[JaZZ]

Quote:

Originally Posted by SmellyGas /img/forum/go_quote.gif The best form of evidence would be a blinded, controlled, level-matched listening test comparing electrostats. The next best form of evidence might be many people who have done A/B unblinded level-matched comparisons. The next best form would be A/B unblinded non-levell matched comparisons. The next best form would be measurements of frequency response and distortion on two amps in order to show that they were equivalent. The absolute lowest/weakest form of evidence would consist of listeners who are unblinded, spent lots of money, compared two amps perhaps on the same sitting (perhaps not), are influenced by peer pressure, and post on an anonymous internet forum.

Are you sure? According to your logic even product sheets are of more use than Head-Fi reviews. And that's called a scientific approach? I guess so.

Obviously I forgot number three: «A/B unblinded non-level-matched comparisons». But with unsatisfying results they seem to be automatically relegated to number five.
.

 

[jcx]

easy enough to find:

"
Specifications of SR-007 MkII (OMEGA II) Electrostatic Earspeakers
●Type: Push-pull Configuration Electrostatic Headphone
● Frequency Response: 6---41,000Hz
● Impedance: 170KOhms (10kHz)
● Sound Pressure Sensitivity: 100dB/100Vr.m.s. Input/10kHz
● Bias voltage: 580VDC
...
Specifications of SRM-727 II Driver Unit
● Solid state output stage
● Inputs: 2 x RCA, 1x XLR Balanced
● Frequency response: DC-115 kHz (SR-007 or SR-404, when using one unit)
● Rated input level: 200 mV / 100 V Outputs ●Maximum input level 30 V r.m.s. / at Minimum volume
● Amplification: 54 dB (x 500)
● Total harmonic distortion 0.01% / 1 KHz, 100 V r.m.s. output SR-007 or SR-404 Signature
● Input impedance 50 KΩ / XLR balance 50KΩ2
● Maximum output voltage 450 V r.m.s. / 1 KHz
..."

100 dB/100 Vrms with 450 Vrms == 113 dB SPL max before clipping

113 dB is way too Loud for even 1/2 Hr if that were the average level but it isn't necessarily enough for Dynamic Peaks when listening anywhere near live music levels

read HeadWize - Article: Preventing Hearing Damage When Listening With Headphones (A HeadWize Headphone Guide) and look at the live sound peak levels

"Rocking Out" out at OSHA allowed 100dB ave SPL level for an hour or less with dynamic recordings having 15-20+ dB peak-to-ave is way different from all day long backgound music SPL requirements

 

[SmellyGas]

Quote:

Originally Posted by JaZZ /img/forum/go_quote.gif Are you sure? According to your logic even product sheets are of more use than Head-Fi reviews. And that's called a scientific approach? I guess so.

Nope. Product sheets contain SPECIFICATIONS, which can be misleading or just plain inaccurate. If you read carefully, I wrote MEASUREMENTS. This entails a third party who MEASURES the frequency response, distortion, etc. of the amplifiers under a given load. Manufacturer product sheets that list a frequency response, often without a tolerance, and without specifying a load (you can assume a "no load" condition) are next to useless. You seem like an intelligent person - you probably could have anticipated my response already without trying to be difficult.

Quote:

Obviously I forgot number three: «A/B unblinded non-level-matched comparisons». But with unsatisfying results they seem to be automatically relegated to number five. .

Unblinded, non-level-matched comparisons are not very authoritative, regardless of the results.

 

[SmellyGas]

Quote:

Originally Posted by jcx /img/forum/go_quote.gif easy enough to find: Specifications of SR-007 MkII (OMEGA II) Electrostatic Earspeakers ● Sound Pressure Sensitivity: 100dB/100Vr.m.s. Input/10kHz Specifications of SRM-727 II Driver Unit ● Total harmonic distortion 0.01% / 1 KHz, 100 V r.m.s. output SR-007 or SR-404 Signature ● Maximum output voltage 450 V r.m.s. / 1 KHz 100 dB/100 Vrms with 450 Vrms == 113 dB SPL max before clipping 113 dB is way too Loud for even 1/2 Hr if that were the average level but it isn't necessarily enough for Dynamic Peaks when listening anywhere near live music levels

Thank you for taking the time to post some numbers. Unfortunately, I don't agree with your interpretation at all.

