Review: Sennheiser HD-558 (modded)

Introduction

Why the Hell would I need another headphone - again? I´ve asked myself this question often enough. But I don´t have an answer. I just like to collect cans. Just as I like to change playback/recording units I love to change headphones, all depending on my current mood. Pure, decadent luxury considering other people can´t afford even one headphone. Well, I received a not-so-small amount of money as a Christmas present and decided to spend a part of it on a new pair of cans. And as the Sennheiser HD-558 had been on my Amazon Wishlist for more than three years, I finally decided that the time was right. Do I like it? Is it able to replace my dearly loved, 9-year old Sennheiser HD-600? Why don´t I just use my Sennheiser HD-448? Or my Superlux HD-668B? And what about the two headphones I reviewed before, the Sennheiser Momentum and the Skullcandy Aviator? Read on and you shall be rewarded with answers, dear constant Reader.

Design & features

Sennheiser HD-558 (with 'NOS4A' by Joe Hill abused as a headphone stand)

The Superlux HD-668B cannot be used anymore since it broke apart (only because I cleaned it!). I won´t replace it with a new unit because... well, I have grown to hate its sound. The Sennheiser Momentum & Skullcandy Aviator did many things well but suffered at others, the worst being comfort-related. In the end I decided in favor of the Sennheiser HD-558, the headphone I wanted to have in the first place. It cost me 129,- Euros, it is a plain and boring looking headphone, manufactured entirely out of plastic in China. Huh? No, I don´t have a problem that it´s been built there, it´s just that Sennheiser doesn´t like to talk about it which feels pretty double-tongued. As if they are ashamed of it. When I think of my beloved FiiO units, it's stupid and not particularly fair to the people who constantly excel at producing highly functional products for the whole world. But enough of that. This headphone is right in the middle of the HD-5x8 line. All of them are engineered to attract cost-conscious people; the HD-518 is the cheapest, costing half of what the top model, the HD-598, costs (roughly 220,- Euros). Their predecessors (HD-5x5) looked almost the same, had the same features and were - as far as I can tell - fairly successful, causing Sennheiser not to change too much. Never change a winning team, eh?

A (too?) proud German company: Sennheiser

The top model HD-598 is finished in a cream-like colour with attached faux-wood veneers, a headband with a fake leather pad and metal grilles covering the outside of the earcups. Sennheiser was obviously trying to improve the inherently bourgeois design of the predecessors; yet all they achieved was making it look tacky. The grilles of the otherwise plain HD-518 arent exactly an improvement either, they look like windows shutters. The incredibly boring and plain looking HD-558 is the one that´s honest, it´s the least flashy of all three models. When it comes to features, it´s equipped with what Sennheiser calls Eargonomic Acoustic Refinement (E.A.R.), which is supposed to 'channel audio signals directly into your ears', enabling all three HD-5x8 cans to have 'outstanding bass and vocal projection'.

Ugh, simple marketing blabla: E.A.R. is nothing more than an angled driver, many headphones are equipped with a technology where the drivers are placed slightly below and in front of your ears. Our ears' stage impression usually reacts favourably to this placement because it mimics how we listen to loudspeakers: those aren´t standing on the left and right of our ears, no, they´re standing in front of us, helping us creating a holographic image of the virtual stage. Is it a good idea to implement something like this in a headphone? Well, so far my experience tells me that it rarely works as intended... with horror I recall my second 'big' headphone, the Sony MDR-CD 570. Because most of the time it just creates colourizations - and not the desired effect of a holographic stage. But we'll see about that later.

Sennheiser HD-558, earcup: comfy velours cushions & 'surround reflector'

The diaphragms inside the drivers are a Duofol design, this principle of constructing membranes was carried over from the HD-6x0 models. They reduce distortions and improve transparency, that's what Sennheiser says. Seems to be true when I think of my HD-600; it´s the least distorting and most airy headphone I´ve ever heard. When it comes to impedance, this headphone is rated at 50 Ohms. The maximum sound pressure level (SPL) is at a very loud 112 dB, it should therefore be perfectly suited for portable units. But no, Sennheiser markets it towards music lovers, listening at home using their big HiFi system. Strange. This will become important again, so more about that later.

Accessories & built quality

Sennheiser HD-558, 3-meter cable

The box the HD-558 comes in, contains the headphone and the cable. That´s it. The 3 meter cable pictured directly above (recall for a moment that it´s targeted at at-home-listeners) comes with a proprietary 2.5 mm plug on the headphones' side. The Sennheiser site lists an additional 1.2 meter cable with 2.5 / 3.5 mm plugs suitable for portable units. However, this cable appears to be a myth as I was unable to find even one place where it was sold.

Sennheiser HD-558, proprietary 2.5 mm connector - what a dimwitted idea

In the end I had to resort to a replacement cable built by Oyaide and distributed by FiiO, the RC-HD1. This took a further 40,- Euros out of my pocket. Frankly, I find Sennheisers' brazenness of not including a shorter cable themselves, insulting. A carrying bag... are you kidding me? Not included, not offered. My cheaper HD-448 came with one, the Skullcandy Aviator had a particularly luxurious bag, the Sennheiser Momentum could be safely tucked away inside a firm, brown bag. Hell, even the super-cheap Superlux HD-668B came with a soft cloth bag. With the exception of the Momentum, all of them cost less than the HD-558 and Sennheiser cannot manage to include something as basic as this? Oh please!

Quite expensive but well built 1.2 meter replacement cable: FiiO RC-HD1

But how is it built? Despite being fashioned completely out of plastic it has been manufactured very well. Everything fits tightly, fins are nowwhere to be found, it doesn´t creak when moved or when you change the position of the earcups. The cushions are firm and they are replaceable... now, that I mentioned it: the possibility of replacing any broken part is one of the reasons why I love Sennheiser products so much. By doing so, you can extend the lifetime of their headphones. My boyfriend owns an ancient Sennheiser HD-250 linear; it´s now 26 years old. With other manufacturers you´d be hard pressed to find even one replacement part, with Sennheiser it is no problem. My boyfriend has been using the HD-250 Linear for many years for DJing, it was beaten, tortured and abused. The faux-leather cushions and the mesh covering the drivers had disintegrated completely, the cable looked like shit too... well, it was inconvenient anyway since it was equipped with a 5-pole DIN-plug! Getting new earpads (including the mesh) and a cable was as easy as ordering a new headphone with the result that this ancient thing now looks like new.

