Mini disk devices and everything about them. Minidisc player Sony MZ-N1 with support for NetMD technology

B. Ya. Meerson, A. Ya. Shcherbakov

Minidisc (MD) is one of the latest formats in the evolution of disc audio media, developed by Sony.

It is smaller than conventional CDs (diameter only 64 mm), but at the same time is not inferior to them in the quality and duration of sound of the music program recorded on it.

This is achieved through data compression (densification), the strategy of which will be discussed below. The small size of MD provides fast access to data anywhere on the disk in less than 1 second.

In Fig. 1 shows an MD player for playing mini-discs.

The mini-disc format provides for the use of two types of media - non-recordable CD-type discs and recordable magneto-optical discs (Fig. 2). The latter make it possible to make repeated recordings on them, erasing previously recorded programs. This “reversibility” of recordable mini-discs makes them direct descendants of recordings on magnetic tape, indispensable for the rapid preparation of radio broadcasting programs and in other similar cases. Both disk formats are placed in cartridges to protect them from mechanical damage. The total weight of such a package is approximately 18 g.

A non-recordable minidisc is very similar to a CD. It is recorded in advance using the usual optical method for CDs (using a laser), replicated in a factory by pressing and intended for playback only. Because this format uses data compression, recordings on these MiniDiscs are not compatible with regular CDs.

A recordable, or, more precisely, rewritable disc is, in principle, not an innovation. This is a magneto-optical medium that was previously used in computer technology (CD-MO). But the magneto-optical media for the minidisc has been improved, the recording device has become much simpler, and the recording itself requires less power consumption.

The principle of recording on a minidisc

There are several methods of recording on rewritable (“reversible”) media, but for mini-discs the magnetic field modulation (MFM) method was chosen as the most reliable and makes it possible to rewrite an almost infinite number (up to 1 million) times. Moreover, with each new entry, the old data is automatically erased.

The essence of this method is that when ferromagnetic materials are heated above a certain temperature, called the Curie point, their magnetic susceptibility curve sharply rises and increases many thousands of times. If such a material is heated at a certain point to a temperature slightly above the Curie point, and the material is exposed to a magnetic field too weak to leave a mark on the cold areas, then the heated point, after cooling to a temperature below the Curie point, will retain magnetization, i.e., its the magnetic state will be fixed.

A recordable magneto-optical mini-disk is formed on a polycarbonate substrate, on which a magneto-optical (working) layer is located between two dielectric layers. A reflective aluminum layer, a protective layer and a silicone lubricant are applied on top of this structure, along which the magnetic head will slide. In Fig. Figure 3 shows a cross section of the disk.

The magneto-optical layer of the mini-disk is a special alloy of iron, terbium and cobalt (FeTbCo) with a very low coercivity of approximately 80 Oersted (6.4 kA/m). This is important so that, despite the fact that the magnetic head does not directly touch the working environment, the magnitude of the magnetizing field would be sufficient and would not require its increase, which would inevitably entail greater heat generation and increased power consumption.

To record information on a magneto-optical layer, it is necessary to influence it not only by the magnetic field of the recording head, but also to simultaneously heat the corresponding point of the carrier to the Curie temperature. This is done using a laser beam.

For the FeTbCo alloy used in mini-discs as a recording medium, the temperature corresponding to the Curie point is approximately 185 C.

And this is a reasonable choice: below there is a zone where there is a danger of an accidental increase in temperature during simple storage of records to a critical value, when destruction of the record is possible.

Higher temperatures are not suitable due to the natural increase in energy required for heating. Therefore, alloys of rare earth elements are used as materials for the working layers of magneto-optical disks. Please note that erroneously erasing data on a minidisk is almost impossible, since this requires simultaneous exposure to a certain (above the Curie point) temperature and a magnetic field.

So, magneto-optical recording systems are built quite primitively.

To do this, the magnetic head is positioned on top of the laser source on the same axis on the opposite side of the disk (Fig. 4). A focused laser beam heats a local area of ​​the disk media, which is exposed to a scattered magnetic field from the recording head that is quite weak to allow recording in the cold areas. However, it is enough so that when the disk rotates in the first moments of cooling of the heated section, information is written on it in the form of magnetization of a certain polarity: “north” - N or “south” - S.

Thus, different polarities of magnetization of preheated spots in the magneto-optical layer correspond to digital logical levels “1” and “0”. The size of such a recording spot, and therefore the recording density on magnetoelectric disks, is determined by the size of the focused laser light spot and the duration of the reversal cycle of the modulating magnetic field of the recording head. For this purpose, a special head was developed that allows rapid magnetization reversal (approximately 100 ns). It is obvious that the surface layers of the disk do not prevent the instantaneous heating of the working layer. Recording is performed by superimposing new records on the previous ones with automatic destruction of the latter.

Reading information from disks

As already mentioned, there are two types of disks and each of them uses its own reading system. A non-recordable disc (MD-DA) is similar to a CD. To read it, the same laser is used as for writing, but at a lower energy level. The reflected laser beam varies in intensity depending on the information recorded in the form of pits (indentations) on the surface of the disk. In Fig. Figure 5 shows the process of reading information from such a disk.

A recordable disc (MD-R) uses a different reading system, since the data is not recorded by a pit system, but is stored in the form of a magnetic layer magnetization polarity that varies from point to point. In this case, information is also read using a laser.

The laser beam hits the disk surface, passes through the magnetic layer and is then reflected from the reflective layer. However, passing through the magnetic layer, the plane of polarization of the laser beam changes depending on the polarity with which this layer is magnetized at a given point. The rotation of the polarization vector of a light beam under the influence of the magnetic medium through which it passes is called the Kerr effect.

In Fig. Figure 6 shows the principle of reading information from a recordable magneto-optical disk.

So, there are two types of reading mini-discs:

  • reading a non-writable CD type disc in which the RF output signal is the same as that of a CD;
  • reading a recordable disc type MO: here the RF flow is continuous, but with changing polarization.

The same dual-function laser is used to read information from both types of disks. However, another element is added to the optical head of the system (when compared with a CD) - a polarization analyzer, the so-called Wollaston prism. The fact is that the Kerr effect is weak.

