What is a waveform diagram?

The waveform represents the shape of the temporal course of the change in the magnitude of an oscillation. The x-axis usually represents the time, the y-axis the level.

What does a waveform diagram show?

In very simplified terms: A waveform diagram shows how loud the signal is at a certain point in time during the period under consideration. The X-axis serves as a timeline. The period under consideration begins on the left and ends on the right. On the Y-axis we see the level excursions. The larger the deflection, the louder the signal at the respective time.

In most diagrams, the amplitude is represented both above and below the X-axis as an oscillation. The Uebervinyl-de diagrams, on the other hand, are composed of two waves for the right and left channels of a stereo signal, whereby the deflections above the X-axis represent the signals of the left stereo channel, the deflections below the X-axis represent what is happening in the right stereo channel. We also choose this “butterfly” view for mono recordings for better comparability.

What can be read from a waveform diagram?

First of all: A waveform diagram says nothing about how good or bad a piece of music sounds. Through visualisation, however, certain characteristics of a piece of music can be recognised. This can be seen particularly well if you compare two waveforms of the same piece from different editions. Here are a few illustrative examples from the uebervinyl.de archive.

Level: In this diagram for a song from the album My Aim Is True by Elvis Costello, for example, you can see that the same track was dubbed with a high level in the upper diagram and with a much lower level in the lower diagram.

The differences between loud and quiet passages can also be read from the waveform. In our example, the song Salt Of The Earth from the album Beggars Banquet by the Rolling Stones. While in the upper diagram almost the entire song seems to be equally loud, in the lower diagram there are large level differences between the first third of the song and the last third.

In the example Sympathy For The Devil from the same album, the level in the lower diagram increases constantly over the duration of the piece. However, this dynamic curve can only be found in one of the two diagrams (above). The second recording suppresses this slow and subtle increase in level. The variant in the upper diagram therefore reproduces the musical performance more realistically than the one in the lower diagram, which shows a relatively constant maximum level.

The rough dynamic range, i.e. the direct sequence between large and small peaks that are very close together in time, can also be read from waveform diagrams. Experts refer to this as the crest factor. This refers to the distance between the level peaks (transients) and the less large peaks. In the diagram of Let Me Out from the album Get The Knack, we should pay attention to the light blue area around the X-axis and the distance to the maximum peaks. In the lower diagram, for comparison, look at the difference between the light grey area and the maximum deflections. The upper diagram therefore shows a lower dynamic range than the lower diagram. The track from the upper diagram therefore has a greater loudness, which is, however, measured differently and expressed in LUFS (Loudness Units relative to Full Scale). 

The waveform diagram for Rainy Day Women #12 & 35 by Bob Dylan from the album Blonde On Blonde reveals two things. Firstly, the track lasts a few seconds less on the upper record than in the pressing below. This usually indicates that the sound engineer ran the master tape a little faster during the lacquer cut.

Secondly, it shows that much more compression and limiter were used for the upper release to achieve a smoother level and greater loudness.

How do we create waveform diagrams?

Before we can create a waveform diagram at uebervinyl.de, we have to digitise the music from the record. To do this, we pick up the signal after the phono preamplifier and digitise the music signals as high-resolution music files with a sampling rate of 24Bit/192Hz in AIFF format. This format works loss-free and without data compression. So we get a very high-quality digital image of the output signal on the record.

The waveform diagram is created by uploading the music file into the audio analysis programme Sonic Visualizer, where the actual waveform diagram is created. To be able to compare the waveforms of two pieces of music, we load the AIFF files of both pieces of music into the analysis programme in parallel.

All music files are created with the same pre-amplification and the same settings on the level controls. They are also not normalised afterwards, i.e. not brought to the same level, so that the level differences are always identifiable in the same way.

We also create the frequency spectograms in the same analysis programme.