I have no idea how effective they were for reproduction, but dang if those don't look cool. 😎
Although i provided additional clarification in my last post (wrt to psychoacoustic reasoning and accurate reproduction) you are still overboasting what i´ve said.
I´ve mentioned two quite contradictionary concepts "german taunus sound" and "british smoothing" and asked in addition "what about the famous BBC dip" .
Why should that mean that all BBC loudspeakers....? It also neither doesn´t imply that all german loudspeakers from that era followed the "taunus sound" idea nor that every british loudspeaker.....
Last edited:
And my point is that NO BBC speakers were designed with a dip.
A midrange dip and its effects are well known and taught to recording engineers, but BBC R&D did not use them. You claimed in your post that this was a conscious design choice of kingswood Warren. You were wrong.
A midrange dip and its effects are well known and taught to recording engineers, but BBC R&D did not use them. You claimed in your post that this was a conscious design choice of kingswood Warren. You were wrong.
Surely then the answer is to do multiple test passes using differing test frequencies to avoid the conflicts you highlight. In passes where you do get test f data going into one of the FFT bins and this masking the IMD/THD data, discard it and only save the data in a subsequent, valid test.
I would assume you could do many multiple passes to avoid this problem.
On a separate note, David Moran in LA Volume 6 page 181/182 berates B&W for their 'decades old and altogether unpleasant audible kink at the mid/tweet crossover . . . '
I think how a speaker is voiced seems to follow national stero-types. I've heard quite a few speakers in my time and can characterize them broadly as 'tizz and boom', 'forward mid-range', 'honky' (a favourite of classical music lovers driving horns with flea power tube amps). I'm not going to suggest which characteristic sound belongs where. You know who you are 😉
Of course, the brits like to think they have it right. I do love my B&W's (now owned by a Californian PE outfit I hear.)
I think how a speaker is voiced seems to follow national stero-types. I've heard quite a few speakers in my time and can characterize them broadly as 'tizz and boom', 'forward mid-range', 'honky' (a favourite of classical music lovers driving horns with flea power tube amps). I'm not going to suggest which characteristic sound belongs where. You know who you are 😉
Of course, the brits like to think they have it right. I do love my B&W's (now owned by a Californian PE outfit I hear.)
Jakob, the "BBC dip" I feel is a general comment on many (but certainly not all) two way speakers, which need to marry a bass driver with a dome tweeter, very different between them.People forget that the small BBC L/S3 was tailor made for report ENG studios. Who wanted better and had room space for it, would be better advised to go for larger BBC type enclosures from people like say Rodgers, who did offer classic three way speakers with fairly classic proportions.
Just as so, as you call them "taunus sound" speakers were indeed, as the Brits used to say, sound bright - but compared to what? Most of their speakers, but in reaity they simply had a very linear power bandwidth in practice. Rather than being bright, they were in fact simply more linear across the band.
A good example I like to remember was a Canton smallish 3 way model 50 speaker, using a 10" compression woofer, and dome midrange and tweeter drivers. The only thing they lacked was the extreme bass lines, as system resonance was 50 Hz, but down to that, they were unusually clear and coherent (for the day). Remember, JBL's "bassmeister" L100 model was specified at 50 Hz as the lower cut-off point and it used a 12" driver in a bass reflex loaded cabinet.
Last edited:
It is my understanding that there are different frequencies for different sample rates. In fact, I remember the AP folder to have separate files with names preceded by 44k or 48k or 96k.
Can't find it anymore..!
Yes I did find it! See attached some sample sweep tables.
Jan
Last edited:
If I read you right what you are saying is that the M2/econowave/Gedlee approaches match your approach with the usual differences of opinion (JBL like 120degree horns, Earl likes 90 with a 22.5 toe in to cross ahead of listener and will trade a dip in on axis for low HOMs)?
Certainly those of us who grew up on minimonitors have had a challenge to get our heads around using a 15" up to 1Khz, but it certainly appears to work.
