Bottlehead Crack Headphone Amplifier Kit Review

[tomtoo]

How about a multioption poll on that one (like 100%right/somewhat right/dont know/somewhat wrong/100% wrong) ?!

Or if you dont feel like it, may i use your words to create one?

Sure do it. My answer is easy to guess.

 

[BDWoody]

What happens if you find a tube amp that doesn't exhibit noticeable distortions? Jump to 1:31 for the measured FR.

To the 'Audiophile' it means your tube amp is likely either a huge disappointment or broken.

To the EE it means it was well designed for the job.

 

[solderdude]

What troubles me is some of the earlier comments to the effect that tubes are garbage

They are, they take time to become fully operational, have a short lifespan, change performance, are prone to be noisy, have limited amplification, aren't linear, are microphonic, radiate a lot of heat, are inefficient, are bulky, have limited current capabilities, are not easy to buy in bulk for production, are expensive, are fragile to name but a few drawbacks.

 

[solderdude]

How about a multioption poll on that one (like 100%right/somewhat right/dont know/somewhat wrong/100% wrong) ?!

Or if you dont feel like it, may i use your words to create one?

You can make a poll.
You will get opinions and not facts though, and you know what opinions are like.

 

[tomtoo]

You can make a poll.
You will get opinions and not facts though, and you know what opinions are like.

Depends on the science you do. Opinions are science. Opinions itself can be scientific facts.

 

[SIY]

One central goal is to understand what is in the intentional distortions typical of commercial tube amps, vs. solid state amps. Why? What is predictable? How can scientific charts better show why one implementation or form of distortion might be preferred over another? A clean tube amp and clean solid state amp would serve as control conditions for various levels of distortion.

So, your answer is, "No, I cannot demonstrate that I can hear the obvious differences I claimed."

I've known a lot of tube amp designers over the years. None of them put in "intentional distortion" other than in fringe things like the Korg amp stages. The guys doing stuff like single-ended triode amps just "designed" according to arcane philosophies, and the distortion ended up being whatever it is; none of them said things like, "Wait, I need to add in a little more third and pull back on the 6th." More competent designers produced amps that are fairly low in distortion and would be likely indistinguishable from a solid state amp with the same frequency response.

 

[tomtoo]

They are, they take time to become fully operational, have a short lifespan, change performance, are prone to be noisy, have limited amplification, aren't linear, are microphonic, radiate a lot of heat, are inefficient, are bulky, have limited current capabilities, are not easy to buy in bulk for production, are expensive, are fragile to name but a few drawbacks.

That are facts!

Non facts are that they sound somehow magicly better.

 

[tomtoo]

But i have to say i love this thread. Even that technocratical @amirm has shown some heard. And that it was hard for him to resist the charme of a well build diy tube amp.
And thats a part of love to music. For me one thing is most importend and thats the love for music. There is the technical part of my brain, but at the end its about fun. And thats it.

 

[Mad_Economist]

I am looking at the AP.com site at their AECM206 headphone test fixture. I am also sketching up a plywood block head with holes drilled through it to place a calibrated microphone where your ear canal would be inside the head phone cup. The side of the plywood block head would be covered with ¼” thick neoprene wetsuit material. A fake ear would be fabricated from the same material. Tuesday I am going shopping for both the AECM206 and the neoprene sheet material. Could be a calibrated block head.

Controversially, I would argue that a blocked canal equivalent measurement - the sort you'll get with an omni capsule at the canal entrance location - can be pretty suitable for headphone purposes. Some of my peers would advocate acquiring one of these cheap Chinese IEC60318-4/IEC711 couplers that're everywhere these days, which together with a 10mm or so canal extension from your pinna would match the load of a real ear.

However, the geometry of your pinna is extremely significant to the transfer function, and it's very unlikely that neoprene has the right mechanical characteristics even if you could match the ITU-T P57 spec for the pinna's geometry. This would likely pose the single largest issue for your DIY coupler relative to a proper system.

Why the AECM206, out of curiosity? You could get a 43AG for less money and get the latest and greatest pinna.

I wonder how the homemade head will compare to the AP AECM206 headphone test fixture for things like mid-bass waterfall plots.

The utility of spectral decay plots for headphones is pretty dubious - at low frequencies, this would be exponentially more true...

 

[Mad_Economist]

If I run a signal through two different gain stages (you pick the gain), one tube and one solid state, and verify the frequency responses as essentially equivalent, then record music through them to generate two level-matched files, differing only in the choice of tube vs SS, would you be willing to post ABX logs showing you can actually detect the "obvious" differences?

If the @generic isn't interested, I reckon I could find quite a few people who'd be interested in hearing that comparison! Can I subscribe to the SIY RSS feed?

 

[SIY]

If the @generic isn't interested, I reckon I could find quite a few people who'd be interested in hearing that comparison! Can I subscribe to the SIY RSS feed?

I'd be happy to put this together, though I pretty much know the results in advance, having done a similar test myself some years back in an attempt to prove Lipshitz and Vanderkooy wrong. They weren't, and that was the day I realized that "tube sound" was an illusion.

It's a pity that the guy making this claim refuses to step up and demonstrate its veracity, but this would hardly be a first.

 

[lashto]

I'd be happy to put this together, though I pretty much know the results in advance, having done a similar test myself some years back in an attempt to prove Lipshitz and Vanderkooy wrong. They weren't, and that was the day I realized that "tube sound" was an illusion.

It's a pity that the guy making this claim refuses to step up and demonstrate its veracity, but this would hardly be a first.

Just record the files & create a poll-like thread. Should be fun.
And btw, how exactly do you plan to record the outputs?

 

[generic]

So, your answer is, "No, I cannot demonstrate that I can hear the obvious differences I claimed."

