Pro vs hifi drivers - pros and cons?

[gedlee]

Originally Posted by vuki

Something interesting: ABX Double Blind Tests: Filters, Fourth Order

Thanks, that is reasonable data, albeit only a single expert listener and the signal source was not stated. I would certainly accept that 15 kHz is the limit of detection for any signal. That IMO would then make 10 kHz a reasoanble point above which the "importance" of the signal content on the sound quality is becoming negligable since by 15 kHz it is irrelavent. None of this is inconsistant with what I have been saying. "Detection" is not "importance" or "relavence".

But thanks for that, its good to see some "real" data and not all the conjecture.

[markus76]

Originally Posted by gedlee

"Detection" is not "importance" or "relavence".

But thanks for that, its good to see some "real" data and not all the conjecture.

There's a paper by Ashihara showing that humans are able to detect signals as high as 28 kHz. BUT the level at the entrance of the outer ear was about 110 dB SPL (they probable heard 28 kHz for the last time in their lives) and listener age from 19 to 25. As a rule of thumb I learned that the upper hearing threshold decreases by 2000 Hz every decade.

Best, Markus

[thoriated]

I have a Sheffield Direct to Disc (Tower of Power) with a close miked horn fanfare that will wake one right up played through my 18khz responding horn HF. Somehow, I don't think the same effect will be given with a 10khz rolloff imposed in the signal path, especially a (blech) fourth order one.

I'm glad to say that my system is not 'bright'. However, neither is it at all forgiving or dull. It, including the two-way Iron Lawbreakers, excels at extracting subtle tonal information in recordings at all dynamic levels up to clipping (of my 60wpc OTL), which the pro sound drivers I'm using significantly enables. It also helps, IMO, that they have alnico magnets & underhung voice coils, and have maximum conversion efficiencies with a system efficiency of 99-100 db/w/m, thus giving many of the advantages of a fully horn loaded system in a 100 liter cabinet. The bass extension isn't too bad, either, being flat to 50-60hz and useable to 32hz, giving sufficient underpinnings for a convincing overall presentation.

[markus76]

Originally Posted by thoriated

I have a Sheffield Direct to Disc (Tower of Power) with a close miked horn fanfare that will wake one right up played through my 18khz responding horn HF. Somehow, I don't think the same effect will be given with a 10khz rolloff imposed in the signal path, especially a (blech) fourth order one.

That might be true in that very special case but a speaker is no musical instrument. How sounds a virtual viola, close miked playing in a horn?

Best, Markus

[adason]

"any particular artist you would recommend? I listen to jazz rarely and don't know any of the band names."

Dave Brubeck has ton of hights on all his recordings. There is significant signal all the way to 18kHz often reaching -3dB in some.
Ed

[pos]

Originally Posted by Borat

But JBL also know a thing or two about sound and they fitted everest with a supertweeter.

i know you will say that was a marketing move. but i would disagree.

Yes they did. According to Greg Timbers himself, they added the supertweeter for the japanese market that wanted a "40khz" extension.
The 1.5" 476Be already goes up flat to 20khz with no issue (other than directivity), and the supertweeter is crossed over at that frequency (with no low pass on the 476Be I think).
But that has nothing to do with the debate at hand since that compression driver WAS able to go past 20khz anyway...

In fact they introduced the 045Be supertweeter for the K2 S9800, because the 435Be (476Be's little browser) could not reach the 20khz mark:
http://www.audioheritage.org/vbullet...787#post109787

[Borat]

Originally Posted by gedlee

Thanks, that is reasonable data, albeit only a single expert listener and the signal source was not stated. I would certainly accept that 15 kHz is the limit of detection for any signal. That IMO would then make 10 kHz a reasoanble point above which the "importance" of the signal content on the sound quality is becoming negligable since by 15 kHz it is irrelavent. None of this is inconsistant with what I have been saying. "Detection" is not "importance" or "relavence".