First of all, nobody in their right mind plays their headphones at anywhere near those volumes. 113dB??!! That's WAY too loud, even if it occurs dynamically.

Secondly, how on earth does the maximum volume achievable on different electrostatic amplifiers translate to sound quality differences? That's like saying my 1000W amplifier sounds better than your 100W amplifier because it can produce 113dB transients and yours can't without clipping.

I'm disappointed. I thought you might have been on to something.

 

[b0dhi]

"but it isn't necessarily enough for Dynamic Peaks when listening anywhere near live music levels"

 

[SmellyGas]

Quote:

Originally Posted by b0dhi /img/forum/go_quote.gif "but it isn't necessarily enough for Dynamic Peaks when listening anywhere near live music levels"

"That's like saying my 1000W amplifier sounds better than your 100W amplifier because it can produce 113dB transients and yours can't"

 

[b0dhi]

Quote:

Originally Posted by SmellyGas /img/forum/go_quote.gif "That's like saying my 1000W amplifier sounds better than your 100W amplifier because it can produce 113dB transients and yours can't"

What can I say - you can lead a horse to water but you can't make him drink it.

 

[royalcrown]

Quote:

Originally Posted by b0dhi /img/forum/go_quote.gif "but it isn't necessarily enough for Dynamic Peaks when listening anywhere near live music levels"

Can you cite a specific piece of music or other such example that has such a high dynamic peak?

 

[SmellyGas]

Quote:

Originally Posted by b0dhi /img/forum/go_quote.gif What can I say - you can lead a horse to water but you can't make him drink it.

Wow. You completely missed the parallel example I gave. Let me make one last attempt to break it down for you.

JCX's contention is:
1) electrostatic amplifiers that have lower output voltages are capable of producing lower headphone volumes
2) for the 727 + Sr-007 combo, the limit appears to be 113dB
3) since JCX listens to music with 113dB dynamic peaks, the 727+SR-007 will not be able to play music any louder without distortion.

HOWEVER, JCX's calculations say absolutely nothing about the sound quality of the amp+phones when the maximum volume is below 113db. I'm sorry, but I don't music at even HALF that volume, EVER, dynamic or otherwise. Thus, for people who do NOT play music at ear-splitting levels, his calculations are nonapplicable and noncontributory.

The analogous example is: a 1000W loudspeaker amplifier can drive XYZ speakers to 119dB without distortion. However, a 100W amp can only drive XYZ speakers to 110dB distortion. Does this mean the 1000W amplifier delivers better sound quality than the 100W amplifier when listening at NORMAL levels?

 

[Uncle Erik]

Quote:

Originally Posted by SmellyGas /img/forum/go_quote.gif Nope. Product sheets contain SPECIFICATIONS, which can be misleading or just plain inaccurate. If you read carefully, I wrote MEASUREMENTS. This entails a third party who MEASURES the frequency response, distortion, etc. of the amplifiers under a given load. Manufacturer product sheets that list a frequency response, often without a tolerance, and without specifying a load (you can assume a "no load" condition) are next to useless. You seem like an intelligent person - you probably could have anticipated my response already without trying to be difficult. Unblinded, non-level-matched comparisons are not very authoritative, regardless of the results.

You have to measure an amp with a dynamic load, just as a speaker or headphone presents. You can take measurements from an amp being fed into a resistor, but the resistor doesn't change value the way that a driver does. Two amps that measure the same being fed into a resistor can measure quite differently when fed to a load that changes its resistance.

I'd also abandon listening tests. I'd feed the same music through two different amps into the same pair of headphones, then use a good mike to record the result. You could then plot the results over each other and get a representation of the difference. No, I haven't done it, but I think this would show conclusive proof that there are differences. It would also show that some amplifiers sound alike, which I am sure exist, too.

 

[anetode]

Q:
Quote:

Originally Posted by SmellyGas /img/forum/go_quote.gif Secondly, how on earth does the maximum volume achievable on different electrostatic amplifiers translate to sound quality differences? That's like saying my 1000W amplifier sounds better than your 100W amplifier because it can produce 113dB transients and yours can't without clipping.

