My ear hurts!

Q: I love using my IEMs, but recently my left ear has started hurting with maybe a little blood when the IEM is in there awhile and I take it out. What should I do?

A:  Pain and bleeding could be from a poor fit, rough IEM shell, or even a problem with the external ear itself.

If you have pain or discomfort from your IEM, that's not normal. And certainly bleeding isn't either. Stop using them. Get something else. Consider seeking medical attention. The bleeding is literally a red flag. 

Aune T1 tube rolling tips

Q: What tubes should I get for my Aune T1?

A: A common-sense rule of thumb is: don't spend more on your tubes than you did on the amp :-)
in general NOS (new old stock) tubes are better choices than reissues, but some NOS especially Amperex and old Genalex tubes can get plenty expensive very quickly.

Many resissue tubes are manufactured by New Sensor Corporation, and their price may be dramatically marked up by tube stores. When it comes to buying tubes, patience is always a virtue, especially when it comes to deals!

Windows sound vs. ASIO vs. WASAPI

There was a time when windows sound drivers sucked. Big time. That's the reason why ASIO was popularized, to get away from the terrible (and I'm being kind here) Windows Kmixer kernel that mangled sound like a testicular vice.

Steinberg developed ASIO (Audio stream I/O) to bypass the truly terrible Windows NT 5.x Kmixer kernel. When ASIO was released the pro audio community jumped on ASIO not only because it was capable of direct bit perfect delivery, but also because it supported 24 bit sampling (Win only managed a max of 16 bit at the time).

But Windows has come a long way since the days of Kmixer, which was discontinued for the Win Vista release. And people seem to have forgotten how clunky ASIO really is. Current iterations of native windows sound are as robust as ASIO, less clunky in implementation, and perfectly capable of supporting "audiophile" setups.

So what's the huge advantage in running ASIO or WASAPI over native windows drivers? Surely it's not because of remarkable improvements in sound quality as touted by some hypemongers in reviews (please don't take a ride on that hype train!). The only advantage is this: if you don't want irritating system noises to intrude when you are listening to your music why then WASAPI exclusive mode bypasses them.

Will a 20 foot extension cable lower the sound quality from my amp to my headphones?

Q: Will a 20 foot extension cable lower the sound quality from my amp to my headphones?

A: If you run really long cable, you have to watch out for two things: [1] the cable may pick up noise, particularly a hum, and [2] the signal may degrade if you run lengthy stretches of cable, usually because the voltage drops over long distances. And by long distances, I mean long distances, not 20 feet.

You can fix #1 by using with better shielding, and #2 by using XLR out instead of RCA.

But in your case, you aren't running cable through the entire neighborhood, just 20 feet, which is well within the tolerances for RCA unless there is a significant EMI transmitter between the your amp and your headphones.

In sum, you will be fine.

PS: Veni's Trivia of the Day: the term "RCA" in RCA cables comes from the Radio Corporation of America, who first used the format to connect their Victrola record players to their radios in the late 40s or early 50s. Prior to that they were using RCA cables to connect their record player pickups to the chassis, but not as interconnects.

Scientifically comparing personal audio equipment

Q; Most comparisons of headphones or amps or DACs are so subjective. Why aren't there many scientific studies comparing personal audio equipment? How can we become better at comparing different headphones or different amps in a more scientific way?

A: Comparative analysis isn't a common or popular modality of testing audio equipment. Even the most quoted "objective" blinded A/B tests (like those for headphones) are methodologically weak, with lots of observer-dependent limitations that significantly detract from any results they manage to come up with.

Rather than trying to randomize the source like a drug trial or a physics experiment, another way to scientifically approach this might be to study the observations of different judges by looking at their inter-rater reliability as a measure of scientific "soundness".

For example, we could look at your data using Fleiss' kappa or a rank correlation coefficient (such as Spearmans, since your data will most likely be ordinal) as a measure of validity, since if our subjects were really under the influence of a placebo effect, there would be a very specific correlation (or not) between their scores, and if there was a true audio difference there would be a different specific correlation (or not) between their scores. So for example, if everything came up with poor correlation, we could make the argument that any sonic difference was attributable to the placebo effect.

Of course, the problem is that these projects are not viable because they are not really attractive to funding by the usual suspects, i.e. industry - after all, which DAC manufacturer would want to conduct an experiment which proved that their $1000 DAC was no better than a $100 DAC, or even the opposite, why would the manufacturer of a $100 DAC want to take a chance that the study would prove that the $1000 DAC was much better?

Can someone tell the difference between lossy and lossless music files?

Q: Can someone really tell the difference between lossy and lossless music files?

A: Why not conduct an experiment.