The rotation of the polarization vector, even under the most favorable conditions, does not exceed one degree, and light receivers do not respond to polarization. The job of the Wollaston prism is to convert the polarization angle into light intensity, which is then directed to two photodiodes (Fig. 7).

The Wollaston prism is a combination of two quartz resonators. The laser beam, reflected from the reflective layer of the magneto-optical type disk, passes through this crystal and is divided into the main beam (the same as the incident one) and its components, side rays. The latter (in Fig. 7 they are designated I and J) are directly related to the polarization of the incident laser beam.

Moreover, when a laser beam passes through the magnetized N-layer on the disk, its polarization will be such that one of the side beams (beam J) will be larger than the other. And if a laser beam passes through the S-layer on the disk, its polarization will change and in this case the beam I will be large in magnitude.

So, the incident beam turns out to be decomposed into two component beams I and J, the ratio of their magnitudes is determined by the Kerr angle or the direction of polarization.

In the event that we are dealing with a non-recordable CD type disc, the laser beam is not affected by the magnetic layer, no changes in its polarization occur and therefore beams I and J will be equal in level.

Next, the read information goes to the mini-disc sensor unit (Fig. 8), which is similar to the CD player sensor unit, but contains additional sensors.

In the case of reading a non-writable disc, rays of the same size appear at the output of sensors I and J. They are then sent to photo sensors and converted into electrical RF signals. At the output of the RF-1 system, after subtraction, the signals cancel each other. At the output of the RF-2, the I and J signals are summed and create a signal similar to the RF signal at the output of a conventional CD player.

In the case of a recordable disc, the I and J signals are different in magnitude. The output of RF-1 produces a difference signal, the polarity of which contains information about the data on the disk.

The remaining signals from sensors A, B, C, D are the focus error signal (A+ C)-(B+D), the automatic gain control signal AGC (A+B)+(C+D), as well as the laser beam tracking signal ( Tracking) from sensors E and F (E-F), similar to those used in CD players. With the exception of the ADIP signal (Adress in pregroove), which carries addressing information and is a signal (A+D)-(B+C).

ADIP

The recordable mini-disc is not full before recording, i.e. it does not contain any information. However, if it did not have preliminary markings, it would be impossible to correctly position the laser beam both when writing and reading information.

Therefore, each MD-R, like a CD-R, has a U-shaped physical groove, or address pre-groove (ADIP), that is stamped into the disc when it is manufactured. The pre-groove is located behind the spiral data track and has a special configuration (Fig. 9) containing a biphasic signal with a fundamental frequency of 22.05 kHz, modulated at frequencies of 21.05 and 23.05 kHz.

Of course, the pre-groove is exactly the same on all discs. Without ADIP, it would be impossible to achieve correct positioning for any reading system. Therefore, ADIP is an indelible addressing capability.

Reading an ADIP is similar to reading a CD, as the laser beam hits the disc surface at some times and hits the groove at others, resulting in RF modulation. By reading ADIP information from sensors A, B, C and D after demodulating the signal, an addressing table is obtained that determines its exact address for each position on the disk. The same address will be used when new data is written.

Track layout

On a non-recordable disc, the track arrangement is similar to that of a compact disc: an area containing the table of contents (TOC), a program area, and an end area. As for recordable discs, in addition to the initial zone containing TOC, they also have a UTOC zone - the user’s table of contents, where the user writes the starting and ending addresses of music tracks.

Thus, the minidisc provides the ability to change track numbers, divide a track into parts, etc. All this is done in the UTOC area. For example, if the user wants to split one track into two, the music data in the program zone will remain intact, but the addresses and table of contents in the UTOC zone will be changed.

In Fig. 10 shows the arrangement of tracks on the disk.

Data Format

The MiniDisc data format (Figure 11) is similar to the CD format, but uses part of the CD-ROM format.

Firstly, when encoding, a noise-proof code known from CD is used, the so-called concatenated interleaved code CIRC (Cross interleave Read - Solomon code). But here this code has been modernized, has a greater number of interleavings and therefore received a new name - ACIRC.

Secondly, a cluster format is used. Each cluster contains 36 sectors: of which 32 are data sectors and four are communication sectors (Fig. 12). The sector format is the same as in CD-ROM. The content of communication sectors depends on the type of disk. In a non-recordable disk, the four communication sectors are fixed subdata sectors that can be used for disk information, graphics, etc.

The use of communication sectors is also necessary in the case of a recordable disc. A cluster is the smallest writable block. Obviously, there must be a buffer zone between any two of the blocks being written to avoid accidental overwriting from block to block.

The first three sectors of each cluster are used as communications, the fourth is the subdata sector. This means that the subdata capacity of a recordable disc is only a quarter of that of a non-recordable disc. However, one must keep in mind that subdata is redundant data. In Fig. 13 shows the cluster format.

Each sector contains 2352 bytes, of which 2332 bytes are data, and the first 20 bytes belong to timing, mode and partitioning. Next come the sound groups; Each of them has 424 bytes, of which 212 bytes are music data for the left channel and 212 bytes for the right channel. A sector contains 5.5 audio groups, i.e. five full groups and one half with only left or only right channel data. The corresponding channel bytes will then be placed in the next sound group.

Adaptive Transform Acoustic Coding (ATRAC) system

This is a key part of MiniDisc technology. To encode and decode an audio signal, special adaptive transformation methods are used, which depends on the structure of the input signal and is based on some well-known laws of psychoacoustics.

The purpose of this conversion is to compress the data, condensing it, while maintaining natural sound reproduction.

The standard CD system (16 bit, 44.1 kHz, two channels) has a data rate of 1.4 Mbit/s. This is before EFM and CIRC processing. In the process of recording music onto a mini-disc, analog signals are also sampled at a frequency of 44.1 kHz and quantized by a conventional ADC, the output of which, like for a CD, is also approximately 1.4 Mbit/s. However, to accommodate this amount of data on a much smaller disk, the bit rate must be reduced. This function is performed by the ATRAC compression system, which reduces the data transfer rate by almost four to five times, from 1.41 Mbit/s to approximately 292 kbit/s. This makes it possible to reduce the disc diameter from 120 (CD) to 64 mm (mini-disc) with the same playing time of 74 minutes in both cases.

