I'm afraid I don't have time to go wading through the thread linked to from the, er, thread linked to in order to find full details of the test setup and methodology, without which the stated conclusions are of limited value. Briefly (and wearily) going through the 5 stated opinions:
1/ Few would disagree
2/ Frequency response dominates (which is not a synonym for 'of sole relevance'), but there is no mention of the test conditions, which in turn affects the relative influence of on & off axis responses. That may have been covered for all I know, it may not, but I'm not about to plough through a 100 page thread to try to find it.
3/ The critical phrase being 'within its mechanical / electrical limits', without directly stating what the drive units in question were, what the stated limits were, what the test conditions were.
4/ Might. They might not. This is opinion sans context or clarity, predicated on a large number of assumptions, not least of them being that said manipulation methods of whatever means are sans issues themselves, which is not in fact the case, or that people even have the facility to implement it.
5/ Yes, they probably will, if they feel so inclined. Especially if the inexpensive units referred to are incapable, even when massive EQ is applied, of doing as desired.
Here's a knotty one for you to mull over. I've got a pair of tweeters on my storeroom shelf. I won't mention the brand-name, they were bought about 5 years ago from a local electronics shop for £25 the pair. FR is acceptable, nothing tragic. The distortion levels are -unfortunate. High across the spectrum, but worse there is giant HD4 and HD5 peaking between roughly 2KHz - 3KHz. Piercing doesn't cut it: shrieks like a banshee does. Perhaps you can explain how EQing the response flat in this region (when it's already reasonably flat and requires no such correction) prevents this? The influence of HD and IMD are often overstated, but the word is 'overstated', not 'of zero relevance', especially when dealing with high levels of high order products.
Last edited:
Well the original poster asked about drivers
And not specifically tweeters albeit they are drivers
But seen as you won’t mention the named drivers how can anyone form an opinion
2-3khz is probably asking too much from them to start with.
Next you will be asking why 15” woofer can’t be flat 20-20 of axis
Yep, that certainly does happen.
The good news is that our ears aren't very good at discerning (relatively) small differences in levels across the frequency domain. That is, it is very unlikely that even the best of us could honestly say that a 1200hz tone is within +-3db of a 1400hz tone. Maybe at 10db but, possibly at 6 but certainly not at 3.
This gives us the privilege of operating on averages rather than absolutes and as I'm sure you know... those averages can sound wonderful.
My last post on this nonsense
And I refer to leaders in this field
Troels
The-Loudspeaker
Secondly danley
Meh synergy
Not a hi-fi speaker driver in sight
The mind boggles 🙄🙄
And I refer to leaders in this field
Troels
The-Loudspeaker
Secondly danley
Meh synergy
Not a hi-fi speaker driver in sight
The mind boggles 🙄🙄
Indeed they are. The same remark could apply to any other type of drive unit.
This coming from the person who posted links to opinion sans supporting evidence. 😉
You couldn't. They haven't been made for years, nor were they supplied with measurements. This is from my own (commercial) data, taken for my own records, given as an illustrative example.
So you now accept that simply EQing a drive unit flat is not all that matters, and they can in fact sound different even when this is done. Excellent, progress is made.
Will I now? And in which fantasy did you invent that?
I don't like to be the one to tell you this, but there are many quality dome tweeters that can handle crossover frequencies well below 2KHz, without excess audible distortion (design context dependent). Be that as it may, the point remains: the wild, contextless statement that 'all you need to do is EQ a driver's FR flat, and they will sound the same' is, bluntly, wrong. Simple as that.
Last edited:
Good-o.
Smashing.
Well, yours might, but mine isn't, and I don't see the slightest indication of anybody else's mind here boggling either. Yes, most are pro-audio units (apart from the Fostex FT96H). So what? Nobody here has said a word against them. Why would they?
The ability to image is related to the “system” to reproduce very small details (ie high DDR), the placement of the speakers, time arrival needs to be very, very close, the ability of the system to keep harmonics within the proper envelope, but also completely dependent of having the critical information available on the recording (rarer than we would like) and the kit between the source & the speaker not remove that information.
Soundstage/imaging is very important to some, me for instance, and all to often wiped out by an XO.
If my system does not do this, it is time for making it better, and i have been quite successful.
dave
"3D soundstage" means height also, how that is suggested by 2 speakers is interesting, ideas on a postcard please... 🙂
A drivers charachteristics are a whole, and individual effects of these things may be able to be generally described, but it is the gestalt of hoe everything comes together.
Some examples… (all from FR drivers, which are amoungst the very hardest to design well).
1/ dustcap. Changing the glue, or the amount can effect things as much as the cap (or anywhere else it is used). In a FR the dustcap often acts as an independent radiator. Smaller it is the better its HF radiation. It does have to be larger than the voice coil.
