advertisement


Why some are stand-mount people and others floorstanders

I’m of the mindset that frequency response actually tell you comparatively little about whether a speaker is good or not. The human ear is just not that sensitive to a db or three here and there, but we really can hear timing or phase issues.


I couldn't agree more.
 
Tony, read the Stereophile link carefully and the comments. It appears the mid tweeter is not low passed much and is overlapping the top tweeter. The change from the original BBC design was that the mid tweeter went much higher.
The result is serious combing between the two which could well be the "phasing" you object to.

No, its not that as the behaviour was common both to the C7 ES2 (one tweeter) and SHL5 (two). FWIW I liked the top end of the SHL5, it was just the crossover region that was wrong to my ears/taste.
 
Stand mounters with cabs that would probably survive a nuclear strike allied with a pair of subs in the 1/3rd/corner configuration and it sounds vast. Subs should only really be apparent when they are reproducing detail that most normal speakers simply can't handle. A really good test for sub integration is a well recorded car door being closed. There should be no smear or overhang yet a certain reassuring weight behind the sound.

My experience is that, if you can mount the mids and treble units in as far as possible an acoustically inert cab, match them with a pair of sealed subs, it's great place to start and gives all the benefits and few of the disadvantages of a large floorstander.
 
I would guess that companies sell far more stand mounting than full size , but make more profit with full size models... remember my first floor mounted quad 7s and celestion 66s
 
You were saying? He-he.
maxresdefault.jpg
 
The reduction in response discontinuities is what I assume is meant by coherence. I have so far never been troubled by this, so correct me if I'm wrong.
Discontinuities in frequency and power responses can be easily measured, and those play a part in perception of coherence. But it is discontinuity of relative phase in the crossover region that makes a bigger impact on coherence. A pair of drivers can still sum correctly and produce a flat FR over their crossover range, but still sound wrong. This is the reason when I design loudspeakers, my first priority is to get the relative phase of driver pairs to be as perfectly aligned as possible, before I try to flatten the FR.
 
I’m of the mindset that frequency response actually tell you comparatively little about whether a speaker is good or not. The human ear is just not that sensitive to a db or three here and there, but we really can hear timing or phase issues.

I'd go further and say that frequency response graphs can be misleading, but there are times when It can show an obvious problem like nasty breakup from a metal coned woofer not being taken care of.

Dips In the frequency response over a small bandwidth aren't usually noticed, even severe ones that look very bad on paper, but over a wide bandwidth even small differences are easy to hear. Increase or decrease the tweeter level by 1db and It will be very obvious.
 
I’m of the mindset that frequency response actually tell you comparatively little about whether a speaker is good or not. The human ear is just not that sensitive to a db or three here and there, but we really can hear timing or phase issues.
No single measurement is enough to characterise performance of speakers (nor electronics), you need a comprehensive set for that.
 
No single measurement is enough to characterise performance of speakers (nor electronics), you need a comprehensive set for that.

Probably very true. Sadly the more vocal online ‘measurement preachers’ only seem to grasp the most simplistic frequency response plots, which for me are all but irrelevant. I’d love to see more discussion of phase, time, dynamic headroom etc etc.
 
Here are pictures fr , phase , group delay and linear impulse response using my target curve.. so what does that tell you my system sounds like...?

Green trace is left , blue trace right

Target curve

34274533_932197850294150_7306529721787875328_n.jpg


Amplitude

34199265_932191953628073_3627565176754012160_n.jpg


Group delay

34268610_932191886961413_7158226236223782912_n.jpg


Phase

34178155_932191883628080_4003918278840811520_n.jpg


Impulse response

34392910_932191880294747_3200375097995034624_n.jpg
 
Here are pictures fr , phase , group delay and linear impulse response using my target curve.. so what does that tell you my system sounds like...?

Green trace is left , blue trace right

Target curve

34274533_932197850294150_7306529721787875328_n.jpg


Amplitude

34199265_932191953628073_3627565176754012160_n.jpg


Group delay

34268610_932191886961413_7158226236223782912_n.jpg


Phase

34178155_932191883628080_4003918278840811520_n.jpg


Impulse response

34392910_932191880294747_3200375097995034624_n.jpg

Room measurements provide information about in-room performance and room acoustics characteristics.
Your target curve is a bit different from what is suggested by B&K but you're not alone, many prefer to draw their own curve to taste:

160503_Blog_AcousticBasisHarmanTargetCurve_Photo_Img3.jpg


I presume that the frequency response plot was measured at the listening spot.
Did you measure one speaker at a time?
How much smoothing did you apply?
 
All at listening position , 1/3rd octave smoothing for the graph display , 1 speaker at a time and a combined one ... Trinnov software and their measuring 3d mic.
Heres a pic at 1/6th octave and then at 1/12th
the target curve you apply is not directly translatable to actual measurement when is is applied , you can see that from the amplitude graph.. my system is capable of profound deep clean crisp bass at ungodly levels so I like it when the bass is a little wound up , if it's tight its right

34177448_932227120291223_8414737060610441216_n.jpg


34138225_932227113624557_2546858788405641216_n.jpg
 
Probably very true. Sadly the more vocal online ‘measurement preachers’ only seem to grasp the most simplistic frequency response plots, which for me are all but irrelevant. I’d love to see more discussion of phase, time, dynamic headroom etc etc.

I go to a lot of classical concerts and my system has to reproduce the feel of live, enthusiastic players when present. In that sense I am more flat earth than High Fidelity. I am not interested in 'trouser flapping' bass so standmounts are fine and more precise?
 
I’d love to see more discussion of phase, time, dynamic headroom etc etc.
True, I would add dispersion vs frequency to that list. Steep crossover between a very directional mid bass and the omni-directional tweeter is going to be very audible off axis and reflected.
 
What intrigues me is why makers of such fantastic sounding standmounts can't translate that into their bigger designs? Lots of people love small Proacs, Spendors and Harbeths for example, but most of their bigger stuff leaves people indifferent at best, why is that?
 
What intrigues me is why makers of such fantastic sounding standmounts can't translate that into their bigger designs? Lots of people love small Proacs, Spendors and Harbeths for example, but most of their bigger stuff leaves people indifferent at best, why is that?

My guess is that some people enjoy the limited low frequency extension which subjectively may produce an apparently "faster", "tighter" bass, and the wide dispersion that increases the "soundstage" effect.
 
What intrigues me is why makers of such fantastic sounding standmounts can't translate that into their bigger designs? Lots of people love small Proacs, Spendors and Harbeths for example, but most of their bigger stuff leaves people indifferent at best, why is that?
Because a standmount is usually a two way with a relatively small mid bass.
A floor stander has the choice of two way with a much bigger or two parallel mid bass units or going to three way with the crossover group delay complications that causes
 


advertisement


Back
Top