Comparative sizes of disks without data compression and with compression are shown in Fig. 14.

Moreover, we emphasize once again that this reduction in data flow should not affect the quality of the transmitted music. After all, it is absolutely obvious that it is absolutely impossible to do without losses, therefore the system software must be such that the deterioration in sound would not be noticeable to the ear even for the most sophisticated listeners. This can be achieved if you thoroughly study some well-known laws of psychoacoustics and use them.

Psychoacoustics

It is known, for example, that the sensitivity of human hearing is not the same at different frequencies of the sound range. Thus, a sound of a certain level is clearly perceived at one frequency and may not be audible at all at another frequency, even at a higher level. Rice. 15 illustrates the psychoacoustic effect - a change in the thresholds of auditory sensation in different parts of the audio frequency range.

Analysis of the music signal entering the recording input allows us to determine which parts of the spectrum lie below these thresholds. They can then be removed and the bit rate can thus be reduced. Moreover, the sensitivity of our hearing system (ear - cerebral cortex) in some so-called critical frequency ranges remains constant and does not change. The higher the sound is located on the frequency scale, the wider the band of the critical range, where the level of hearing sensitivity remains unchanged. In the range of 100 Hz, the width of the critical range is approximately 160 Hz, and at a frequency of 10000 Hz it is 2500 Hz. 25 such critical ranges were identified using the method of subjective examination. But in the ATRAC system, for the sake of greater transmission accuracy, even more of them are used - about 52.

In Fig. 15 another very important psychoacoustic effect is noticeable - masking of signals. If two sound sources are playing at the same time, but at different levels, the louder sound will mask the quieter sound. For example, if during a conversation between two people an airplane flies in the sky, then the conversation will have to be stopped for some time due to the complete masking of voices by the noise of the engine, which is significantly louder than them in volume.

In addition, the effects of so-called pre-masking and post-masking are observed.

A louder sound occurring just before or after a quieter one (within 3 ms) (within 200 ms) masks the quieter sound.

By analyzing a certain number of input samples based on patterns known from psychoacoustics, it is possible to determine which components of the input signal will be inaudible due to masking and can be removed with minimal sound degradation. The effect of tone masking is illustrated in Fig. 16.

Block diagram of coding in the ATRAC system

The psychoacoustic patterns outlined above are used by the ATRAC system to reduce the input stream rate from 1.4 Mbps to 292 kbps and are carried out in a specific sequence.

However, it should be noted that the ATRAC system allows you to change the system software, i.e. use different calculation algorithms without creating compatibility problems.

First, processing is carried out in the time domain. From the digitized input audio signal (16 bit, 44.1 kHz), at the first stage of analysis, a band of 11.025 - 22.05 kHz is isolated using a high-pass filter. At the second stage, a crossover filter separates the middle and low frequencies (from 5.025 to 11.025 kHz and from 0 to 5.025 kHz, respectively). This delays the higher frequencies to maintain synchronization between them and other bands.

The coding block diagram is shown in Fig. 17.

Subsequent analysis of the signal is carried out in the frequency domain. Based on the Modified Discrete Cosine Transform (MDCT) algorithm, essentially similar to the Fourier series expansion of the signal, the spectrum of the input signal is analyzed to determine its frequency components and their corresponding levels. However, this analysis is performed block by block (in separate sides - signal segments). Therefore, before moving from the time to the frequency domain, separate time/frequency blocks are selected, in each of which the input signal will be analyzed to convert it in accordance with the ATRAC algorithm. The number of blocks and their duration are determined.

As mentioned earlier, the number of critical frequency bands is established experimentally, but ATRAC uses a much larger number of bands, the timing of which is determined adaptively according to the structure of the input signal. The duration of each block that will be analyzed must be a maximum of 11.6 ms (or 512 samples, at a sampling rate of 44.1 kHz). The reason why the time interval of the time/frequency block has to be taken into account is also explained by the laws of psychoacoustics. The fact is that with a short pulse sound we can only notice fairly large level jumps. On the other hand, with smooth music, when the sound level changes slowly, the ear is able to register very subtle volume nuances.

Based on the assessment of these differences, the ATRAC system selects both the number of blocks and their optimal duration. When the selection of blocks has been made and MDCT calculations have been carried out in each block taking into account the principles of psychoacoustics, the actual transformation of the data stream is performed. In this case, the data transfer rate can be reduced as redundant information is removed. In Fig. 18 shows the ATRAC encoding process.

Finally the bit allocation is done. Each time/frequency block is analyzed. Its level is determined. The compressed digital data obtained using the ATRAC algorithm is not directly audio data. They merely describe the dynamic range of that signal block in variable bit word lengths from 0 to 15, as well as the relative signal level, which is a scaling factor. The remaining bits can be removed because they are quantization noise. Thus, only the significant part of the data is encoded. In Fig. 19 shows ATRAC data.

Digital data passed through the ATRAC encoder, as with CD encoding, is processed by the ACIRC circuit and subjected to channel coding using the eight-by-fourteen modulation (EFM) method.

ATRAC reading

When playing mini-discs, as in the CD system, EFM demodulation and ACIRC decoding are performed, after which the received data (word lengths, scale factors of this spectrum) enters the ATRAC decoder, where reverse calculations are performed, as a result of which the original data is restored ( Fig. 20). Because the MD recorder's input and output data adhere to the 16-bit, 44.1 kHz standard, it can be directly digitally copied from the MD recorder to CD-R and other digital media. However, the MD standard includes a sequential copy management system (SCMS), which allows only the first copy, disallowing the second. In other words, you can't make a copy of a copy.

Shock resistance

An obvious advantage of the MiniDisc system is the increased resistance of MD players to vibration and shock. This is because the audio data is read from the MiniDisc much faster (1.4 Mbps) than required by the ATRAC decoder (292 Kbps). Therefore, the read data is first entered into the preliminary buffer, from where it enters the ATRAC decoder. Rice. 21 illustrates the principle of shock resistance.

