. I wouldn't worry too much about it to be honest, it was just information. You just wouldn't want to be trying to run a pair of 300w RMS speakers off it. Any box type that is being discussed in this thread is not really in that bracket.
As it so happens ... I haven't lol, and I don't even own any I'm too far down the road Dynaudio and Quested wise now but if I had a total loss or was starting again I would most likely go PMC. I do however have some small Bose I would out...
User... what did you end up with? Also, what was / is your budget... you havent really specified.
So unless you prefer the sound of second order distortion (as Ttamasle may) and the vintage warm glowing light that tube driven electronics provide, I suggest a very simple and minimalist solid state system (preferably purchased used). Content from various digital sources (iTunes, Playstation 3, cable box) are fed to the television via HDMI and the television is connected to the HiFi via optical "toslink" cable. By the way, all digital connections are transparent as they are made up of 0s and 1s. Either the signal is transmitted or not (usually due to a physical failure in the cable or connection).
I would suggest the following:
1. Benchmark Media DAC 1 HDR. This is a brilliant device as you have top quality components in one box. You get a DAC to convert digital to analog (from various digital sources including USB). You get a remote controlled pre-amp for switching between sources and adjusting volume. But the best part is...the device is sonically transparent. It will add nothing to the sound of your original music! Therefore it is important to have high quality original sound files. A system that does not color or distort the signal will be very revealing of audio compression used in making mp3s, etc.
2. A high quality 2 channel amp. There are thousands of these available. Find one that has enough (but not too much) power to drive your loudspeakers. I'm very happy with my Parasound Halo A23, but I purchased it mainly because it has the added ability to work in 220v/50hz or 110v/60hz if I move back to the USA. Most amplifiers do not (the Benchmark DAC also is voltage switchable).
3. Speakers. This is where you should invest your time (and money). Make sure to audition speakers on the same electronics and more crucially at the same volume level and on the same track of music. It doesn't matter if they are more or less efficient as long as your amp can power them sufficiently. All speaker manufacturers tell you what the maximum watts per channel are, so it's fairly simple to determine whether your amp will drive them properly. B&W is quite a good brand for speakers. Myself, I have a pair of ADS (Braun) from 1986 and I love them...purchased for $175. There are many of these on ebay.de but I won't say they are the best speakers or the right ones for you. I bought them because my father had a similar pair and I grew up loving their sound quality.
4. Room. The room will determine how big your speakers need to be, how loud you will need them to play, how much power your amp will need, and in the end...if you are living in a completely concrete and glass space with metal furniture...expect bad sound reproduction no matter what!
5. Beware of the HiFi salesmen. Don't get into discussions about the "sound" of electronics (unless you are comparing solid state with tube) and make sure you don't buy expensive speaker cables...it's the biggest scam going.
Have fun, best of luck and don't get swept up in the voodoo B.S. of audiophile culture!
Essentially, humans tend to interpret louder as better in hifi tests, even with subtle differences in volume. I guess it hasnt been beyond some hifi dealers to push certain brands simply by playing them a little louder.
Nowadays it is simple to download a decibel meter to your iphone to check the sound levels to reasonable levels of accuracy.
I can recommend this to the serious listeners among us.
In today's world of digital media, people have become increasingly accustomed to hearing compressed music files. If one does not address and favors the use of high-end equipment, then you would get the standard: crap in, crap out. Im certainly not an audiophile, but its quite obvious how different a 128kbs file (MP3) sounds against a 1,411kbs CD file (PCM,Wav,Aiff).
I agree with you completely that speakers are something that one should not compromise on. BUT...since the topic of transparency in digital audio has come up, I thought it might be a good time to have a look into that to help clarify my position (not that any of this actually matters as this is an internet forum designed to help us discuss issues like audio and taxes while we wait for the sun to shine again).
In the early/mid 1980s when formats were changing from cassettes (analog recordings) over to CD's (obviously digital)...the engineers at the time developed a way to convert the digital signal to an analog signal. Hence the birth of the home-audio DAC. Even back in the 80's (and still to this day) vinyl is still being offered. Every wonder why that is? Its not a matter of nostalgia...its a matter of quality. A vinyl record is a full analog copy of the the original recording (effectively, something that is, in equivalence, recorded at an infinite bit rate). Granted recording technologies over the years have too gone to digital formats as well, so an argument can be made that the recording itself may not be as perfect as it should be. However, that would most certainly be a theoretical conversation.
In any case, CD's when they were first being produced utilized audio compression methods like WAVs and AIFFs (go find a an old CD and check it out for yourself). The audio bit rates of early digital media were in the neighborhood of 1,100kbs. Do the math (2channels x 44,100(hz/samples per second x 16 bits per sample = 1,411,200 bits per second (or 1,411kbs). This, by today's standard is a fairly high bit rate. Now...check out some digital media gathered over the past 5 years. What will you see? You will see other compression technologies like MP3, MPEG-4, MPA, WMA. The majority of these compressions result in bit rates of 128 to 356kbs...usually. Check out files from a standard digital library and you will have even lower bit rates.
So, take digital media from 20-years ago and look at the media of today. Let us assume that we had a time machine and we produced the same audio 20 years apart in the formats of the time. By simple math you get this = 1,411kbs - 356kbs = 1,055 kbs differnece. To an untrained ear, you can easily tell the difference in quality between these two bit rates. Now, compare a 128kbs file against a 356kbs file. The difference will almost be indistinguishable thanks to the effects of psychoacoustics: our brains have trouble decoding as fast as we can hear. Thanks to this effect we can remove actual information from audio files and not really perceive a difference in quality. However, we can still easily hear the difference between a 1,411kbs and a 128kbs file. This is where the modern DAC comes into play.
