Sound Quality Vs. Measurements

Status
Not open for further replies.
Hi,

However, now that you mentioned it, I seem to remember that in an interview, Mr Johson did mention that he and his colleague Mr Conrad did actually design some of their first products ...

Is that too hard to bear? :p

Not at all, rather, if the very early stuff I had to fix was designed by economists and bankers it would explain the "interesting features".

Ciao T
 
T, at least you pointed out WHY we, in the PA industry, used limiters. We needed them to PROTECT the loudspeakers! The idea of needing them for home hi fi is almost a joke. Just get a slightly bigger amplifier, OR bi-amp. That will fix your problem, properly.

No, it's not a joke. Home users often don't have a slight idea how often their equipment clips. They don't understand that they don't have a headroom they want. They just think that equipment is bad. When my limiter works it flashes overload LED on the front panel. Compression sounds less nasty than clipping, and if the user does not want it he/she would follow the advice to decrease signal level when bright red light starts flashing.
 
diyAudio Member RIP
Joined 2005
T, at least you pointed out WHY we, in the PA industry, used limiters. We needed them to PROTECT the loudspeakers! The idea of needing them for home hi fi is almost a joke. Just get a slightly bigger amplifier, OR bi-amp. That will fix your problem, properly.

When I worked on the electronics for OEM computer speakers, the primary objective was to avoid having a customer call and complain. So if the typical source, the computer soundcard, had an output sufficient to drive the amplifier into clipping, this could prompt calls unless there was some way of limiting. Somehow the knowledge possessed by most everyone in the home hifi arena about turning the volume down when you heard unacceptable distortion just went out the window in the computer domain.

There was usually not enough power to damage the actual drivers, although with process and design issues for these one can indeed get into trouble.

And of course, any whole system had to be dirt-cheap. However, the volumes justified spending a good deal of time optimizing things. If you are going to ship a few million units, saving a few pennies is warranted even if it takes a few days of work.

My favorite design managed to get a full-wave detector with complementary polarity outputs out of one transistor, which in turn drove a four-transistor current-steering cell wrapped around a power amp which incorporated a two-pole underdamped highpass, whose low-frequency boost got attentuated preferentially when things began to compress. A little bit of mid-dip EQ compensated for the 2Pi-4Pi transition in the midrange, and overall it was a pretty decent pair of speakers to flank a monitor. One of the nicest parts was that the industrial design was, for once, something good for acoustical directivity, with smooth, rounded features that minimized diffraction.

I can listen to music through them without fretting, casually, although obviously with limited low-frequency extension and overall SPL. Anything but high-end though! The bill of materials cost was shockingly low.


Brad
 
Last edited:
Hi,

It turns out that economic systems can exhibit behaviour analogous to phase transitions. To some extent this explains the world crash, which most economists did not see coming.

Absolutely. Economics as now generally practised attempts to throw a corrective feedback loop around a non-linear system, with considerable stochastic non-linearities, "memory distortion" and it does so not in any direct way, but attempts to do so by adjusting power- ... ooops... money-supply, creating both massive lag (phaseshift) and leverage (loop gain)...

So, if anyone wants to know what is "wrong" with negative feedback, take the current economic troubles as an illustration what a badly designed feedback system does.

Indeed, one might argue that rather than trying to control the economics systems perceived non-linearities via global feedback systems, it may actually yield better results to simply abandon such feedback systems and where corrective action is needed to apply this on the most local level possible, in favour of maximising the accumulated non-linearities and to then use massive leverage to try to correct things when they really get off track.

But perhaps this is where the analogy breaks down.

The Economy is actually not a simple mechanical system (as theorists would an Amplifier be) or even highly complex mechanical system (as the practicians find real amplifiers to be), but involves large numbers of both irrational (pun intended) and easily influenced elements which are so irrational that in the trivial pursuit of any short term gain they can be easily made to act deliberately against their direct own long term interests...

