Some Hi-Fi Myths

Firstly, here is my full system:    select thumbnails at the bottom or hover over to pause the slideshow

Secondly, there are many myths floating around about Hi-Fi, so I'll try to explain a few here:

• Firstly some definitions. What's what? A basic system comprises of the source component, it's interconnect, the amplifier, speaker cable, and the speakers (inc. stands if needed). Upgrades are when you change an essential part of the basic system (i.e. what you need, bare minimum, to get any sound) like the cd player for a better one. Upgrades usually always cost money. Tweaks are things you buy to try to enhance the existing sound by adding to what you already have, broadly speaking anyway. Most tweaks cost money, with the more absurd ones being more expensive and the best ones being free. Optimisation is altering things like positioning of speakers to get the best sound. It doesn't cost anything (if it did it would be a tweak) and you aren't adding anything to your existing system (again, a tweak)...you're just getting the best of what you already have.
   
  
• Different analogue cables make your Hi-Fi sound different. Maybe. I upgraded my interconnect from Ixos 1002 Gamma Encore to Kelvin Audio K2 and found an improvement in sound quality. Upgrading to Chord Co. Carnival Silver Plus was an even bigger step up. People usually cite the issue of interference....that's why 'figure of eight' cable is normally not used due to it's unshielded nature but then again, shielding is pretty easy to get right, so it would be a surprise if even a cheap cable wasn't up to the job here.
  However analogue cables convey analogue signals that are still subject to variation and degradation and that are propagated differently depending on their resistance, inductance and capacitance properties. This is important because cables should aim to be neutral, i.e. not let the signal between the components be altered in any way. For example, an analogue coax that doesn't convey a lot of treble will sound good in an already bright system. This is because it is artificially taming the treble. However, this does mean that the cable itself is altering the sound quality (whereas my set-up relies on the components to sound superb - the cables should only be there to complement the system, allowing it to shine through, not subtracting anything from it). The same goes for (analogue) speaker cable, it really can improve the sound quality of the system, again not aiming to diminish anything, merely allowing the amp and speakers to sound as they should do in the first place (i.e. the cables should be 'transparent' for want of a better word). There are many different types of analogue cables to choose from (most of which sound similar) so you'll have to try them and see. Also remember that many claims by cable manufacturers end up being "snake oil" and not in any way verifiable so YMMV. You can get the best info on cables from Audioholics.
   
  
• Electrical coaxial digital cables are better than optical digital ones. False if running short lengths. What do digital cables convey? 1s and 0s. That's it. Oh, and very good error correction at the other end that eliminates the chance of mistakes. Coaxial cables send electrical signals and optical ones send light, but they're sending the same signals. So in practice over short runs of wire, it makes no difference at all, with digital you either get a 1 or a 0 no matter what cable you use, so they sound identical.
  However, if you need longer runs of wire over 10 meters, all metal wires are subject to amplitude loss and signal degradation and will "reveal" this degradation because the signal is running through a long enough wire to lose integrity/amplitude, theoretically resulting in signal dropouts and waveform distortion (plus you might get ground loops due to the metal construction). On the other hand optical cables will theoretically run all the way around the Earth and still give no loss of quality (light won't degrade in a sealed glass conduit like electrical signals can in metal) which is why they are the cables supporting most communication infrastructure around the world.
  So for <10m metal coax and glass optical cables will result in the same digital signal with no differences between them, but for >10m I'd stick to optical cables where needed.
   
  
• A £600 analogue cable will sound better than a £60 analogue cable. Probably not. It has nothing to do with price. The £600 is for people with more money than sense. It may sound better, it may not. There is a law of diminishing returns which means that you have to spend larger amounts of money as you go up-market to hear smaller differences in sound quality. Besides, the wiring inside a component is made of 50p wires. Do you really think that by adding an expensive wire it will magically transform the system?...the wiring inside the components is still the Rate Determining Step and a cable will only make a difference up to a point (a.k.a. generally speaking, think carefully before spending above £100 for your cables - you usually don't need to).
   
  
• A £600 digital cable will sound better than a £1 digital cable. Absolutely not. It's digital remember...1s and 0s? It is either built to spec and works or it isn't and doesn't. If it doesn't, sound will drop out entirely. If it does, the data will transmit identically regardless of price, so it will sound "perfect" already...it cannot "improve" with a more expensive cable if that cable is of the same EIA/CEA-861 spec. Even jitter is accounted for because most modern DACs have associated clocks and buffers along with robust error correction which means you must be getting all the correct data. Again, it has nothing to do with price and the £600 is for people with more money than sense.
   
