This Audio-GD R-8HE, with its extremely careful design, takes up the key elements of its predecessor, the R-7HE, whose audio quality it inherits, while seeing its performance revised upwards and its price reduced. In the R-8HE, there are 4 Ladder DAC modules, this time in version 2, as well as the regenerative power supply. These two elements, at the heart of the device, allow it to offer great transparency, high dynamics and incredible sound richness, with a reproduction close to what can be offered by analog.

The R-8HE also includes surface components, making it possible to reduce the price of the device without degrading its audio quality, and therefore making it more accessible to buy a DAC with a regenerative power supply.

In addition, the R-8HE has completely separate power supplies. The digital circuit and the analog left and right channels each have a transformer. The signal circuit and transformers are separated by 5mm thick Aluminium plates, ensuring a high level of insulation. This improves SNR, isolates harmonic interference and reduces noise. The left and right channels are placed completely symmetrically so as to maintain the distance travelled by the cables equal on each side. The R2R DA conversion modules are placed between two Aluminium plates to avoid RF interference.

Seamless components and connectivity

Like many Audio-GD products, the R-8HE includes an Amanero USB interface, supporting PCM resolutions up to 32bit 384kHz and up to DSD 512. The device has 5 inputs: Coaxial, Optical, HDMI I2S, USB and XLR. Each input has interfaces adapted to provide the highest sound quality. On the output side, the R-8HE offers XLR, RCA and ACSS.

R2R 4x DA-7 V2 Design

The R-8HE, like its big brother, is a DAC R2R, a design that has become popular thanks to its undeniable sound qualities, especially when integrated into high-end products. While there are several methods for designing a DAC R2R that impact both performance and design complexity, Audio-GD continues to choose the highest quality as usual.

The design chosen by the brand switches ultra-low tolerance resistors in parallel and an ultra-fast FPGA chip controls and corrects the Ladder R2R. This design corrects each bit to achieve incredible performance. In addition, with this parallel structure, a single clock cycle allows all data to be processed, where a series connection would require 8 to 24. Audio-GD also corrects imperfections in basic R2R assemblies, i. e. those caused by resistance tolerance, through the use of the integrated FPGA.

Main advantages of the DAC R2R Ladder discrete:

• The R2R does not convert the clock signal into the output signal.
• R2R is not sensitive to jitter, unlike Delta-Sigma.
• R2R provides a more accurate output signal than Delta-Sigma.

Ultra fast signal processing

An FPGA chip is a programmable component that is now regularly found in high-end DACs. It allows total control of the internal hardware via a complex software layer. A huge advantage of using an FPGA is that its software layer can be easily updated to add new features or improve the overall performance of the device. This allows the device to be upgraded without necessarily having to replace components. It is both a high-performance and flexible solution.

In R-8HE, in this case, the FPGA is responsible for :

• Excellent performance of the SPDIF interface, replacing less efficient traditional interfaces such as DIR9001, WM8805 or AK411X chips.
• Total re-clocking process with FIFO design applicable to all inputs. In this way, the output data is always fully synchronized with the clock signal, thus limiting jitter.
• The integration of 2X, 4X and 8X oversampling filters, digital filters and 4 different NOS modes.

Discrete symmetrical analog output stage

After the digital-to-analog conversion by the dedicated R2R modules, the analog signal is transported to the discrete symmetrical stages using paired discrete transistors. The analog sections have no surface components, only discrete components. The output buffers are asymmetric FETs. Two parallel stages provide very low impedance and all output stages have a pure Class A design so that there is no non-feedback. From a sound point of view, this translates into an absolutely disconcerting neutrality. Four AOPs work like a DC Servo, so that no coupling capacitors are required and the DC output is automatically polarized. The output stage also has ACSS outputs for even greater detail and definition. Finally, the circuit does not integrate any relays or switches after the digital/analog conversion modules, in order to maintain the purest possible reproduction.