Oversampling DAC Vs. Non Oversampling?


What are the advantages/disadvantages of oversampling vs. non oversampling DACS? I have a Squeezebox 3 and I love it, but I want to buy a DAC and I am not sure which to buy. I listen to Internet radio on the Squeezebox most of the time, which is usually around 128 KBPS and sounds good. However, I want to use the DAC for two-channel home theatre as well, and I will route the DVD through the DAC via a digital cable to listen to music from time to time. What to do? Thanks!!
regafan_1972

Showing 3 responses by kijanki

Matt - "forward and aggressive" is an issue of system matching. Your Dali speakers would not like SS amp with bright DAC. My older Paradigm speakers were a little forward with class D amp and Benchmark DAC but new Hyperion 938 speakers sound laid back and relaxed (but not warm) delivering even more detail.

There is an important difference between oversampling and upsampling DACs. Upsampling DACs do asynchronous resampling suppressing jitter (noise in time domain) in the process. I like sound of my upsampling Benchmark DAC1 and get additional benefit of working equally well with any transport (even cheapest DVD player or Airport Express)or digital cable. Benchmark is very revealing and neutral sounding - not for people who look for warm sound.

Matt - post your experience with WEISS DAC2. I'm interested since it might be improvement to my Benchmark.
Matt - I'm sorry to hear it didn't work for you. No excitement I can understand, as some DACs like Benchmark sound less exciting (or too clean), but poor soundstaging is disturbing (for expensive DAC). Benchmark brings tons of low level details and according to review decent soundstaging but Audionote imaging is a little deeper. I'm working on taming my room acoustics that, as you might remember, is far from being perfect. I bought twelve 2'x4'x2" sound absorbing panels and plan to buy 6 more. I was in a demo room where all walls were made of this material and even sound of voice was incredibly pure and directional.

So far Hyperion speakers were the greatest improvement to my system and brought musicality and warmth to neutral (unforgiving) Benchmark + Rowland 102. I've noticed that new generation of DACs use FPGA (programmable logic) ICs to execute filter algorithm. Non-apodizing filter in new Merdidian CDP, with different step response, suggests possible improvements in filtering (ovesampling or upsampling schemes). The main purpose of executing filters in FPGAs is speed. FPGA can do everything required in one cycle (clock tick) that uP would have to do in many machine cycles.