
Sound under pressure

Computer sound has long since emerged from that embryonic state, when it was only present for the show, unable to compete with specialized equipment. Today, many sound cards, even middle class, are far above their rivals in the face of not so mediocre hi-fi. Recently, there are also fewer and fewer problems with acoustics; At such a rate, in a few years, the entire breeding road will finally turn into a true hi-hi. But in addition to the path, there is also a file format, in the choice of which, due to ignorance, users are often really limited. The purpose of this article is to get rid of these limitations.

Gone are the days when it was impossible to distinguish a violin from a cello in computer acoustics, but the saying “CD Quality – MP3 128 kbps” has remained, and for some it is not so archaic. Meanwhile, the most common formats are Wav (also known as CDA) or MP3. However, uncompressed PCM (Wav and CDA) has too large a file size and MP3 is compressed with loss of quality. But there are alternatives and more than one. Let’s take a look at the most popular and high-quality formats / codecs.
First, let’s divide all codecs into two groups based on compression: lossless and lossy. The former operate on the principle of filing cabinets, for example RAR: a file compressed in this way loses weight by up to 50%, and the entire original is reproduced during playback, before compression. Lossy compression algorithms exclude “unnecessary” information from the original signal and then compress it, which is why the original signal cannot be fully restored; JPEG compression is an example. Now in more detail.
At a loss
MP3
Perhaps, as with the most popular ones, let’s start with MP3, also known as MPEG-1 layer 3. We compress test snippets with lame, the highest quality of all MP3 codecs. In the snippet with the classics, we see that there is no upper cutoff frequency as such, that’s great! However, in the most “powerful” (noisy) places, the upper part (from 18 kHz) is consumed. Such dynamics processing is quite strange for hearing and auditory logic, but for a computer it is easy to explain: just as the overall signal density increases, it grows almost throughout the entire range and therefore one more channel is required. wide to pass all the flow. But since the channel is fixed (we compress in CBR, with a constant bit rate, 320 kbps), for normal encoding of the mids and bass, you need to reduce the treble. A snippet with modern music is practically indistinguishable from the original, just a small cut in the frequencies for which real high fidelity is needed, in which you will definitely not remember the MP3, in addition to the frequency response, there are still many characteristics spoiled by compression. Compressed file size: classic – 6.11 MB, modern music – 6.11 MB.
WMA 9
The format promoted by Bill Gates is not yet popular: firstly, it supports DRM (copy protection), and secondly, of all the lossy codecs considered here, it provides the weakest sound quality. With almost the same dimensions as with other formats, making a hard cut at the top above 20 kHz, as well as cutting the top based on the overall level (similar in effect to Dolby noise suppressors) is not very good in our opinion. Compressed at 320 kbps, WMA 9 (non-professional). Compressed file size: classic – 6.14 MB, modern music – 6.12 MB.
OGG Vorbis
Open source codec with good sound quality and safely taking second place. It has a floating cutoff frequency (but within reasonable limits) of the order of 20 kHz. It is true that in this case we set the bit rate at 350 kbps … It has one more drawback: longer encoding time. Compressed file size: classic – 6.70 MB, modern music – 6.65 MB.
The winner of this nomination is the one with the closest sonogram to the wav file sonogram. Therefore, MP3 is still the winner. However, it should be noted that the codec is different, and even those that work according to the same algorithm. For example, the same MP3 is lame, there is Fraunhofer and Xing, the latter being the fastest, but also with the most terrible sound quality (the cutoff is 16 kHz).
No loss of quality, no loss
FLAC (Lossless Audio Compressor)
One of the most popular formats for lossless audio compression is the FLAC codec. The main advantages of this audio codec are its constant updating and, of course, multiplatform: FLAC is ported to many platforms.