Upon deciding whether I should get the HD-558 or not, I also considered the famous Philips Fidelio X1 (gorgeous sounding headphones). At first glance, it appears to be perfectly manufactured and built to last. But any impression of reliability is destroyed once you know that its cushions and everything inside it is glued together. Even the cushions are glued to their frame, they aren´t replaceable. Perfect example of planned obsolescence if you ask me and crucial for my decision not to buy it. EDIT 17.03.2015: The successor Fidelio X2 fixes that problem. Well done, Philips!

Comfort & quirks

Velour cushions & headband: comfy and replaceable

When referring to new Sennheiser cans, people usually talk about the 'clamp of death'. It´s true: the clamping force of the HD-558 is quite strong, your head is tightly sandwiched between the two earcups. Even after wearing it almost every day for more than two months, this force hasn´t diminished. The earpad and the headband cushions are firm, but not uncomfortably so. Their one redeeming aspect is their size; while the cups of the HD-600 are even bigger, the earcups of the HD-558 are certainly big enough to accomodate almost any ear.

There´s just one thing inside perhaps posing a problem: a so-called 'surround reflector'. You see, the fabric covering the drivers is applied to a thin plastic foundation which itself protects the membrane underneath it and also forms the reflector. The latter sticks out and, for me, has been a nuisance. It causes me to re-adjust the position constantly because it tends to touch my upper earlobes. This doesn´t hurt at first but after a few hours (yes, I wear headphones that long) it starts to hurt, even when it´s only slightly brushing my earlobes. Wouldn´t be any problem at all - if the surround reflector would be softer. But the hard plastic underneath the mesh averts this. So before you buy it yourself, try it on and see if your ears fit.

The foam mod & Burn-in

Did you know that you can modify the HD-558 to almost sound like the top model of the line, the HD-598? It´s easily said and done, just look above at the video. But please, be more careful than the creator of the video, the cables soldered to the drivers are very, very fragile. They break if you blow on them. According to what people say online, by removing the piece of foam you´ll get more punch and air. I cannot attest if this is true or not, I never listened to my HD-558 before it was modded. Whatever it changes, this tweak won´t completely transform your HD-558 into the HD-598. To achieve that, you´d have to grab pliers to cut holes into the plastic structure covered by the outer felt-mesh grills. The HD-598 sports very fine metal grills which aren´t dressed up with some fabric, apparently they help to achieve a more airy sound. This might be true; before I got my HD-600 I owned its predecessor, the HD-580, which is (apart from a different colour scheme) exactly the same headphone. The only difference are the driver grills: on the HD-580 they are plastic, on the HD-600 they are metal, causing the latter to sound different. But I don´t want to remove anything that can´t be re-attached again, I´ve therefore skipped this part. Below you can see measurements by Headroom where I´ve compared the frequency response of the HD-558 against the one from the HD-598. They are extremely similar, the HD-558 emphasizes frequencies around 3,000 Hz while the HD-598 slightly accents everything above 10 kHz. The latter presumably is responsible for the HD-598's stronger amount of 'air' while the un-modded HD-558 will sound a bit more aggressive. The foam mod will bring the HD-558 much closer to the response of the HD-598, in the end you have to decide yourself of course if you actually want the sound of the top model.

Frequency response differences Sennheiser HD-558 & the top model HD-598

Whatever you decide, you simply have to burn in the HD-558 (this goes for the HD-518 & 598 too). I listened to it once I had removed the foam and in spite of it already sounding very well, burning in really brought out its strengths. I admit that I´ve never experienced ANY device reacting so strongly and positively to a burn in. My Sennheiser HD-448 remained essentially unchanged in terms of sound, the Superlux warmed up considerably but these cans turned from undifferentiated into 'differentiates-like-hell'. Spatiality improved a lot, some unpleasant high frequency sibilance was removed too. So please, burn it in. For at least 48 hours.

Tech section

Impedance response HD-598 vs. HD-558 vs. HD-448/449

I know you probably hate to read measurements of any kind. They are boring, you cannot understand them or maybe you´re simply too lazy to even want to understand them, right? Yes, in my opinion, people who ignore measurements willfully do not deserve to experience sonic bliss. So stay with me and let me lecture you, for a few minutes only, I beg you. In this case it´s important because they tell you if the HD-558 needs to be amped or not. Because the answer is YES and NO.
NO: It´s loud enough for many portable devices. At this moment I´m listening to music played back by a recently acquired Sony NW-HD5 (review will follow) and with the HD-558 it gets loud enough to shatter my ears. The same is valid for many of my portable MD recorders/players. Not to mention the FiiO E07K and the FiiO X3 (review will follow), they can kill the HD-558 with ease. Sensitivity isn´t any problem, the impedance is. YES: The HD-558 is rated having an impedance of 50 Ohms. Look above at the measured impedance response and you can clearly see that it reaches a resistance of almost 300 Ohms at 95 Hz. Meaning: a headphone output with an output resistance of more than 6 Ohm will produce severe frequency distortions.

Well suited for portable units - but not for stationary units: HD-558

Example: to monitor an MD I was recording with a Sony MDS-JE 530 (review to follow), I plugged the HD-558 into the MD-recorders' headphone socket. The resulting sound was so bass-heavy that it obscured everything else. Mids and treble were drowning in bass, it was unbearable. Beats by Dr. Dre territory? Think bigger, more blown-up and hollow. What my problem is, you ask? I´ll tell you: many big amps, CD players, network players or TVs are equipped with headphone outputs featuring a fairly high output impedance. With all of them the HD-558 will produce this bass-heavy, highly coloured and droning sound. But if all of them are playback devices for usage at home, then why was the HD-558 engineered in a way that will produce a bad sound with exactly these devices? Remember: Sennheiser targets these cans at at-home-listeners... why the fuck did they construct a headphone that will play with superior quality on half of all portable devices instead?
Thankfully I´m accustomed to amps with a low output impedance. In case I want to use it with one of my vintage portable CD players, I simply connect either the FiiO E6 or the E07K. Or I make use of the FiiO X3. Due to its fairly high sensitivity you might want to listen with it once you step outside... you'd look ridiculous though. And don´t forget that it is a very open headphone, leaking a lot of sound to the outside world. What you´re hearing, other people will hear too. So what is it for? Portable use? Electrically, yes. Isolation? Forget it. At home? Yes - but only with sources having an output impedance of less than 6 Ohms.