2/ cone profile. A shallower cone will generally have a wider dispersion character. The most radical i have seen are the new 4 and 6” Mark Audio Alpair 7/11ms where the cone curves outward to a max before it reaches the surround and then dips back down again. Aside: besides giving these quite good dispersion, and superb HF, it is necessary to move the centre of mass to a point where the spider can be eliminated, removing the fight that always goes on between 2 different suspension parts with different characteristics leading to higher DDR. A curve also lends itself to the holy grail of FR where the cone, in a very controlled way, reduces in radiating size as frequncy goes up and the outside of the cone decouples from th einner part of the cone. Cruder ways of doing this include cncentric ribs/rings, and even a seperate “surround” mid cone (ie bi-flex). I expect a whizzer cone could be considered as an extension of this, smaller, with its own curve and various ways of terminating the open edge.
3/ phase plugs. Reduce the cone area, and the dustcap as a source of HF. Acts as a heat sink for the motor (if metal), dramatically reduces side-to-side resonant structures across the cone (actual moves them up in frequency where it doesn’t redirect them outward. I have found them almost universally required in drivers with whizzer cones, but the most dramatic changes in the Visaton B200. Stock these have laser beam like HF that has you either saying to yourself, ouch or what would happen if you added a tweeter depending on where your head is. The cone is very deep and it has a significant rise in FR as it goes up. Remove the dustcap and that removes some of the HF, add a phase plug to divertthe HF radiating from the steepest part of the cone outward, greatly improving the flatness of the FR and making the driver WAY more listenable and a considerably better driver. Of note is the official sensitivity of 95 dB, but a real one of closer to 90 dB, a result of the spec cherry picked off teh rising response.
Tarted up Visaton B200:
An externally hosted image should be here but it was not working when we last tested it.
And a common whizzer cone driver (Fostex FE206eN2)
An externally hosted image should be here but it was not working when we last tested it.
dave
It is certainly true of some, but given the inevitable deviation of natural human hearing capability, and the very significant role of traing & experience, there are those that can have the level of capability alluded too. A quite small number at the top of the curve of the 2 factors convoluted with each other.
So not to be completely dismissed.
dave
Paul, of Paul & Stan. I stil remember meeting these guys (1972) in the original factory in Santa Maria, where i saw my 1st Acoustat 1+1, and it was memorable when they apologized as they dug a bag of pot out of a big black garbage bag, the dregs from when it was full, it was kick-butt.
An astute man, with lots of experience, and many important things to share.
dave
It actually has nothing to do with the equipment in your room.
You have two separate speakers, an amplifier with two independent channels, a preamp, dac, cd player, turntable all also having just two independent channels. Nothing in any of that electronics does anything to create the illusion of a continuous soundstage in front of you...
The entire soundstage is an auditory illusion. It's not really there. It is an artefact of proper mixing in the source recording.
So what you hear is a question of how deeply you buy into it...
What can a driver do to affect soundstage, and considering we've covered fixing the response and knocking out the breakup region.
I came across a summary of DDR written by Allen Wright. This seemed confined to minimum phase issues. Is there other information?
Height is, according to all the hearing texts i have read, an artifact of the way humans (tend) to hear different frequencies. Studies to date have still been crude, we really need to bypass the what i heard, translated to vocalization and instead directly read what is happening in the brain. Tools to do this are just coming to market.
dave
A driver itself has to be able to reproduce very, very small detail.
I was Allen’s computer tech guy for years and perhaps had more exposure than most, DDR was coined by him primarily for his work in electronics, but is even more applicable to loudspeakers where it is, even in the best cases, much less capable. Drivers can have significant differences in this kind of capability.
While very controversial, the prime thing EnABL does to a driver, is increase its DDR, revealing (or better said, not killing) very small details. When i first did extensive listening to the 1st pair i encountered it took me many hours to tease the differences out (ie listen for the right things). Now in a double blind test, i choose a suitable recording, and within 1 or a fes more switches i can accuartely pick the EnABLed driver from one not. The clue is a blossoming 3D soundstage/image as one switches to the EnABLed driver, usually less than 30 sec.
My initial descriptive analogy was: you are flying over a valley, there is a thin cloud cover hiding the detail of the surface, an EnABLed driver is like having a wind come along and eliminating the clouds with the surface revealed in much higher levels of detail.
dave
Believe it or not ... practically nothing.
As I explained to scott, your system is two independent channels, with nothing about any of it the creates a "soundstage". That comes from the audio mix on your source recordings.
Now it is possible to mess it up environmentally, by poorly placed speakers, a listening position not centered or objects between speakers and listener...
I've heard $150.00 Best Buy stereos produce a perfect soundstage through screechy little 3 inch speakers. But I've also heard $10,000 systems that never get there because of the way they are set up.
Last edited:
That seems how it can be implied in a mix, simple as higher frequencies are up, lower frequencies are down, makes sense in the natural world?
Nope,
during listening sessions we replaced one component, a DAC or amplifier and sometimes the soundstage depth was reduced or gone.
Source, amplifiers, speakers all can have an impact on the soundstage.