Thankfully Amplifiers are generally rather devoid of such elements, lest you live too close to a Training centre of the Church of LRH and your Amp is pre-clear. In which case it is entirely possible it was taken over by one of the Body Thetans that got shed by the Clear's progressing to OT Level VIII, who now thinks it's a transistor, capacitor or resistor in your amp - in that case all bets are off...

Ciao T
 
I guess I just use bigger amps, because my hi fi does not clip very easily or often. The smallest amp that I use is over 100W/ch.

Pyramid prototypes I made are used with Magnepans that are not efficient, and according to one user 80W / Channel is not enough for his taste, but he keeps it because it sounds better than all he tried before. Though, the same amps driving line arrays on campgrounds were adequate.
All depends on efficiency of speakers, room sizes, and individual tastes. If I can add some fool-proof so the user would not blame on me for own mistakes, when such mistakes are common, I will do that.

The only case when such protection was considered as bad, when one DJ tried to get "right sound" from my amp. I helped the guy pulling down master faders on console, pushing up faders on channel strips that have sharp diode limiters. He was satisfied. :D
 
Last edited:
diyAudio Member RIP
Joined 2005
Hi,

[snip]


Thankfully Amplifiers are generally rather devoid of such elements, lest you live too close to a Training centre of the Church of LRH and your Amp is pre-clear. In which case it is entirely possible it was taken over by one of the Body Thetans that got shed by the Clear's progressing to OT Level VIII, who now thinks it's a transistor, capacitor or resistor in your amp - in that case all bets are off...

Ciao T

;)

They are NOT getting their hands on my amplifiers!
 
Hi John,

I guess I just use bigger amps, because my hi fi does not clip very easily or often. The smallest amp that I use is over 100W/ch.

I guess I just ratchet up the Speaker Efficiency instead.

The highest efficieny speaker I have used long term was 104dB/1m/1W (not 2.83V, but 1W) the lowest ones often hover in the lower 90's...

Of course having a really quiet 150WPC or bigger Amp on a pair of near 100dB efficiency Urei's or similar main studio monitors is seriously silly grin inducing.

I occasionally clip my 35W Tube amp into my current 91dB/W/m speakers, but it's a tube amp with no global looped feedback so it can be overdriven quite hard before it goes from sounding "peak compressed" to "distorted", probably on the region of 6dB peaks above nominal power...

When the Amp goes from compression to obvious distortion I usually fear the neighbours in my Condo will pull my fuses.

Ciao T
 
Hi,

My favorite design managed to get a full-wave detector with complementary polarity outputs out of one transistor,

Cute. I usually can afford an 084 in the BOM if I need that though.

which in turn drove a four-transistor current-steering cell wrapped around a power amp which incorporated a two-pole underdamped highpass, whose low-frequency boost got attentuated preferentially when things began to compress.

Yuck. Fets are really not THAT expensive...

A little bit of mid-dip EQ compensated for the 2Pi-4Pi transition in the midrange, and overall it was a pretty decent pair of speakers to flank a monitor. One of the nicest parts was that the industrial design was, for once, something good for acoustical directivity, with smooth, rounded features that minimized diffraction.

At a guess, the **** Fader design from Just Brutally Loud?

I can listen to music through them without fretting, casually, although obviously with limited low-frequency extension and overall SPL. Anything but high-end though! The bill of materials cost was shockingly low.

Yes, it's amusing just how low one can go and still enjoy music casucally. My office has a Linksys "Wifi Radio" with some modest tuning# that streams music from my main Media Server or internet radio, I prefer to have it on, rather than off...