  
• Speaker stands are needed for smaller (bookshelf) speakers. True. Most (probably all) non-floorstanding Hi-Fi speakers are designed and developed with correct stand placement in mind. Weighted stands allow the speaker cabinet to remain immobile whilst all the drive unit(s) energy goes into producing sound (not making the speaker wobble - an inefficient use of sound waves). It also places the tweeter at the correct height (eye level). Stands should be filled with something heavy - lead shot or sand will do fine.
   
  
• Stabilising discs, equipment racks and squash balls improve CD sound quality. False. As long as the equipment itself is immobile, the laser used to read the information (from the CD) is reading a 'correct' set of data (see technical explanation in the next section below). I repeat that once a CD transport is physically immobile, it's physically immobile so it will be reading the disc accurately enough anyway. Thus a £1000 rack does the same job as a £250 rack or a well made £25 solid table. The money you're spending on a fancy looking Hi-Fi rack is going towards only that...the fancy looks.
  Squash balls and isothorbane feet are also meant to isolate decks, but actually add nothing to the sound if it is already on a good rack (it's already immobile anyway).
  Stabilising CD platters tend to largely be snake oil as the CD laser reading mechanism doesn't gain any benefit from the extra damping. Put another way, if the laser was having trouble reading a clean set of data, you would hear sound dropping out. You don't have a situation where the sound "improves" with CD platters. It's digital folks, you either have accurate sound or you have none.
   
  
• Demagnetising discs and anti-static mats improve CD sound quality. No. Audiophools make me laugh....how on Earth can magnetism or static affect an OPTICAL PICKUP SYSTEM? People who hawk these snake oils and people who 'believe' in them simply have no grounding in basic G.C.S.E. Physics and really should spend their money on remedial courses in college or something rather than wasting it on crap that simply cannot have any effect on signal transfer/sound.
   
  
• How do CD players work then? Warning: techie talk. Here I'll try to explain how the CD player decodes the CD. The 1's and 0's are actually a "snapshot" of the complicated waveform at one point in time (basically the numerical value of the amplitude of the waveform at that point). Each of these values is encoded as this binary number (base 2 rather than base 10). If the CD were encoded as 2 bit, the waveform can only have values of 0, 1, 2 or 3. So increasing the number of bits to 16, allows us to encode any whole number value between 0 and 65535 (which, including 0, equals 65536 values (= 2 to the power 16, hence it is a 16 bit signal)). So you get a string of numbers per value, for instance, the value 15 would be 0000‌0000‌0000‌1111, which is a word (the binary representation of that specific amplitude). Now, 44100 snapshots take place every second (hence a CD being encoded at 44.1kHz) so you have each word overall 44100 times per second - that's a lot of numbers for just 1 second of music, enough to mean that you will inevitably get a few errors now and then. So the CD itself has a set of built in error correction systems.
  At the end of each word is a parity value that tells the DAC how many 1's or 0's it contains, so any missing values can be "filled in" by it. The "filled in" values are not just guessed, they are obtained from tracks of data that are not normally used and are, at their simplest, merely 'backups' of the audio signal that can be used to fill in the blanks if original data is lost (redundancy). By increasing the number of parity bits, we can create quite complicated systems that allow us to correct and detect more accurately. Secondly, the data is not set out in a continuous string of numbers like words in a book. They are, in fact, all jumbled up - interleaved. This means that if you had the equivalent of an error in the phrase WHATISYOURNAME, it might end up looking like ISYOURNAME. Whereas, with interleaving, the phrase is coded as AIAORMEHTNSYUW so the mistake creates RMEHTNASYUW which recodes as WHT S YUR NME, which is easier to correct (it is easier to correct small missing "bit errors" than longer tracts of lost data called "burst errors"; so interleaving tries to create manageable bit errors from more difficult potential burst errors). We can also use a system by which the data is arranged so that there are fewer transitions between 1 and 0 (or vice-versa), which is where errors can occur. This allows pits on the CD to be grouped in such a way as to only trigger certain possible frequencies in the laser pickup, so therefore making it easier to decode correctly - modulation.
  Finally, if these initial stage corrections (built into the way the CD is coded) do not work, the DAC has two more tricks up its sleeve: 1 It can interpolate (guess) the correct missing values based on what it sees either side of the missing figure (so it's sound will improve if either the amount it is interpolating gets reduced or if what it does interpolate is a more accurate representation of the waveform) and 2 if this fails it will mute the sound altogether over the mistake - so you will hear sound dropping out from the speakers.
   