Not at all. There is no reason to frame it this way or try to play gotcha. Demonstration would be part of structured and controlled (double blind) empirical testing. I could never answer that type of question one way or another without setting up a double blind test. I, you, and everyone else is vulnerable to testing biases.

Perhaps some amp designs can avoid the characteristics common with many commercial products (i.e., my personal experience to date). This is a self-described science site after all. Science requires data. Rules have exceptions. Subtle interactions require assessment. To claim "science" one must follow standard scientific methods and address observed, measured, and (human) reported phenomena, or admit defeat and incompleteness.

There are hard sciences and soft or human sciences. Human scientific data is routinely fuzzy and imprecise (i.e., gaussian, probabilistic, bell curves). Some observers are indeed able to hear things that others cannot hear. Just because I, you, or another doesn't hear a difference doesn't mean that someone else cannot hear it (e.g., 2% could hypothetically be found to be statistically valid "golden ears"). This has been throughly demonstrated 1,000 different ways across a century of psychological and biological research. I'm merely saying to test human perceptual characteristics and limits in the tube/distortion context. Scientifically. There's no debate that some audio fans are "bass heads" while others love Grado/treble. This may indeed follow from biological/neurological individual differences in the human auditory system. Similar considerations apply to distortion levels and distortion production methods (i.e., tube/non-tube).

I'm trying to avoid the pitfalls of isolating the science of "high fi" without considering the interactions with distortion and human perceptual limits/differences. This is not a controversial concept or a remotely new idea, but it gets lost sometimes. It gets wrapped up in ideology and absolutism.

Be a science site and use documented, integrated, and established scientific processes or don't.

 

[SIY]

Not at all. There is no reason to frame it this way or try to play gotcha. Demonstration would be part of structured and controlled (double blind) empirical testing. I could never answer that type of question one way or another without setting up a double blind test. I, you, and everyone else is vulnerable to testing biases.

Perhaps some amp designs can avoid the characteristics common with many commercial products (i.e., my personal experience to date). This is a self-described science site after all. Science requires data. Rules have exceptions. Subtle interactions require assessment. To claim "science" one must follow standard scientific methods and address observed, measured, and (human) reported phenomena, or admit defeat and incompleteness.

There are hard sciences and soft or human sciences. Human scientific data is routinely fuzzy and imprecise (i.e., gaussian, probabilistic, bell curves). Some observers are indeed able to hear things that others cannot hear. Just because I, you, or another doesn't hear a difference doesn't mean that someone else cannot hear it (e.g., 2% could hypothetically be found to be statistically valid "golden ears"). This has been throughly demonstrated 1,000 different ways across a century of psychological and biological research. I'm merely saying to test human perceptual characteristics and limits in the tube/distortion context. Scientifically. There's no debate that some audio fans are "bass heads" while others love Grado/treble. This may indeed follow from biological/neurological individual differences in the human auditory system. Similar considerations apply to distortion levels and distortion production methods (i.e., tube/non-tube).

I'm trying to avoid the pitfalls of isolating the science of "high fi" without considering the interactions with distortion and human perceptual limits/differences. This is not a controversial concept or a remotely new idea, but it gets lost sometimes. It gets wrapped up in ideology and absolutism.

Be a science site and use documented, integrated, and established scientific processes or don't.

Long version of "no, I won't do the actual experiment to validate my claim of obvious differences."

Let me know if you are ever ready to do the experiment. It's easy, I generate the files, you do the ABX test and post the log.

 

[generic]

It's a pity that the guy making this claim refuses to step up and demonstrate its veracity, but this would hardly be a first.

See my reply #293. I'm proposing a full human factors testing model. I am in fact trained in a related discipline and deeply aware of past pitfalls (i.e., what I'm seeing from some at ASR). There are human individual differences and there may be some biological/neurological process that causes some (and only some) to perceive things that others do not perceive. Humans (including myself) can indeed be wrong about their perceptions too -- to include being wrong one way or another about tube amps. Or mislabeling what they perceive. Or, or, or...

Human factors methods have been routinely used for many decades to get around differences in objective mechanical measurements. This starts with something as simple as human differences in reaction times to seeing a light turn on. And these factors pervade human perception.

I'm not at all ideological here. I'm suggesting the use of really, really old-hat scientific methods. References to 1,000,000 published psychology, biology, and human factors publications are available online.

Perhaps this should be called "Limited Parts of Audio Science Review" or "Incomplete Audio Science Review"?

 

[generic]

Long version of "no, I won't do the actual experiment to validate my claim of obvious differences."

jAka "I'm not open to using established outside scientific methods."

 

[SIY]

Just record the files & create a poll-like thread. Should be fun.
And btw, how exactly do you plan to record the outputs?

Using the RME ADI-2 I have on hand.

 

[SIY]

See my reply #293. I'm proposing a full human factors testing model.

You claimed "obvious differences." So you can either hear them without peeking or you can't. Declining to demonstrate that speaks loudly no matter how furiously you wave your hands or move the goalposts.

"Obvious differences." Put up or shut up.

 

[generic]

Have a nice day.

 

[Mad_Economist]

I'd be happy to put this together, though I pretty much know the results in advance, having done a similar test myself some years back in an attempt to prove Lipshitz and Vanderkooy wrong. They weren't, and that was the day I realized that "tube sound" was an illusion.

It's a pity that the guy making this claim refuses to step up and demonstrate its veracity, but this would hardly be a first.

I'd be quite keen! Generally, I reckon I know where the answer will be too - but I've "known the answer" before a listening test before and been wrong too! The unhappy part about testing your beliefs is that you often find out you have been wrong - the happy part is that it gives you a chance to stop being wrong, eh?