But thanks for that, its good to see some "real" data and not all the conjecture.

that data is consistent with my own experimentation ( not blind ) with a 36db/oct adjustable low pass filter. i could hear the difference in the way music sounded with the filter and without up to about 16 khz. i can hear sinewaves up to about 17 khz.

of course if the filter is shallow and you don't know what the music is supposed to sound in the first place you could probably be listening to a 10 khz low pass and think there is nothing wrong.

but if you A/B with a speaker that is flat to 20 khz the difference will be apparent.

it's all about what you expect to hear. if you expect to hear sizzling, shimmering heights - you will be disappointed. if you expect to hear something laid back and easy on the ears then you will not be.

[Borat]

Originally Posted by pos

045Be and ~50kHz ultrasonic hearing - Page 2 - Lansing Heritage Forums

great link. let me paste some relevant parts of his post here:

The plain and simple reason for the existance of the 045Be and 045Ti is that the 435Be and 435Al (2430 and 2435 are similar) have a substantial response problem around 10 kHz on many horns and are ceratinly dead by 15 kHz" ...

... So there actually is a valid Engineering and Marketing reason for this device to exist ...

... Does it really matter that we go to 40 kHz or even 60 kHz? Good luck in figuring that out because I'm not going to touch that discussion with a 10 foot pole.

so the Japanese requested 40 khz minimum and JBL developed a driver that goes to 40 khz off-axis, 60 khz on-axis. he agrees that this kind of extension is pointless although he will not say it out loud.

but he does at the same time say that it is valid engineering to address a driver that begins to have problems at 10 khz and dies by 15 khz.

he also admits that they could have used a much cheaper supertweeter to do the job but they didn't want to sacrifice the image of the product with a "toy" device.

i guess you can always put one of these:

http://www.madisound.com/catalog/pro...roducts_id=293

on one of Geddes' speakers and fix it

[kevinkr]

Originally Posted by Borat

<snip>
but if you A/B with a speaker that is flat to 20 khz the difference will be apparent.

My JBL annular ring tweeters are good out to about 16kHz based on in room measurements, but really nose dive above this. More to the point my hearing nose dives around 12 - 13kHz so while I take your point, I suspect in my case at least I could not hear a difference solely attributable to FR above my range of hearing.

I've had my hearing tested a number of times and I can attest to what Gedlee states as being true in my case. Some of it is the inevitable effect of time on the clock, but some of it is probably not. We now live in a world of excessive and constant noise that we were NOT designed to handle on a continuing basis, this has to take a toll.

Friends in the Navy in the past have told me that they are having a great deal of difficulty find young men with sufficient hearing acuity for training as sonar operators. No computer has yet been developed that does this job better than a seasoned human. The problem seems to be the loud world most youth live in today - the problem started to appear about 20yrs ago shortly after the incidence of devices like the Sony Walkman.

I've noticed the most perceptive rarely make much noise their sensory acuity or lack thereof, you simply come to realize that they can "hear" through the occasional well placed comment relating to some aspect of sound reproduction you are working on. One of the most perceptive people I know has told me that he cannot hear beyond 8kHz in one of his ears, but he notices everything. Clearly an ability to listen critically plays a big role here.

[gedlee]

Originally Posted by Borat

on one of Geddes' speakers and fix it

I think that you have a misunderstanding. My speakers go to 15-18 kHz. What you see in the data is the anti-aliasing filter. My point is that I do nothing above 10kHz in the design, because it is so unimportant. I could boost the heck out of the response and get above 20 kHz, but that would simply be the wrong thing to do just to be able to say that I can. Lets not get these things mixed up.

The JBL driver is a bigger throat and will have a great deal more trouble that the 1" throat at those frequencies. I have yet to see a 2" throat driver that can even get to 10 kHz on a CD horn, usually about 7-8 kHz and its suffering. The ones that show that high a response do so only on axis because of serious beaming in the diffraction horn. The 1.5" dies about 10-12 kHz and the 1" can get to 18 -20 kHz, which is IMO just fine.

What is not being considered or understood here is the vast difference in sound quality between CD at 10 kHz and the typical beaming at this frequency.