  
• Up/Oversampling a number of times is important. Yes. Firstly, as explained above, sampling is the 44.1kHz rate at which data is stored and read off the disc. The filters in the DAC will produce time distortions and artefacts that are audible (especially during quantisation - the actual conversion from bits to an analogue waveform).
  OVERsampling involves increasing the sampling frequency from 44.1kHz to, say, 88.2kHz in an effort to reduce those distortions and result in them being a smaller overall percentage of the signal as they’re spread out over a much larger bandwidth. More importantly this will also have the effect of increasing the signal:noise ratio which should improve the sound which is why pretty much all DACs do this (as an example, my Naim CD5si and my Arcam irDAC use different DACs from Burr Brown but they both utilise 8x oversampling). The process itself is referred to as oversampling when it takes place within the Digital to Analogue Conversion itself within the DAC and is usually an integer value (eg 2x, 4x, 8x or 16x oversampling).
  UPsampling OTOH takes place in the digital domain only (i.e. before the conversion to analogue) and is an arbitrary increase not necessarily an integer (e.g. upsampling to 192kHz from my Naim and Arcam DACs is 4.353x the standard 44.1kHz signal). It has gained popularity recently as you can feed a standard optical output from a CD player into a separate outboard DAC and that can upsample to whatever it wants. The goal of upsampling is the same - increase bandwidth to reduce noise to increase signal:noise ratio.
  Both of these methods involve interleaving new data values in between the 441000 data points/sec of the existing waveform and therefore interpolating a more accurate analogue waveform from the digital data.
   
  
• Increasing the DAC's working bit level is important. Yes. Hand in hand with sampling is the opportunity to increase the quantisation level as well, to maybe 20 or 24 bit (again, using my CD5si & irDAC as examples, they both use different Burr Brown 24 bit DACs). Now even though the raw data is only ever 16 bit in a CD, any processing of a 16 bit signal will be more accurate if done at a higher bit rate (as I've explained in more detail wrt video processing ). So a higher bit depth is the equivalent of having a higher resolution and will result in a more accurate waveform reproduction.
   
  
• So if it's all digital and foolproof, can DACs still sound different? Yes. DACs themselves have inherent properties that will alter their performance including their Resolution (see above), Bandwidth (see above), Monotonicity (the ability of the output to only move in the direction of the input - so no output 'dips' initially when input rises), Total Harmonic Distortion/Noise (distortion and noise introduced into the signal by the DAC - this should be as small as possible) and Dynamic Range (difference between highest and smallest signal).
   
  
• Re-clocking the player's DAC will improve it. Possibly. Firstly, a CD player needs two signals, one is the binary one, the other is the individual timings of those values, needed so that the DAC can place those signals in the correct (temporal) sequence. Thousands of 1's and 0's are being read off the disc, but need to be decoded by the DAC at regularly spaced intervals (i.e. it needs to know when to decode the next word). So the DAC has a "word clock" which is a timing signal input that tells the DAC to decode the next word every 1/441000th of a second (for a 44.1kHz sampling rate). It is here (and only here, at the DAC's word clock input) that jitter can cause a problem by causing the words/values to not be correctly spaced. If this were to happen, a pure sine wave, for instance, would end up looking slightly irregular as certain numbers are closer to their neighbours than others, rather than equally spaced. Now if the sampling frequency were only 1Hz then a normal quartz watch oscillator would be fine, but here the word clock is running at 44.1kHz so you only need inaccuracies in the order of nanoseconds in a 16 bit DAC to possibly be audible and induce jitter, therefore it has to be very accurate (jitter is very inducible in the path between the transport and the DAC, so to fix this samples are read off the disc and placed into a RAM buffer and then passed out regularly, which means that samples are reliably fed out independently of the transport or reading mechanisms).
  Re-clocking involves trying to tackle jitter where it matters...yes you guessed it, at the "word clock" input of the DAC by having timing so accurate that it introduces no jitter, however, if the DAC already has a properly made and extremely accurate master clock then things may not improve at all.
   
