Audiophile Quality: Separating Snake Oil from Quality

Since long ago I have been amazed by stories I have heard about extreme audiophiles  spending obscene amounts of money on silly gear.  Examples include "Special" speaker cables, HiFi setups than cost more than a decent car, and things like special power conditioners and downright silly things wooden knobs to replace the plastic ones on their equipment - all in the hopes some some small improvement in sound quality.  

It's easy to make fun of all of this, but the reason it persists is because there is some validity and truth in some of it.  Sadly, there are unscrupulous companies that prey on consumers and try to sell all sorts of things with questionable benefits for obscene benefits.  It may seem harmless to part a rich fool from his money (and it does usually seem to be males), but given that there are people out there who legitimately want to get the best sound possible on a reasonable budget there is real harm being done.  For some reason I visualize a gadget addicted audiophile who is obsessed with constantly buying and upgrading every component of his system, while spending money that should have gone to his daughter's college fund.  

On the one hand, most people would be pressed to tell the difference between a 320k VBR MP3 file vs. a FLAC file, and likewise might not be able to tell the difference between a $500 headphone vs. a $2500 headphone, especially when connected to something like a smart phone.  I would bet good money that almost nobody could tell the difference between any given two paid of speaker wires.  

So it's all bunk, right?  well...

Not exactly.  

On the other hand, if you listen to a qood quality source that was mastered well, on a good pair of headphones plugged into good equipment, and compare that with tlow end consumer qauipment, you will find that sometimes the difference is like night and day.  

The reality is that there might not be one magic component, but the small differences do add up.  

Let's think about the average listening experience today:

First, there is the production side:

The musician plays music, and this audio is picked up by microphones.  The type, quality, and location of the microphones all matter.  No microphone can pick up all of the audio perfectly, and there are always trade offs.  So there will be some loss of the original audio at the microphone itself.  

Modern recording is typically done digitally, so the audio from the microphone must be converted to a digital signal at some point.  This is of course imprefect owing to multiple factors, and even a perfect analog to digital converter would still be converting to a digital signal with limited bandwidth.   So some of the signal will be lost by digitizing it.  

Next, the producers mix and edit the audio, which involves adjusting the volume, merging different tracks together, adding effects, etc.  This process is hopefully done with a higher bandwidth data stream than the one that is finally used for mixdown, but there is still likely to be some degredation as it is mixed down into (usually) stereo for the final press.  

Finally, the data is compressed with something like mp3 or AAC for consumer use these days, which involves further quality loss.  

All of the above is before you even get your hands on the music as a consumer.  

Next there is the playback side:

First the compressed song has to be decompressed.  Thankfully this normally shouldn't have any further quality loss unless there are issues with the decoder.  

Once the data has been decoded to PCM audio, it needs to be converted back into Analog audio to be sent to your speakers.  This is the job of the Digital to Analog converter.  Again, these devices are not perfect - infact, some are downright bad.  The DAC on the sound card in your computer, or chip on your phone might suffer from interferance from the power supply, radio signals from your processor, or more.  

Thus the signal coming out of the DAC will never exactly match the signal that went into the digitizer exactly, even if there were no compression in-between.  This sort of loss is why we moved away from Analog in the first place - but sadly our ears are analog, and so are speakers.  

Next, you need to amplify the sound that will actually power the speakers.  Most amplifiers aren't even close to perfect.  There is always noise, and that gets amplified too.  We all know the hissing static sound when you turn your stereo up too high when nothing is playing.  It's less noticible when you are playing music at a lower volume, but it is still there.  Besides amplifying things they shouldn't, amplifiers also amplify different parts of the sound spectrum with different efficiency.

I don't personally believe that any reasonable speaker cables will have any amount of loss that would matter, but let's give the benefit of the doubt here, and say that we lose 0.5% of the signal with the worst cables through crosstalk, phase shift, or whatever.  

Finally we have what is probably the most important component in most setups, the actual speakers (or headphones, or IEMs).  A speaker has a tough life.  It's expected to reproduce sound from 20hz all the way up to 20khz at a minimum.  

These sound frequencies have very different characteristics, and cannot be efficiently produced with the same components - which is why you will often see speakers with different tweeters, sub-woofers, etc.  Using these different drivers inside the speakers does improve things, but at the cost of added complexity and expense.  The signal needs to be separated so that the lower frequencies go to the sub-woofer, the higher frequencies go to the tweeter, etc., and this means additional cross-over circiutry, which also won't be perfect.  

Wrapping it up

Each of these stages or components may not affect things much, and you might not always be able to even tell the difference by using A/B testing - but the real issue is that these distortions are cumulative.  If your microphone only captures 90% of the original sound, then the best you can ever hope to get when listening to the recording is 90%, even if every other part of the process is 100% perfect.

Sadly, things aren't perfect in the real world, so if compression cuts off a few percent, and your DAC cuts off a little more, and then your speakers aren't perfect, well at the and of the day, what you hear may be a pale imitation of the original sound!

To see what I mean, take a look at the cumulative effect column below:

Note that the losses don't just add up, they are effectively multiplied by each other!

Note: The actual percents used here are completely made up, and the real situation is infinitely more complicated, as there are different types of losses and distortions.  

To further illustrate my point, let's assume you are stuck with the music you already bought on CD, so there is no way to get a higher quality recording.  On the consumer side, you can switch from MP3 to FLAC, so that the compression loss disappears.  

In our scenario let's say you were listing using your computer's built in sound card which isn't very good, and you switch to using something like the SoundBlaster X4 instead.  For argument's sake, this cuts the distortion of the DAP in half and improves the amplifier performance somewhat as well.  No more hiss.  

While I don't think cables really matter, again let me give the benefit of the doubt here and say you upgrade from the $5 Amazon Basic's cable to the $250 oxygen free gold plated silver coated braded cable or whatever, and it cuts the already minor cable distortions in half.  

Finally, you upgrade your speakers or headphone to something a bit better.  

Each one of these things improves your quality a little bit.  Probably the DAC/AMP and Speakers are the most important, but each one improves the signal going onto the next part of the chain.  

Using my made up numbers, we get an increase of over 4%!

My real point is that even if none of these changes were really notibile by themselves, the combination of all of them together probably would be.  

This is probably an obvious concept in the audio enthusiast community, but something that a lot of orginary people don't think about much.  

If you try these things out one at a time, then you might come to the conclusion that there is no difference and they are a waste of money.  

For Example: You might buy a pair of fancy headphones and plug them into your phone instead of the ones that came with your phone.  Maybe the pair that came with your phone actually aren't too bad, and your source music isn't great anyway, so you don't notice much of a difference.  

Or maybe you do upgrade the sound card on your computer, but you are still using those dinky $20 speakers that came for free with your the Packerd Bell computer your bought 20 years ago, and you don't hear much difference at all.  

Another thing to consider is that some items, like the speaker wires, are not likely to have a big impact, while others, like the speakers themselves, will have a larger impact.  

Before spending money to upgrade anything, you should take a look to see what the weakest link in your system currently is, and how much bang for the buck you can get with different components.  Fancy isn't always better.  

p.s. If anyone has ideas on realistic numbers for the tables above, I would be very interested to hear them.