-handling speaker can end up at the top of the list (eg if sorted by "Watts & Db") is if its sensitivity was amazing, like 103dB. Examples of this are the Klipschorn, Cornwall, and Fostex LS/2. You also have cases of low sensitivity and high power handling, such as the ElectroVoice ZX-1 and Energy RC-10. And there you see the purpose of the "watts to reach 105db" and "100 watts db": I wanted to easily highlight to people that a speaker might be capable of reaching 105db within its power-handling capabilities, but it's going to require a bigger amp than what's in the typical 100 watt receiver. More educated viewers should be able to make their own judgement that "Oh, it requires 300 watts per channel? I've heard this is a great speaker, and I happen to have three 300wpc amps."

Questions
What do y'all think of the ranges, the coloring?

What do y'all think of comparing these speakers in terms of reaching 105db? For instance, we could instead pick 100db, but then you'd always have people coming in saying, "Hey, I thought reference level was 105?!"

What can be improved about the coloring?

I hesitated to colorize the price and scoring. The price, because as a number by itself, I don't know that it should be treated as bad or good. Although I do incorporate it to various extents in the scoring. And the scoring itself, well it's too in flux and subjective. Which is another topic.

Thanks!

Definitely. I remember noting them, and I had one started in a hidden row. I think I back-burnered because they were active, or difficult to get the specs or pricing for?

Definitely on my list for ones to add. Thanks for steering me to appropriate model lines. That helps a lot!


I've added on in Google (not published in PDF yet...I save that for every day or two).

Where do you think it should go? After the model? At the right-end of the sheet? Or would a color coding on the price work better?

So, I added a few Genelec speakers (I know there are more to add), and I have some questions about them.

Genelec manufactures an active speakers, which means they come with amplifiers that are matched to the capabilities of the speaker, and of each driver.

Genelec Conundrum
I'm wondering how to translate their SPL output and watts numbers into the spreadsheet list. And this applies in general to active speakers, as well as questions about interpreting continuous power, program or "music" power, and peak power ratings, too.

Let's look at the Genelec 1038CF:

• Maximum short time sine wave acoustic output at 1 m on axis in half space, averaged from 100 Hz to 3 kHz: >= 118db
• Maximum peak acoustic output per pair with music material: >=124db
• Amp Power: Bass: 180w
• Amp Power: Midrange: 120w
• Amp Power: Midrange: 120w

Other company's speakers I was able to extrapolate like I'm trying to do here, and came up with a sensitivity, using their amp watts, that worked for their SPL output ratings.

Is Output Limited by the Bass Amp / Woofer?
I think it's the woofer (bass) that has the most difficulty in handling power and achieving higher sensitivity (is this generalization true?! Or is it true only for compression drivers / waveguide / horns, and rather soft dome tweeters may be the "weak link" in other designs?). Thus, I'm only looking at the bass amp for watt specs for these SPL calcs.

Which "Watts" Are These?
I don't know if their 180w is "continuous," "program" (which seems to be usually 2x continuous, or 1/2 peak?), or "peak" (which seems to be usually 2x program).

Watts/Sensitivity Don't Work for Both Published SPL Ratings
Using the Peak SPL Calculator, if I plug in 3.3 feet, 90 watts (assuming the 180 watts was "program" and they're using half that, continuous for this spec), I need to use a sensitivity of "96" to get to 118db. And that's "near wall" (I switched to this in case it helped, but it didn't).

Then, to achieve their 124db at 6.6 feet, with 2 speakers, I can't get loud enough unless I choose 720 watts. Which is twice what their peak watts of 360 should've been.

If instead, I make it work first for the 124db, I can choose 720 watts, but the 118db can't be achieved with that sensitivity. 180w is too much power, and it gets 3db too loud (121db).

So, what's y'all's interpretation of their numbers? Can you come up with a sensitivity that works for both 118db (1 speaker 1 meter) and 124db (2 speakers 2 meters)?

Thanks!

Actually, my PMs with cmryan821 about his ElectroVoice ZX-1 led to his suggesting I contact you about using pro speakers, due to your experience with pro sound and their speakers. I was about to reach out to you anyway for advice on such speakers and this list, so I appreciate you stopping by!


I think any speaker, given the sensitivity, or power handling, sufficient to be considered a "high dynamics" theater speaker should be on this list, so I appreciate this suggestion.

Can you point me to what the Pharos' power handling specs might be? While Salk does briefly mention such for other speakers, I didn't see it for this one.

And I suppose I need to start including info about the HF component of these speakers, like soft dome, compression driver, ribbon!

Thanks!

I've been torn about what the bottom threshold of this list is.

On the one hand, if it's labeled "Reference-Capable Speakers," it suggests that the list contains just those speakers, however that's defined:

Can hit 105db with its "peak watts."
I use peak watts because we're talking about sound track peaks, not sustained sound.
Or, would you define the cut-off (if there was one) at some other point? Include up to 95db (-10 from reference level, 12 feet)?

On the other hand, I find it very edifying to see where other speakers fall by including more marginal ones. For instance, Paradigm is excited about their new Series 7 Monitor line of speakers they rolled out last year. What's not so exciting to me is that low power handling and low sensitivity. It's a big drop from past versions of the Monitor line. What gives? Do they hate high dynamics home theaters now?