  
• Do all accessories always improve a Hi-Fi? No. A lot of reviews in Hi-Fi magazines have one thing in common - PLACEBO. Medically speaking, placebo can account for even up to 90% of improvements in symptoms in some studies. Now if a person can lower their symptoms by BELIEVING they are receiving medicine, it's entirely plausible that a lot of so called improvements in Hi-Fi are because people believe they will work. When you review a cable in a shop, the fact that you know you are listening to interconnect X allows for the effect of placebo. The way to eliminate placebo and truly assess the merits of an upgrade would be to blindfold yourself and let the shop assistant RANDOMLY connect different upgrades WITHOUT SAYING ANYTHING OR WHAT THEY ARE and let you truly listen for yourself with no clues as to what you're listening to. This is called a single blind trial. I don't think any Hi-Fi magazines conduct blind trials and in the absence of any of them saying they do, you have to assume that placebo and personal preference (as well as advertising revenue) plays the largest part in their comments. The only scientific way of truly testing a hypothesis is to conduct a double blind trial in which both you AND the cable swapper have no idea what is being tested.
   
  
• Adding a sub is not 'High-Fidelity'. False. This is Hi-Fi snobbery at it's worst. Without a sub, there aren't many speakers around that go down to 20Hz (lower limit of our hearing) on their own with the same appreciable loudness as the rest of the frequency curve (i.e. where the lower bass is at the same level as the mid/upper bass, mids and treble). Sure, most speakers (even stand-mounts) will output something at around 35ish Hz, but it's usually a wimpish fart-like sound; something that you can only just about hear...the same speakers may only give good levels of SPL from about 50Hz upwards (regardless of what their on paper specs try to convince you of). To achieve that lower limit you need both the speaker with the capability of doing it (e.g. my current Tannoy Turnberry’s) sitting in a room with the correct dimensions/composition to allow it (e.g. a ground floor, brick walled living room). You also can’t beat physics so in an age where most speakers are slimline/small and “family lounge friendly” it’s hard to get the combo right.
  To fix this problem, you can add a sub to your stereo pair setup that, because it's being fed by (and outputting) the same stereo signals that the main pair is getting, you're still listening to a pure "2 channel" system (unlike home cinema where the sub is actually being fed as an extra "separate channel"). A sub should integrate seamlessly with your stereo pair to extend the usable frequency range (i.e. that which has the same volume) down to basement-like levels. Without my BKsub, my old stereo standmounts ran out of puff at around 45-50Hz as expected. I could just about hear a low frequency warble below this but this doesn't count because as I said, it needs to be a similar volume to the rest of the sound. With the BKsub added, the frequency response of the system went clean down to 20Hz all with matched SPL. On top of this, you should also set up your sub so that it is not heard as a separate box (i.e. so that the bass sounds like it's emanating from your stereo pair only).
  As a final thought, very well respected Hi-Fi companies such as Martin Logan contain built in powered subs which are powered separately to the electrostatic panels. The very well recieved Anthony Gallo Reference 3 floorstanders have an extra bass voice coil per speaker that you can (if you wish) power as an extra driver...in effect acting like add-on subs. Wilson Watt have speakers with the bass driver housed separately to the mid/bass and treble cones, again, similar to having a separate sub in a separate enclosure. So why are these numerous examples (as well as others) "allowed" but other separate standalone subs pooh-poohed? Bizarre to say the least (then again, if you're happy with your Hi-Fi not being able to reproduce/extract everything that's encoded on the CD that's fine...it's just that ignoring the lowest 1-2 octaves of encoded information is not my idea of High-Fidelity).
  If you wish to check wether you have usable volume at low frequencies, I suggest you go to AudioCheck. Alternatively, I have a set of 20sec WAV files (the 1^st 10sec are a warble tone and the 2^nd 10sec are a pure sine wave) available for you to download. They are a zipped file of 20Hz-20kHz wav Test Tones (69.4Mb zipped folder 31 WAV files, 1 AAC and 1 Readme file).
   
  
• What about Headphones, are they Hi-Fi? True. Headphones are just as essential to a complete Hi-Fi system as any other component (and some cost even more)! Have a read of some of my reviews at my Headphones Page
   
  
• Personal players, mini/micro's, multiroom systems, MP3, T.Vs, videos, dvd players, BT Speakers, PCs... etc constitute Hi-Fi. False. This is all very well for people wanting "lifestyle" systems and convenience but true High-Fidelity is based solely upon SOUND QUALITY. It's like trying to say an MPV is a Drivers Sportscar. MPV's may well be convenient and comfortable but they are not designed for "the ultimate driving experience".
   
