List of Reference Level, High Sensitivity & SPL Speakers - Page 2

OK, that makes it easier, thanks, I'll treat it as a "purchaseable" product.

I'm not adverse to including DIY builds. I think I'd want to wait until I could find (or people tell me) what the specs and price are. And then I could add another column for whether it's a DIY, link to flat pack / plans, etc.


Thanks for finding that.

Well, the more errors found, the closer I am to calculating sensitivity in the spreadsheet! : I guess I've reached that point. And then I can treat distance as a variable. It will be that much closer to being a web app (although this list has moved rapidly in a couple days, and I can't see this becoming a database-driven web app until maybe the summer).


Yeah, I intended to add those at some point. Means a lot of reading through the JBL fans' posts. It seems to get a little complicated when there are several modules being stacked. Maybe in those cases I should treat them as one product, as an amalgamation of two or three models?

Can you make it easier on me and point me to the top three JBLs people would like to see (like a budget level for cheaper and smaller space, a mid-level, and overkill-but-not-IMAX)? If used are a common option, those are fine too (esp. since we have the discontinued Klipsch in there).


Good point. I realized last night that the "link to spreadsheet" I was publishing wasn't really a spreadsheet, but rather the HTML output of the sheet when I clicked "Publish." Today I'll link to the sheet as intended, and then (I imagine) everyone will be able to sort, etc.

Value Score
In my initial publications (prior to making the whole spreadsheet available), I have sorted alternatively by "Value Score," which takes several performance and subjective parameters into account, and by "SPL with 1/8 peak watts"

Performance Parameters (subjectively weighted)
100 watt decibels, peak power handling, sensitivity, % of peak to achieve 105db [always from listening distance], and db with 1/8 peak watts due to power compression. While these are objective numbers, the weighting one puts on them is very subjective. eg, a person who knows he'll only have a 100 watt receiver is very interested in what the speakers can do with 100 watts. Someone else who uses powerful amps is not as concerned with stratospheric sensitivity, and they do care about peak watts handling (which some of the pro speakers do well, even if their sensitivity is ho-hum).

Review/Sentiment, Price
Along with all the above, I included price ranking. I tried to weight the pricing so that it didn't matter as much if a speaker was $100 or $200, because at that low level the speaker is suspect anyway being so cheap. This is due to the economics even in an internet-direct model, where there are minimum baseline costs. ie, if a speaker is listed at $30 (and is supposedly high performance), there's probably something wrong.

However, price is something that is more important in the $200-$2,000 range. Above that, it'd less of an issue, and I didn't want to subtract massive points just because a fantastic speaker was $4,000, and to an even lesser extent if it's $16,000. Thus, the ratio of price increase versus points subtracted is not linear.

I also have a column for Review/Sentiment. Some speakers don't get reviewed that much, but there might be a powerful tide of fans. For instance, even if there was never a professional review of the JTR Triple 8, I'd still give that speaker a maximum rating for sentiment. I don't ever read anyone say, "Oh, yeah, JTR, that's okay if you want to get loud, but the imaging is sucky and its freq response is all over the place." Rather, it sounds like people rank it with the best of the best. Some speakers have a mixed following. Some have very little data. Some have negative.

More Granular Scoring
I realize now that lumping so much into a score might be satisfying in that it yields a magazine-like simple number. But the downfall is that it's either too general (weighting too many aspects equally) or too skewed.

I think the answer is to do what Ricci did with his data-bass: provide several types of scores (his "< 20hz," "10hz to 40hz," 20 to 80hz," etc.).
If I tack on to the end of each row a few columns I can put the scoring there.


Yes. I just have to unleash the sorting and it'll be more useful.

I also added a Directivity Index, but that data is available in the minority of the speakers. Hopefully people will be able to contribute other data to fill in the gaps.

Now using formulas for calculating all SPLs. Bume! This has eliminated copy-paste type errors.

I'm now sharing the spreadsheet. However, people with read-only access also can't sort (annoying, I wish Google would make a distinction between data and data presentation!).

I read that List View does allow for sorting. And I changed the link to use List View, but when I browse to it not logged-in, I cannot sort.

It's getting there though!
Google spreadsheet link: https://docs.google.com/spreadsheet/...er=false&gid=0

 

Reference-Capable-Speakers-1_4.pdf 386.4306640625k . file

I had to find formulas for:
Given a speaker's program or peak watts, and D distance and its sensitivity: calculate SPL.

How many watts, with D distance, and this speaker's sensitivity does it take to reach X SPL?

And then....I noticed that on the oft-linked SPL Calculator, he lets one download his spreadsheet as an Excel. That woulda helped!

Thanks! Fixed in the spreadsheet. Will show up in next published static versions.


I appreciate the input! The anecdotal information is critical when we're working in a relatively new realm of using pro speakers for home theater, and there is little to no internet "buzz" about a brand or model, or no reviews in the context of critical "medium-field" listening in a home.

