Thread: Cd Formats

Re the RCO free down loads that we all must have tried by now. These are the first d/l that sound ok when put on CD and played on my HiFi system, I had given up with getting decent results from the net. I would guess the d/l rate of 320 kbps is the reason
This now poses a couple of questions for me:

1. The splitter that I am using “Wave Pad Sound Editor” can convert to a heck of a lot of formats I have chosen WAV under the impression that this is the best to use,
using Windows Media Player I am told the RAW format will be the default format,
So is WAV nearer to the original than RAW ??

2. What format do the pros use on the commercial CDs ??

3. Can you decompress the d/l [mp3] to get the original “Pro CD” format, I realise that each transfer looses a bit of the quality

Hi Andante

Let's go through your points one at a time.

1. WAV is short for waveform, and CDs use what's called PCM (Pulse-Code Modulation) WAV. But you don't need to worry about such technical stuff. All you need to know is that WAV is essentially the sound, as it is (i.e. uncompressed) on a CD.

Now, I've never come across RAW, so I can't comment on this, but what I will say is that converting lossy formats like MP3 (it's called 'transcoding', BTW) which have had digital information stripped out in the encoding process is not a great idea. Far better to just let your chosen program make an audio (as opposed to a data) CD, and accept that while the end result won't be up to the quality of a shop-bought disc, it'll nonetheless be as close to this as you can get given that you're making the CD from 320kbps MP3 files.

2. I've answered this in the above point: PCM WAV.

3. In short, no. You can NEVER recover lost information from a lossy format like MP3 or WMA. All you can do is burn these files as a standard audio CD and enjoy an approximate replication of what the original CD sounded like.

Hope that helps a little.

FK

Thanks Kuhlau I should have answered earlier but I am still trying to work it all out, with the better results obtained with the 320kbps I wonder why it is not used more for classical d/l, even allowing for the longer time involved it would serve the receiver better but may be a pain to the sender?

Many download sites are now switching to 320kbps MP3 files. It gives them the credibility to claim 'highest quality' (for the format, anyway), and is a convenient compromise between storage space requirements on their servers and relatively quick download times for the majority of customers.

Personally, however, I prefer FLAC these days. Not only does this technology allow EXACT replication of the digital information on a CD by using an intelligent compression method that reduces filesizes to about 60%, but it also eliminates the gapless playback issues that have long plagued the MP3 format, and which seriously impact on the enjoyment of classical music in particular.

FK

Please enlighten on FLAC sounds too good to be true what is the catch??
regarding the gap less play back I have the Wave Pad Sound Editor which allows you to make the division between movements but I am still learning how to refine it lol.

FLAC stands for Free Lossless Audio Codec ('codec', incidentally, is techie shorthand for 'encoding', or making the compressed format, and 'decoding', or 'unpacking' it, so to speak, ready for playback).

Best advice would be to Google the following terms, read carefully, experiment and come back here with any queries:

FLAC

EAC (Exact Audio Copy)

foobar2000

MediaMonkey

Winamp


The last three terms are all names of media players which can handle - and create - FLAC files. EAC is the best and most accurate freeware music ripper in the business. And FLAC ... well, you'll see.

Good luck!

FK

Just to get one thing clear, if I run one of the RCO 320kbps d/l through say MediaMonkey
(which I have done as FLAC but not yet burned to CD) will it convert the mp3 to FLAC? I have noticed that the file size has doubled, I ask this because you said ""In short, no. You can NEVER recover lost information from a lossy format like MP3 or WMA.""

The important thing to note is this: an MP3 loses information during the encoding process which can never be recovered. If you want true FLAC files, you need to create these by ripping and encoding from the source CDs.

What typically causes confusion for those new to all these technologies is, as you've discovered, that filesizes inevitably increase during any transcoding exercise. Although sizes do get bigger, this doesn't, however, indicate that information originally lost has been recovered (sorry).

FK

So we can never improve the audio quality of the mp3 down load ??
A friend has requested that I ask you this question : Where does the extra information on the CD come from (larger file size) when you decompress the compressed file format and why does it sound much better when you do so? 

The 'extra information' is simply a result of burning an MP3 file to CD. I don't know the technical ins and outs, but I do know that despite whatever your ears may tell you, an MP3 file burned to a disc is IDENTICAL in terms of its sound quality.

To prove this point for yourself, make an MP3 file of a piece of music using whatever program you like, and make sure the encoding bitrate is 96kbps (which is very low quality). Next, burn that file to a CD, then listen to both the MP3 file you created and the CD burned from it.

Notice anything? You should be able to hear clearly that both the MP3 file and the CD sound equally poor. But - and this is the curious bit - look at how big the CD filesize is after this low-quality MP3 file has been transferred to disc. Still large, isn't it? Yet the quality is, as I say, identical to that of the MP3 file from which it was burned.

Don't trouble yourself too much about this. There are those who can't tell the difference between a 128kbps MP3 file and the source CD. There are also those who can clearly hear compression artifacts in a 320kbps MP3 file. We're all different in terms of how we perceive sounds, it seems.

