Will MP3 Replace CDs

Great info guys. I am really more concerned that I will need to buy that stupid “Meet the Beatles” album in another format (yet again) [I have the record, tape, cd) :slight_smile:

Arken,

panamajack already took on your misconceptions about reconsturcted waveforms from digitally sampled data, so I’ll address your distortion statement. You’re stating that the sampling error causes distortion in digital but not in analog. It’s true that there will be a very small amount of distortion because the digital signal only knows about 65536 possible amplitude values. But the distortion from this error is at the -96 dB noise level, which is way below what the noise from other sources, such as your amplifier, the microphone, and lots of others. This theoretical distortion noise will be buried in a sea of noise from other sources.

With an LP, the noise from the record itself is so much higher, that “buried in a sea of noise” would be a gross understatement. You couldn’t measure whether the distortion would be there or not, because it would be so much smaller than the random noise on the LP.

You can record at different qualities, and if you record from cd you can still have cd-quality sound, with just a few differences.

I think that they will help the CD industry. Except for my 1-2 friends with cd-burners, everyone I know that listens to mp3s say that they have helped them buy more cds. I’ve bought over a dozen because I listened to them on mp3 and then bought the cd because I liked the mp3s.

There are always going to be the people who don’t care about the artist and just burn their copies, but stats have shown that CD purchases have been going up in the past few years even with mp3 technology.

For me, they already have.
I never got a CD player, since I had a lot of tapes.
Now that tapes are rare, I buy MP3 files to play at parties, on WinAmp.
Yes, my computer has a CD read/write player, but it’s only held software so far. I may use it to transfer music to my next car, which will have a CD player.

Thus sprake CurtC:

Two problems here, Curt.

First of all, sampling error cannot possibly cause distortion in analog for the simple reason that analog does not use sampling technology.

Secondly, while many records do have an initial noise floor, this is not necessarily true of all records. It really depends on multiple factors including how clean the signal is going into the cutting end (and also the quality of the lathe) and also how clean the signal is going out, how good the needle is, if the needle is the right size, if the arm is properly weighted, etc.

I have worked with LPs which are virtually noise free.

You also are confusing my statements about distortion. Distortion is most prevalent in digital when the dynamic range goes above digital zero (when it gets too loud). Again, the advantage in analog is that it can get as quiet as a digital recording device (i.e. -infinity dbm) but it can also get to around +6 dbm without much distortion and around +15 dbm without ridiculous amounts of distortion. The loudest a CD can get is digital 0 which is around +4 dbm. Therefore, the increased dynamic range in analog aids in the quality.

The nicest thing about digital though, IMHO, is what you were saying which is that in the ideal world, digital is always free of an initial noise floor (hiss, rumble, etc). However, that ideal world would really require a room soundproofed to near -infinity dbm, and a digital microphone (they exist these days!) going into a digital preamp to your digital recording device. Otherwise, you risk the chance of of electrical interference (one of the most common causes of analog noise) and noise degredation through analog transducers (another common cause of noise).

So… the bottom line?

  1. In the perfect world, analog will always sound better because of all sorts of previously discussed reasons.

  2. In the perfect world, digital will always sound better because it’s a cleaner signal.

  3. In the real world, it’s a toss-up.

Arken wrote:

Uhh, I don’t think I said it did.

This doesn’t make sense. There’s no such thing as no noise. The CD medium has a noise of around -96 dB. Of course, other noises will be high enough that you’ll never hear this. An analog LP has a noise floor much higher, also with other noise sources added in. A very good LP, with its own noise below the other noise sources in the recording and playback stages, might sound as good as a CD if there are no pops.

I guess I don’t understand what you mean by “digital zero”. I think of the digital data as a set of numbers between -32767 and +32768, with a perfectly quiet signal being at digital zero.

Again, your terminology is throwing me off. The measure of “dBm” is a measure of power, not dB’s on their usual relative scale. But I think you might be saying that a CD has a hard clip limit when the digitizer is saturated. True, but no (professional) recordings go above that limit. The CD medial itself introduces no power-sensitive distortion. The analog-to-digital and digital-to-analog converters could possibly introduce some, but with any reasonable quality encoder and player these will be completely swamped out by the distortion from your power amplifier, which is the main source of distortion for both LPs and CDs.

I still haven’t heard a single valid reason why an LP could sound “better”, unless you’re saying that your ear likes more noise and pops.