Yeah, it struck me, too, that there was no ZX-2. Seems like a hole where there should be a 10" for $300.

I used formulas here, some of which I needed to string together to combine distance and watts: http://www.live-audio.com/studyhall/dB_calculations.pdf
They involved logs. The numbers I got in the spreadsheet matched what I got here with this [rudimentary, in that it doesn't account for high freq roll-off with distance; see below] often-referenced calculator: http://myhometheater.homestead.com/s...l#anchor_13115. For instance, here's the formula for SPL at 12' with 100 watts: =F5+(20*log(3.3/Calcs!$B$2))+(10*log(Calcs!$B$3/1)).


I'm so glad to get input on this and get feedback on assumptions, or mistakes I may have made, or the utility of the project or even specific numbers and calcs!

My understanding is that when one is aiming for "X" potential peak decibels capability at each of the 5/7 channels, one is not considering things like what the total dB will be when two, three, or five speakers are all playing at the same time.

For instance, going by your question of two speakers...why consider only two speakers? Wouldn't it be off by even more when you include the center? Or the right surround?

Besides, the speakers are often not playing sound that's correlated between channels. If there's a gunshot, it might be coming from hard right, and would be (at reference level) 100db front-right and 90db surround-right.

The specification is just per channel, so yeah, people will experience louder often than that. But I don't know that it matters here.


Right, so we're not calculating for, say, hearing damage, or what one will read on an SPL meter when one, two, or five speakers are all playing at once.

This is more about, "Can this speaker do what I need it to do?"

While it might be a crazy few who have enough proper design/treatment to support the amps/speakers for reference level...

It's a "crazy-a lot" who approach reference levels at -15db or -10db and are experiencing a lot of distortion.

So the point isn't "Are you at reference level, or not?" but rather "Which speakers will do better at reference level, and thus also better near reference level?"

This realm of home theater has not made its way into popular knowledge. It's more difficult to discern than, say, the visual world.

One analogy might be video. Picture a television where it's nicely focused, with okay contrast, at low brightness levels. If you have the brightness down to 10 out of 100, and you watch it in a pitch-black room, it looks okay. But if you want a punchier picture, or have sunlight or light in the room, the picture's washed out, so you would want to turn up the brightness. But this TV gets blurry as it gets brighter. Instead of resolving 1080p, you get 480p. And the contrast gets terrible. The inky blacks become grays.

This is what happens with audio dynamics and speakers that can't reproduce them at higher levels. An 88db sensitivity speaker might do fine at -20 on your receiver, but turn it up to -10, it sounds terrible (like a blurry picture). Also, the speaker probably won't output 20db greater peaks, so now the dynamics are squashed (like the TV whose contrast got worse).

If you can't experience these dynamics, you're missing what makes the soundtrack exciting.

Take for instance the new Paradigm Series 7 Monitor 7. This is a floor standing speaker, and their Monitor line has been fairly well-regarded price-performance wise for years.

Sensitivity is 88db, peak watts are 180. Since power compression sets in at a fraction of the peak power (I've read 1/8th), one could say that any more than 23 watts being delivered to the speaker, you'll get distortion (what comes out doesn't match as well with the input). The more watts, the more distortion. 23 watts would yield 90db as a peak, which is 15db below reference level. If you're happy with -15 as a level in your room, and those 100 watts were truly not distorted, then you should be a happy camper. You should be getting clean, undistorted dynamics on the peaks of your movie soundtrack.

You'll be less happy if:

• The receiver really puts out 70 watts, and why you try to put out 100 watts, you're just sending massive distortion to the speaker.
• When you have two or three channels fully driven at the same time, the amp can't put out 100 watts, so it really puts out 50. Again, at best, the loudness will not be the peak it was intended to be. Worse, it'll be distorted (sound grating).
• If you want to be closer to a good commercial cinema, like at -5db. That's 10db louder, and would require 10x the watts. I sure can't afford a 1000 w per channel amp!

If you did feed it 100 "good" watts from your receiver, you'd should get 97db, except due to power compression the speaker won't produce that level, and there'll be unpleasant distortion.

Thanks for stopping by. When Gamer suggested "Procella!" with a link I just spent a bit of time and included one speaker. I'll revisit and add more. Your contribution helps put them in context. Are you saying the P8 should not be included (as a speaker one would use with subs in a theater)?


I think sensitivity is either more important, or more neutral. But it can't be a non-factor, because an 85db sensitivity speaker would require thousands of watts of power per channel. And could more sensitivity, all things being equal, be worse? I do score with sensitivity and watts being more equally weighted, and other columns where I rate more on price, more on watts, more on sensitivity.


Great point, thanks. The speakers should be vetted for those with an up-tilt response intended to maintain high frequencies at distance. The speakers that are suitable for monitoring should not have this problem, right?