Each 1 Mbit of buffer RAM chip placed between the pickup and decoder stores information for 3 seconds of music in real time. When the buffer RAM is full, the cartridge pauses reading data until the buffer is full again. If playback fails due to impact or vibration, the pickup has sufficient time (up to 3 seconds) to return to the correct playback track. This is achieved by re-positioning the sector. If a pickup fails, the system detects the incorrect address and returns the pickup to the correct position. Immunity to significant shocks, vibrations and shocks during playback is especially valuable when the MD player is not used in a stationary environment.

Conclusion

The MD format provides random access, durability, portability, ease of use, shock protection and rewritability. Analog digital tapes are inferior to MD in terms of durability and because they do not have the random access inherent in disk media. This explains the gradual displacement of conventional tape recorders by equipment for recording and playing mini-discs wherever the speed and efficiency of creating sound programs is a determining condition for successful work.

The truth about minidiscs

One of the impressive inventions of the 20th century is the mini disc (MD). High-tech media that holds up to 5 hours of music, up to 140 MB of data; smaller in size than a 3.5-inch floppy disk, well protected from damage... A mini-disc is not only in many ways similar to a regular music compact disc (CD), but also surpasses it in a number of indicators.

From the very beginning, Sony developers (namely, they have the honor of inventing the MD format) set themselves the task of creating a “small-caliber” media for music files, capable of displacing the recognized favorite of the audio market - the compact cassette, and bypassing it in the most important parameters - sound quality and size. Managed! The result was a plastic disk enclosed in a flat miniature case, coated with a special layer, on which music is recorded (and then read) using a laser.
The advantages of the new product over the “honored veteran” are obvious and undeniable. Imagine: a lightweight, durable and compact mini-disk can withstand a million (!) rewrites. A cassette is not capable of even a thousandth of this. Each time it is played, it degrades the properties of the audio recording on magnetic tape...

Bypass CD

It should be remembered that Sony itself was instrumental in the development of the CD format, so it was especially difficult for its engineers to develop a medium that could compete with the well-established compact disc in the market.
How is a mini-disc different from a CD? Firstly, the ease of recording. By the time the first MDs were released, the ability to independently record a music CD was practically inaccessible to the average music lover - for this there were only computer peripherals, and such devices cost a lot of money (in addition, access to them, like to any professional equipment, was strictly limited) . And a mini-disc can be recorded directly in a player, both portable and stationary. A computer is not needed for this simple procedure.
Secondly, MD is a “reusable” media. You can selectively erase information recorded on it (say, a song you don’t like, one from an entire album). For a “disposable” CD (CD-R), this is impossible in principle, and a “rewritable” (CD-RW) must either be cleaned completely or use special preparation for recording - formatting, which eats up a sixth of the useful volume and is quite labor-intensive for an inexperienced person. user. And again - all this only with the help of a computer.
Thirdly, you can record both music and data on a mini-disc (albeit with the help of special equipment). Different CDs are produced for audio files and computer data (universal CD-Rs can be used to record music, but only in computer drives; household audio CO recorders refuse to read such discs).
Fourthly, in terms of sound duration, MD is not only not inferior to CD, but, thanks to recent improvements, even surpasses it. Fifthly, a mini-disc is 2.5 times smaller than a CD. And finally, the durable housing and sliding MD curtain reliably protect its working layer from accidental damage. Very often they write about comparing the sound of a CD (a kind of standard of sound quality) and its copies on various media. This is true. Recordings on compact cassettes are most often a copy of a CD. It is logical that owners of MD players save their favorite tunes from CDs to a mini-disk.
In order to record the same amount of musical information on a smaller medium as on a CD, a special sound compression algorithm was developed taking into account the characteristics of human hearing. It was called ATRAC. The sound recorded on MD is slightly different from the original, but not enough to be felt by the naked ear. Each re-recording from one mini-disc to another leads to the accumulation of errors and distortions, which at some point become noticeable to the ear. But the sound quality of the first MD copies from a CD is practically indistinguishable from the original.
It is argued that even a non-professional music lover will always distinguish an MD recording from a CD recording. What about the fact that well-known companies - Yamaha, Kenwood, Denon - have long been producing stationary Hi-Fi-class mini-disc equipment intended for use as part of a home audio system?
MD has been adopted not only by hi-fi companies, but also by professional musicians working in the entertainment business. Today, not a single serious band going on tour can do without MD equipment. Musical accompaniment, effects and even “plywood” are recorded on the mini-disc. MDs are also widely used on radio stations, the extensive music library of which is collected mainly on mini-discs. They are much more convenient than any other media: they allow you to quickly find the desired piece of music or composition, do not get dirty, do not scratch, do not break, do not deteriorate over time, and take up very little space.

Mini Discs for Hobos

But still, Hi-Fi class MD equipment is not a “consumer product”. Demand for it is restrained both by its low price and the need to have a home audio system.
Portable mini-disc players have proven to be much more popular. They are much smaller than cassette players and have many advantages in terms of control - more and more models are equipped with a remote control that is simply clipped to the lapel or collar. A miniature indicator displays information about the track being played, the artist, and the duration of the musical fragment. (By the way, the mini-disc contains detailed information of this nature. This is another blow to the reputation of the old cassette tape. And if you also take into account the almost instantaneous access to any fragment, accelerated playback, the ability to loop playback of a piece of music and even the entire disc... )
All portable mini-disc players are divided into two categories: simply players - playback devices - and recorders that can independently record mini-discs from an external source. Among them there are recorders with both an analog input, to which you can send an audio signal even from a microphone, and with a digital one (also called an S/PDIF input). With the latter, it is possible to transfer music from a CD player equipped with an S/PDIF output almost losslessly. You can even use a computer with a sound card equipped with such a “digital” connector for this.
It is quite natural that there are also mini-disc players with a built-in radio on the market. As a rule, they even provide recording from the receiver to a mini-disc.