Since it is a numerical fact that there is information missing when comparing files of differing bit rates, the role of the modern DAC is to put that information back...(by a process of interpolating the gaps in the original file and to then fill those gaps with calculated information). Lets look at the numbers of a DAC like the ESS Sabre 9023. This provides 2 channels of up/over-sampling at 96,000hz/samples per second at 24 bits per second. Final bit rate: 4,608kbs. And yes, when you play a 128kbs file through this you could hear the difference with your ears filled with water.
Since analog audio has a bit rate that is effectively infinite, getting your digital audio as close to infinity would give you the best chances of hearing what was actually recorded. That is just a fact (think: is it real or is it Memorex? (also recall that memorex was an analog thing))....but of course this only means something to those who use digital media. One could spend all the money they wanted on listening equipment and set up a room with perfect acoustics and imaging...but if one uses sub-par digital sources, you can expect to only to hear well reproduced garbage.
Sorry that was long winded.
Here is another interesting set of numbers since I was on a thought train. Lets assume that in 1985 there were global sales of CD's to the tune of 100 million units. A disk weighs about 15g and with a jewel case this weighs about 90g. For CD's alone that is about 1.5 million kilograms of polycarbonate. At the jewel cases and we have a total of 9million kgs of material on hand. That is equivalent to 53 Boeing 777-300ER's. As a manufacturer who would need to produce purchase and process raw materials...it would seem in the interest of that business to remove packaging from the expense list. Especially when you consider the reality that in 2000 (alone) global CD sales were reported at 3.5 BILLION units). Ponder that for a moment...that amounts to 315million kilograms of material used just for packaging in 2000 alone. Distributed/downloadable media most certainly solves that expense issue.
Wow...that was off topic.
They have a shop to go listen to the different speakers and gave me a lot of good advice. I also went to media markt and felt the service no where near as personal and advice wasn't as good either.
I bought the KEF Q500's, really really nice bit of kit!
You will find an explanation here: http://en.wikipedia.org/wiki/Nyquist...mpling_theorem
Some record labels still produce music in analog. Daptone for example.
New stuff is almost exclusively recorded digitally; tape manufacture for analog is dwindling.
Tom
A critical ear will hear the differences quickly.
However, your quoted article is having quasi-scientific explanation about DAC and sampling/reconstruction. And not true explanation, by the way.
It does not matter if the analog recording on vinyl is "perfect waveform" (which can't be, but let's say for the sake of academic discussion). However, if you sample any waveform with the 40 000 Hz sampling rate, the reconstruction (assuming DAC is doing its job properly) will produce perfect reconstruction for any frequency under 20 000 Hz (half the sampling rate). It can not perfectly sample and reproduce frequencies above 20000 Hz, though they are not relevant for hearing.
That is well explained in the link I posted.
Analogue distortion is very kind to the human ear, and often produces a sound that can sound warmer and more pleasant than the equivalent sound on undistorted digital, which can sound quite clinical and cold. Hence this is why a lot of people prefer vinyl.
About twelve years ago I did an experiment. I bought 2 copies of the same album, one on cd and one on vinyl. After I played the vinyl through a couple of times just to get rid of the inevitable imperfections I then recorded it directly to a CD recorder I had in my system. Subsequent listening to both the vinyl and the recorded CD showed no difference, both were warm and really quite enjoyable. However the comparison with the original CD version showed that that version was not as warm or pleasant, but nevertheless more detailed.
I must confess that the system I was testing this on was not state of the art, but I had invested a few thousand pounds into it, and was quite well respected amongst the HIFI community at the time.
Personally I enjoy both, I like the detail of digital, but sometimes that nice distortion which analogue provides fits the bill.
I agree that a sampling rate of as demonstrated by the Nyquist-Shannon theorem is sound and obviously correct. However, its necessary to look at this from the other direction.
The Nyquist-Shannon proof dictates that a sampling rate of double the audible range is sufficient to replicate the sound as true as it could reasonably be. No argument there. When you take the streaming bit rate of a CD (1,411kbps) and break it down backwards you get the following:
1,411,000bps / (2 channels * 16 bits/sample) = 44,100hz (sampling rate)
Now, take a 128kbs file:
128,000bps / (2 channels * 8 bits/sample) = 4,000hz.
or more commonly
128,000bps / (2 channels * 2.902 bits/sample) = 22,050hz
Sampling rates of 4,000hz or 22,050 do not meet the definition of "perfect theoretical reproduction" as provided by the Nyquist theorem. Lets go further.
Lets take 1 second of audio. In the CD format, a 1 second sample is comprised of a left and right channel sampled 44,100 (quantity) times with each sample composed of 16 bits (fatness) of information. Another way to look at this is that there are 44,100 16-bit slices of information. Conversely, in a common MP3 format there would be 22,050 2.9-bit slices of information for the same one second of audio.
What does this mean? It means that there are gaps in the data. See below. More bits make fatter red zones. A higher sampling rate means there are simply more red zones.
Our ears make up the difference of what is missing when the files are decoded and played. Considering that digital audio is playing back sliced-up information, the ear is hearing linear changes in tone that would otherwise be rounded (as in analog audio, live sounds, etc). When files of insufficient "fatness" and "quantity" are used, the linear connections between each tone will begin to be heard. This is often heard in cymbals that sound raspy or in strings that sound flat.
No quality speaker can cover this up...they can only make these artifacts more apparent. This is why there is a need to correct the digital signal to restore the missing information. The more information there is, the richer the sound. Period.
Again, please let me reitterate, all of the above (as well as my previous posts) ONLY apply to individuals that rely on their digital libraries or who stream audio from online radio stations.