Ciao T

# to wit cheap bamboo chopsticks liberally applied to speaker surfaces and the 4" Driver spamped basket with 2-component epoxy, some Elna Silmic Cap's and around 0.03F in extra PSU capacitance for the Amp's (TDA2003), tweeter crossover cap changed from bipolar 'lytic to Film, most RCA and 3.5mm Jacks bypassed and hardwired, some Os-Cons in the digital section [which has a fairly high end AKM 24 Bit DAC actually] and an AD8066 instead of the original horrible Op-Amp as analog stage
 
Of course T, there are exceptions, AND I often forget that most people have low power mid-fi amps, and low efficiency speakers. I have done this combination (with some success) with an Electrocompaniet and a pair of WATT PUPPY's, and as well driving the same WATT 1 loudspeaker with a DYNA MK4 (40W on a good day). Here, I have to draw the line, and admit that these combinations or even worse ones, exist, AND clipping is possible, but I never found a need for limiters, even then, because the Electrocompaniet or the Dyna handled peaks very well, AS they are both low feedback designs.
I can certainly see how a 10W (on an exceptional day) solid state power amp built into a sound card, can clip, and clip badly!
 
diyAudio Member RIP
Joined 2005
The powered speaker (not a soundcard amp) was actually an "aftermarket" product sold in the Apple stores and otherwise online, branded JBL. The power amp was a very tricky-to-make-stable TDA2007A class AB amp, which has, I understand, now finally been discontinued.

The transducers were Clayton Williamson specials with remarkable x max for their size, and some pleated surround invention that enhanced linearity a bit (but there is still a boatload of IM distortion that will show up when pitched transients are mixed with significant low frequency energy --- but that's where the frequency-dependent compressor helps a lot).

The four-transistor cells, using unmatched 2SA1162GR PNPs (but usually from the same wafer) worked very well except for the approximation to a properly translinear Gilbert circuit, in that it lacked the predistorting diode-connected parts. So there's a little propagation of control voltages, but not reproduced with the limited LF extension to speak of. 1162s were about 1.2 cents then in high volume, an incredible bargain; not sure how much now. FETs are problematic for adjustment-free circuits owing to the huge spreads in production on pinchoff voltage etc., although one can use them as choppers. Many are the ways :)

But I really think the thing I like besides the circuit tricks (I look at the schematic and each time forget how it works for a while!) is the smooth curved baffle. Within the limitations of a single-transducer system, it has very acceptable dispersion.

Harman shipped about 800k of them the last time I checked, and made an obscene amount of money. If I had only been smart enough to work out a licensing deal :( They have also been remarkably reliable, although I don't know the exact numbers.
 
Any decent economist would be quite good at maths. The weakness of economics is that when creating their equations they make false assumptions about how markets work etc. They then find a correct solution to the wrong equation. Physicists are now beginning to muscle in on economics, using either better assumptions or (shock!) the results of actual experiments or simulations of how people really interact. It turns out that economic systems can exhibit behaviour analogous to phase transitions. To some extent this explains the world crash, which most economists did not see coming.

You have a clear oxymoron in this statement. "Decent economist". If you put 12 economists in a room and ask them what happened yesterday, you will get at least 13 answers. Economics are nothing but some tools to describe one small aspect of sociology. BTW, my minor was econ. :p
 
Any decent economist would be quite good at maths. The weakness of economics is that when creating their equations they make false assumptions about how markets work etc. They then find a correct solution to the wrong equation. Physicists are now beginning to muscle in on economics, using either better assumptions or (shock!) the results of actual experiments or simulations of how people really interact. It turns out that economic systems can exhibit behaviour analogous to phase transitions. To some extent this explains the world crash, which most economists did not see coming.

What you said is very near the truth. Macroeconomists in particular tend to regard the world and its markets as a set of equations, which is very foolish because you can't pack people into equations. Motives are often other than economic, politics play an ever larger part in the working of the whole world. Since politics are rather prone to sudden changes and swings, and disregard laws of economics all too often, their assumptions are very often based on questionable trends.

Another problem I've noticed is that virtually none of those macroeconomists has even one second of time spent in an actual, real world company. What to them appears to be statistically insignficant may actually lead to the ruin of companies who are unable to profitably implement what was envisaged by some smart *** in a mahogany office. But, that's also the key problem of politics the world around, I think.
 
Hi John,



I guess I just ratchet up the Speaker Efficiency instead.

