tomdee:
(on flat frequency response, low distortion/noise, `criteria of good sound')
Quote:
This is true, but its not all.
<font size="2" face="Verdana, Helvetica, sans-serif">I also did mention directivity in the context of loudspeakers but I ought to have been more specific: flat frequency response over a wide coverage angle yielding good power response.
What else am I forgetting?
Quote:
In fact it seems to me, that most new speaker designer consider this as most important,
<font size="2" face="Verdana, Helvetica, sans-serif">Researchers in the field of psychoacoustics have determined this long ago, which is why engineers - old and new - have channeled their efforts into attaining flat frequency response and low noise/distortion.
Again, what is it that all these researchers and engineers are neglecting?
Quote:
in line with high SPL.
<font size="2" face="Verdana, Helvetica, sans-serif">...to increase headroom and decrease distortion at normal listening levels to track the significant improvements in dynamic range and signal-to-noise ratio with improving media...
Have a look at Altec literature from their halcyon days of the 1950s-1970s...
What attributes are highly prominent? High efficiency. Low distortion. High sound pressure level.
In fact, Altec was one of the few companies that prominently featured `maximum sound pressure level at maximum rated electrical input power' on specification sheets.
It would not be too far of a stretch to say that Altec initiated the `SPL Wars'.
Quote:
They measure, they dont listen.
<font size="2" face="Verdana, Helvetica, sans-serif">Oh dear, this statement exhibits utter ignorance of how audio products are designed and manufactured...
Yes, listening tests are the final evaluation of any audio product, particularly transducers (microphones, loudspeakers). However, listening evaluations are the last step in the process.
Normally, we set performance specs based upon what marketing desires or less often, what engineers aspire to achieve. We design. We measure. We revise. We measure. We revise again. We measure.
This cycle continues until the performance specs are met or exceeded. ONLY THEN do we even bother to listen, and nearly every time, meeting performance criteria based on lab measurements is reflected in subjective review.
Floyd Toole, former researcher at the Canadian NRC and now vice-president of Acoustical Engineering at Harmon Int'l has a now-famous statement regarding audio engineering:
`Science in the service of art'
His research culminating in an excellent keynote address at the 111th AES Convention regarding the commendable effort to standardise loudspeaker listening evaluations.
His findings concur with traditional criteria of good sound: people prefer loudspeakers that have flat frequency response, low distortion and good, even power response (flat frequency response off-axis). He also noted that the European Broadcast Union Tech Document 3276-1998 regarding loudspeaker system frequency response tolerance is not stringent enough.
This excellent address is featured in The Audio Critic, issue no. 28. An EXCELLENT, rational publication I might add...
As an aside, based upon this article I retired my Yamaha NS-10 monitors to the closet and purchased an excellent set of Mackie HR824 powered monitors which have advertised (and measured) flat response - within +/- 1.5 dB, 40 - 20000 hz. Dr. Toole is absolutely correct: there's nothing quite like flat response!
The subjectivists' retort about equipment that `measures well but sounds terrible' is an extremely rare occurance; in my experience, it is usually caused by improper measurement techniques or subjectivist reviewers `listening' with their eyes.
Quote:
Altec was made for those who listened.
<font size="2" face="Verdana, Helvetica, sans-serif">I certainly hope so; I reckon it would be quite difficult to market loudspeakers to the deaf.
Quote:
Thats why the almost unmeasurable part of dynamic behaviour while reproducing complex sounds of different instruments is so well mastered by altec speakers.
<font size="2" face="Verdana, Helvetica, sans-serif">Pray tell, what is this `almost unmeasurable part of dynamic behaviour'???
I've not heard of this, nor have I seen it in JAES, JASA, Proceedings of the IEEE, etc. I'm quite fascinated and madly desire to learn what this is.
And since many of the Altec (Anaheim) engineers moved on to competing audio manufacturers, why haven't these other companies ever attained mastery of this elusive characteristic?
Not to say that this doesn't exist; perhaps psychoacoustic researchers haven't found it yet. Still, it's quite odd that Altec would have exclusive mastery of this some thirty years on...
Quote:
High wattage, overhung voice coils may have their charisma, but they will not bring that realistic sound altec had with lightweight underhung voicecoils.
<font size="2" face="Verdana, Helvetica, sans-serif">No, actually they provide even *more* realistic sound, particularly since many of these new, high-wattage drivers have an order-of-magnitude lower distortion than vintage Altec drivers at the same sound pressure level.
There's something to be said about flux control, extensive heat-sinking and computer-optimisation of magnetic circuits, something that was not available to engineers during the reign of Altec.
Quote:
Altecs were fabricated with precision and love, with the goal to sound as realistic as possible.
<font size="2" face="Verdana, Helvetica, sans-serif">Altecs were definitely fabricated with precision prior to the mid-late 1970s; not so certain about love but it's a statistical certainty that there were more than a few fetishists at any given time working for Altec.
Quote:
The realism of playback behaviour of a loudspeaker cant be measured (yet?), it must be heard.
<font size="2" face="Verdana, Helvetica, sans-serif">Yet. At least in one measurement.
However, we can assess the quality of a loudspeaker system with an array of measurements and guess what? We measure for flat frequency response, low distortion and good power response!
Now is this realistic??? For a reference monitor loudspeaker???:
Altec 604E frequency response curve
Remember, the vertical scale is compressed to 20 dB...
BobR