And now: the sound

Sennheiser HD-558: felt mesh grills

Does it sound well? Of course it does, otherwise I wouldn´t have kept it. Once it´s modded, burned in and fed by low impedance sources, you´ll be rewarded with a stage like you´ve never heard before. Only one word fits: spectactular! For decades I´ve been dreaming to hear a soundstage like this one and until the HD-558 arrived, this desire had never been fulfilled. My beloved HD-600, already featuring a realistically wide and deep stage, feels shallow in comparison. Yup, the HD-558 easily bests the HD-600 in spatiality. It´s like being able to 'see' around instruments, effects, soloists, anything. The sound is fantastically layered, not only from left to right but from the front to the rear. Every instrument, group of players, synths and effects, even reverb, is defined precisely. It doesn´t sound artificial, it feels completely organic. Very well done, Sennheiser!
The impression of virtual stage depends on the recording you´re listening too, of course. The device used for playback and amping plays an equally important role, too. For example, these cans revealed my FiiO X3 to sound a bit distant, compact and slightly constricted (I knew that before, the HD-558 only made it all the more apparent). The more recent FiiO X5 (loaned to me by FiiO, review to follow) was worse, sounding flat, confused and blurry. 'Stage kings' like the Sony MZ-R 37 or the MZ-R 55 are unmasked to sound either artificial or congested while the MZ-R 50 improved. Recordings produced using several groups of microphones profit as well, recordings with less mics will extend their dimensions. Binaural recordings... superb. It appears that the Sennheiser marketing experts haven´t promised too much when advertising the HD-558 as featuring an improved spatiality.

The fantastic stage of the HD-558 is... hard to match

Frequency response is a bit more problematic. There seems to be almost no sub-bass (below 50 or 60 Hz). Yep, deep bass is lacking badly, causing a "loose" sound signature. I´d guess 6-8 dB amplification below 60 Hz might do the trick. Frequencies above however are fully present and accounted for... which is also the reason why an impedance mismatch like the one I described above will produce a hollow and bloated sound. But when properly amped, bass is punchy when needed, dynamic and reasonably tight. Everything above low frequencies sounds balanced with a tendency towards warmth and charm. Voices have a correct chest size and won´t ever sound sibilant. Overall, I´d call the frequency response fairly leveled out from 100 to 10,000 Hz without too much colorization. My HD-448 sounds muffled against it, the Superlux... well, when it was still able to sound... no, I don´t want to compare it, it´d be insulting to the HD-558. Still, just as it would profit from more deep bass, a stronger amount of air / crispness would be benefitial as well. Sometimes it sounds rather dry; this helps differentiation and precision of course, the HD-558 will always be able to tell things apart with ease, better than the HD-600 even. But air is missing and a too dry sound appears abrasive to some people. Though 'abrasive' is the last word I´d use to describe the HD-558. But there´s no denying that the HD-600 is superior when it comes to air.

Could have more 'air': HD-558

Another area the HD-600 easily surpasses the HD-558 is the amount of details. Due to its slight warmth and missing air, the HD-558 drops tiny details. Revealing errors isn´t its forte, no, it´s a headphone concentrating on presenting music pleasantly. But don´t make the error to think that it sounds boring or that its timing sucks! Far from it, this is one of the most vibrant and snappy headphones I´ve ever heard. One minute, it´s explosive and calm the next, it sounds much more engaging than my HD-448. It can be lovely and aggressive, all depending on the recording. Will it ever sound congested, plastic-y, confused, coloured, veiled or boring? No.

Now a warning: all of this depends on how it sits on your head. The position is extremely important for the sound and no, I´m not exaggerating. Moved too far back on your head it´ll become more airy, stage will be wider but shallow, moved too far to the front it´ll sound coloured, veiled and hollow with a strange stage. This is the moment where the angled driver design rears its ugly head. All headphones more or less depend on being correctly positioned to be at their best, this is equally true for the HD-600 and the HD-448. In case of the HD-558 however the effect of a wrong sitting position is much more detrimental than I expected, to find the golden mean thus is vital for a perfect sound.

Conclusion

Put it on your head, grab a good book and dream away... crap, that was corny :)

The Sennheiser HD-558 (and when it comes to sound, the HD-598 too) is a wonderful headphone: well built, important parts are replaceable. The latter will extend lifetime of this headphone far beyond of what people are nowadays used to. It isn´t a perfectly neutral headphone, it´s a bit warmer and more charming than necessary. More air would surely improve its sound; more deep bass would too. The impedance response sucks - big time. It´ll only sound perfect with many portable devices, yet Sennheiser exclusively addresses home stereo listeners. Unwise of them to create a headphone that cannot decide which personality it´s supposed to embrace. Additionally, the bourgeois design could deter people from taking it seriously. If you can accept all that, you´ll be rewarded with a stage performance besting every other headphone I´ve ever heard (and there were many). A precisely defined rendering of objects on the virtual stage combined with a fantastic sense of space surrounding these objects... as if you would take an auditorally magnifying glass to zoom in on something without ever loosing the bigger picture. It doesn´t sound boring but vibrant and apart from the warm tilt it´s neutral enough to objectively judge sound quality should you so desire. Add to that the comfort and you'll end up with a headphone that´s wonderfully suited for hours and hours of aurally pleasure. Just use it with amps having a low output impedance, adjust the position on your head until it sounds best and you´ll be having fun for years to come. Recommended.