Double... quadruple!
Not so long ago, the younger brother of MD was born - the MDLP format. Two additional letters in the name hint at the possibility of slow playback - Long Play. MDLP has two modes - LP2 and LP4 (the playing time of the mini-disc increases by two and four times, respectively). A “double” modification of a standard mini-disc for 74 minutes will “hold” more than two hours of sound, and a “quadruple” modification will last about 5 hours; but there are also 80-minute discs... This, of course, affects the sound quality, but not enough to talk about noticeable losses. It’s just that some specificity appears in the sound, but that’s all.
The advent of MDLP made it possible to halve the time for rewriting from one MD to another (until now it was necessary to wait for those same 74 minutes). But this only applies to the standard recording mode; for slow-motion LP2 and LP4 you will have to wait two hours (or, accordingly, five) until the process is completed.

Alexey Adamenko

In March 2013, Sony finally ceased production of digital mini-disc recording and playback equipment. Two facts are surprising about this event: how monstrously outdated this format is perceived today, although it is only twenty years old, and how long it has managed to exist in the world of the Internet and MP3. Let's remember how MiniDisc, which was very promising at one time, was born and why it did not become as popular as a CD.

The development of the MiniDisc standard began in 1986, when the world had already begun to embrace the first mass-produced digital audio medium - the CD. The CD format was introduced in March 1979, and the first mass-produced players went on sale in April 1982. Compact discs were well received by music lovers, who quickly appreciated the new medium's wide frequency and dynamic range. What was especially striking was the absolute silence during the pauses, against the backdrop of the characteristic crackling of the gramophone record, which the CD replaced.

The mini-disc was conceived as a successor to the compact cassette, only more convenient to use and storing recordings in digital form. Work on MiniDisc began in the mid-eighties - during the “golden age” of audio cassettes, when the “ceiling” of the capabilities of this medium was reached and its fundamental limitations became obvious to all leading manufacturers.

The new medium had to remain as compact and protected from external influences as a compact cassette, which would allow it to be used not only in home audio systems, but also in car and portable players. In terms of sound quality, it had to surpass an analog cassette and be as close as possible to a CD with a significantly lower capacity. Finally, an indispensable condition had to be the possibility of repeated independent recording on such a medium - with the functions of erasing and editing.

As a result, MiniDisc implemented several advanced technologies at that time, protected by many patents. The recording technology chosen was magneto-optical, in which a laser is used to store data, heating the ferromagnetic layer of the disk to the required temperature, and a magnetic head, changing the magnetization of the heated area. To play back the recording, a laser of lower power is used: when the beam is reflected from one or another area, depending on its magnetization, the plane of polarization changes, which is recorded by an optical sensor.

Magneto-optical technology gave the mini-disk important advantages: high reliability of recording storage and increased resistance to magnetic fields, fast random access to any fragment of the recording, as well as the ability to repeatedly erase and rewrite. The only significant drawback at that time was considered to be the high energy consumption of the laser during the recording process, which limited the use of burner drives in portable devices. However, it later became clear that MiniDisc is just as susceptible to demagnetization as regular film.

The disk, with a diameter of 2.5 inches (64 mm) and a capacity of 140 MB, was installed in a hard plastic case with a sliding shutter - almost like a computer floppy disk. To encode the music, Sony's proprietary Adaptive TRansform Acoustic Coding (ATRAC) algorithm was used, based on the psychoacoustic characteristics of human hearing. In its ideology, ATRAC is close to MP3 and other lossy audio encoding formats: information that is inaudible or barely audible for most people is removed from the file, and the remaining is compressed to the most compact volume possible. As a result, 74 minutes of recording on a 650 MB CD was compressed to 140 MB of mini-disk capacity. The base bitrate of the first version of ATRAC was 292 Kbps.

The MiniDisc format and equipment for recording and playing minidiscs were officially introduced on January 12, 1992, but the new product was received rather coolly. In terms of sound quality, the first players were inferior not only to DAT tape recorders with a lossless recording algorithm, but even to the now completely forgotten DCC digital cassettes, which also used a psychoacoustic compression scheme. At the same time, the equipment turned out to be comparable in cost to CD players, which provide much higher sound quality.

The fourth version of ATRAC, released in 1996, made it possible to further improve sound quality: all internal data processing was carried out with 24-bit precision while maintaining a 20-bit ADC. Version ATRAC 4.5 was used in the flagship models of the ES series and featured improved encoding quality at high bitrates. The ADC capacity has increased to 24 bits. The latest variant of the classic ATRAC DSP Type-R appeared in 1998 and, among other things, provided high-quality high-frequency encoding.

However, the MiniDisc already had a bad reputation among audiophiles, so Sony was forced to once again draw attention to the MiniDisc with high-quality decks of the top-end ES series - MSD-JA30ES and JA50ES, which received 20-bit analog-to-digital converters and excellent-sounding DACs. At the same time, it became possible to produce inexpensive devices with high sound quality. A good example is the MDS-JE500 deck, released in 1996 and built on a CXD2650R chip with the ATRAC 4 algorithm. These devices easily competed in sound quality with mass-produced CD players, and were ahead of them in terms of service functions.

The sixth generation of ATRAC, which appeared in 1999, received its own name ATRAC3 (without a space). In this implementation, MDLP extended recording formats were added: 2LP with a bitrate of 132 Kbps and LP4 with a bitrate of up to 66 Kbps, allowing you to record up to 324 minutes of audio on an 80-minute disc. The ATRAC3 DSP Type-S modification combines the ATRAC3 codec with improved playback quality of low-bitrate MDLP recordings and the top-end first generation ATRAC1 Type-R codec.

At the end of 2001, the MiniDisc format finally made it possible to copy recordings to a minidisc not only in real time. The NetMD extension made it possible to send data in an accelerated mode via the USB interface from a computer running Windows through the proprietary SonicStage program. At the same time, recordings in SP mode could be copied at speeds up to 1.6x, in LP2 mode - at speeds up to 16x, and in LP4 mode - up to 32x or even 64x compared to the actual duration of the recorded program.

Finally, in January 2004, the latest incarnation of MiniDisc, Hi-MD, and the ATRAC3Plus codec came to market. At the same time, 1 GB Hi-MD mini-discs were released, on which you could record up to 45 hours of music with a minimum bitrate of 48 Kbps, up to 94 minutes of uncompressed music with a bitrate of 1411 Kbps and a sampling frequency of 44.1 kHz in the format Linear PCM or up to 980 MB of computer data. In 2005, the production of Hi-MD players with built-in MP3 support began, and in 2006, for the first time, it became possible to transfer digital files from MD players to a computer (with the exception of discs recorded via SonicStage or OpenMG).