  
• Do I need five or six speakers in order to have good sound? No. True Hi-Fi usually consists of separate decks/components allied to stereo (i.e. one pair of) speakers - after all, the soundstage is meant to be in front of you, not whizzing around your head. It's a shame that most Hi-Fi mags previously devoted to Audiophile sound quality have now gone down the convenience (non Hi-Fi) multichannel road.
   
  
• I should buy based on what other reviewers think, I don't need to listen to it. False. My components have been rubbished by some reviewers but applauded by others. It depends on what they reviewed it with and what music they played on it. The important thing is that YOU think the system sounds amazing. Some speakers (my old Cerwin Vega's for instance) can sound stunning at first listens but tend to fatigue you with their treble after 20mins. It is therefore extremely important to have an EXTENDED LISTEN before you buy.
   
  
• I'm not afraid of Placebo. True. As an example I think analogue (not digital) cables can make a difference because I can repeatedly hear differences between an ixos gamma and chord co. chameleon, say, in my system. Having said that, it is subtle enough to be within the sphere of "placebo". So it may just be my mind. Do I care?...no.
  The reason is that as long as I'm happy with the sound, I'm also happy with the idea that placebo may or may not take a part in my reactions/perceptions...
  ...but hang on, why is this? Well I am a human being (last time I checked). It is inevitable that placebo takes part in a lot of my perceptions. That's how nature built me. If you believe placebo plays no part in your everyday experiences in some way or another (Hi-Fi or not), you may not realise how big the placebo effect is. The problem really stems not from the idea that various accessories may or may not be different but from the fact that people equate placebo with the phrase "being duped". Which means people take offence at the idea of "succumbing" to placebo ...are you afraid you might be branded a pleb or something? This should not be the case. You aren't a pleb...you're normal, and you should be happy with your equipment no matter what (which is also why you should only spend sensible amounts of money on accessories as some can be placebo and therefore you won't end up wasting money on things that don't alter the sound).
   
  
• Are double blind tests essential? Yes they are. Firstly, a double blind test is one in which an item is tested against a known control to see if it has an effect big enough that a statistically significant proportion of people can hear it. ALL other variables are constant and therefore nullified (so it DOESN'T matter that the ancillary equipment may be different to yours - as they are constant between tests they are effectively neutral to the result). It's important to remember that what you're seeking to prove is that if there is even a slight difference with any bit of equipment, then a difference is at least possible. This will lend credence to the claims of said item that it does what it does and the so called "effects" aren't simply wishful thinking on behalf of the person who spent money on it.
  So regardless of whatever rubbish is spouted by Hi-Fi (so called) "journalists", double blind tests are important to sort fact from perception. It is dishonest for these haϲks to try to spin the idea that they aren't needed in Hi-Fi and conversely, it would do their cause a lot of good if they backed up their claims by allowing double blind testing. This keeps you, the consumer, well informed and ensures that you are not being ripped off by the numerous "snake oil" remedies out there in audiophile world.
   
  
• I don't understand what these Hi-Fi magazines and reviewers say anyway. Neither do I. However, if you want more examples of the rubbish these guys spew, you can try my Hi-Fi Journalist Bullshit Generator for Audiophools →
   
  ...to peruse more drivel spouted by the industry lackeys look at Wat Hi-Fi.
   
  
• Hang on...surely Hi-Fi magazines have good reasons to ignore science? Yes they do. Do you know what would happen if magazines scientifically PROVED that most analogue and all digital cables sound the same? Not only would their advertising revenue dry up but they run the risk of alienating all the folk they duped into buying hundreds of pounds worth of cables and gizmos over the years that had no effect on the sound. It would be commercial suicide so they will never ever do this as it would jeopardise their followers belief systems.
   
  
• Digital coax interconnects transmit digital signals differently to normal analogue phono ones. False. A digital interconnect is only digital in the sense that it conveys a binary code. A coat hanger used instead will ALSO convey a binary code so the idea of it being digital doesn't make it "special". Also, for digital signals, physically a data stream of ones and zeros is being sent. If you use a single phono lead (50Ω) instead of a digital coax (75Ω), this change in impedance of the cable won't all of a sudden cause twelve's and twenty-eight's to appear. You're still getting a binary code of ones and zeros. Any differences between the two are due to the cable's interference rejection / signal purity (see the second point above - 75Ω wires are meant to reduce reflections within the wire, but there is no discernible evidence that this makes any audible difference at all). The fact that the makers call one "digital" doesn't give it special powers (a properly made 50Ω analogue cable will functional perfectly well in transmitting digital signals/data, and will end up sounding no different to a "dedicated" 75Ω "digital" cable).
   