A:
Quote:

Originally Posted by Dan Lavry /img/forum/go_quote.gif Why bother with more gain then one needs? Can one just “throw it away”? You do not want to just “throw it away”. Both tubes and transistors have too much non linearity. The mirror is curved too much and what you see (or hear) is just too distorted. Some of the extra gain capability is utilized not to amplify, but to linearize the transfer function or in our analogy, to make the mirror flatter.

 

[royalcrown]

Anetode: You're completely misquoting.

The original quote, when read in full, is referring to negative feedback to make the output of an amplifier linear. This is a completely separate issue from what is being discussed in this thread. Negative feedback involves trading gain for linearity, but it doesn't follow from this that excessive voltage is needed to provide a linear response, especially because the decrease in gain from negative feedback implementation would already be taken into account with the output specification.

In fact, the quote is completely out of context, as the original post was actually referring to solid-state vs. tube amps, and how tube amplifiers have a non-linear response because they do not have enough gain. Hardly surprising or controversial, and has nothing to do with the discussion.

 

[anetode]

I understood it as saying that using an inefficient method may sometimes be beneficial in avoiding frequency drop-offs. Then again I know jack crap about the mechanism feedback in an amp and was quoting something that I thought explained the discrepancy in power consumption. Sorry. For future reference, doesn't the efficiency rating of an amp directly correlate to the max output? Are you saying that the standardized power consumption rating takes things like negative feedback into account? I must be thinking of two different things here.

 

[SmellyGas]

Quote:

Originally Posted by Uncle Erik /img/forum/go_quote.gif You have to measure an amp with a dynamic load, just as a speaker or headphone presents.

That would be nice. I'm sure someone has done it with RMAA. I'd be interestded.

Quote:

You can take measurements from an amp being fed into a resistor, but the resistor doesn't change value the way that a driver does. Two amps that measure the same being fed into a resistor can measure quite differently when fed to a load that changes its resistance.

This is a common "belief" that is propagated here. I don't necessarily think it's true, even though it would "make sense" that it is true. There is a lot of audio mythology that "makes sense" but doesn't pan out to audible differences. In fact, if you look at loudspeaker amplifier data, when The Audio Critic tests them with stressful inductive, capacitative, and low impedance loads, only rarely does even the most stressful load substantially influence the measurements of interest to audibility.

Quote:

I'd also abandon listening tests.

One of the basic tenets of experimental design is to directly measure the entitity of interest. If you want to know if XYZ change is AUDIBLE, you need a LISTENING test. Simple. However, sometimes if you have validated measurements, they can be used as a surrogate. For instance, if you know that frequency response variations within 0.2dB from 20-20k in a large population is not audible, this can be measured. But what if it's 0.3dB? What if it's a 1dB peak, but located at 10kHz?? You don't know. That's why you need listening tests.

Quote:

I'd feed the same music through two different amps into the same pair of headphones, then use a good mike to record the result. You could then plot the results over each other and get a representation of the difference. No, I haven't done it, but I think this would show conclusive proof that there are differences. It would also show that some amplifiers sound alike, which I am sure exist, too.

Someone should do this. The best design would involved a standardized dummy head with binaural microphones.

 

[royalcrown]

Quote:

Originally Posted by anetode /img/forum/go_quote.gif I understood it as saying that using an inefficient method may sometimes be beneficial in avoiding frequency drop-offs. Then again I know jack crap about the mechanism feedback in an amp and was quoting something that I thought explained the discrepancy in power consumption. Sorry. For future reference, doesn't the efficiency rating of an amp directly correlate to the max output? Are you saying that the standardized power consumption rating takes things like negative feedback into account? I must be thinking of two different things here.

Power output does take negative feedback into account, but that's an awkward way of phrasing the question. In very simplified terms (because I don't have that good of a handle on the math), negative feedback trades off gain for a linear frequency response - however, that NFB is built into the circuit, and the output that is measured is of the finished amplifier, once the circuit is finalized and the actual product built. Two amplifiers may have an output of X, but they may utilize NFB or other such things differently, but at the end of the day things like NFB are used in designing a circuit. Once the circuit is completed, however, having additional "headroom" or excess power output is not going to make your frequency response more linear.