Current Mode Amplification has its roots in a paper written in 1975. It has been used in high frequency (microwave) broadcast and it has been used by extreme high-end amplifier companies such as Halcro and Krell to link between their preamps and amplifiers, but Questyle is the first to use it as a sole form of amplification, and they hold a patent on the design.
To understand the advantages of Current Mode Amplification, you need to understand a few terms; voltage, current, modulation and transient inter-modulation distortion (TIM or TIMD). Voltage and current are the two components that make up the amplitude of a signal (a Watt or Voltamp). Voltage is the pressure and current is the volume and are interchangeable in an electronic circuit. 10 amps Current at 1 volt of Voltage is equal to 10 volts of Voltage at 1 amp of Current. Modulation is the variation in signal amplitude that represents sound waves in an electronic circuit. TIMD is a distortion caused by the use of negative feedback to reduce Harmonic Distortion (THD) in an amplifier circuit when the feedback loop is too slow to accurately trace the signal. TIMD is the root cause of the sound difference between a Tube Amp and a Solid State Amp (it causes a metallic ringing). Oddly enough, the slew rate (the speed at which an amplifier is able to modulate a signal) in a tube amp, is actually too slow to produce TIMD.
Conventional amplifiers modulate voltage to amplify the signal (which is relatively slow) producing high levels of TIMD. Questyle’s Current Mode Amplifiers modulate current which is significantly faster (producing bandwidths in excess of 650 KHz), and negative feedback loops which are fast enough to not produce TIMD. The result is an amplifier with the musicality of tubes and the detail and authority of transistors.
The one drawback of Current Mode Amplification is that it does not deal well with changes in impedance. To correct this issue, the Questyle amplifier converts the signal back to voltage mode at the output (making the amplifier compatible with the 600 ohm Sennheisers as well as the 20 ohm Audeze headphones). To do this, they employ an Op Amp, it is important to stress that this Op Amp performs no signal amplification (the signal amplification circuit employs discrete transistors in a current mode configuration and is Class A).
There is one other company making a Current Mode headphone amplifier, Bakoon, other than being significantly more expensive than the Questyle, they differ in that they do not employ negative feedback (oddly enough, as it is the whole point of Current Mode Amplification) producing high levels of THD, and they do not convert back to voltage mode, limiting headphone compatibility. The Bakoon amplifier is also battery powered (only) severely limiting playback time.
Direct Stream Digital (DSD) is a different form of digital recording from conventional PCM (Pulse Code Modulation). PCM like most digital data is composed of a series of data packages called “words”. The size of these words (Bit Depth) determines the dynamic range (96db for 16bit, 144db for 24bit), and the frequency response is determined by the sample rate. The typical bit depth/sample rates of PCM are 16/44.1 KHz, 18/48 KHz, 24/96 KHz, 24/192 KHz, 24/352.8 KHz (DXD) and 24/384 KHz. Using PCM nomenclature DSD is 1/2.8224 MHz as sampling rate 64 times that of a standard CD (16/44.1 KHz).
Until recently, DSD was confined to SACD which is heavily copy protected. DSD is now available for download (though it is still not possible to “rip” an SACD directly to your computer) for playback on a computer or DSD music server.
There are three ways to output DSD from a computer over USB to an external DAC. The most common is to convert the file from DSD to PCM and output it in the normal way as a PCM file. The problem with this method is that it combines the deficits of both systems while removing the advantages of both systems (this ironically applies also to converting PCM to DSD).
The second most common method is called DoP (DSD over PCM), this involves breaking up the DSD bitstream into 16 bit words compressing them into 24 bit PCM words, sending them over USB or SPDIF, decompressing them into a buffer and rebuilding the DSD bit-stream at the other end to send to the DSD DAC. This entails a lot of processing overhead as it has to happen in real-time. The weakest link in this system is the computer, USB and the associated cables. DSD requires a lot of bandwidth, this means full spec USB 2.0 (Win7 or newer) or USB 3.0, and the DoP stream requires a 3rd more bandwidth. A dedicated (running nothing else) state of the art computer should be able to do this, but a budget or older computer that is being used for other applications may not. Not to mention that most DACs will not have sufficient buffers to contain more than a few seconds, so there is likely to be overruns and drops causing sonic degradation.
The third way to output DSD, is as a True Bit-stream over USB. This requires special drivers and software (co-developed by Questyle). This is what they call “True DSD”, and is what they consider the proper way to process DSD