And still i was just using the known phrase "BBC dip" to describe another frequency response curve consideration; as you might have read in my post, Harwood didn´t talk about a workaround for membrane resonances but about a psychoacoustic reason.
Again the LS3/5a prototypes were flat, and have been reported to sound way better than any production units. Production variability on those was horrendous, even with the complex crossover they. Designed flat, ruined by rogers.
EDIT: Harwood may have talked about the effect, but he did not design that way. Proof is in that BBC paper I linked. It has his name in it!
EDIT: Harwood may have talked about the effect, but he did not design that way. Proof is in that BBC paper I linked. It has his name in it!
Marked in bold as you seem to have forgotten.
Fact is BBC research did not design speakers with a dip. They certainly knew about how it could change the perception, but chose not to use it in monitor designs.
Is it so hard to accept that you might not be right and there is more to it that a glib statement?
Bill
There is no such thing as a flat loudspeaker. They are all three dimensional. In more ways than just the case. 😉.
ES
There is no such thing as a flat loudspeaker. They are all three dimensional. In more ways than just the case. 😉.
ES
I haven´t forgotten, but fail to see mentioning of any BBC loudspeaker in it.
Did i say anything different?
Again, i used the phrase "famous BBC dip" as an example for the difficulties in defining accurate response (because, as pointed out quite a long time, within reproduction we are not dealing with original sound fields, not even close to that).
And in fact i thought the idea originated from BBC research.
And,as i cited Harwoods comments on ideas, concepts and that not every colleague necessarily agrees with, i´m still wondering why you insist that i _must_ have meant that all or some BBC loudspeaker had it built in.
No, i´ve no problem to accept if you´re right and the "famous BBC dip" did not originate from BBC research but instead was long known by recording engineers.
But you should direct this question also to the "i´ve-heard-different-and-remain-silent-Billshurv" 🙂
The average British 'salon' appears just large enough to fit seating for four, in which the lord of the manor leap jumps over the furniture in the absence of visitors to entertain. (aka, my home is my storage room)
Jakob, you are flailing now. It seems the 'famous BBC dip' is an invention of the audio press. You have not found any speakers that display it. In fact the also referred to 'swayback' shape attributed to the LS3/5a could be claimed to be the 'german sound' as appears in many of the Audio Physic designs http://www.audiophysic.com/download/spark/hifi_news_and_records_bright_spark.pdf. Joachim may be able to comment on this.
Thank you John, wise information.
As per JC's experience noted above, perhaps useful set of test signals would be digitally generated sets of three frequencies/wave shapes with sets of stated sampling frequency and bin numbers that satisfy all of the above ?.
Scott, you have the smarts ?.
To be expected.
The question of course is the subjective importance of these products.
Dan.
The only Harbeth I found measurements on has it.
Harbeth Super HL5plus loudspeaker Measurements | Stereophile.com
Better not to look at the FR, but at the polar. Off axis, it soon goes dipperdydip around 2.5kHz. This is a speaker with a built in tone control. By increasing the listening angle, you can deepen the Britdip.
John Atkinson's assessment is that the "Super HL5plus's measured performance is beyond reproach". Old school, even not a complete yes, but he must have missed all modern thinking about the importance of controlled directivity. Plus that he proves by measurements that the double tweeter is a flawed concept, but not a word about it.
If I have a chance and some brain power, I'll write up a Python script that is compliant with the concept. No need for multiple passes--simply bin width, which is: FFT_W = Fs / N
FFT_W is bin width (in Hz)
Fs is sampling frequency (in Hz)
N is samples taken
From this selection criteria you can generate primaries and keep track of all harmonics up to Fs/2 (you can go past that, but no point really). Then just keep reiterating up to a selected number of primaries and eliminate those primaries that bin-collide with another already established harmonic.
Edit: looks like Jan attached files that do my work for me (other than show an algorithm that would find said set of primaries).
- Status
- Not open for further replies.