This review page is supported in part by the sponsors whose ad banners are displayed below

To address our collective audiophile mind, I first had questions: What can you tell us about the R&D leading up to this launch? This was a complex ambitious project that much went into. Relative to digital volume, could you explain what exactly happens with your 1.5MHz upsampler and 32-bit bit-depth expansion? How does this eliminate resolution decimation, bit stripping and compression of dynamic range at high attenuation rates? How much attenuation is possible before resolution suffers? Relative to DSD, does Sabre have a DSD-direct mode without any further processing like the Cirrus Logic 4385/4398 and Wolfson 8741/8742 chips? There's much confusion about what exactly native DSD playback implies when pure 1-bit chips are no longer made except for the Xilinx FPGA coded by Meitner and Playback Design to operate as 1-bit converters. The better delta-sigma chips today are 2.5-bit '5-level' designs. How does this relate to native DSD streaming? Relative to amp-direct drive, what is the output impedance on RCA/XLR, slew rate, current delivery and max voltage-swing potential?



"The ARK MX+ was our first export product based on the domestic ARK MX model already selling in mainland China. In 2009 USB DACs were still new to most audiophiles here. Our market was dominated by Firewire products from Weiss. The only asynchronous USB DAC Chinese customers knew of then was the very expensive Ayre QB9. Into this scenario we launched our original ARK MX as an affordable 32-bit/192kHz asynchronous USB converter. It enjoyed great local success but the shortage of digital inputs did limit sales. We thus summarized the experience gained during its design, improved the circuit, added digital inputs and had the ARK MX+. As a product it still focused on USB. It thus avoided an input switch since we felt most people would use USB. It also lacked a sample rate indicator because we thought people would see that in our driver panel. But we did know that the functionality of the ARK MX+ was still too limited for today's high-end systems.

"The first prototype of the Vega almost coincided with the ARK MX+ release but still was very different from what we have today. Too many things changed in the intervening two years. We had to revise the design several times, a main reason being the new DSD-over-USB standard. The first Vega prototype used the AKM AK4399 with our Sanctuary interface, Orfeo class A modules, customized digital filter and present inputs. Between first prototype and final production we had altogether nine different versions. During this R&D evolution we changed to the ES9018 to support DXD and DSD128; and adopted XMOS for USB even though we still use the Sanctuary chip for our ActiveUSB buffer stage. XMOS has better compatibility with various computer operating systems. We also added the Femto master clock and beautiful OLED display.


"The DAC's power supply was revised several times to meet true bit-resolution requirements. The most difficult part during this development was figuring out how to best harness the ES9018. Its datasheet is an error-riddled nightmare with lots of vital information missing. We would like to openly thank everyone on diyaudio.com for their help on how to wring the most performance from this freaky chip. Another big event during development was the added support for DSD. It's actually no big deal to modify software code to support the format but hardware—especially analog circuitry—had to be completely redesigned to suit DSD's sonic character. From that perspective I very much doubt that any DAC which adds subsequent DSD support via a basic firmware upgrade will offer comparable sound. The Vega was designed from the ground up to optimize DSD playback.


"By using the Femto master clock, we also added clock options, a very important function. As the machine warms up, we can enable the additional 'fine' and 'exact' levels for clocking precision in the menu. This can create huge sonic improvements particularly over USB. Over the past few months we upgraded our internal firmware several times to create the perfect balance between secure data lock and best sound quality. With retail firmware version 1.3 and higher it allows most modern computers to run in exact clock mode without any dropouts. We thus consider 'exact mode' the Vega's true advantage though most people already were shocked hearing it in 'auto mode'.