Pick a song you have on CD that you know well, and rip it to FLAC, 320 kbps mp3, 256kbps mp3, 128 kbps mp3 and 64 kbps mp3. Now start at the bottom and slowly listen your way up the quality ladder. It's easier to listen "up" and appreciate nuanced differences than to listen "down".

At what point do you stop hearing a difference between a file and the next one up in sequence? That's your bar. If it is really low, you can always practice to become a more discerning listener (develop those golden ears!).

Having said that, you should probably be able to tell the difference between a low bitrate mp3 file and a high bitrate mp3 file, but at bitrates of 320 kbps, it becomes astonishingly difficult to differentiate lossy from lossless, even with high-quality equipment.

This experiment works best when it is lubricated by a glass or two of single malt (or your choice of beverage).

Do it for science!

What sound experience do real audiophiles want?

Q: Do real audiophiles progressively move toward the most neutral experience they can get?

A: Not all audiophiles, it may be better to use the word "some" instead of "real". Rather than call it "neutral" vs "colored" or "objective" vs "subjective" I think of the listening experience more as being "analytical" vs "fun". I'll explain what I mean.

Some times I want to get the source recording, the whole source recording, and nothing but the source recording. While you may think this desire only extends to classical and/or live performances, in reality there are times when I ant to hear music the way the artists/ producers conceptualized it. So for example when Daft Punk won the "best engineered album" award at the Grammys, I wanted to go back and hear RAM exactly as it was on my source file, ni plus ni moins. And the best way to achieve it is with a neutral DAC, a solid state amp that does not sound too bright or too dark, and something like the HD600.

But at other times I want a completely different experience. I want the violins in Mendelssohn's Concerto in E minor Op 64 to come out of the orchestra and smack me across the face, I want Christina Novelli to breathe in my ear when I hear Garth Emery's "Concrete Angel", and I want that in-your-face Southern rock stance when I listen to 38 Special's "Caught Up In You" and perform right alongside them on stage with my air guitar. And frankly, for any of those scenarios (classical, EDM, rock) when you want a more "fun" experience, the analytic set up fails. So an alternative set up with a warmish DAC, a tube amp and a HD650 or Fidelio X1 will give you a very different (and probably more pleasing) experience than the one I outlined in the previous paragraph.

So what headphones like Grados offer is a "fun" experience. And while "fun" is good, there is a totally different "analytical" experience out there waiting to be discovered. Some may enjoy it, some may not. And for everyone who enjoys an "analytical" experience, there is a "fun" experience waiting around the corner to surprise and thrill them. That's why there are so many headphones out there.

I just realized that everything that I said about headphones also applies to your spouse/girlfriend/boyfriend/significant other. Sometimes you want fun, and sometimes you want analytical.

What song do I demo my new headphones with?

Q: What song do I demo my new headphones with?

A: I would say the most important criterion for demoing headphones (apart from the song being of good quality) is that you know it very very well, so that you can pick up nuances sooner rather than later. Typically my demo playlist comprises of one song from each genre I love that I know like the back of my hand.

It's hard to pick one song, my current demo playlist starts with New Order - Regret (Fire Island mix). Lots of tonal variations, both "cold" and "warm", instrumentation allows you to focus on bass, mids and highs in turn, tests tightness of bass rather well, brings out coloration, vocals have good intonational characteristics that can be used to advantage when A/Bing, positional nuances are subtle so a reasonably good test of soundstage, too.

Bent headphone connector

Q: Help, my headphone connector is bent!

A: It is probably best to replace the connector end. Once the connector bends, you run the risk that it will break off inside the 3.5mm jack, which is much more complicated to fix than replacing the connector. Your local Radio Shack or electronics parts store should have the connector. There are lots of guides on YouTube to walk you through the process.

One way to straighten it out is to place the connector at the end of a flat surface like a table, place something flat and reasonably heavy over it like a hardback book, and then move the book back and forth so that the connector rolls between the book and the table and straightens. But once those TRS connectors are bent it is usually only a question of time before they fail.

What makes really expensive headphones so expensive?

Q: What makes really expensive headphones so expensive?

A: Essentially there are 4 aspects to the cost of any product [1] R&D [2] materials [3] manufacturing and [4] marketing. These costs may be either fixed (eg payroll) or variable (eg sourcing raw materials).

R&D is variable, so calculating the cost of R&D for an individual headphone model is difficult. Newer models still haven't seen a return on ROI costs, while older models (for example HD600) have probably recouped their R&D costs, and then some. R&D costs may be hard to figure out accurately because one technology may be used in multiple products, or may be responsible for the development of other innovations, for example the design of the HD580 enclosure and drivers directly influenced the design of the HD600 and 650 enclosures and drivers. Another great example is the Q701 -- it is in essence a K701 with modest changes. Most of the R&D cost was for the K701, and the Q701 reaped much of the benefit.