These are on my mental list of ones to add. It's interesting how my perspective has me focusing on a lower budget level, with a sprinkling of some other speakers and price levels to start off with. Others have wanted certain speakers added either because they were ones they owned, or like to see them on the list like the "Top 500 Supercomputers List" etc. And your perspective, probably the best one for this list's purpose, is "Where are the main go-to speakers that should be on any short list of great theater speakers for dynamics."

However, most everyone does, either in online calculators, or back-of-napkin calculations.

This isn't a tool for, say, an acoustic designer to use to definitively tell the client, "This is how loud this channel will be, peak." It's a speaker comparison, it's an educational device to see, "Oh, here's how these speakers are stacking up." Which incidentally is why I've added speakers that clearly should not be on the list.

I understand sensitivity to be an average of a range, such as 300hz to 3,000hz. I can see how if high frequencies are attenuated more significantly as distance increases, a simple calculator might show a a 3db drop, but if you factored in, say, a 5db drop in frequencies 2,000hz to 3,000hz, the aggregate rating should rather be a 4db drop.

So, a question: is there a general rule of thumb on how much high frequencies are attenuated versus low frequencies? If so, can it be universally applied to all speakers? Or do some, like horns, attenuate the highs less than a direct radiator?

I think you're saying,

'If their room isn't treated, so they don't lose as much volume; they keep more sound from reflections. So therefore, they shouldn't worry as much about sensitivity?'

While it's true that without treatment you can get the dB louder, I hear it's also more difficult to listen to as you get louder [without treatment]. The acoustic energy of the reflections swamps the direct sound more.

So, if they don't want to treat the room, it'll sound not-ideal when the volume isn't up, but it'll sound even worse when turned up.

In other words, they can't go very loud, and the ~$3k for audio equipment was wasted, to an extent.


I've noticed in commercial theaters that were not that nice, definitely did not play at reference levels, that still the dynamics were much higher than at my house. The dishes clattering on the table. The sudden roar of the engine when it accelerates, the gun-shot, the yell to "Look out!" Everything seemed more alive, more exciting.

In actuality, all those dynamic peaks were probably lower than they would be in reality. The engine roar, the jet, the gunshot. The latter especially are 30db on up more than reference level peaks.

That's why the dynamics are more engaging...the sound, while still not perfect reality-fidelity, are more like life. Your adrenalin will response more. You grin and think, "This is how it's supposed to be." Just like moving from VHS to DVD to BluRay. Or getting that display that can portray inky blackness, instead of grays, of deep shadows, or outer space.

If you ONLY want to hear the micro dynamics of softer sounds, you should get equipment and a room with a very low noise floor, so you can turn down the volume and still have the dynamics peaks within capabilities.

But even then, room reflections will blur the accuracy of the sound. If you want the pin-drop accuracy, you need to treat the room to mitigate the reflections until after 20ms. If you can't, at least horn-loaded compression drivers can keep more of the sound pointed at your, and less at your walls.


I added the Sierra-1. There are always trade-offs in speaker design. I think a lot of the time the speakers are targeted for two channel music, where people are looking to play without a sub. For home theater, I would've much rather they sacrificed some extension I don't need, in return for more sensitivity. 85-86db is pretty low! 89dB (middlin' sensitivity) with extension to 60hz would've been fine.

This is more of a "play at -14" speaker. Which is fine if that's what one is after.


Well, the funny thing is, when people complain about "loud" (especially women, I hear), they're really complaining (without realizing it), "This system sounds terrible loud in this room." If Dennis Erskine had designed the room, it'd sound better, loud or not. And if the equipment was properly spec'd, it'd sound even better. The dynamics would be present, and you wouldn't be hearing the distortion.

Imagine you're driving in a car more suited for in-city driving. On a poorly-banked road that has some loose gravel over the asphalt. You try pushing it to 95 mph. You try to accelerate and it takes forever. It feels out of control, and when you hit some curves the car feels like it's about ready to fly off the road. It tilts dangerously. You pull off at an exit to end this nightmare, and exclaim, "I'll never drive even close to 95 mph again! It's so dangerous!"

Now try a try a sports car that has great handling, acceleration, and top speed. The same road is just as rough, but the car handles the acceleration, the curves, and 120 mph with aplomb. Completely different experience!

The first car may have had a great sound system, climate controls, insulated cabin, turning radius, and fine leathers from purpose-bred animals on a remote mountaintop. It does fine driving at less than 70 mph, and not accelerating quickly, which is 90% of your driving.

But for fun, adrenalin, excitement, isn't the second car more satisfying, for those other 10% occasions?
And how much better would it be with a great road (room)?

Remember, the point isn't "Can I reach reference level" or "Should I reach reference level." It's that most of us are probably listening to movies with lots of distortion, without dynamics that the sound designers intended.