EDS TOOLS software for analyses, syntheses and processing of audio files is also worth mentioning. It is designed for well-forgotten MS-DOS and has got an excellent use's interface in Windows 95 stile. EDS TOOLS enables statistical analyses and processing of 8.. 32-bit floating-point data with 1024-channel equalizer, digital infinite and finite impulse response filters. Moreover one can easily realize any sound processing algorithms on built-in system of commands. EDS TOOLS functions of 3-dimension spectrum analyses based on FFT (Fast Furie Transform) and modern LPC methods are missing in the other software packages!

Legends concerning the specific digital distortions destroying sound quality and not registered by any devices are so dubious as telepathy or a "transistor" sound. The essence of a "transistor" sound consists in different speed of decline of the amplitudes of non-lineal distortion harmonics and insignificant quantity of even harmonics. Fast exponent harmonic amplitudes decreasing for increasing frequencies is characteristic of tube amplifiers. In transistor ones harmonic amplitudes decrease slowly (according to 1/x ratio). Additionally, in tube amplifiers almost all the following higher frequency harmonics are masked with few first ones. Such psycho-acoustic phenomenon is laid into the foundation of MPEG sound compression standard. So, as I see it, a tube amplifier adds some first even and odd harmonics to a sound signal. Usually a Hi-End tube amplifier has got a non-lineal distortion co-efficient from 0.5% up to 3.0%. I should mention that the same principle of adding the first few harmonics to the original sound signal is adopted in exciters - studio sound processing devices. I would say that a tube amplifier is a kind of an exciter. That is why the tube amps with very little non-lineal distortions are not popular with audiophiles. They produce the sound very close to that of the transistor amps with minor non-lineal distortions. In transistor amps mask effect shows itself significantly less. Due to that we begin to hear "grange" and "sand" in the sound. A non-lineal distortion co-efficient must be considerably reduced to get an approximately "tube" quality sound. It is a complicated technical task. Presently, its decision is not always economically justified. For that reason, a tube amp made in Asia can be much cheaper than a North American or European Hi-End transistor amp with subjectively the same sound quality. Therefore, not all the American firms on the Hi-End market survived the crises brought about in the beginning of 1998 (see March, 98 "Class A" magazine).

Sometimes tube amps are employed to make a sound "alive" in the final record. I this case, a digital record is converted to an analog one, processed by tube equalizers (like TL Audio G400) or amps, turned back to digit and recorded to CD-RW. If such a record is played back via a transistor amp you might hear a positive result of the procedure. Otherwise, the double use of tube amps (on the stage of recording and playing back) can completely "kill" the sound. There were attempts to manufacture the digital model of a tube amp. Though, in my opinion, RedValve plug-in for WaveLab performs a sound comparable with that of an inexpensive tube amp. Generally saying, the advantage of a tube sound is especially distinguished for the middle frequencies diapason (200..8000 kHz) and dissipated for the high frequencies (8..20 kHz).

BOSS GX700 is a very interesting sound processor from the point of view of the imitation of a "live" sound with purely digital methods. It fully emulates a typical virtual audio recording studio in digit in real time. At first, an audio signal (from electric guitar and so) goes to a 20-bit high quality ADC. Then a digital signal is processed with the imitation of a tube power amp and equalizer. Moreover, you can choose them from the wide list of the devices really existing in the audio market. Then the signal proceeds through a speaker simulator (and again, there is a whole list of them) playing an important role to make a "live" sound. After that it goes to a reverb imitating the acoustic properties of the sound recording studios chambers. The dimensions of the room and the reverb decay can be chosen from the list and fixed up as desired. Flanger, chorus, phaser, harmonizer, pitch-shifter, delay effects are also available. Then the signal goes to a virtual microphone of any type accessible in the list. It is also possible to regulate the location of the microphone in the virtual studio. After a tube microphone preamplifier the signal, finally, passes to BOSS GX700 output. And moreover, it all works in real time! I have not yet come upon a software analogue of such gear.

A standard music CD comprises audio signals recorded as a sequence of 16-bit numbers with 44.1 kHz sampling frequency. Theoretically, it allows recording and consequent playing back of the audio signals with frequency components from 0 up to 22.05 kHz. In practice, most modern DACs enable to play back the frequencies up to 18-19 kHz without significant distortions. Digital and analog interpolation filters influence on the higher frequencies (about 22 kHz). They put them down on 40 dB and more. It brings about lineal, non-lineal and inter-modulation distortions. Such cut off frequency choice (about 18-19 kHz) is stipulated with economical reasons. Complexity and therefore the price of a digital interpolation filter increase when cut off frequency value approaches the half of the sampling frequency value.

For a long time amid Hi-End fans and studio engineers it has been accepted to bitterly criticize 16-bit music CD sound quality. For all the faults of hi-fi amps, speakers and CD players it was digital format of audio signals to blame. But in the numerous articles appeared lately experts comparing equally expensive modern digital CD players and analogue players of "vinyl" often make choice in favor of banal 16-bit 44.1 kHz sound. So, let us leave fruitless attempts to prove the theorem by Nyquist wrong and try to see the recent development of Hi-End equipment. There is no notable progress in the scheme technique and element base of tube power amps but one exception. Ten years ago when I was keen on tube amps construction I could not even dream of an output transformer that would have passed frequency components from 10 Hz and up to 50 Hz. Now it comes true! Such a fantastic transformer "transforms" practically any tube power amp into a Hi-End gear (for example, Old Timer for $4700).

A CD player is gradually changing into something like a PC with CD-ROM and a sound card. For instance, the transport part of Audio Research CD2 CD player is based on computer CD-ROM hard ware. Crystal CS4329 micro scheme of 20-bit DAC is frequently used in the sound cards. Note that 16-bit audio signal processed with noise shaping, dithering algorithms and digital filters can be turned to a more bit signal. Besides more bits enable loudness control in digital domain without analog pre-amps. It also provides a sound quality improvement. Audio Research CD2 developers have paid much attention to a jitter problem. But initially they have created it themselves. Audio Research CD2 is connected to DAC with S/PDIF. Here it is the jitter problem to fight! And a skilled special processor does that. But it seems to me that it would be more reasonable to follow the ideology of PC structure and connect a CD transport component to DAC via IDE, SCSI, ISA, PCI or Fast Ethernet. It would totally eliminate jitter. It is an evident decision for computer hardware developers but revolution for Hi-End engineers! After all such CD players like Audio Research CD2 can boast with quite Hi-End sound quality. And all that - within the limits of 16-bit 44.1 kHz!

To draw the line I should say that your home PC can be easily turned to a full-featured professional recording studio capable of handling all aspects of the recording process, from tracking, editing and processing to mixing and mastering.