explain bits.

Ezmeralda11,

I have a background more in digital imaging than in digital audio, but here goes.....

I believe a 20 bit or higher converter on a 16 bit source alone, all other factors aside, would result in a smoother presentation upon output of the original digital signal available at the source. It does not add any new information, it just does a better job presenting the information that is available in the original 16 bits.

This helps alleviate the "graininess" issue that many audio enthusiasts commonly associate otherwise with digital CDs.

If this makes any sense to anyone......

Hueske,

Nice explanation! Thank you.

go to yahoo, type in hifi and 3 or 4 hits down should be a link something like "hifi on the www" and it'll be a page full of links and down in the list should be an article on how the cd player works. You questions all focus on the Digital to analog convertors, obviously all the shiny discs have "16 bit origin." To address your fascination with one-bit, they do exist. They aren't great. The impression I get having talked to some more knowledgeable on the subject is they are quick and dirty engineering to get good upfront specs. I know very little about digital and couldn't tell you anything more, I wouldn't necessarily base a buying decision on the type of convertor, alot depends on how well that type is carried out (although i suspect alot of it is simply for market fetishes). hayenga.com has a massive links page that may have a good article or two for you. But when it comes to digital that's probably the one area where most people really don't understand a thing on (me included), I don't understand the point in a 20 bit (I've even seen 45 bit) convertors on a 16 bit source, much less the upsampling techniques and what not. I've noticed other threads of a similar nature never get much response.

A sixteen bit CD player is a standard DAC. A DAC works to convert digital to analog by inputing the binary word (in this case 16 bits long) into a resistor ladder, which converts the information into a voltage. The longer the word length, the more information that can be extracted. There are 2^16 possible inputs with a 16 bit DAC. The higher bit DAC "upsample" the input from the transport, because all CD's (well, most) are encoded in 16 bit format. The upsampling adds high frequency air, and can sometimes reduce transport induced jitter. A 20 bit DAC, for example, has 2^20 possible variations in input, which smooths the waveform (think of having to chop a sine wave into little rectancular pieces vertically...the smaller the chop, i.e. the more bits, the smoother the reconstruction would be). Dual DACs are generally used for balanced operation, that is, the left and right channels are seperately decoded.

A one-bit DAC is a different animal. These are more correctly called noise-shaping, Delta Sigma, or Upsampling DACs. One bit DACs don't use a resistor ladder (which has different resistors for different bit weightings)...they simply have two states: on and off. Ladder resistors have to be VERY accurate in order to precicely recreate the analog signal (which is why good DACs cost so much). A one-bit DAC doesn't have this problem. But, one-bit DACs have to operate at a MUCH higher frequency in order to produce the analog signal in real time. Consequently, the clock is much more important in a one-bit DAC. Older one-bit DACs tend to lack bass response, because there is a lot of stuff going on in order to convert the bits to a voltage (it's quite complicated, and this is going on too long already). But nowadays, they are very very good. IMHO, a one bit DAC can perform as good (or better) than most ladder DACs on the market. But just listen for yourself. Execution is key in the digital game, not numebrs. Don't judge the purchase of a CDP or DAC based on its conversion alone. Listen to a good batch from both crops (also compare 16 bit to 20 bit and beyond). It's makes for an interesting listening experience. Hope I haven't confused myself! Cheers!!!