The ATRAC Advanced Lossless codec made it possible to record music in ATRAC3 and ATRAC3plus formats, accompanied by a service stream to completely restore the original signal. At the same time, the player could decode both compressed and uncompressed formats, which ensured compatibility of such recordings with outdated equipment.

Hi-MD drives began to appear in computers manufactured by Sony. Such drives were displayed as removable drives with the FAT32 file system. File transfer was carried out using the operating system, and only to record an audio disc played on an MD player, it was necessary to use the SonicStage program.

The Hi-MD standard has finally eliminated many of the “copyright” restrictions of classic MiniDisks that are ridiculous by today’s standards - for example, intermediate transcoding into PCM of files transmitted via NetMD or the impossibility of digitally re-recording sound recorded “live” from a microphone or linear input neither via optical output nor via USB (Serial Copy Management System technology).

All these restrictions were originally designed to prevent “bit-by-bit” copying of DRM-protected files, allowing only analogue rewriting with loss of quality. In the latest versions of SonicStage, the only limitation remaining is the inability to edit tracks on a Hi-MD recorder recorded to disk using this program. There are no such restrictions when recording via optical or line input.

While Sony was perfecting algorithms for hardware encoding and decoding of compressed audio, the 21st century had arrived - the century of the Internet, MP3, Napster, file-sharing networks, iPods and flash players, CD, DVD and Blu-ray burners. It seems that all this time the creators of MiniDisc lived in some other reality and returned back only to 2004. The heroism with which Sony carried this long-outdated format until 2013 deserves a separate discussion.

Support for the MiniDisc format in Europe and North America was discontinued in 2008, as was access to the company's online music store Connect. However, mini-discs were not particularly popular either in the Old World or in the New. The fashion for pocket MD players passed back in the late 90s, and after that mini-discs were used only at radio stations and in music studios, where quick access to different fragments of the recording and ease of editing were required. In Russia, the last time I saw blank mini-discs in stores was at least ten years ago - right around the same time the boom of MP3 and CD burners began.

However, in Japan itself and in some other Asian countries, the minidisc has become much more widespread. It initially turned out to be a more affordable and convenient alternative to CD, which came in very handy for a country that spends a lot of time on trains. Japanese youth still happily buy singles from fashionable artists on mini-discs and record entire “live” concerts on them. The Japanese version of the Connect online store is not going to close either. It is not surprising that some other companies, including Onkyo, are not yet planning to abandon the production of MD-players, although Sony stopped producing portable MD-Walkman two years ago.

We can only say goodbye to another format, the fate of which turned out to be very difficult only because its creators tried to combine the incompatible: the digital technologies of the future with the habits of large corporations to ignore the surrounding reality and impose their archaic rules of the game.

MD technology was introduced by SONY as a digital replacement for the analog compact cassette. The first MD players, released back in 1992, did not receive recognition. In the first years of the standard's existence, MD equipment was used mainly by professional studios and concert venues; phonograms for concerts and performances were recorded on minidiscs.

SONY, the developer of the MD standard and the main manufacturer of the corresponding equipment, took a wait-and-see approach and over the course of several years brought the technology into a more usable form. Most of the complaints about MD concerned the sound quality: in the early 90s, chips encoding sound using the ATRAC algorithm did not do their job well, and the sound they produced was too rough. In just a few years, the digital processor industry has undergone a real revolution, and MD devices have new DSPs at their disposal that process audio in real time using floating point numbers, consume a tiny amount of power and sound better.

In the late 90s, most high-end minisystems had MD decks built into them. New portable MD players SONY MZ-R90 and MZ-R70 gained popularity due to their meager weight, long battery life and absolutely chic capabilities for those times: recording to MD from a microphone, line and digital inputs. The sound quality of these players was not inferior to other models of portable CD players. The disadvantages of MD devices were the inability to digitally copy data from minidiscs and the need to rewrite audio in real time, just like on a good old compact cassette.

And finally, in 2002, a kind of revolution took place in the world of MD: the new line of SONY players supports NetMD technology, which allows you to transfer sound to minidiscs via USB at a speed of up to 32x (compared to traditional MD devices).

About minidiscs

Before moving directly to the description of the new player, we should briefly talk about what a minidisk is.

The MD is housed in a protective case like a floppy disk, which greatly reduces the possibility of mechanical damage. It is known that both writing and reading are carried out by one laser head on magneto-optical disks using a laser beam with a wavelength of 790 nm.

Sound on MD is recorded using the ATRAC algorithm, which works according to an algorithm similar to MP3: “excessive” sound information is eliminated. Many people believe that the compression quality of the latest generations of players is superior to MP3. A regular disc holds 74 or 80 minutes of audio.

The minidisc has a special zone called TOC (Table Of Contents), which stores information about track addresses (similar to FAT). With its help, you can search for tracks, which works on most new players almost instantly.

The new players also feature the MDLP function, which allows you to record sound with a higher degree of compression (obviously by reducing the bitrate). LP2 mode allows you to record twice as much music (148 minutes on 74-minute discs and 160 minutes on 80-minute discs), LP4 four times more (296 and 320 minutes, respectively). Mono recording mode is also available.

Despite the presence of moving parts, it is almost impossible to disrupt playback using sudden movements and other things in a modern MD player. In this parameter, MDs, despite popular belief, are not much inferior to MP3 players.

One MD can withstand an average of 5000 rewrite cycles. The cost of a minidisc fluctuates around $1.5–3.

Sony MZ-N1

The box with the Sony MZ-N1 contained a huge number of various accessories. In addition to the player, the following were discovered:

  • Network adapter;
  • USB cable;
  • Optical digital cable;
  • Remote control with LCD display;
  • Unnamed SONY headphones;
  • Cradle stand with connectors for USB and adapter;
  • Flat battery in a plastic box;
  • Holder for connecting AA batteries to the player;
  • Soft leatherette bag for carrying the player on your belt;
  • CD with OpenMG software;
  • A bunch of different instructions.

