[ultrabike]

Quote:

 
Well, obviously measurements don't mean very much to an objectivist that can't hear a difference.

 
LOL!

 

[Steve Eddy]

Quote:

Can someone either confirm or deny the assertion made earlier in the thread  that full Class A amps take the power out of the equation (assuming you have a reliable 120V/230V source, of course) ?  Does that mean owners of class A amps can dispense with conditioners and expensive power cords etc ? 

 
Depends on the amp. A typical push-pull tube amp when running in class A will draw a constant current from the power supply. However a typical solid state complimentary push pull amp running in class A will draw current from the power supply that's modulated by the signal. But if you take two of those complimentary push pull amps and bridge them together, when running in class A they will also draw a constant amount of current from the power supply.
 
As for power conditioners and whatnot, that's another story. And although the amp may be drawing a constant current, the main reservoir caps in the power supply are still being drawn down and refreshed every 1/120th of a second so without regulation you'll still have power supply ripple on the supply rails. Just that by drawing a constant current, you won't end up with the signal on the supply leads (by way of their resistance and inductance) which can potentially feed back into the amplifier causing oscillation.
 
se

 

[Magick Man]

which can potentially feed back into the amplifier causing oscillation.

I take it that's a bad thing. Is it audible? What does the oscillation sound like?

 

[Steve Eddy]

Quote:

Actually it does, because there are people on this forum whose ears I trust, and both sources of information are worth more together than apart.
 

 
I'm sorry, but without controlled listening tests, it doesn't challenge or support anything  because there's too much ambiguity with regard to purely subjective experience.
 
Sure, in the purely subjective realm, you can trust whatever you want. Someone's ears, the alignment of the planets, whatever. But it ain't science.
 
se

 

[Shike]

Quote:

...based on a misunderstanding of the application of CSD plots, the primary purpose of which is not to measure decay (although that is a secondary consideration), but to see driver resonances and to a slightly lesser extent, internal enclosure reflections. But of course the astute objectivist would already know this.

 
I know speaker builders that would argue saying it's more beneficial for finding cabinet resonance than driver issues unless they are manufacturing the driver themselves.  In other words, driver resonance probably wouldn't even register on their usage assuming the driver is decent (how I've used it various times in the past).  There was an original usage for CSD, but it's useful for a lot of things obviously and as such the initial or primary usage isn't really important to the discussion.
 
I'm perfectly familiar with the traditional usages of CSD, but many decent headphones don't have enclosure resonance worth mentioning.  As for driver resonance, that's part of decay inherently (in other words, time to settle).  Unless of course you're talking about exciting harmonics, which I guess is at least semi-interesting (still comes back to decay, in other words how bad is said resonance).  So I don't really get what you're trying to go on about there.  Either the decay is bad enough to be audible, or it isn't - this includes resonance/ringing obviously.
 
PS: Nice try with the backhanded astute objectivist comment BTW, but it's tired if you use it twice.  Want to make another assumption?  It's very becoming of you.
 
@ultrabike:
 
 
Quote:

More importantly, issues conveyed through CDS plots have been correlated to inaccurate sound reproduction. So I wouldn't be too quick at dismissing their importance.

 
I'm not saying they're entirely worthless, but rather a good eyeball to make sure nothing is largely of concern.  Compared to speakers above average headphones tend to perform very well.  I'm saying there's people that scream over a hundredth or tenth of a millisecond in areas over other comparable headphones and, as such, I grow tired of it as I've yet to hear a decent headphone where it was a severe enough of an issue in the first place for me to hear it.  For example the "Stax" speed I keep hearing about.  Bought some Stax . . . and wasn't impressed in the end.  In fact, none of the people that I had try them picked them over any other headphone I had.  They sounded like my K601s more than anything else, and were promptly sold within a week.

 

[Steve Eddy]

Quote:

I take it that's a bad thing. Is it audible? What does the oscillation sound like?

 
In some cases it may not sound like anything because it's not severe enough. In other cases it can sound like high levels of distortion as the amp will often clip the input signal. In other case still it may not sound like anything because the amp's destroyed itself.
 
se

 

[purrin]

It could be science if he had sufficient data points and tabulated the data on a excel spreadsheet. Heck, in medical studies for the efficacy of painkillers, they do ask people to rate pain on a smiley face / frowny face scale.

 

[rhythmdevils]

Quote:

 
I know speaker builders that would argue saying it's more beneficial for finding cabinet resonance than driver issues unless they are manufacturing the driver themselves.  In other words, driver resonance probably wouldn't even register on their usage assuming the driver is decent (how I've used it various times in the past).  There was an original usage for CSD, but it's useful for a lot of things obviously and as such the initial or primary usage isn't really important to the discussion.
 
I'm perfectly familiar with the traditional usages of CSD, but many decent headphones don't have enclosure resonance worth mentioning.  As for driver resonance, that's part of decay inherently (in other words, time to settle).  Unless of course you're talking about exciting harmonics, which I guess is at least semi-interesting (still comes back to decay, in other words how bad is said resonance).  So I don't really get what you're trying to go on about there.  Either the decay is bad enough to be audible, or it isn't - this includes resonance/ringing obviously.
 

 
Fail post of the year. 

 

[Shike]

Quote:

 
Fail post of the year. 

 
Since you're so enlightened I'm sure you will highlight said inaccuracies?

 

[rhythmdevils]

no

 

[purrin]

Quote:

 Want to make another assumption?  It's very becoming of you.
 

 
Yes. I still think you lack some understanding. Everything you've said only confirms this with me.
 