  
• Do different USB cables between source and DAC sound different? No. Again, they're carrying digital signals so a £100 USB cable will carry data identically to a £1 USB cable made to the same spec.
   
  
• Do different Ethernet cables used in streaming sound different? No. As long as they're made to the correct (universal) spec, they're also carrying digital signals so a £100 Ethernet cable will transmit data identically to a £1 Ethernet cable.
   
  
• MP3 sounds as good as CD, so why buy a CD deck? Depends on the compression. CD does not use a compression system to work and therefore a well-mastered CD will sound better on a Hi-Fi than low bitrate (highly compressed) MP3 files. Generally speaking, the usual 128kbps tracks available for download sound pretty poor on a good system. These tracks were designed for portable iPod listening as opposed to home Hi-Fi listening. The higher the bitrate, the better a track should sound. In fact, some folk say that 320kbps is 'as good as' CD, but I would argue they need to get their ears cleaned or invest in a good rig. The only digital tracks that sound as good as CD are those that are uncompressed...meaning using some form of lossless codec such as WAV, AIFF, FLAC, ALAC or OGG etc (of course, the source material needs to be at a higher quality from the get go because quality only ever goes down after each conversion).
  On the other hand, tracks which will sound better than Red Book CD would be those mastered higher than the standard CD value of 44.1kHz, 16 bit. This can include those sampled direct from analogue sources as well as DVD-Audio or SACD sources.
   
  
• How much should I spend on interconnects? Up to you. A general rule would be about 10% of the value of the component it's used on. Overall, all accessories should cost no more than 15% of the TOTAL rrp cost of all the components, maximum (spend too much and you're effectively just throwing your money away).
   
  
• Biamping/Biwiring improves sound quality Maybe. First of all, you can only do this if you have two pairs of binding posts at the back of each speaker. Running a separate cable to each was thought to keep the delicate signal of the tweeter separate to the less delicate bass signal for longer (published papers have proven that sending low and high frequencies down a single wire introduces more low frequency noise than if sending that same signal down 2 separate wires). However, you are actually hearing the pistonic motion of air from the cone not electric signals, so intermodulation between drive units may not be the only factor here. The beneficial effects are actually more probably achieved because you have effectively doubled the gauge of the wires going to each speaker (thus reducing their overall impedance which then improves the frequency (and phase) response of the system). It also removes the need for the (poorer) quality bridging links. This can at best, give you a tighter more accurate bass with a crisper treble and better soundstage.
  The next step up is biamping which involves amplifying the drive units completely separately (this requires two power amps, one for stereo high and one for stereo low frequencies ...monoblocking, on the other hand, is one amp for the left (full range) channel and one amp for the right (full range) channel).
   
  
• Can I use both sets of taps from my amp? It's not ideal. It's important to remember that, despite what the manual states, stereo amps tend to run better when driving one stereo pair of speakers. Using both taps from the rear will double the load on the amp which is fine, if you have a 100W behemoth, but the amp is at risk of clipping if overdriven.
  For instance, bi-impedancing is a variation on biwiring and works if you have drivers within a speaker cabinet that have different impedances. The Mid/Bass cones in the old Quad's have 4Ω drivers, so these can be connected to the 4Ω taps from the amp. The Tweeters, however, have 8Ω drive units, so these can be separately connected to the 8Ω taps from the amp. This way, you would think the drivers would be correctly impedance-matched to the source amp, but it actually means that you are really just messing around with the x-over (and the respective cone-EQ curves), so will effectively be running the speakers under non-ideal conditions!
   
  
• Vinyl's always better than CD. False. At the same price point and set up correctly, they sound equivalent. Vinyl tends to sound more lush whilst CD has a wider dynamic range (ave. 93dB vs. vinyl's 70dB) at the expense of being considered clinical (although one man's clinical is another man's accurate/truthful).
   