Pro:

• incredible stage performance; surpasses every other headphone I know
• vibrant, yet pleasant sound
• neutral from 100 to 10,000 Hz...
• ... with warm & charming character
• good differentiation
• good precision
• no distortions at all
• good timing
• loud enough with portable units
• very good built quality and haptic
• comfortable to wear for extended amounts of time
• earcups big enough for almost any earform/size
• can be modded almost to top-of-the-line model
• most parts can be replaced

Contra:

• sound depends a lot on correct sitting position
• no bass at all below 60 Hz
• could be more airy above 10,000 Hz
• highly problematic impedance response
• 'surround reflector' inside earcups might hurt upon being touched for longer periods of time
• comes without accessories like bag or additional, shorter cable
• short cable comes extra & is nowhere to be found
• conservative, boring design

Last update: 17.03.2015

Mini reviews: Sennheiser Momentum & Skullcandy Aviator

Left: Sennheiser Momentum (photo copyright: Sennheiser) / Right: Skullcandy Aviator (photo copyright: Skullcandy)

Failed first attempts of buying a new headphone:
Sennheiser Momentum & Skullcandy Aviator

Everyone, and I mean EVERYONE has been talking about the Sennheiser Momentum. Unleashed to the market in January 2013, it has been reviewed by every major magazine, even the ones usually not dealing with audio stuff. Judging from countless sights of people wearing it on the streets, one gains the impression, that it is the momentary beacon of hipster culture. Fascinating; Sennheiser isn´t exactly a company I´d equate with being 'cool'. Whatever, people describe it as well sounding, perfectly manufactured, etc. But only thing is certain: it´s expensive with a price tag of 300,- Euros. My Sennheiser HD-600 now costs the same... and you´d be getting A LOT more headphone. According to some reviews (for example by German magazine AUDIO) the Momentum not only costs the same, no, it´s also on the same sound level. WTF?!?

The Momentum sounds warm. Not as warm as my HD-448 but still bloated on occasion. In fact, wouldn´t there be the lovely, slightly coloured treble response, the Momentum wouldn´t sound so different from my HD-448. Main differences: the stage is too wide, ripping apart the stereo perspective. My dear Sennheiser engineers: wider isn´t always better; the stage not only goes from left to right but from front to back too. Bass is deeper but slightly soft. Some good can be found as well: treble sounds fine. It features characteristics my HD-448 sometimes lack: precision, gloss and attack, even though it feels slightly artificial.

Built quality is 'Meh'. Sure, real leather, aluminum, glass fiber and the whole shebang... yet it all feels so flimsy and fragile. The cable - typical for Sennheiser - feels extremely thin. Good for suppressing cable-noise, bad for reliability. Thank Heavens, it can be replaced. The bag the Momentum comes in, is well built but too big for small cans such as these.

In the end I didn´t keep it: I didn´t like the stage, I didn´t like its' expensive price, I didn´t like the built quality and - most important - I didn´t like that the earcups are so small. My head is larger than average, my ears are much smaller. The Momentum still didn´t fit comfortably... and I´ve tried every setting possible. Adding insult to injury: it´s manufactured in China (not a bad thing but Sennheiser is ashamed of it) and I´m sure, that Sennheiser would still make a buck if they sell it for half the price. To make a long story short; I sent it back. And ordered the Skullcandy Aviator.

Comparison frequency response Skullcandy Aviator / Sennheiser Momentum

The Skullcandy Aviator is beautiful - if bling is your thing. Made out of metal, leather (albeit artificial), comes with ample accessories... it´s a bargain considering the official price of 120,- Euros. I got it for only 60,- Euros, who´d be able to say no? Built quality is fantastic considering the price: the metal parts look shiny & solid, the leather feels real enough, the replaceable cable is beautifully braided. Everything apart from the plastic cups is attached with screws. You can take it apart easily should you want to! As accessories you get a very nice carrying bag, looking a lot like sponge bag. Everywhere possible Skullcandy placed their logo, the skull. You can find it on zippers, the mesh covering the headphone drivers, inside the bag, on the cable plug, everywhere. From its looks, the headphones literally screams 'DESIGN-OBJECT'; the amber-coloured, transparent cups are supposed to be evocative of those sunglasses worn by pilots, glasses us Germans refer to as 'Porn-glasses' (because male pornstars used to wear them in the '70s). Through these transparent cups you can easily see what's inside the headphone; a nice touch. It´s foldable too, so summarizing, there shouldn´t be anything left to desire.

Oh wait, it´s a headphone, it has sound! Know what? It sounds well. And I was just as surprised as you are now. Look above at the frequency response, it shows pretty much how it sounds compared to the Sennheiser Momentum: less deep, but more upper bass, recessed mids, a lot more treble. Sounds bad, yet it isn´t. Mids are recessed, yes, but not as much as I´d have expected. Bass doesn´t feel bloated, treble isn´t piercing but most of the time too glossy. Upper frequencies are in fact the biggest problem of the Aviator because they aren't differentiated enough, they lack precision. Stage? As wide as the Momentums', but equally shallow - and blurred. This blurriness can be removed completely by removing the amber-coloured, transparent cups. But that doesn´t look beautiful, nor is it safe to use: you could easily break the needle thin cables on the now accessible interior parts of the headphone.

The comfort... sigh... again, a too small headphone. It´s supposed to be an over-ear headphone, yet it feels a lot like an on-ear headphone. The headband is on the brink of being too small for my head (it´s not THAT big!), the cups are still too small for my tiny ears. The cushions are very soft and comfy... but they don´t help because all the time my earlobes touch the inner mesh covering the drivers. What to do about it? Sending it back of course. Might be a nice headphone for other people though.

Review: FiiO E6

EDIT 16.01.2013: I´ve received a FiiO E07K a few days ago and the experience I made with it prompted me to alter this article. I´ve therefore updated / changed it considerably.

I bought the FiiO E6 in November 2011 and since then this tiny, little headphone amp has made me extremely happy. I´ve always loved its pleasant sound. Some weeks ago I started to think about this and asked myself: "Why do I like its pleasant sound so much? Should it really be this pleasant? Is it really neutral?" Should you be a faithful follower of my blog you already know that I´ve been using it for almost every review I´ve written for this blog, I´ve used it for my portable CD players, I´ve used it for my Sansa Clip+, I´ve used it at home, I´ve used it when I was outside, I´ve occasionally even used it with my ASUS Xonar Essence ST. The FiiO was the only thing during those months that was able to make me truly happy without showing any flaws whatsoever. And that´s where I have been wrong because it sadly colours the sound, neutrality really isn´t an advantage it has. My original conclusion to this article was:

"By now it should be clear to you, dear reader, that you won´t necessarily need an ultra-expensive headphone amp in order to experience sonic bliss, the FiiO E6 will do just nicely with headphones ranging from 32 to 150 Ohms. You shouldn´t trust someone who describes the FiiO being inferior to other amps because of its low price. I cannot imply you to buy it, you have to make that decision yourself. Will you like the additional little box attached via another cable to your portable player? I myself don´t mind but what about you? I´m afraid you have to find that out on your own. You might be in for a sonic surprise, a surprise you maybe won´t like: bass you were expecting from your iPod may now be absent just because the problem of mismatching output impedances has disappeared. In my opinion however you should try it out anyway, $ 25 won´t really hurt and you might be eternally rewarded with something you´ve never experienced before with portable players: almost perfect neutrality, nice design and more gain than ever before."

