|
Audio Asylum Thread Printer Get a view of an entire thread on one page |
For Sale Ads |
72.186.82.177
In Reply to: RE: upgrades exist outside of one aspect of a xover topology. posted by Edp on November 19, 2007 at 12:32:47
Even worse, the drivers are non-coincident.
The only true (measurable) 1st order designs that I know of are Thiel. There may be others, but I really don't care, for the reasons as mentioned.
cheers
Fermentation may have been a greater discovery than fire
Follow Ups:
GMA is committed to and executes the 1st order phase coherent (over a range), time coherent (at a specified distance) very well. There should be no doubt that GMA products are well engineered for the choices and commitments that they find important. They have the chops and commitment.
Their spin is a bit heavy on "this is difficult and you should thank god we are here to make it happen" stuff, almost denying others doing now and in the past to same level of execution/commitment. In his Prime, I'd take 3 to 1 on John Dunlavy taking Roy Green to the mat, technically and maybe even physically though! ;^)
I'd bet on Roy's hair to beat up John's hair even on a windy day, but wouldn't want to take a punch from either one of them.
Let me toss out a bit of speculation on the observed sound quality of GMA, Dunlavy, Vandersteen and Theil:
The ear is good at hearing phase issues below about 1500 Hz, but not as good at higher frequencies. This is because the nerves in the inner ear are incapable of firing faster than that, so the ear hears those high frequencies by another mechanism (placement of the hair cells) that is better at detecting intensity (as partially revealed by the Fletcher-Munsen curve). The dramatic and obvious waveform deformation shown in graphic depictions of a phase shifted signal is of much greater visual than audible significance because (according to published research) the ear isn't highly sensitive to such issues. I'm not saying that phase coherence or linear phase shift aren't desirable, but it's my understanding is that the ear is more sensitive to other things like power response, resonance, and diffraction.
Diffraction?? But doesn't diffraction only have a very minor effect on frequency response? Yes, but that's not where its significance comes from. You see, the ear has a characteristic called "masking" that will conceal the presence of a low-level signal when there's a simultaneous high-level signal at or near the same frequency. The ear is very good at masking in the frequency domain, but very poor at masking in the time domain. Because of the path length difference, diffraction arrives later in time than the main signal - and therefore is not concealed by the ear's masking characteristic. So diffraction is picked up by the ear more easily than it is typically picked up by test instruments (unless they are very precisely time-gated), and the diffracted signal is distorted and tends to impart a harshness to the sound. The diffracted signal is also prone to skewing the image, as it can be interpreted as a (false) source. Note that many loudspeaker systems are perceived as sounding more harsh at high volume levels, when in fact the loudspeaker is still operating linearly (there has been no sudden rise in measureable distortion) - this may be related to diffraction in many cases. Recent research indicates that the ear's perception of non-masked distortion - such as diffraction - is non-linear; that is, we don't hear it at low or medium volume levels but we do hear it at high volume levels.
So to get back to GMA, Dunlavy, Vandersteen and Theil, notice how each of these designs goes to great lengths to eliminate diffraction. Presumably they did this to improve the impulse response, but in doing so they minimized a source of audible distortion that is under-appreciated and neglected by many designers. I wonder which really makes the greatest audible contribution - the first order crossover, or the elimination of diffraction.
Now Roy and John and Richard and Jim will probably want to take turns pummelling me.
Duke
Sharp corner felt removed Spica TC 50 still maintains its minimum phase change (and not 1st order only) over the 200 to 6000 range, so don't know that diffraction is a masking issue. Imaging although down still is fairly impressive.
If you look at the Europa crossover, you'll see a bunch of funky impedance equalization in addition to the base crossover. I would be surprised if this wasn't at least partly the case in all of those designs.
Note that the real proof is in the pudding. Regardless of technology, how does it sound?
I got to hear the GMA Calypso again at the RMAF and was pretty impressed. They just sounded like music and were quite listenable. I spent a fair amount of time with them. What surprised me is the lack of lobing when changing my seating height. What surprised me more is that a woman actually complimented the looks in the hallway outside. My wife begs to differ on the styling.
Bill
Cos I thought s'phile had published some GMA reviews with step shapes that look as good as those in their Thiel reviews....
- This signature is two channel only -
FAQ |
Post a Message! |
Forgot Password? |
|
||||||||||||||
|
This post is made possible by the generous support of people like you and our sponsors: