I thought the 'BBC dip' was intended to correct a power response problem in loudspeakers, i.e remove some of the excess energy which results from a widening of the radiation pattern above crossover?
Harwood on the BBC dip:
An externally hosted image should be here but it was not working when we last tested it.
Many driver have quite signficantly long decay in this region. I would say, if all the proper impedance smoothing in this region is done, and the driver has good decay characteristics, with a transient perfect XO design, the dip would not be necessary.
I have to regect with the idea that the problem is with crossovers and tweeter excursion. I say this because the probelms with mass vocals and massed srtings is present with ELS speakers with no crossovers above 120 Hz. Now, a large panel ELS has minimal excursion problems. There is just no doppler induced IM distortion to speak of.
Though I completely agree “in general” I’d like add one or two things
It might well be that mass choral pieces simply are not your thing. Mostly these are at the very spot of western culture, spirit and religious.
So, you possibly may admire and enjoy but simply get not “triggered” by this sort of music – nothing serious – there's a lot of stiles out there to choose from – no single stile opens the door into heaven for each and everyone.
On the other hand there must be more about mass choral presentation than low IM.
Years ago I used to listen with one of this ancient stereo combos – you know - single EL84 and Alnico whizzer cones firing to the sides.
(though not exactly that one !)
This (heavily modified) system for sure had plenty of all kinds of distortion you can think of – especially at insane levels I used to crank it up with favourite pieces of mine - BUT – it was definitely an *experience* with such mass choral material I rather rarely listen to.
IMO there is something in the lower department the speaker has to be exceptional in addition to the vocal range.
Big chorus' are basically “big” – meaning there must be good rendering (and differentiation) in the lower department - certainly it isn’t alone a question of IM in the vocal range.
My background with such less than perfect (in technical terms) ancient device made also absolutely clear to me that good rendering is something totally different than technical perfection.
In that sense – whatever the outcome of this “beyond thread” will be – I’m absolutely sure about one thing – the resulting system will most possibly not be technically revolutionary but for granted be of great inner harmony .
In the end - that’s what we are *all* after !
It's worth to await – and mess around in the meantime with all the topics at hand - even if it will take some more month' or years - possibly until all of us walking with a cane then
Having said all above allow me to perform an other sacrilege in recommending a webstrem I sometimes manage to listen to.
If you already have digital integrated into your system its worth to join Peter Togni at CBC2 (Choral Concert - Sunday 9 a.m. (9:30 NT))
http://www.cbc.ca/radio2/program/choralconcert.html
If non of his tracks goes subcutan – forget about mass choral in future.
High source quality (Studer, SAC) certainly is fine to have - but if a system can't handle low source quality (FM or MP3 that is) up to the point where you can enter myth of music – forget about your "dream set"
Michael
It might well be that mass choral pieces simply are not your thing. Mostly these are at the very spot of western culture, spirit and religious.
So, you possibly may admire and enjoy but simply get not “triggered” by this sort of music – nothing serious – there's a lot of stiles out there to choose from – no single stile opens the door into heaven for each and everyone.
On the other hand there must be more about mass choral presentation than low IM.
Years ago I used to listen with one of this ancient stereo combos – you know - single EL84 and Alnico whizzer cones firing to the sides.
An externally hosted image should be here but it was not working when we last tested it.
(though not exactly that one !)
This (heavily modified) system for sure had plenty of all kinds of distortion you can think of – especially at insane levels I used to crank it up with favourite pieces of mine - BUT – it was definitely an *experience* with such mass choral material I rather rarely listen to.
IMO there is something in the lower department the speaker has to be exceptional in addition to the vocal range.
Big chorus' are basically “big” – meaning there must be good rendering (and differentiation) in the lower department - certainly it isn’t alone a question of IM in the vocal range.
My background with such less than perfect (in technical terms) ancient device made also absolutely clear to me that good rendering is something totally different than technical perfection.
In that sense – whatever the outcome of this “beyond thread” will be – I’m absolutely sure about one thing – the resulting system will most possibly not be technically revolutionary but for granted be of great inner harmony .
In the end - that’s what we are *all* after !
It's worth to await – and mess around in the meantime with all the topics at hand - even if it will take some more month' or years - possibly until all of us walking with a cane then
Having said all above allow me to perform an other sacrilege in recommending a webstrem I sometimes manage to listen to.
If you already have digital integrated into your system its worth to join Peter Togni at CBC2 (Choral Concert - Sunday 9 a.m. (9:30 NT))
http://www.cbc.ca/radio2/program/choralconcert.html
If non of his tracks goes subcutan – forget about mass choral in future.
High source quality (Studer, SAC) certainly is fine to have - but if a system can't handle low source quality (FM or MP3 that is) up to the point where you can enter myth of music – forget about your "dream set"
Michael
Last edited:
I got a set of WoodStock CDs a few weeks ago. Initially, it did not sound so good, however, after a polarity swap, it was quite pleasant. I have also found this trend on music of the 50's, 60's, etc. There are so many factors that may influence reproduction that we should really keep an open mind and look at all possible data and try to relate it with what we hear.
There are many things that cannot be described in words. Just about a month or so ago, I put the speakers shown in my avatar on display at a local CD/High End Audio store. The owner confessed that when he just saw the speaker, there was no expectation in what it would even sound acceptable to him; but after listening to music using them, he felt that he was moved by the performance. That was the exact response I was hoping for.
There are many things that cannot be described in words. Just about a month or so ago, I put the speakers shown in my avatar on display at a local CD/High End Audio store. The owner confessed that when he just saw the speaker, there was no expectation in what it would even sound acceptable to him; but after listening to music using them, he felt that he was moved by the performance. That was the exact response I was hoping for.