To reach this conclusion I´ve beforehand spent precious time on explaining what a headphone amp is, gave examples why you won´t need a costly headphone amp, why the FiiO E6 would also work for a headphone as expensive as for example the Beyerdynamic T50P... blablabla. But everything changed with the arrival of the FiiO E07K: To review that headphone amp / USB DAC I wanted to find out how it would behave against a reference, something that appeared to be impossible for me before. When I wrote the original review for the E6 I couldn´t think of a method to reliably compare it to an ideal reference and I´ve always hated that I seemed unable to find out how it really sounds, if it colours the sound, if it introduces errors... when all of it could have been so easy.

FiiO E6

This feeling of being unsure about my rating for the E6 and the aforementionend original conclusion has been nagging me since publishing this article in June 2012. But now I´ve found a way to come close to an objective-as-possible review. My E-MU 0202 USB broke down during the last days of December 2012 which has robbed me of my basic ingredient for my testing methodology. It not only forced me to devise a new method for comparing units like portable CD players to my reference files, it also prompted me to alter my testing chain for such an occasion so that I could finally perform listening tests with headphone amps. The solution for latter cases was the Creative Soundblaster X-Fi HD USB: by recording its lovely output with my ASUS Xonar Essence ST I would create a new reference against which the FiiOs would have to be compared. Therefore my usual reference files for the FiiOs wouldn't be the sources for the testing material themselves, instead I´d use the playback from the X-Fi HD recorded with the Essence ST. After that I would insert one of the FiiOs into the signal chain between the Creative and the ASUS to record the same material again. Have a look:

Reference:

Creative Soundblaster X-Fi HD USB -> ASUS Xonar Essence ST = Reference files

Compared to:

Creative Soundblaster X-Fi HD USB -> FiiO E6/E07K -> ASUS Xonar Essence ST = FiiO results

Having done both steps I now would be able to compare the results from the FiiOs to my reference files. The sonic characteristic of the X-Fi HD USB embedded in the reference files would ideally be completely retained by the FiiOs and in case they´d colour the signal it would be incorporated as well. There are some obstacles which again might colour the results: the line-in of my Essence ST is electrically not comparable to a headphone which is what the FiiOs were designed for in the first place. From what I´ve read a line-in is a much easier load for a headphone amp than a headphone with an impedance of for example just 15 Ohms. Which means that when one of the FiiOs is driving the line-in of the Essence ST it could theoretically be 'easier' to 'hide' possible colourizations. But I nevertheless think that this method enables me to reliably spot differences compared to unplugging/plugging headphones from one amp to the next. To do it this way also made it feasible to create crude measurments; I wanted to see how these would compare to the professional measurments NwAvGuy did for his wonderful article about the FiiO E6.

FiiO E6 from below

To shorten this article I´ve partly removed the long & boring explanations about headphone impedances, gain and possible advantages of having an extremely low output impedance. Some of those sections can now be found in the article about the FiiO E07K; albeit shortened much. Reason: there are much better articles to be found anywhere else, much better than I could ever write them. Just a few words about the features the FiiO E6 offers: its high input impedance gets rid of every impedance mismatch error a headphone amp can possibly produce. It sports an output impedance of just 0.25 Ohms (according to NwAvGuy´s measurment) which should be perfect for any headphone and certainly doesn´t pose any problem for my Sennheiser HD-448 or my Superlux HD-668, even though these aren´t very susceptible to high output impdances anyway. I bought the E6 to rid myself of any possible impedance problem I could ever encounter with portable players. For example: most of the headphone amps inside my portable CD players or MiniDisc recorders have a fairly high output impedance and other obstacles not advantageous for good sound. Then there is the amp inside my X-Fi HD USB which supposedly has an output impdance of 36 Ohms and the one built onto the Xonar Essence mainboard (11 Ohms) - both are not suited for difficult IEMs. Connected to their headphone outputs the FiiO E6 simply renders this problem moot. But this tiny headphone amp can also be used with line-outs from stationary sources, for this purpose it is equipped with a special low-gain setting (2 Volt mode). However, you might run into problems should you be using it with one of those louder-than-usual sources I described in this article: instead of the quasi standard 2.0 V (which the FiiO adheres to) they sometimes output with 2.2 or 2.5 V, supposedly leading to severe input distortions when used with the FiiO. But now off to my own measurments and audio rating, yes?

FiiO E6, 1 Volt Mode (High Gain)

FiiO E6, 2 Volt Mode (Low Gain)

FiiO E6, Total Harmonic Distortions, 1 Volt Mode (High Gain)

FiiO E6, Total Harmonic Distortions, 2 Volt Mode (Low Gain)

FiiO E6, Intermodulation Distortion, 1 Volt Mode (High Gain)

FiiO E6, Intermodulation Distortion, 2 Volt Mode (Low Gain)

Click on one of those pictures and scroll trough them using your mousewheel, you´ll find that differences between them are not that big. In low gain mode the FiiO does have less high frequency distortions but I really don´t know if they´re audible or not. But low frequency distortions seem pretty strong to my layman eye, these could be responsible for its pleasant sound by adding something that doesn´t belong there. Everything else looks fine though, especially when you consider how small this amplifier is and that it is selfpowered - there must fit a battery somewhere into the small case. What I couldn´t find however was the overload problem NwAvGuy described in his article about the E6... strange. Well, I have to consider the possibility that my X-Fi HD USB simply isn´t loud enough on its line-out (I have to change the volume by +2.5 dB after recording anything that has been played back by it if I want to have 0 dBfs) to distort the input of the FiiO. It´s also possible that his unit behaved a bit different, who knows.

FiiO E6 power/EQ/voltage mode button

At this section you formerly could find sentences like "...seems to warm up the sound ever so slightly (...) sometimes it sounds a tad boring...". Being now able to compare the FiiO to my reference files good and proper I have to say that those sentences have been the understatement of the century because the biggest problem of the FiiO E6 are dynamics. It removes much of the impact transients have, bass punches disappear into nowhere and high frequency attacks loose their bite. This is very evident when the FiiO-derived results are compared to my reference tracks from The Girl with the Dragon Tattoo or Episode II soundtracks; both show an extreme amount of very high frequency attack diminished quite a bit by the E6. Deep bass impact is decreased as well, all of this reduces the impression of dynamics, snap and punch. Madonna's track "Skin" sounds mellower than usual, its timing is noticeably slower when amplified by the FiiO. Well, at least speed is reduced consistently throughout the whole frequency band. People call this lack of dynamics a 'laid back' sound, I call it wrong. Resolution and definition are other areas where this amp fails: cymbal hits by an orchestra are close to being reduced to simple noise, the characteristic noise of a bow striking the strings of a violin quite audible on the reference tracks sounds unintelligable through the FiiO. Rendering the depth of a virtual stage correctly is equally problematic because it flattens the stage considerably, the impression of a holographic recording disappears. On the other hand, sonic balance and character of a recording are retained decently. While deep bass and highest frequencies are diminished, this isn´t very noticeable because it is compensated by adding a small amount of upper bass even though this also thickens up frequency areas that shouldn´t be thickened up. This is very audible on the organ recording I always use (Vierne, Michael Murray)... in fact, the organ almost sounds as if having distortions (there are none of course on my reference files). I´m very sorry but here are my reconsidered ratings for the FiiO E6:

Sonic Balance:

Dynamics:

Resolution:

Stage / Ambiance:

Character:

FiiO E6: black beauty

As it turned out in the end people deriding the FiiO for sounding boring were right - but I wouldn´t see it. Maybe because I didn´t want to or because I lacked a convincing method for comparing it. I´m not ashamed to rectify this error but have you noticed how many of my articles have been rewritten recently? First the one about the X-Fi HD USB, then the Sansa Clip+ and now this one. And then the articles about USB cables which in hindsight feel like a hoax... I really have to be more careful in the future when reviewing some audio gadgets. I have to measure what I can measure and I have to get rid of any error that could possibly happen. And I have to find and maintain valid methods for objective comparing before I´m writing a review. Anyway, sonic signature of the FiiO E6 is not as disadvantageous as it sounds because it can be used to create exemplary synergy effects with audio players or headphones. Take my Sansa Clip+ for example: combine it with this FiiO to achieve much more gain and you´d have a balanced sound you wouldn´t expect when listening to the Clip+ alone. The explosive dynamics of the Clip+ are a perfect fit for the mellow sound of the E6, its sonic disadvantages will compensate the ones from the FiiO. During the last days I´ve tried this combination and it indeeds sounds very well. Besides: both of them are tiny and don´t cost much. On the other hand, a combination of the FiiO E6 and the Sony NW-A 1000 has to be avoided; you´d drown in mellowness and boredom. This little amp would also be a good choice for overly dynamic or 'sounded' headphones (from Beyerdynamic for example). The FiiO doesn´t fit the signature of my Sennheiser HD-448 - but the opposite happens with my Superlux HD-668B. In the end it all comes down to your preferences: if you like a dynamic, very crisp and bassy sound signature you´ll have to shun it. But if you know what you are doing and if you plan to compensate for an aggressive sound signature the FiiO E6 will be perfect for you.

Last update: 16.01.2013

An attempt at "Remastering" - the score to "Hook" serving as an example

EDIT: During the last few days I´ve received numerous complaints about this article. I came across (here and on the FSM forum) as an arrogant and vulgar jerk and I´ve needlessly offended everyone; Bruce Botnick, John Williams, Americans, LaLaLand Records and score collectors, I even uttered racist sentences. I am very sorry for this. I created problems I didn´t want to create and also ridiculed what I was trying to say. The version of this article you´re about to read has therefore been revised to a much less - I hope - offensive state and while editing after publishing is an example of poor form I still had to do it. I would like to thank David Sones who moderates the FSM forum and also my best mate, both made it clear to me that I have been unable to see the bigger picture and that I was navigating myself into a corner I never wanted to be in.



Hello my friends! Since my last article exactly one month has passed and while I´m sorry that I needed so long to write another one I tend to think that I used this time quite well. I´ve acquired several additional portable CD players (including a legend and a surprise - review to follow), did something incredibly stupid (will also be described in the CD player article) and I´ve purchased some new CDs, soundtracks / filmscores to be exact. Today I´d like to talk about one particular score, the one to the Spielberg directed "Hook". As usual with Steven-Spielberg-directed movies master composer John Williams wrote that score and among fans it rightfully ranges as one of the best efforts Mr. Williams achieved during his very long career. It often pops up in "Best-of" lists, you may remember that I too mentionend it in my own "Best sounding scores of all time" list. I included it because the original CD from 1991 sounds so wonderful and perfect, it truly is one of engineer Shawn Murphy's (many) outstanding efforts. Still, this CD wasn´t complete and for us score collectors a lot of the music we heard in the movie was missing, regardless of its more than generous 70 minutes runtime. You see, John Williams had written roughly 130 minutes of music, not including alternates, songs and early versions for the ultimately unrealized musical version that was planned originally.