The highest efficieny speaker I have used long term was 104dB/1m/1W (not 2.83V, but 1W) the lowest ones often hover in the lower 90's...

Of course having a really quiet 150WPC or bigger Amp on a pair of near 100dB efficiency Urei's or similar main studio monitors is seriously silly grin inducing.

I occasionally clip my 35W Tube amp into my current 91dB/W/m speakers, but it's a tube amp with no global looped feedback so it can be overdriven quite hard before it goes from sounding "peak compressed" to "distorted", probably on the region of 6dB peaks above nominal power...

When the Amp goes from compression to obvious distortion I usually fear the neighbours in my Condo will pull my fuses.

Ciao T

I seem to have something not quite, but approaching that. My speakers are rated at 92 dB/2.83V/1m, and are often powered by a Karan Acoustics KA-i180 integrated amp, delivering nominally 180/250W into 8/4 Ohms.

On top of which my room is rather small, about 4.5x3.5 m (app. same in yards), so basically I never really needed much power. So, theoretically, I have 22 dBW in tap, which should give me a theoretical (92+22) 114 dB SPL, not counting the fact that there are two speakers at work, or the room reverberation, etc.

In practice, I have never managed to go past the 10 o'clock mark on the volume control because by then, my window panes start to boogey with the music, they get into it, man. :D
 
Thorsten, this world is no longer ruled by econonmics, it's ruled by out and out greed. Examples are so many, it's hard to pick any one out.

But think of this. We hear day in, day out, how USA has an incredible national debt, how Greece does as well, as do Spain, Portugal, Ireland, etc, etc, etc. Now, when you use "maths" to simply add up all those numbers, you have to ask yourself - if one country borrows, surely it borrows from another country which has a surplus it is willing to lend, right?

But, all those borrowings, when added up, total the world's three year planet Earth production. And since just about everybody owes, one needs to ask: where does all that money come from?

The answer is painfully obious - from national money printers. That "surplus" in effect does not exist.

Which means that you are out of economics and head on into politics; in economics, if you are serious about it, you deal with real resources, not imaginary. In the end, as the theory of accounting dictates, the balance sheet must equal zero, and in our case, it has long stopped being zero. The entire system has been hijacked into imaginary assets and as such, it can no longer be called economics.

A much more appropriate name would be a con.
 
A much more appropriate name would be a con.

I've found this world 10 years ago, it is Recyclonomics.

If ROHS caring of recycling industry dictates what to use in manufacturing, it is Recyclonomics. Need for faster recycled goods dictate needs for investments that are borrowed from the future profit from sale of goods being recycled.

Edit: economists obviously keep out of equation the fact that resources when recycled loose their value, and Mother Earth can't replenish them with the speed they anticipate.
 
Last edited:
if one country borrows, surely it borrows from another country which has a surplus it is willing to lend, right?

No , it s lended from private banks who are allowed to lend 20$
for each 1$ they have as deposit.
As such the private banks , who had been given the right to create
money out of thin air , did in fact lend even the money that they
didnt detain , that is , your money , my money , everybody s money.
But, all those borrowings, when added up, total the world's three year planet Earth production. And since just about everybody owes, one needs to ask: where does all that money come from?
As said , it comes frome nowhere , it s pure book entries , that is
0 and 1 on hard disks...
The amount of derivative contract in notionnal value is currently
about 900 000 bn$ , fifteen times the annual world production.

Which means that you are out of economics and head on into politics; in economics, if you are serious about it, you deal with real resources, not imaginary. In the end, as the theory of accounting dictates, the balance sheet must equal zero, and in our case, it has long stopped being zero. The entire system has been hijacked into imaginary assets and as such, it can no longer be called economics.

A much more appropriate name would be a con.

That s it , what is called the financial world create false money
and use it to buy all existing ressources.
The only scientific trait is that it s the application
of the law of least effort , that is , optimisation
of financial preadation...;)
 
Status
Not open for further replies.