Material costs vary over time, often wildly, and are subject to supply and demand constraints. For example the price of copper, used in headphone voice coils and wires, has changed dramatically over the last few decades, and copper futures are still often volatile.

Production costs also vary, for example Q701s were made in Austria, but AKG wanted to save costs and moved production to China. But moving production to China does not necessarily make things cheaper, for example some Audio Gd amps probably cost more to make than equivalent Schiit amps, even though Schiit makes its stuff in the US and Audio Gd in China.

Many companies OEM some or all of production, which again affects costs depending on who they OEM to. And it's not just a case of folks using cheap Chinese OEMs -- some Denons for example were OEMd by Fostex, for example. And keep in mind that OEM companies typically conduct their own R&D, have their own design costs, have to retool equipment for specific products, etc, which changes cost calculations.

And finally marketing: some companies spend more on marketing (e.g. Apple, Nike, Beats) and make huge profits, while others spend very little on traditional marketing and still manage to do very well for themselves. One way to get around substantial marketing costs is to engage in direct marketing (eg Emotiva, Schiit) or to depend on word of mouth (like HiFiMan depends on HeadFi).

But eventually, keep in mind that companies will try to make as much profit as they can, so the price may be determined by what the market can support. For audiophile products, there may be a substantial markup in price just because companies know they can get away with it. A classical example of this is the Grado RA1, which has been substantially marked up when you take into account its design and materials/manufacturing cost.

USB vs optical vs coaxial inputs

Q: What is the difference between USB and optical/coaxial inputs? Is one better than the other?

A: In essence they are the same -- a digital signal is a digital signal, no matter how you transport it. Having said that, there are a couple of differences:

[1] if you use USB, you can plug it into any computer. With optical or coax you need a source with an optical or coax jack. Not all computers have them, for example.

[2] At one point until not so long ago, USB technology was limited to 24 bits/48 kHz. At that time, it was much more advantageous to use optical or coax, which went up to 24/96. That technological disadvantage is no longer the case with current generation of USB chips. However, the technology has swung the other way -- many DAC manufacturers limit optical/coax thruput rates to 24 bits at 96 or 192 kHz. If you want to go higher beyond 192 kHz, for example 24-32 bit/352.8kHz-384kHz, USB may be the only way to do this. Keep in mind that your ears probably cannot tell the difference between 24/96 and 24/192.

[3] If you want DSD you may have to have USB input (of course, if your DAC does not support DSD then this is moot)

[4] if you have USB line noise from EMI (because USB carries power in addition to the audio signal) then optical/coax may get help rid of the noise (because you are not using USB)

Personally, my favorite connection is optical out, and not for any musical reason -- I just like the idea that my music traveled along a fiberoptic cable as light before ending up between my ears.

When lossless isn't really lossless

Q: Are all lossless audio file formats truly lossless?

A: Many like FLAC or ALAC are. But others can be somewhat deceiving. For example Sony's ATRAC Lossless audio files.

ATRAC Lossless is actually an interesting format. It works differently than FLAC or ALAC -- the way I understand it is that ATRAC lossless is actually a lossy ATRAC3 file with additional information ("correction data") that allows the lossless sound to be reconstituted. Kind of like packaging the water and the Top Ramen separately in the same box to reconstitute the whole noodles. Somehow that isn't the same as cooking the ramen from scratch.

The sweet spot of compressed versus uncompressed audio files

Q: At what point does compressed vs uncompressed audio files really become an issue?

A: You can have compressed lossless files (eg. FLAC files) that are sonically identical to uncompressed lossless files (eg. WAV files). The only issue you would have here is if you compress FLAC files very tightly, you will need a powerful computer to uncompress on the fly.

Now with respect to mp3 vs WAV or CDs; that's more of a lossy vs lossless question rather than compressed vs uncompressed question. For example, with mp3, the lower the bitrate (measured in kbps, an index of the file's audio quality) the more obvious the quality deterioration is appreciable with good headphones.

For streaming music, uncompressed isn't really viable because it takes up an extraordinarily huge amount of bandwidth to stream, so all radio stations / streaming services typically offer some level of lossy streams. For me, the 320 kbps Orbis bit rate that Spotify uses for streaming its Premium service is not an issue, even with really good headphones.

For local files, what you really get with compression is the ability to save some space. Typically I rip my CDs to FLAC files, and then rerip to varying mp3 rates when transferring to my phone or mp3 player because, well, like real estate, there is only so much you can fit in limited space, and either you fit a lesser number of quality files or more files but of lower quality.