[aurelius]

Hi Wayne & others,

Please don't take this the wrong way... I am merely trying to justify expense, and I'm much more comfortable if I understand how the benefit works...

Specifically, I don't understand how a better power supply for a "transport" digital output section could possibly improve the sound downstrem.  My understanding of digital is that we are trying to resolve 0's and 1's... either we do an adequate job and the result is excellent, or we don't and we introduce random bit errors that make the whole system sound shyte.  The notion that we can retrieve more detail (as one might with a better analog supply) just doesn't wash with my theoretical understanding.

(BTW, I do understand jitter and its affects).

Any understanding you could provide to enlighten me would be much appreciated.

If, however, you do what you do by educated trial and error and don't understand why your products work as they do, then you are not the last (in fact, my system is dominated by equipment following this paradigm - AKSA amps & preamps, Peter Daniel DAC etc.)

Regards,

Mark

[Wayne1]

There are no such thing as 1s and 0s in digital. This is a very common misconception.

What you have are square waves (hopefully) that change between low (theoretically 0 volts) and high (3.3 or 5 volts). How the processor translates this information will depend on how quickly the voltage changes from low to high. If there is overshoot, ringing or noise on the low state. There are many things that can affect the digital signal and I hope you can now see that what the digital signal is actually the power supply being changed from low to high by the associated circuitry.

The "better" the power supply (less noise, quick recovery time, etc) the better the end result will be.

This is the straight, simple engineering part. There is also the "art". I have found, by listening, that different capacitors used in a regulated power supply produce a different "sound" from the Squeezebox. Some are "dry and analytical" some are "full bodied and fat". While the basic circuit for a power supply can be designed and measured for low noise, the actual final result needs to be listened to.

The standard switching supply that comes stock with the SB does tend to have a very"harsh" sound to it.  Most of the inexpensive linear supplies being tried in place are much "smoother".

[eg]

Hi Wayne

Also to this point, what sort of audio equipment must I be hooking a "modded" SB3 such as yours to in order to get any benefit from it? Which is to say that I doubt my wife's crummy Technics system would be a good candidate.

Curious,

Colin

[Wayne1]

Hello Colin,

I can hear the benefit and changes of different levels of mods and power supplies in my bedroom system. It consists of a pair of Onix X-LS "mains" and X-Sub. A total cost of $400.00 for speakers. A modded SB2 is plugged into the digital input of a stock Panasonic SA-XR55. That receiver can be purchased for about $230.00.

You do not need to spend a lot of money to get VERY good sound. The improvents to the SB will come through even more so on more revealing gear.

[CornellAlum]

Wayne,

Are you saying every argument I have ever read in regards to digital that contains 1 and 0's like all of them do, is essentially, BS?  If so, where does the whole 1/0 stuff originate in digital speak if you will.

Daryan

[kfr01]

Wayne,

Are you saying every argument I have ever read in regards to digital that contains 1 and 0's like all of them do, is essentially, BS?  If so, where does the whole 1/0 stuff originate in digital speak if you will.

Daryan

http://en.wikipedia.org/wiki/Digital

0 and 1 are just two characters to represent two different binary states.\

Historical links regarding digital systems are included in the Wiki.

[brj]

In discussions about virtually any digital device, references to 1s and 0s correspond to two different voltage states.  In theory, you can transition between them perfectly and instantly, but in real life, it is difficult to avoid overshooting or undershooting the target voltages.  In addition, the timing that dictates when to transition between the two states can be affected by all sorts of outside influences.  All of these effects and others typcially show up in discussions regarding jitter.

You might want to browse some of Steve Nugent's technical papers on his website, as well as some of his interviews on Positive Feedback Online.  This article, in particular, is especially relevant.