Excellent package. From the very beginning, everything you might need to work with the player is included. Except, perhaps, a blank minidisc. (It is included only in the American version of the player).

Connectors and power

Cradle stand is a new product for SONY MD players. It is made of durable plastic and has a fairly massive base, which also has sticky rubber feet attached for stability. A USB cable and an AC adapter are connected to this stand.

The following connectors were found on the player itself:

  • connector for connecting a remote control and headphones (also a linear output);
  • microphone input with phantom power;
  • linear/optical input;
  • contact block for installation in a cradle;
  • connector for connecting a network adapter;
  • screw fastening and two contacts for connecting an external module containing an AA battery/accumulator to the player.

Pay attention to the last point. The player comes with a special plastic structure with a screw, which is screwed to the player and allows you to place a standard AA battery inside. Thus, the player can be powered simultaneously from a flat battery and an external battery.

The characteristics declared by SONY for the duration of the player's operation in various modes look like this:

Recording (hours)

Playback (hours)

As you can see, the numbers are quite impressive. However, it doesn't hurt to check them. I conducted the test in the two most common modes: playback in a compromise LP2 and recording in a “stripped-down” LP4 for recording lectures and concerts on MD (and where else might you need long-term autonomous recording?). Playback was checked by playing various discs in full; time was calculated by multiplying the playback duration by the number of discs played. Recording is lectures written for two days in a row at your favorite institute. Testing was carried out with a fully charged battery from the MZ-N1 kit. The results were as follows:

Playback, LP2, NH-14WM coin cell: 30 hours

Recording, LP4, NH-14WM coin cell: 13 hours

As you can see, the numbers are different, but the result is still quite satisfactory. As for the recording, you should keep in mind that the recording came from a microphone, to which the player may supply phantom power. In linear recording mode the numbers will obviously be higher.

Ergonomics and controls

Like all top models of SONY MD players, the MZ-N1 contains a minimum of controls on the front panel. On the side there is a Jog-Dial wheel, used to navigate through the player's tree menu. On the other side there are volume control buttons Vol+/Vol-. Below the screen there are END SEARCH buttons (go to the end of the last recorded track), a RECORD slider and a GROUP button, which turns on/off the group playback mode. Also hidden on the side is the T MARK button, which is responsible for placing track end marks during recording and playback.

Playback is controlled using a rectangular joystick located next to the screen. Pressing the edges of the button works as going to the next/previous track, moving it up/down as pause/stop, and pressing in the center as playback. To say that he is uncomfortable is to say nothing; this thing is a real joke. It is almost impossible to use due to its rigidity and frequent inappropriate response to user actions. Compared to it, the cheaper models of NetMD players SONY MZ-N700 and MZ-N505 have much more convenient controls.

Player display three-line. In playback mode there is enough space for the track name, time and signal level line. The menu is organized tree-like and very convenient; The abundance of functions is simply amazing at first. You can customize everything from screen contrast to graphic equalizer modes and microphone sensitivity. Using the jog-dial, navigation is carried out in a matter of seconds. During playback, the miracle wheel allows you to quickly “jump” through the list of tracks or groups.

For each MD disc, you can record individual settings, such as sound, playback programs, etc.

The remote control is also equipped with a backlit screen. From the remote control you can access all the player settings through the same expansive menu. Playback control using the remote control is organized in a very original way. On the edge there is a rotating wheel, which, when turned, causes the next/previous track to be played, respectively. The “Stop” button is located on the side of the remote control. The wheel can be moved forward; then turning it will be responsible for adjusting the volume. I must say that this solution, in my opinion, is not the most successful; it almost eliminates one-handed control of the player. The fact is that the wheel is quite rigid, and when you try to turn it on a remote control attached, say, to clothing, it begins to desperately resist and slide off the mount. Therefore, you have to hold the remote control with your other hand. In the previous generation remotes and the most inexpensive NetMD player (Sony MZ-N505), the remote control without a screen provided much more convenient control.

In a word, inexpensive models of NetMD players definitely outperform the leader in ease of control.

Sound

The Sony MZ-N1 uses the latest generation ATRAC DSP Type-R DSP chip, which Sony says is used in high-end stationary MD decks. Type-R operating mode is activated only when recording in SP (standard) mode and only through optical/line inputs.

The listening sessions were carried out using professional Sennheiser HD200 headphones; the player’s power was enough to fully pump them up. A very important observation: the new range of SONY MD players sounds much louder than all previous ones, which were characterized by rather quiet sound. The culprit is probably the new Class D amplifiers used by SONY in modern portable devices. In the subway and with intense external noise, even not too loud music will be clearly audible.

A few words about the included headphones: the manufacturer was embarrassed to write the model name on them, and that says a lot. Although, I must say, they are not so bad: of course, the bass is almost completely absent, but they match the sound very well with the player. They give a noticeable peak in the mid-high frequencies, which, combined with the harsh nature of the player, sounds quite harmonious.

Subjectively, the sound of the MZ-N1, compared to models of previous generations, has become somewhat harsher and even rougher. Perhaps the fault for this lies with the DACs used in the player, and not at all with the new Type-R chips. Using the RMAA program, we decided to check the characteristics of the MZ-N1 DACs. The input signal from the Audiotrak Maya44 professional sound card was supplied digitally via an optical cable, and was taken from the output of the player, turned on in LineOut mode, back to the Maya44. The results turned out to be very interesting.

Maya44 SPDIF Out Sony MZ-N1 Line Out Maya44 Line In

Frequency response (from 40 Hz to 15 kHz), dB:-0.65, +0.00 Good
Noise level, dB (A):-90.8 Very good
Dynamic range, dB (A):88.5 Good
THD, %:0.014 Good
IMD, %:0.078 Good
Stereo crosstalk, dB:-90.5 Excellent

General performance: Very good

As you can see, the characteristics of the analog part of the SONY MZ-N1 are quite up to par. Extremely low noise, significant dynamic range. The decline in the low-frequency region is probably a consequence of the ultra-low-frequency damping circuit, which is used in players to reduce power consumption.