So, I was including a few other speakers that were popular, were promoted a bunch, or even my own, just to see how they do on the list, and to provide some point of comparison.

If a speaker isn't seen on the list, is one to assume "It was evaluated, and didn't "make the cut"? Or would they assume that someone hasn't evaluated this 89db 150 peak watt speaker yet, and it may be a contender? Both are valid assumptions!

The problem with this method (which I've sort've used) is that then one says, "Oh, this speaker's 'on the list,' it does well for this application. Rather the list is "Some selected speakers evaluated for how they perform for home theater high spl high dynamics." But that sound kludgey.

I hoped that being colored in red, at the bottom of the list, was enough indicator that while the speaker is listed here, it performs poorly [by these criteria/ranking].

Thoughts?

I actually have to headroom-ish specs:
"% Watts Peak" [to get to 105db]
and "1/8 Peak Watts (start cmprsn) dB"

These are poorly named and probably poorly positioned columns.

% Peak Watts to Output 105db
The first, percentage of peak watts required to get to 105db [at 12 feet], indicates the extent of headroom. I used peak watts, because that's what we're talking about: peak db.

Let's pick on the Paradigm Series 7 Monitor 7. This is a tower speaker. Should be great for a dynamic home theater, right?

• Sensitivity: 88db
• Peak Watts: 180

Hmm. Would needs 670 watts to reach 105db from 12 feet away. We show that'd be 362% of its rated peak watts of 180. It's a red number. The speaker is at the bottom of the list. While I'm sure it sounds very nice at middlin' volume for music and movies, not good.

Let's try another. Ascend Acoustics' CMT-340 SE.

• Sensitivity: 90db
• Peak Watts: 400

Now we're doing better. This would only need 423 watts, which is 106% of its rated watts. So, you have slightly negative headroom: it can do it, but it's straining to do so. This should result in distortion.

Last, let's look at at Triad's InRoom LCR Gold.

• Sensitivity: 95db
• Peak Watts: 500

Not only is it more than twice as sensitive, it can also handle a few more watts. It needs 134 for watts to reach reference level from 12 feet. I'm not sure how the dynamic power of an amp and its capacitors work, but this sounds like it may even be within reach of a properly spec'd 100 watt receiver! Its Percent Watts Peak is 27%, and its color is a light blue. It's loafin!

Keep in mind, too, that if I change the "desired db" value from 105db to 100db (if I want to target -5db as my listening level), the Ascend CMT-340 only needs 134 watts, which is 33% of its peak watt handling.

Paradigm Monitor 7: 212 watts, 118% of its peak watts handling. Capable, but we know it's going to be strained, with some reduced dynamics.

I bet, given enough time, we could get them hammered out!
I've been begging for help on this kind of stuff, so I am glad you stopped by.


Well, when the spec is given as "2.83 v 1 meter anechoic," I used that spec. If the only spec is "sensitivity (half-space)" I dropped it by 3db. Except sometimes other models of that manufacturer did specify both, and the half space was only 2db more than the anechoic, so that'd make me revisit those numbers.

If the company has a line of speakers whose marketing "spec-ers" are known to be lying liars who lie...*cough*...klipsch...*cough* then I dropped all their sensitivities down to match what people say anecdotally and as measured in reviews. Notice I didn't indict the entire company (I hear their heritage line and THX line can be trusted), just the Reference line. I'll probably trust their Cinema line, too.

If a company wants to be vague and coy about their numbers, I hope that the errors are on the side of punishing them!

The great thing is, people can help contribute to correcting these sort of errors, whether my interpretation errors or the company's marketing. "This speaker was tested by the blah blah, here, and you'll see its measured sensitivity anechoic is X."


Can you explain what you mean by this and what the implications are? Is this what Dennis was saying above, how the loss in db is not linear across the frequency range [due to more loss of high frequency]?

Is this addressed by the fact that the speakers are all being treated equally in this respect (good for comparing them with each other, not accurate for definitively knowing that "given this test signal I'll get exactly this many db with this many watts in my home")?


It's probably good you gave up there. I have enough questions and issues swirling around this week. But I do want to do as good a job as I can, given time!


As I suggested above, I don't care so much about this being an absolute reference to be used by an acoustician in designing a room. I'll even take "somewhat misleading" over:

• Many enthusiests being totally unaware of how compressed their sound is, using speakers that otherwise sound good and review well
• Manufacturers playing spec games where the parameters change even within one speaker, let alone from speaker to speaker, line to line. I say, make a good product for what its intended for, and tell people about it, honestly.
• The industry's general unwillingness to address this. It's like a world where most of the car manufacturers only make cars that go 0-60 in 10 seconds and max out at 100 mph. And the auto reviewers only review cars by driving them like timid grandmothers.

Eep! I must've had a couple Triad pages open (I often have 20-30 browser tabs open in two monitors, esp. lately!) and copied the wrong numbers.

Corrected in spreadsheet. And I added the Platinum. Although, I do see that the Platinum is reviewed / test as lower sensitivity by a couple db.

What do you guys think I should use as its sensitivity? 92db or 94db?