  
• What about Sonic Memory? It's important. Our brains are always comparing things, which is why ABX testing is so popular. However, our Sonic memory (ability to accurately remember every nuance of a musical piece/sound) diminishes very rapidly with time. When discussing tweaks, the differences (if any) are usually very small, and will be lost/forgotten by our brains within a minute or so. So unless we switch between different sounds very very quickly, we cannot hope to accurately remember all the subtleties that went before and thus make a true comparison. In fact, this is how most snake oil functions...by allowing a sufficient time to elapse whilst you 'set it up' that you wouldn't be able to notice that it's sound is unchanged even if it came and bit you on the arse. Also, we aren't necessarily aware of how much our sonic memory degrades over time because even though the subtleties may have gone, we retain the rough overall picture; so we continue believing our memories are 100% accurate long after they have ceased to be.
  Switching between comparisons should ideally be (almost) instantaneous if you want an accurate picture of what's going on (a minute at most).
   
  
• Should I buy second hand? Up to you again. Second hand goods can be bargains, but usually are 'sold as seen'. However, older speakers can sound better. For example, the mid/bass driver in speakers used to be 8" as standard. Nowadays, most models have 5" or 6.5" drivers to look slimmer and more 'lifestyle'. These don't produce as good bass as an 8" driver (such as those in my old Tannoys), and don't therefore sound as rich (also see note below).
   
  
• Floorstanders are usually better than Standmounters. Sometimes. Actually, Floorstanders use their cabinet size to produce deeper bass. The downside is that the bass is amplified by the enclosure and ported out, which results in a deep but less controlled bass than that directly from a mid/bass cone. Standmounters on the other hand, even if ported, may have less cabinet colouration and are therefore generally easier to design and 'get right' than floorstanders, but with floorstanders you don't have to worry about correct ear height when it comes to placement. Remember, speakers are probably the largest hi-fi component market out there so you'll find all different types and sounds on offer to suit your ears.
   
  
• Should I listen with the grilles off?. Generally yes. Grilles are designed to protect your tweeters/cones from prying fingers as well as keep them dust free. However, they also result in slightly less treble (my Tannoy Prestige grilles give a 1.5dB cut in treble when in place) as well as causing a diffractive effect which alters soundstaging resulting in more diffuse/less precise imaging than if they were off. As a result, for most Audiophiles, grilles are generally removed for serious listening but it varies from speaker to speaker so YMMV.
   
  
• Metal sounds sharper than wood. Oh dear! There is the general opinion of many hi-fi mags that such and such hi-fi rack made of glass and steel imparts a "brightness" to the sound. This other wooden rack sounds "warmer". Cases made of a natural material makes the sound more "natural". All of these statements are complete and utter tosh. The perceived property of a material that is not in the signal path is not imparted to the sound.
   
  
• Displays that turn off improve sound. Maybe. The display will not affect the parts of the player dealing with digital signals. However, the analogue sections of older players (anything in the signal path after the DAC) may be susceptible to interference from noisy displays that are emitting RFI (interference). Newer CD players have well manufactured displays that emit less or no RFI and players are constructed such that the analogue sections are routed away from the display anyway, so you may hear no improvement despite the theory.
   
  
• Do supertweeters work? Well Yes... in that they will produce high frequency waves well above 40kHz but you won't be able to hear that though. What you can hear is all the sound produced 20kHz and under. Now even if you cross-over at 20kHz, the slope isn't an abrupt cliff, there is a gentle slope below this i.e. in the audible spectrum. Mine are crossed over at 16kHz resulting in a slight lift of high treble in the sound. As mentioned above, if I listen to my speakers with grills on (which you have to do sometimes in a home environment) not only is the treble reduced slightly but soundstage width contracts a small amount and height and depth a bigger amount. By adding a supertweeter, that small lift in the audible spectrum rectifies the treble somewhat as well as improving the soundstage back to closer to what it should be (mainly higher and deeper and slightly wider). It's still not as good as listening "as is" with grills off but it's an acceptable compromise if you do actually need to keep the grills on. Everything above 20kHz is only audible by your dog.
   
  
• Spikes act as one-way diodes wrt sound propogation. False. Ridiculous idea really. Tests on the component casings with various spike positions (point up/point down) show there is no difference to the intensity/pitch etc of transmitted sound, regardless of which way the spikes are pointing.
   
  
• It's about fine detail, harmonics, soundstage, separation, and depth then? No. It's all about just sitting back and enjoying the sound (NOT analysing it).
   
  
• If I cover myself in special "hi-fi" quality jam and wrap my head in "audiophile clingfilm" (proven to work, EISA award winner @ £700/m), my system will sound better. Haven't you been LISTENING...?
   