By the way, a feature of the Sony MZ-N1 compared to previous models is that when a signal is supplied and removed from the player in real time, the sound supplied to the output is uncompressed. That is, the sound is monitored by the player “as is”, without undergoing ATRAC compression. That is why we were able to arrange a quality test of the player's converters. Older models of players immediately supplied ATRAC-packed sound to the output.

NetMD

Now we come to the most interesting part. What exactly is NetMD technology?

Immediately upon connection, a new device (NetMD) was detected. The drivers installed without any problems. The first step was to install the OpenMG Jukebox program.

OpenMG is a very powerful jukebox software that supports ripping from audio discs, playing WAV/MP3/WMA/ASF tracks and playing from different playlists. But the most important thing is that the program allows you to fully control the MD player when it is connected to USB and, of course, transfer audio files to a minidisc. Let me make a reservation right away: the NetMD players that exist today CANNOT transfer tracks from a disk to a computer via USB.

Therefore, if you want to rewrite something that is recorded on a minidisc, you will have to use old-fashioned means: an analog cable and a sound card. (It is clear that this will not add quality.)

OpenMG reads the contents of the MD disc in the player and produces a list of tracks. You can do whatever your heart desires with the tracks: rename, move, delete, play. In general, the OpenMG developers tried to make sure that the user has the complete illusion that he is working with ordinary files, and not with sophisticated records on a minidisk. Tracks can also be combined into groups - analogues of folders on regular MP3 players.

Recording sources on MD can be either WAV/MP3/WMA files or tracks from AudioCD, and track names on MD are imported directly from ID3 tags or CD file/track names. Russian, unfortunately, is not supported.

The principle of operation is as follows: you import the selected files into the internal OpenMG playlist, where the files are packaged using the ATRAC algorithm in the mode you choose (SP, LP2, LP4). Yes, yes, that’s right MP3/WMA are not recorded directly to MD, of course. Therefore, it turns out that the packaged “empetrishkas” go through another stage of compression using the ATRAC algorithm. Even though the thought of double compression seems a little monstrous, in reality it’s not that scary. Firstly, when classically transcribing music directly (via the linear output of the sound card), the sound is subject, in addition to ATRAC, to the action of imperfect converters on the board and the player itself. If you record tracks via an optical cable, the sound will still be ATRAC-packed. The beauty of this algorithm is that in SP/LP2 modes it does not introduce audible artifacts into the sound, as MP3 likes to do. ATRAC files have a soft sound, somewhat limited in detail - and only in LP2/LP4 modes. Therefore, double transcoding of files is not at all as scary as it might seem at first.

However, traces of double packaging are very clearly audible when rewriting MP3 with a low bitrate (So, let’s take a dozen MP3 files with a bitrate of 192 kbps and transfer them to the player. SONY declared transfer speeds via USB 1.6× for SP, 16× for LP2 and 32× for LP4. First, let's try to rewrite files in SP mode.

First, the file is converted. When you subsequently rewrite the same track, conversion will not occur, since the converted files are stored in a special format in the OpenMG folder. Start of recording: the player flashes the “RECORD” light, the recorded time is displayed on the player screen. It runs quite slowly: file 03:01 was recorded for exactly two minutes, which corresponds to a speed of 1.5×. Well, not much, but not too different from the figure announced by SONY.

And finally, LP4. The file was rewritten in exactly 6 seconds, which corresponds to 30×.

Test number two: let's try to rewrite an Audio-CD using OpenMG. Austin Powers 2: The Spy Who Shagged Me OST, 42:28. We will rewrite everything in the same proletarian LP2 mode.

OpenMG behaved very wisely: while the first ripped track was being copied to MD, the second one was being digitized from the disk. Saves a lot of time. And as a result, the entire process of rewriting AudioCD to MD took 3.5 minutes. By the way, before the grabbing process, OpenMG went to the Internet and downloaded the names of the CD tracks, which were completely copied to the minidisc.

And finally, a traditional spoonful of unappetizing substance. To protect against piracy, OpenMG has built-in “advanced” security technologies. So, the process of rewriting to MD is called check-out. You can make no more than three of these checkouts for one file; that is, you can only hold one song on three different MDs at the same time. You are also free to perform a check-in process, which will delete the track from the MD and give you the opportunity to rewrite it to another disk.

In principle, this in itself is not so scary. Few people would think of keeping one recording on more than three discs. Moreover, as a last resort, you can copy one file under a different name... But here many obstacles arise. Firstly, tracks recorded via OpenMG cannot be deleted from the player yourself. To delete a track, you will have to connect it to the cradle and perform operations in the program. Non-standard situations may also arise: for example, the system crashes or something unpleasant happens to the files on the hard drive. In this case, nothing can be done with your MD until you reinstall the program. What if you lose your minidiscs? You will not be able to check in and the files will not be writable.

Conclusion

It must be said that NetMD technology has fully justified itself. Everything works exactly as promised. However, the ghostly fighters for the rights of record companies could not help but spoil the matter: we received ill-conceived protection, which, in the Russian tradition of reinstalling the system every Tuesday and Thursday, could ruin all the pleasure of working with minidiscs. Another serious drawback, which was also given to us by the same fighters, is the inability to record audio digitally from a minidisc to a computer. In this regard, MP3 players are still invincible.

Today in Moscow, a player with NetMD technology can be purchased for a little more than $200, which is not too expensive compared to MP3 players on flash memory, even without taking into account the cost of the flash memory itself. Obviously, the price/quality ratio of cheaper SONY NetMD players is higher than that of the specific MZ-N1 model. Otherwise, this is a very successful device.

Pros:

  • increased speed of music rewriting due to the use of the USB bus;
  • high-quality sound, high volume level;
  • ability to record from microphone, line and optical inputs;
  • miniature dimensions (77.7 × 71.4 × 16.4 mm);
  • light weight (87 g without batteries);
  • long battery life.

Minuses:

  • high price ($330);
  • inconvenient controls;
  • ill-conceived copy protection;
  • loss of quality when rewriting MP3/WMA due to double compression;
  • lack of Cyrillic support;
  • cannot be used to transfer files
    (property of all MD players).