Design for design, how much has average speaker sensitivity increased since the 70's?

How much have modern day manufacturing techniques and materials along with more advanced transducer design and technology increased average speaker sensitivity?

If the average 10" acoustic suspension bookshelf speaker of the 1970's (say small Advent) put out 85-86db SPL with 1 watt input at 1 meter, how much more sensitive would a modern version have for same size cabinet and bass cutoff?

To put it another way, if JBL were to make the best possible version of the original L-100 configuration today, would the sensitivity be higher or the same holding all other parameters the same?

I have sold audio daily since the early 1970's. My estimate is an average industry increase of about 3db.

My guess is that since JBL was always pushing boundaries for high sensitivity, their drivers or speakers have not increased that much. Certainly they have smoother response, lower distortion, greater power handling and wider bandwidth though.

4312E is more or less a recent version of the L100. Perhaps your answer can be found here.

http://www.jblsynthesis.com/productdetail/id_4312e.html

dPrice: Good suggestion. 3-4db is the increase. JBL L-100 literature usually states 78db at 15 feet which would be around 89-90db at 1 meter. The 4312 is spec'd at 93db.

Speaker efficiency has gone down- this is because of the change towards smaller systems that can be fit into the places men hide from their wives. Because bass is an inflexible requirement for most users, Hoffman's iron law dictates that with the decrease in box sizes, efficiency had to give. Not a good thing for sound but that's the world we live in.

At the high end of efficiency (pro), driver efficiencies haven't changed terribly much in that time.

If you make the cabinet smaller, all other parameters being equal, you would be right that sensitivity goes down and also right that systems have become smaller often due to WAF over time.

What I was asking, perhaps clumsily, was whether improvements in driver construction and design have allowed sensitivity to go up while maintaining all other parameters the same.
Over a 40 plus year period of selling audio day after day, I would say that it has (my guess was 3db). But that was more of a seat of the pants estimate.

dPrice's reply pointing out that the JBL 4312 is 3-4db more sensitive than the very equivalent L-100 it replaced years later, is the best objective evidence I have seen yet that sensitivity has or is at least capable of going up without sacrifices elsewhere.

Quote:

---

Originally Posted by jpw

If you make the cabinet smaller, all other parameters being equal, you would be right that sensitivity goes down and also right that systems have become smaller often due to WAF over time.

What I was asking, perhaps clumsily, was whether improvements in driver construction and design have allowed sensitivity to go up while maintaining all other parameters the same.
Over a 40 plus year period of selling audio day after day, I would say that it has (my guess was 3db). But that was more of a seat of the pants estimate.

dPrice's reply pointing out that the JBL 4312 is 3-4db more sensitive than the very equivalent L-100 it replaced years later, is the best objective evidence I have seen yet that sensitivity has or is at least capable of going up without sacrifices elsewhere.

---

Except that those are just stated specs, which don't mean much (particularly the 93dB- 93dB anechoic to 40hz and under takes a pretty good sized box, 6+cubes). There were 97-98dB pro 15"s and 101-102 guitar speakers in the 70s- we haven't departed from that meaningfully nowadays.

I see a point in your remark regarding stated specs vs actual measurements.

The problem with sensitivity measurements is that they can vary a lot based on the parameters selected. Does a 1970's speaker measured at 90db sensitivity, but with a 3db peak at 2khz inherent in it's design, offer as much real sensitivity as today's 90db model that has flatter response?

JBL has done a far better job than most speaker companies documenting their specs and insisting on some sort of measurement standards, at least internally. At least for a multi-way design with crossovers, which can effect overall system sensitivity, I would think the 4312 vs L100 comparison likely to hold some water if today one measured them against each other.

Your idea of comparing musical instrument speakers is interesting and perhaps the most direct one we have as they are almost always one way designs. I think it likely that the latest designs would maintain their stated sensitivity number over a wider frequency range.

Honestly, you're vastly overestimating the changes in loudspeakers over the years. Some substantial refinements have taken place but sensitivity and bandwidth possible aren't meaningfully changed. Clever tricks around thermal management, overload management, motor linearity, better modeling/measurement, improvements in durability have all taken place. But they really aren't that different for efficiency or bandwidth. There's only so much flux you can get into only so tight a gap and only so much tweaking of a cone you can do.

I may add that in the pro industry JBL and others have about pratically the same efficiency but linearity is greater, power handling have been increase by big factor, and are better desinged for their intended purpose and bandwith...

Interesting topic. I think low end drivers, woofers and sub-woofers, have dramatically changed since the '70s and efficiency has gone down. Partially because power amp watts are much cheaper now than in the 1970's and so designing in efficiency is less of an imperative. Not to mention that subs and content to drive them really didn't exist back then (unless you consider turntable rumble as "content" :-). My 1980's technology 2245H comes in at 95db sensitivity (600w continuous) while a modern Aura Sound NS-18 runs 85db (800w cont.).

Good points all. Thanks for chiming in.

In general it went down because of the rise in popularity of smaller acoustic suspension with direct radiating domes and the corresponding decline in much more sensitive large vented boxes with cone tweeters or horn loaded compression drivers. ADS made a 1900 gauss tweeter in the late 70s/early 80's for use in their L1530/L2030 studio monitor (also used in the early Wilsons) that was rated at 95 dB, and that's about as sensitive as I've seen in a direct radiating dome tweeter. Unless you use multiple tweeters, the sensitivity of the tweeter limits the overall sensitivity of the system. My old L1530's were the only "conventional" speaker that came close (but short) of reproducing dynamics like a compression driver/horn equipped system. I've never heard L2030's but I bet they come even closer.

Quote:

---

Originally Posted by Val

My 1980's technology 2245H comes in at 95db sensitivity (600w continuous) while a modern Aura Sound NS-18 runs 85db (800w cont.).

---

Big difference in maximum excursion and maximum power handling. I haven't tried to model it but I bet the NS-18 offers similar performance in a much smaller box...with much higher input power required. Space efficient, not power efficient. Sometimes you need to save something other than watts.

Again, back to sensitivity being up or down in like designs where other parameters are held the same..........

As I mentioned in my first post, compared with other manufacturers, I wondered whether JBL drivers achieved as much of a sensitivity increase due to how sensitive they were to begin with. As to Badman's point, they already had very high manufacturing tolerances, the best of their time in the 70's and earlier, so improving would be difficult. On the other hand I remember first hand how incredibly cheap tweeters and woofers were from any number of popular manufacturers in the 70's like Advent, AR, EPI, BIC etc. There was far more room with the vast majority of brands sold for improved manufacturing tolerances to have increased sensitivity. At least for awhile into the 80's and 90's I believe it did. At some point cheapness began to reign supreme which would have probably mandated crappy parts again.

It may be a small point, but new drivers that can maintain the same sensitivity as their older brethren with less peaks in the response effectively have higher useful sensitivity.