  
• My experience of legitimate upgrades along with various accessories foisted upon us as improvements:
  

  upgrade from £200 CD player to £400 one		✓ definite improvement
  upgrade from Arcam CD player to a Naim		✓ even bigger improvement
  upgrading Naim to Dual Power (FlatCapping)		✓ slightly larger soundstage and deeper bass
  change from solid state to valve amp		✓ big leap forward in sound quality
  change from AVC to PrimaLuna valve amp		✓ GIANT leap forward in sound quality
  metal stands filled with sand		✓ stands £100, sand £3 - big improvement by keeping tweeters at ear level
  upgrading analogue interconnects and speaker wire		✓ in my case, yes, but does depend on what you buy
  taking speaker grilles off for listening		✓ improved dispersion and thus imaging and focus with them off
  upgrading to Tannoy M15 speakers		✓ more refined sound than Cerwin Vega
  upgrading to Quad 12L speakers		✓ more laid back than my M15's but also great midrange and overall detail
  upgrading to Tannoy Dual Concentrics		✓ simply superb and blows my Quad's away
  adding digital motor drive to turntable		✓ improved speed stability thus reducing wow & flutter
  changing all SMPSs to LPSs instead		✓ less "hash" being injected into the mains = improved transients & less grainy/harsh treble
  placing marble slabs (£2.50 each) under speakers/sub		✓ massively tighter/more focused bass on my suspended floor
  putting 20kg wieghts on the sub		✓ even tighter bass with more impact (cabinet not wobbling - only the cone)
  adding Dynamat (£12 / 2 pieces) to the underside of the sub		✓ absorbs some bass going down to the floor (useful if on suspended floor)
  internally rewiring speakers with QED Silver Anniversary (£7.50)		✓ improved clarity and bass in my Tannoy M15's with this thicker gauge wire
  adding a felt/cork/acrylic/etc slipmat to the TT platter		✓ yes but YMMV depending on how well it works with the platter material
  IsoTek Isoplug (£30) anti-'mains noise'		✓❂ If you have noisy mains gives improved clarity & wider, deeper soundstage & imaging
  using valve dampers (£35) over your valves		✓❂ if your valves are microphonic can give a better 3D image and less inaccurate top end
  biwiring speakers		✓❂ might improve clarity and definition due to doubling the wire guage (see above)
  Ikea furniture rather than dedicated hi-fi rack		❂ works just as effectively (neither better nor worse sonically) as an audiophool rack
  using Brasso and ethyl-alcohol to clean plug pins		✗ no change in sound at all
  CD lens cleaner		✗ no change in sound at all & may well scratch the laser lens...*CAUTION*
  turning the CD display off		✗ no change in sound at all
  Ferrite Rings placed over speaker wires at either end		✗ no change in sound at all
  high quality banana plugs		✗ no change in sound at all ...try using the bare wire ends instead
  Ferrite Rings over power cables		✗ no change in sound at all
  squash balls under components		✗ no change in sound at all
  adding a Sound Improvement Disc (£18)		✗ no change in sound at all
  cryogenically freezing things		✗ no change in sound at all ...LMFAO!
  using Densen DeMagic (£12) once per month		✗ no change in sound at all
  using anti spike/surge mains blocks		✗ worsens sound quality if you add extra fuses/switches/circuits prior to components

If you're having problems with the acoustics of your room, this Room Mode Calculator might help (it's an 'xls' file that needs excel or numbers to view).
I can also show you how to bias your valve amplifier and how to supercharge your Naim CD5i by DIYing a FlatCap into it.

Budgeting for a new Hi-Fi? Here is a simple rough budget to follow to get a well balanced 2 source system (works regardless of total budget). Copy the following list to figure out how much to spend on what:

🎵 12% on CD Player or chosen front end #1 🎵 1% on CD Interconnect (~10% CD Player cost) 🎵 12% on Record Deck (inc. cart+phono) or chosen front end #2 🎵 1% on Deck Interconnect (~10% Vinyl Deck cost) 🎵 22% on Amplifier 🎵 2% on Speaker Cable (including connectors if used ~10% Amplifier cost) 🎵 42% on Speakers (floorstanders OR speakers+stands+sub) 🎵 2% on Equipment Rack (remember, once immobile, it's immobile) 🎵 6% on Others (e.g. mains stuff)

The budget is deliberately slightly higher as you move toward the speakers. Of course this is a rough guide, only meant as a starting block...happy shopping!