The first release (1991, Epix Soundtracks, copyright by Sony Music)

While it´s fairly common not to release everything that has been written for a filmscore (because of redundancy or musician's fees) fans were disappointed nevertheless by what had been left out. Not surprisingly roughly ten years later or so a bootleg surfaced, which - as is common with bootlegs - suffered from very bad sound quality. Finally in 2012 the pleading of us score fans was answered and an official and completely new release was unveiled by the label LaLaLand Records with a lot of help by Sony Music. They took the greatest care in assembling this 2-CD-Release, even John Williams himself participated in its assembly - and he rarely embraces expanded releases of his scores and it´s even rarer for him to take part in their creation. Everyone involved seemed to make sure to create the best version possible which probably wasn´t easy at all. I recommend to read the interview with producer Didier C. Deutsch, it´s very enlighting in regard to the problems producers of these releases sometimes have to face. In any case, I was very excited since "Hook" has been one of the first scores I ever bought and it has been partly responsible for musical preferences I have today. But not only my musical tastes were shaped by it, no, it also shaped my sound preferences because its sound shows true reference qualitites. You can imagine that I was very excited to have the possibility to own the expanded version myself - more wonderful music, more wonderful sound. I was eagerly awaiting the day of its arrival and because it was that important to me even my boyfriend took part in my decision to purchase it (he isn´t interested in filmscores and is sometimes ennerved by the scores I listen to with my headphones :)), he subsequently decided to present it to me for the Easter Holidays - a very welcome surprise.

The second release - beautiful, isn´t it? (2012, LaLaLand Records, copyright by Sony Music)

Another surprise awaited me on the day of its arrival: while I adored the additional tracks I´ve until then only known through the movie (I have seen it many times; I own it on DVD) I became irritated about their sound. The tracks I knew from the first CD sound like their old pristine self whereas the new tracks sound a bit different. For sure, their level of sonic perfection is equally high but at the same time they have more bass, treble and an increased amplitude. Example: one minute you get to hear the full orchestra playing Hook's theme in "Presenting the Hook", the next minute an assumed smaller part of the same orchestra follows with higher volume level (roughly 2-3 dB) combined with a slightly different sound signature on the new track "Pirates!". To make a long story short, I was disappointed because I was expecting the perfection I am used to from the first release - what I got instead was sonic inconsitency. An even bigger disappointment were "The Ultimate War"-track: this piece also known from the first CD has been expanded into three seperate, considerably longer tracks with help from a less than pristine source; the producers obviously had to use the score mix prepared for usage in the movie alongside sound effects, dialogue etc., I assume that the original masters for these tracks couldn´t be located. But why does the usage of this score mix pose a problem? For one, this score mix is a copy of the original master, here in this case an analogue copy as evidenced by the drop-outs and slight channel imbalances. Secondly, this mix has to play alongside the rest of the movies' soundtrack with the result that the music changes amplitude constantly; it gets dialed down for dialogue or turned up for action sequences, sometimes in a fraction of a second. Another problem of these tracks is their amount of editing: in an effort to closely mirror the music we knew from the movie the first, formerly pristine half of the "The Ultimate War"-track has been edited, effectively creating loops and awkward jumps in continuity and amplitude. Remember, this was done on a part of the track that formerly was perfectly allright. Additionally they adapted the amplitude of this first half by increasing the volume level in order to match it to the mixed-in moviemix, creating an example of the "Loudness War" as a result and making the whole track another 2-3 dB louder than the rest of the score. Two examples are below:

The whole orchestra playing at full force on "The Ultimate War"
 - brickwall limited in combination with 0 dBfs distortions (clipping)

"Remembering Childhood" - note the part 8:30 to the end: this is the complete orchestra playing
again at full force - with roughly 5 dB less RMS level though. Peaks and dynamics are intact.

While this seems to be bad it isn´t: the three seperate "Ultimate War"-tracks are the only "Loudness War" tracks, the rest is perfectly fine. But I´d like to know if you, dear reader, were disappointed like me or if you don´t mind something like this. Does brickwall limiting combined with clipped 0 dBfs peaks disturb you? I´m of the opinion that maximizing the RMS level doesn´t make sense with purely orchestral scores, it´s not like this music has to compete with Lady Gaga's newest release on the radio. I guess what mystifies me the most is how a release in the year 2012, published by two very experienced and professional companies, with a pristine mastertape as a source, happens to be mastered like this. Usually one of the goals of mastering involves matching the sound of one particular track to the sound of another, possibly different sounding track. This process consists of bringing their respective frequency responses (or sound signatures) and their overall RMS levels with respect to the overall dynamics of the complete work closer together. It seems strange to me that it wasn´t done here, especially when considering how much work went into a spectacular release like this. Possible reasons? I don´t know, I´d be very glad if someone with information would be able to comment on this article in order to shed some light into this.

But what can we do about "Hook", assuming you were disappointed like me? We can try to improve it further by turning it into a more coherent sounding soundtrack CD. To make it easier from now on we`ll use the term "Remastering"; plain wrong and misleading of course since we´ll be doing nothing more than adapting the sound to our own personal taste or what we presume to be correct; what we are certainly not doing is creating a new master because only that would justify the term. But for the sake of convenience let´s just use this imprecise expression, shall we?

In the rest of this article I shall attempt to show you how I "corrected" the shortcomings I´ve talked about, if you want to do the same let me remind you that you should work with good Equalizers, resamplers and DAWs only. I also advise you to work within a bit depth of 32 Bit Floating point; while the guys at hydrogenaudio.org might tell you that it´s unnecessary I disagree since you will be doing lots of processing; working with this bit depth helps to preserve the original resolution and dynamic as close as possible without introducing too much errors. Additionally I also work with a version I upsampled to 176.4 kHz before using EQs or other DSPs because A) I keep the resulting 96 kHz files and B) signal processors are using oversampling - with a higher samplerate I can avoid additional oversampling close to the passband.

These are the tools I used for "Hook":

iZotope RX Advanced, iZotope Ozone 4, Sound Forge Pro 10.0
You don´t have to use them, Audacity is very good too. 
For up- and downsampling I´d also recommend SoX
 - it´s a very good resampler for use within foobar2000

My configurations for iZotope Ozone 4 are following below. I attempted to change equalization and stereo separation beside level changes, both of them are intended to change the new tracks into something that closely resembles the original tracks already known from the 1991 release; their frequency response has not been changed for the new release (apart from noise-shaped dither) so they are serving as the ultimate reference to which all other tracks have to be adapted. My configuration should also be applicable to other equalizers... and don´t worry about using them: they tend to be described by audiophiles (people like me) as coming from the devil himself when in reality they are used on 99% of all releases out there, with them it´s like with all things: it doesn´t matter if you use them, it only matters how. So, still with me? Want to see a world you´ve never seen? Then take my hand for I now will take you down the rabbit hole.