FK

This is his reply, and I will make it the last as it is getting beyond my understanding

Don’t want to get too technical either – but basically when audio is stored in digital form the signal is broken down or converted into bits of information – these bits can be reconverted into the analogue signal. The formats that result from converting the analogue signal into digital uses various algorithms – these are often “proprietary brands of formats”, each one using different algorithms (mathematical sets of rules) – there are also different ways and algorithms for the processing of the digital information in the CD player – you may remember the debate about single bit vs. multibit processing – but what we are talking about here is the digital storage format of the music.
Now the more attributes of the music you store of the original music, the bigger the storage file.
A .wav file stores a large number of attributes and therefore is high quality. To save on space mainly but also I guess to promote your unique brand of software various formats with different processing of the original .wav high quality digital information have evolved. Generally they compress the information or leave out some attributes of the original digital file and make it smaller.
If you have a high bitrate it generally means a lot of attributes of the music is stored in the digital file
Lower bitrate is lower quality signal.

Now the question is what happens when you reverse the process – can you recover a higher quality signal form a lower quality file.
The answer I guess is – it depends.
If the algorithm is simple to reverse you should be able to re verse the process and recover the quality of the original (unless the algorithm is applied in a faulty way)
If the algorithm removes so much or reduces the quality of the original so much that it can’t be reversed or you need a very complex different algorithm to recover or restore the original you may be unable to improve the quality – a bit like photocopying a picture in black and white and then putting the black and white copy on the photocopier and hitting the colour copy button. You may be surprised to know this but there are algorithms that can try to do this but they are complex and fraught with problems. We all know that poor quality originals can be enhanced and brought closer to the original signal – for instance algorithms used by geographers and the military to improve satellite pictures. Or the “remastering” of old recordings.
So I think ( and I might be wrong) that if it is possible to convert something from a say 192kb/s bitrate back into wav format with say 1412kbit/s rate by reversing the process of the algorithm I have a higher quality signal (and my ears tell me this) You can, in fact, store the low bitrate format on the CD – it doesn’t automatically change back into a larger file when you burn it unless further processing is applied - check it out.

When it comes to sound ultimately we have to listen and judge it on that – and as said before it depends on the reproduction system that converts digital to the analogue sound we hear also.


Kahlau, I am interested in your response but being the intermediary is awkward

Well, Andante, that's quite some reply from your friend. I can find fault with almost nothing in it ... although I am curious (and mildly suspicious) about the suggestion of algorithms that can replace lost digital information. I don't know whether or not this is true. I just know that everything I've ever read about lossy formats being burned to CDs indicates that the digital bits of information that were discarded during the encoding process can't be recovered. Beyond that, it's beyond my understanding, too.

As an aside, ask your friend to download and set up foobar2000. It has a double-blind listening test .dll component as standard (called ABX Comparator). If your friend truly believes that he or she can accurately hear the differences between compressed and uncompressed sound files, this test will prove how good is his or her hearing. It'll also calculate the probability of guess work. Trust me, this test sorts the men from the boys.

FK

I will pass on the message , I have d/l 3 of the programs that you mentioned, foobar included but am still figuring out how to use them, one point that if I have understood correctly is: that when compressed to say mp3 certain digital info is removed and not just compressed so it could never be recovered, so why does the file size double ?? it must be adding something

Originally Posted by Andante

... when compressed to say mp3 certain digital info is removed and not just compressed so it could never be recovered ...

Correct.

Originally Posted by Andante

... so why does the file size double?

No idea.

FK

Original digital recordings can be made in a number of different formats, all of which can be and are saved as wav files.
For instance I recently received the master wav files of an album originally recorded on 2" 24 track analog tape.
These had been recorded into Pro Tools, a high end digital audio workstation, and burned to DVD as 24 bit 96k audio wavs.
They could have been saved as 8, 16, 32, 44.1 (cd standard bitrate) 48,96, or 192 with obvious impact on the file sizes involved.
In order to get the best possible recording onto a cd it is (due to the Red Book standard as to how CD`s should be burned) only possible to use 16/44.1 rate.
Cd`s do not support anything higher, so that the resultant CD will play on a standard audio CD player.
So by definition CD quality is far less than is technically possible & indeed these days most recordings are made with higher sample and bit rates.
And this is where the tricky part comes in.
When the high quality recording is dropped or "dithered" down to standard CD format, there is inevitably some loss of signal information, but bear in mind this is information that your home system is physically incapable of reproducing even if it were on the CD in the first place.
So you are not actually losing big chunks of music unless you have a system capable of reproducing the higher rate files.
Funnily enough, a lot of what you perceive to be missing is our old friend dynamic range.

And in conclusion sample rate and bit rate are both part and parcel of the same situation.
A CD is by definition limited to the standard bit and sample rates, so the only variable is how efficiently the algorithm used to dither down the higher quality original recording file manages the job.

And no, as far as I am aware there are no reverse engineering programs that will allow the revesing of the lossy compression methods used to encode mp3, ogg vorbis etc.

It`s sort of like those bottomless Guinness bottles in the Irish jokes.
And if you ever find one, I`d like two as well.

(grin)

Ivan