Absolute top tier DAC for standard res Redbook CD

http://www.computeraudiophile.com/section/ca-academy-2/

at this site there are several articles on Ethernet input and advantages/disadvantages compared to a music server.

I have on loan a unit from Waversa from South Korea which has both inputs (USB and Ethernet). Speaking with two friends over there, they believe uniformly Asian audiophile market has moved away from USB for the reasons Steve Nugent mentioned previously on this thread. In essence I found the Renderer (Ethernet input) is a magnitude better than USB.

Setup is more complicated and playback options are somewhat limited. This is the downside I found so far. Maybe others could chime in here.

The renderer (dac with ethernet) has an IP address. You use a remote that can be a portable device app or another computer. The remote only tells what files are sent from your source (HD or NAS or Internet) to your DAC. Some dacs have volume control, like this one. However, if used, the DSD over PCM will no longer work as the bit length is altered and DSD signal cant be determined in the PCM 176 sample rate.

With the IP address, you (or the manufacturer) are able to make processing adjustments remotely, alter other settings or implement firmware upgrades.

Wish I understood more of this. Think I'll be sticking with USB servers for the OSDE as Steve said the USB is the best input on his device. So many choices, lol...which is good.

I am not sure that Dac with Ethernet port is a good idea if the barrier from the crud that packetized data like Ethernet (and USB) brings gets thru.

The idea behind the SSR I mentioned above is to stop that cold and get to spdif (OK) or better yet i2S. The Transport has been the neglected stepchild for far too long.

Please read the post by Barrows I linked above very carefully to understand the KEY point I am making.

Winson, I learned from your post, A very good point I must say.

The problem with USB is two-fold:

1) common-mode noise on the cable that is not completely rejected by the USB receiver

2) poor software on the server device

#1 can be mitigated with filters or galvanic isolation. I have done both.

#2 can be mitigated by using a Linux-based server properly configured with good playback software and CODECs

Short of doing both of these, you will have some sonic issues with USB

If you do both, it will be identical in performance to Ethernet renderer, assuming both are implemented similarly. Implementation is everything.

Steve N.
Empirical Audio