CSDs are of little concern to me as a speaker builder because speaker drivers (woofers, midranges, etc.) are crossed over well before they ring. As long as I can bury the resonance with a deep enough crossover slope, I am fine. Also, speaker measurements are much easier than headphone measurements because they lack issues with head/ear interactions. If we see a significant spike or null on an FR graph on a speaker driver, it's always going to be resonance (hence the CSD is not really needed.) This may not be true with headphone FR measurements, which are really difficult and tend to be screwy. Bottom line is that most decent speakers I've heard don't ring. So needing to see CSDs are not an issue.
 
A serious limitation of headphones is that they use a single driver. Every driver is going to ring*. Good engineering can get around it somewhat. Most above-average headphones measure very poorly to above-average speakers in CSDs. We are not talking about "speed", which you keep mentioning, but rather driver resonances at specific frequencies. Driver decay is inherent, but how that energy is released is critical. If the energy is blurted out at specific narrow sets of frequencies, the effect can be rather nasty. 
 
*It has nothing to do with exiting harmonics (that would be non-linear distortion measurements). It has to do with membranes wanting to resonance at a specific frequency or frequencies - much like a drum.
 
The CSD plots are objective data after all. Not all people can hear ringing or do they mind it. Some people even like it in certain areas of the audio band. Just because you can't subjectively hear ringing doesn't mean that it isn't there for most other people. From my experience measuring almost a hundred headphones, some multiple times, most headphones in general excel in non-linear distortion compared to speakers, but come very short in FR and CSD measurements. As far as stats, for me, it's not necessarily the speed thing, but rather that they in general decay smoothly without any ringing.
 
It's funny how you are all of a sudden a subjectivist when the measurements don't suite you!
 
At the end of the day, I like to see FR, CSD, and non-linear distortion. Those three give me a really good picture.

 

[ultrabike]

Quote:

@ultrabike:
 
I'm not saying they're entirely worthless, but rather a good eyeball to make sure nothing is largely of concern.  Compared to speakers above average headphones tend to perform very well.  I'm saying there's people that scream over a hundredth or tenth of a millisecond in areas over other comparable headphones and, as such, I grow tired of it as I've yet to hear a decent headphone where it was a severe enough of an issue in the first place for me to hear it.  For example the "Stax" speed I keep hearing about.  Bought some Stax . . . and wasn't impressed in the end.  In fact, none of the people that I had try them picked them over any other headphone I had.  They sounded like my K601s more than anything else, and were promptly sold within a week.

 
Many headphones, including some TOTL, exhibit severe notches in the higher audible frequencies. In linear systems, deep notches come hand in hand with different levels of phase distortion (note notches are not the only indication of phase distortion - just one likely cause.) Moreover if you take a look at CSD plots, FR notches usually correlate to relatively long trails of energy across the time axis (though these may appear even in the absence of notches.) All these issues, exhibited in a CSD plot, may manifest themselves as inaccuracies in sound reproduction described as edgy sound among other terms. 
 
Moreover, closed headphones usually have a harder time than open ones in terms of sound quality. It is believed this is because they have to deal with the cup being right behind the driver, and damping of the cup has been shown to improve sound quality. Specially in orthodynamic headphones. While speakers have to deal with room interaction, headphones have to deal with cup and head interactions.
 
As far as the Stax, I really liked the 007 mk2 out of a T2. Don't know about the K601, but from memory that setup sounded better than everything I've heard so far (including PS500, DT990-600, LCD2, HD650...). It was more open and airy with out drilling my ears. A lot of detail and dynamics. Quite pleasant!

 

[purrin]

Quote:

 
No problem.  Inaccuracies/misrepresentations in bold below.
 
 
many decent headphones don't have enclosure resonance worth mentioning
 

 
The enclosure thing is a secondary consideration, not primary, at least to me. What I meant was more enclosure reverberation or more precisely sound bouncing inside the enclosures. Enclosures will cause certain frequencies to excite because of standing waves and this will exhibit as cause ringing behavior on CSDs.
 
The Grado cups are a good example. The cups dimensions themselves will produce an excitation at 2-3kHz. (This is a very audible effect and one part of the charm of Grado's.) The wood cups on the RS series actually helps to dissipate (in the form of a wider band excitation) of what would otherwise be nasty-sounding narrowly focused energy release at 8kHz from the driver to something much more pleasing. (I've actually taken Grado's apart and measured and listened with various cups, mods, naked drivers, etc.) Still the most serious ringing for the RS1 is at 5kHz.
 
The HD800 is yet another example. I believe HF'er arnaud did some sort of supercomputer simulation showing how the HD800 cup design contributed to the HD800's 6kHz peak. The Anax mods for the HD800 were partially a response to the results of that simulation; and indeed, shoving felt or foam into areas inside of the HD800 cup did help to tame that resonance.

 

[Magick Man]

I'm sorry, but without controlled listening tests, it doesn't challenge or support anything  because there's too much ambiguity with regard to purely subjective experience.

Sure, in the purely subjective realm, you can trust whatever you want. Someone's ears, the alignment of the planets, whatever. But it ain't science.

se

My cousin is the head of a food lab for Nabisco, PhD in organic chemistry and a BA in culinary arts. He tests samples of all the food that facility produces, and a taste test is an important part of it. Why would he do that? Because it's food and its taste is one of its most important qualities. Why do they trust what he's tasting? I can't think of anything that's much more subjective than that. Why do fragrance companies trust their chemist's noses? I'll guarantee companies that make skin cream wouldn't sell an ounce of it until they've slathered it on someone.

Audio companies can run all the objective tests they want, and they are valuable, but they don't mean as much until someone gets their ears involved.

 

[Magick Man]

In some cases it may not sound like anything because it's not severe enough. In other cases it can sound like high levels of distortion as the amp will often clip the input signal. In other case still it may not sound like anything because the amp's destroyed itself.

se

I guess we wouldn't know until we listen to it...