Last edited:
Chiming in on the topic of recordings for testing systems:
Ray Kimber's IsoMike recordings seem promising (I don't own any yet, but will soon). The IsoMike concept is basically a Jecklin disk writ large. Most recordings are minimalist 2-mic stereo (or 4-mic surround). He records straight from mic to DSD with no compression. Ray has made quite a number of albums with tracks covering a fairly broad range of music. I imagine his marching-band recordings would be another one of those albums that would get you ejected post-haste from most show demo rooms. It would take very special loudspeakers indeed to pass that test with all that brass treble energy.
This from the recording notes found here: http://www.isomike.com/iso2005b.html#about
I don't know know if the recordings are still made this way, but I remember reading Ray's description of a slick trick he used to keep from throwing away resolution on recordings with real-world dynamic extremes. He had custom mic preamps built with dual outputs--one normal, and one at -6dB--so each mic channel records two digital tracks. When setting levels, this allows him to run ~6dB hotter to gainfully employ more bit depth. On peaks where this results in clipping of the normal output, he overlays a splice from the -6dB track, raised 6dB and matched exactly with timecode. I thought that was a pretty ingeniously simple way to maintain low-level digital fidelity without compromising on dynamics--sort of an audio equivalent to how HDR photography works.
Ray Kimber's IsoMike recordings seem promising (I don't own any yet, but will soon). The IsoMike concept is basically a Jecklin disk writ large. Most recordings are minimalist 2-mic stereo (or 4-mic surround). He records straight from mic to DSD with no compression. Ray has made quite a number of albums with tracks covering a fairly broad range of music. I imagine his marching-band recordings would be another one of those albums that would get you ejected post-haste from most show demo rooms. It would take very special loudspeakers indeed to pass that test with all that brass treble energy.
This from the recording notes found here: http://www.isomike.com/iso2005b.html#about
I don't know know if the recordings are still made this way, but I remember reading Ray's description of a slick trick he used to keep from throwing away resolution on recordings with real-world dynamic extremes. He had custom mic preamps built with dual outputs--one normal, and one at -6dB--so each mic channel records two digital tracks. When setting levels, this allows him to run ~6dB hotter to gainfully employ more bit depth. On peaks where this results in clipping of the normal output, he overlays a splice from the -6dB track, raised 6dB and matched exactly with timecode. I thought that was a pretty ingeniously simple way to maintain low-level digital fidelity without compromising on dynamics--sort of an audio equivalent to how HDR photography works.
Last edited:
I definitely agree. My fullrange speakers showed this behaviour immediately and the graphs did not show much of a linearity issue, while energy storage.
I daresay that the issue is more pronounced in fullrange and wideband dynamic drivers.
I tend to ascribe the problem to the micing techniques that destroy any real time domain aspects of the original venue. When there are some twenty different mics all mixed together there is no way that any resemblance of a time-domain event is going to be preserved. The more dense the number of instruments the worse this effect becomes. That's why studio made recording can sound so good since there are so few mics for the individual instriuments and little to no cross talk between different microphone signals. This preserves the original time-domain aspects and some sense of realism is possible.
John - I agree with this as well. What I sense is a poor recording technique in most cases. Knowing that the recording does not have the flaws that we attribute to the speakers or electronics is very difficult to determine, but entirely critical to the validity of any judgements.
Last edited:
Yes it is. Very clever.
Good recordings do exist on "compromised" formats. Like CD and even vinyl. It's amazing what lies in those grooves if you can get it out. That has always surprised and impressed me.
I've heard big symphonic works done right on huge 4-way horn systems, double stacked ESLs - and even on more conventional (but $$$) cone type systems. There was one at T.H.E. SHOW in Vegas this year. From the Netherlands, IIRC. Strain gauge cartridge on the table, too!
My comments were in regard to over excursion and IMD or other nonlinear distortions which are generally very low for large format ELS. You are referring to features which are in the category of linear distortion.
Frankly, I see this as somewhat anecdotal. High quality after high quality system has problems with mass voices/strings. Then a system is stumbled across which seems to handle them well. Statistically it's more likely than not that the systems which seem to perform poorly are actually the more actuate where as the one that sound correct (or better to the ear) is actually the system with a complimentary error. A purely subjective evaluation without any data to back it up, even a simple frequency response measurement.
Again John I completely agree.
I really disagree with classical music being the only venue from which to judge sound quality. I'm not a real fan but I have been to more than a hundred live classical performances, so I'm not a novice either. But the recordings just seem so bad to me, while many other recordings sound fine, even great. It all seems to be source related from what I can tell and if the recording IS bad then it should SOUND BAD and any speaker that makes it sound better is inferior.
I don't know whether anyone has documented what for of problem should sound like what, but normally when the speakers are driven into the non-linear region, the more noticeable thing would be impression of sound being compressed. Most of the time when we consider distortion, unless the distortion in continuous, in the same form for some time, we would consider it as some short term sonic irregularity, and is normally not irritating. Combine this with a long CSD time, the suddenly it is more noticible, and probably will show up as harmonics of some sort. But in a listening test, we would probably not be able to identify it as distortion, and if you currected for this kind of non-linear effect, it may not provide audible improvement if the CSD characteristics is not good enough.
A bit OT but shomwhat related. I attended a meeting on 3D displays recently, and as someone addressed the issue about 30% of the people cannot differentiate between 2D and 3D, I could not help but think how it also applies to audio as well. I further remembered that fighter pilots go through additional visual tests to determine how good distance differences can be interpreted. What this means, is that in reality, very few people have good capability to interprete 3D image. This makes me wonder what percentage interprets 3D capabilty of sound image.