Quote Originally Posted by RGA
You may not recognise the fact that "low diffraction" and small narrow baffles are mutually exclusive, because this is governed by laws of physics that neither I nor anyone else can change, whether we like it or not.

All modern designers try to do is to circumvent the laws of physics in order to get an edge in appearance stakes, manipulation of the measurement techniques are widespread and are now more used as marketing tools than as guides to whether the speaker is actually any good.
This guy can repeat himself a billion times, and his devoted followers can revel in every word he speaks, but it doesn't make his statements true all the time.

I don't even by the fashion argument he's presenting. Wide baffles with pretty wood finishes wood look every bit as good. I don't understand why he thinks baffle size is related to fashion at all. If anything, cabinet depth is every bit as intrusive and annoying, who wants their speakers jutting out 4 feet into a room?

Just use some common sense here. Severity of primary baffle diffractions is directly related to the surface area of the baffle. Given a fixed height, the narrow baffle has less surface area. Less speaker to diffract off of. The wider you make the baffle, the wider the range of frequencies that will experience some diffraction. This will be mitigated some by lowering the frequencies for which edge diffraction becomes a concern (larger wavelengths). Again, the few extra inches of baffle width will impact the imaging ability of a speaker some....but it shouldn't be a deal breaker. The trade-off is the benefits afforded the lower midrange/bass region. This has been well documented for over 50 years too. You can't just pick and choose which laws of physics you want to apply...that's what marketing departments do.


Dear John,
Firstly, how many speakers have you actually designed and successfully marketed??
I get the feeling this John guy must have pissed Peter off? I don't see any relevance in his question. Dr. Bose has successfully designed and marketed more speakers than Peter. Does that make him more of an authority?




1.) There several ways of achieving low frequency response and your contention that only a large diameter driver can do low frequencies is easily disproven, but rather than talk the talk you need to walk the walk and listen to a speaker which provides low frequency from a small woofer in a setting it was designed for.
Peter's quite right here...but then again, all things equal, diameter size is probably the best way to extend bass - then it becomes a question of which side-effects do you want to tackle. I think that's more of a personal taste thing.

2.) I don't understand why you would measure a speaker at 180 degrees and I have not measured or seen measurements on the NHT you mention, but I can tell you that I would have no real problem putting the 90 degree off axis response of any of our speakers up against any other forward radiating speaker, we would come out well there.
Who the hell listens to speakers at 90 degree angles? If I was Peter, I wouldn't waste my time responding to nut bars like this guy.
3.) We go one better that simple time/phase alignment, we individually adjust and match the woofer's behaviour to the tweeter at the points where they both reproduce the same frequency, this is far far more important and sophisticated than the primitive practice of sloping the baffle a bit to "compensate" for the tweeters earlier and shorter response time.
I think Peter is out of date with current practices. The "sloping the baffle" trick is rarely done anymore.

The ear is far more sensitive to incorrectly matched start - stop anomalies than it is to minor static differences in frequency response, a fact which is neither well understood nor practiced by the loudspeaker industry.
Yeah, to a point, there's precedence effect where very short delays won't be perceived at all. But I agree with him here. Poor transient response and sloppy driver integration is worse IMO than a +/- 4 dB response.

To help you understand what I said in the next paragraph, basically in a 3 way speaker finding drivers which work together in such a way that it is possible to align their timing differences at two crossover points rather than one immense complicates the problem of adjusting the above behaviour.
It doesn't have to...besides, I thought they had developed a super great 15 year old method of doing this?


Dear Greg,
Quite correct, which is why we cross over at below 2.3 kHz and use specially designed units to tailor their response to each other.

Like you say, there are VERY few tweeters who work this far down.
I wish I knew Greg's question...but I disagree with Peter's last statment. I have in my possession 3 tweeters that can be crossed over below 1600 Hz, one as low as 1200 Hz.
There's plenty out there. Maybe it's a more recent development though? Morel and Usher make a bunch.

Dear RGA,
The market perception (supported by most magazines and manufacturers as it is) is that if there is "more" or "deeper" bass then you are getting something for your money.

This is not the case, in fact it is rarely the case that a sub qualitatively improves the music reproduction, generally the money would be better spent on bigger speakers or elsewhere.
Again, just because Peter says it, doesn't make it true. Some people probably don't like subwoofers. My 15" sealed sub is easily integrated with my speakers, is very musical, and greatly improves the qualitative element for music reproduction. Maybe it's because it's not long-throw, high excursion, gimmicky design?


Serious dereliction of duty and care to their readers, much like the stock analysts who recommend shares they privately denigrate.
Was this for my benefit? Any charted stock analyst wouldn't publicly recommend shares they privately denigrate. That would be in violation of their Code of Ethics (most of us do adhere to these, out of fear of the brutal fines and impossible re-entry to the industy). Any recommendations should be supported by substantial analysis and risk measurement.

The magazine reviewers and speaker manufacturer/marketers would be well served adopting a similar Code of Ethics.

There are almost no speakers that have a flat off axis response at say 30 degrees either side.
Guess it depends by the definition of flat...if +/- 3 dB can be considered flat, then I disagree. If +/- 3 dB up to 15 KHz or so can be considered effectively flat, then I very much disagree...
Is it possible Peter got so frustrated with speakers that he's just completely ignored the market in recent years?

I never said power response was new, but almost no-one uses this today, and it is a good indicator of performance.
Yes, Yes, Yes!

I think the reason we see neither of these measurements any longer is that modern speakers with their shallow deep cabinets and resulting poor and very uneven off axis response do badly when measured at listening distance.
Those measurements aren't a trait of the cabinet shape. I think it's fair to say there's just a lot more bad narrow baffle designs than there are bad wide baffle designs these days. Don't blame the baffle, blame designer.

50 inches is about 1 meter 25 centimeters, that is close enough to one meter, 2 - 3 meters are not really valid either, as you need to take into consideration the room reflections and their influence on the sound, as what you hear is always in the listening position, so why not measure where you sit?

Paul Messenger from Hifi Choice generally measures at a distance similar to listening distance and his overall measurements tell you more about the sonic balance of most of the speakers he measures.
I agree that a 1 meter listening position is not indicative of the speakers application. The problem is moving out further into the room will result in measurements being influenced heavily by the room itself. Subtracting this influence from the equation is extemely difficult. Worse, much like a speaker and a cabinet interact differently with each other, a given speaker interacts with a room differently than another speaker beside it. It's not even fair to say that both speakers are in the same room. Problem is we design speakers to fit in rooms, we don't include the room in the design. This is first compromise for all speakers.
Second problem - have you ever heard a speaker that didn't measure as well as nother speaker, but sounded better than it? I have...this tells me there's a few measurements missing.

I aim to make a speaker with a virtually perfect hemispherical dispersion behaviour, which is why I like wide shallow cabinets, in my view a speaker should have an even non jagged drop off as you move from 30 to 60 to 90 degrees off axis in all directions in order to be able to present the room with an even energy waveform.
Just an afterthought...you can eliminate the jagged drop off he describes in the crossover. He's probably doing this anyway with his designs, the wider baffle doesn't eliminate the drops, it just shifts them lower (kind of like what he doesn't like to do with energy by damping).

Likewise, it is important that the speakers within the pair are acoustically identical, otherwise it is difficult to reproduce stereo.
My father listens to more music than anyone I know. I've built him 2 nice systems, but for whatever reason he still spends a great deal of time enjoying music on some lesser equipment he has - ghetto blaster, FM stereo clock radio...
Sometimes I think we get carried away here...if he can get into music with such obviously crappy systems, what is he hearing that we aren't?