EQ setting for the new parts (except "The Ultimate War") - called "NewTracksEQ"

Stereo separation for the new parts (except "The Ultimate War") - part of the "NewTracksEQ"-setting

EQ setting for the filmmix tracks ("The Ultimate War" etc.)
- called "FilmMixEQ" (yes, these settings are quite severe)

Stereo separation for the filmmix tracks - part of the "FilmMixEQ"-setting

After my suggestions for equalization and stereo separation I´ll now show you where and how they are going to be applied. I came up with the augmented tracklist below during my initial spotting session of the score, I tried to find out the differences while at the same deducing what needed to be done to flatten them out. The tracks I assumed to be in need of change / processing / corrections are marked in bold letters. A warning: some of my suggestions seem - even to me - completely arbitrary because I cannot know the differences exactly, it´s a bit like solving a puzzle. My obsession with perfection however caused me to be as precise as possible by using the "Matching EQ" feature of Ozone in combination with my own experience to work out frequency response differences. Anyhow, I think that my reworked tracks still sound slightly different even though my result now seems to flow very nicely sonically with less audible differences than before. This isn´t an exact science (not in my case) but done by gut feeling only. The level changes for example were "calculated" only by how I think how loud an orchestra utilizing certain players should be at some particular point. Granted, because of years of exposure to orchestral recordings I like to think that I´ve gained some experience but I can never be completely confident, therefore you should regard my suggestions with reasonable doubt. Another point: while I strive to create the perfect version you mustn´t necessarily do the same, if you´re lucky with something else it´s perfectly ok and probably more healthy anyway.

DISC 1

1. Prologue 1:30 (adjusting level for -2 or -3 dB)
2. We Don't Wanna Grow Up 1:50  (adjusting level for -1.5 dB)
3. Banning Back Home 2:25  (adjusting level for -1.5 dB)
4. Granny Wendy 2:57
5. The Bedroom 1:07 ("NewTracksEQ")
6. The Nursery 1:38 ("NewTracksEQ")
7. The Watch :56 ("NewTracksEQ")
8. Hook-Napped 3:56
9. A Portrait Of Wendy 1:06
10. The Arrival Of Tink/The Flight To Neverland 6:03  (from 5:24 "NewTracksEQ")
11. Presenting The Hook 3:01
12. Pirates! 2:41  (adjusting level for -3.0 dB & "NewTracksEQ")
13. Hook Challenges Peter 7:50 (adjusting level for -2 or -3 dB & "NewTracksEQ")
14. From Mermaids To Lost Boys 5:13 (from 4:08 adjusting level for -2.3 dB & "NewTracksEQ")
15. The Lost Boy Chase 3:32
16. Smee's Plan 3:25 (adjusting level for roughly -3 dB & from 1:42 "NewTracksEQ")
17. Pan Is Challenged 1:20 (adjusting level for roughly -3 dB & "NewTracksEQ")
18. Hook's Lesson 3:08 (adjusting level for roughly -3 dB & "NewTracksEQ")
19. The Banquet 3:10
20. The Never-Feast 4:41
21. Hook's Madness 4:00 (adjusting level for roughly -2 dB & "NewTracksEQ")
22. Follow That Shadow 2:38 (adjusting level for roughly -2 dB & "NewTracksEQ")

DISC 2

1. Remembering Childhood 11:04
2. You Are The Pan 4:03 (from 2:26 "NewTracksEQ")
3. When You're Alone 3:16 (adjusting level for -1.5 dB)
4. Tink Grows Up 2:20 ("NewTracksEQ")
5. The Ultimate War: To War 9:45* (up to 2:23 adjusting level for -3 dB, from 2:23 onwards adjusting level for another -2.5 dB - from beginning to 7:50 "NewTracksEQ" & from 7:50 onwards "FilmMixEQ")
6. The Ultimate War: The Death Of Rufio 2:36* (adjusting level for roughly -5 dB & "FilmMixEQ")
7. The Ultimate War: Sword Fight 5:32* (adjusting level for roughly -5 dB & "FilmMixEQ")
8. Farewell Neverland 11:15* (up to 1:01 "FilmMixEQ" & from 1:01 "NewTracksEQ")
9. End Credits 6:08 (adjusting level for roughly -2.5 dB & "FilmMixEQ")
10. Prologue (alternate) 1:35 (adjusting level for roughly -2.5 dB & "NewTracksEQ")
11. Banning Back Home (film version) 3:14* (adjusting level for roughly -2.5 dB & "FilmMixEQ")
12. Presenting The Hook (film version, extended) 4:03 (adjusting level for roughly -2.5 dB & "NewTracksEQ")
13. Hook's Blues 2:17 (adjusting level for roughly -2.5 dB & "NewTracksEQ")
14. Wendy Tells Peter The Truth (partly unused) 2:24 (adjusting level for roughly -2.5 dB & "NewTracksEQ")
15. Exit Music (unused) 1:42 (adjusting level for roughly -2.5 dB & "NewTracksEQ")

* Those tracks show numerous level jumps / distortions - these cannot be corrected with one singular setting, they can only be corrected individually by hand.

All in all it took me four days (or eight consecutive hours) to do this, not counting the final listening session right before the finishing touches (downsampling and/or dithering) in order to spot curiosities I might have overlooked or introduced myself. Luckily, I couldn´t find any. The most daunting part for me was the spotting session and averaging out the level jumps on the filmmix tracks. The latter had to be done entirely by hand, including going back to other parts of the score consisting of (an assumed) similar orchestration & overall amplitude to compare my changes with them as a reasonable reference. My experience came in handy so I think that my result is worthwhile and a slight improvement over the original CDs.

"The Ultimate War" now looks like this - quite different, eh?

So, after all this talk you must be interested in hearing some of my results, aren´t you? Well, lucky you because now I´m going to present fully legal samples from my version compared to the tracks in their original state from my own CDs. As I´ve said, these samples are fully legal, they are under 30 seconds but are nonetheless able to show what I did. Before I forget: my versions are lower in volume (roughly 3-5 dB) so you might want to adapt the volume of your stereo system in order for both versions to be roughly at the same volume level.

Here I removed a loop that wasn´t there before

Corrected volume jumps, overall level and EQ - not perfect but improved

Corrected volume jumps, overall level and EQ - not perfect but improved

Corrected volume jumps, overall level and EQ - not perfect but improved