makes sense on the defender.
were you able to ab test the 3 dacs?
when i had the ares and the soekris, in my ab tests it was very difficult to tell a difference between two.
curious what makes the audio gd stand out most for you ? was it a night and day difference ?
A/B not and each time it was in different configuration. For comparison were available only Delta-Sigma DACs. To be clear, I came out from long time resisting CD players. PCM-63 implementation with Pacific Micronics filter was the first accepted with some reservation. Didn't look back after doing some DIY mods. When it was stolen, I thought I would be happy with current Delta-Sigma offers and made few mistakes not realizing of being R2R fan.
I found Soekris in the semi-private showroom. It was clearly superior against other offerings and I thought that it would be an ideal choice, as it had some reverbation on a decay, and a hint of a natural sound I like very much. However after 15 minutes I noticed digital ?traces? in the sound, missing a grass during fast transients. It is the same things that keeps me away from DS DACs I heard. A dominant tone takes over and a background dissappears for a moment. Ares had no such problem, it had smooth sound and completely fatigue free, we had a whole evening, chatting on unrelated matters... and drinking a beer. It has a similar character to my portable Nobsound 8xTDA1387 (which I also used on desktop), but a class or two above in refinement. Having engineering background, I picked up it was a passive output giving a difference.
I would definitely buy Ares if I had no occasion for a second hand R2R-11, received by mail order, no prior testing. A clear advantage over Ares is NOS. Not smooth, but a raw hard slam character, not limited by electronic buffer working in pure Class A with 0 interstage feedback. The I/V conversion is made with ACSS link, the same technology used in Audio GD upper level DACs. It is worth to mention that this hard slam is fully textured, it is full body, but it doesn't hide details, similar to a good tube amp, it should pair well with, BTW.
My English is poor, not to mention audiophile terminology, so please excuse. If you want to compare sound of Denafrips with Audio GD, the best description is
here. Both companies have consistent house of sound across models, a difference is just in refinement.
Ehh.. sounds like you're way in deeper than I am. Can you clarify at say difficulty level 2/10 please?
Not sure I will explain it clearly enough without pictures and tutorials. D/A conversion produce a required sound in the audio spectrum (0 to 20kHz in the case of CD) and also a repeating mirror images starting just above a half of the sampling rate. These images are fuly correlated with a sound, so when this high frequency content intermodulate on the downstream equipment, its products leak to the audio band, it is heard as a very unpleasant harshness.
CD NOS images start a 22.05 kHz, it is very difficult to filter out, as images have high energy very close to the audio band. Therefore there is a strict requirement on the amplifiers to have distortion-free range range at least 100kHz. In other words, a top notch downstream amp is required.
A second approach is to upsample few times, by example if 4 times, then the first image starts above 88.2 kHz. It is much easier to filter such unwanted energy, leads to less itermodulation and allows to use common integrated amplifiers with satifying results. If we apply the same audio amplification to the NOS DAC, results can be worse than with OS. A prime example is Soekris. It offered a firmware with NOS filter, I didn't hear any positive feedback amoung NOS fans.
A digital signal processing shape a signal before conversion to further limit unwanted spectrum. Problem with this approach is a limited power of the processor, leading to digital artefacts, timing inconsistency and untrue complex harmonics. Oversampling is also sensitive to the inter-sample overloads, poor recordings are clipped.