I had my tech perform a square wave reproduction -*test*-('") on my Luxman D-109 CD player. My tech used a Philips CD that had a number of -*test*-('") tones recorded on it so we could view the output on a scope. The -*test*-('") results were surprisingly poor. There was moderate to severe overshoot and ringing in the lower frequencies. As the frequency increased, the waveform lost definition so that it no longer resembled a square. No doubt, this output is amplified down the line to the speakers. I was very surprised as I always thought my D-109 to be a sweet sounding CD player. My tech mentioned that most CD players look like this because the sampling rate somehow interacts with the output. This is a fundamental weakness in the digital audio conversion process.

pure conjecture on my part as i dont really understand the process, but is this why transports and DAC combo's generally sound better than stand alone players?

Imo, it all depends of the type of the dac...
I have a friend who said that "Happiness is a good ‘scope".... He can actually see jitter in a `scope....
Also it can be very depressing. You think you have a good source with all the bla bla bla audio bs like tranpanracy, dynamics, soundstage, accuracy, and suddently you put a `scope on it and....
Thta's why i never go to the doctors. I feel sick all of a sudden...

Well said and very true. Testing my favorite CDP and seeing the flaws on the scope is very depressing but at the same time an opportunity to fine tune my main system. In all honesty, I always thought my D-109 was one of the better sounding CDP's in my collection but that throws into question what I am actually hearing AND what I should be hearing. With this discovery, I feel like I am back to square one with CDP's. This search is on!

The other question that comes up from this is CD the best format? My tech mentioned that vinyl does not have this problem because it is much simpler, i.e. no DAC. My tech has an LP with square waves recorded on it so eventually, I'll get him to show me on a scope.

I also heard the same thing about seeing jitter on the scope. I wonder if the jitter is the overshoot and ringing I saw?

Some people believe that -*test*-('")ing is the end all for sonic quality. I believe that some -*test*-('")s are important but the ultimate arbiter is how it sounds to me.

As an example of sorts my speakers sound good to me and the brand has lots of fans.
However, when you look at how they measure you'd think they were absolute junk. So I think measurements can be misleading when used exclusively.

When it comes to CD players I think there is a lot of bogus information out there. One that tickles me is the idea that a single disc player is inherently better sounding than a changer. I have a changer that uses the Nak stacker and it sonically bests my Sony 707ESD, and several higher end Denons I have and a Roksan Kandy MKIII (to add a newer machine to the list) the MCD-205 sounds better to me than the others by a wide margin.

Very true.
Imo, if designers build their audio components based only in graphics and -*test*-('")s in the end all sound the same and all sound bad.
Even tough, i find with relativily importance how a component behaves in some -*test*-('")s, and most important what to do to minimize some bad results, and when minimized what's the final result in terms of sound, of course.
So, socal sam keep us posted.

My tech mentioned that he built a circuit to address this issue. I asked if he could re-recreate this circuit and he said yes but it would be time consuming and expensive. I'll eventually have my other CDP's in for a square wave -*test*-('").

Test results can indicate better sound. IMO, there is a strong correlation between -*test*-('") results and sound. I've seen enough of my amps on the scope and then taken them home and heard them to strongly believe in the correlation. The other thing to consider is -*test*-('")s are used by the industry to compare against the competition and to improve there offerings. I'm satisfied the numerous engineers have found subjective -*test*-('")ing methods that yield useful results.

OTOH, our ears are our only sensual link to sound and so cannot be discounted. If you say your drug store receiver sounds great, I'll respect your enthusiasm but I will remain unconvinced.

see the stereophile review of the playstation 1. This was interesting. apparently if you smere things enough, you wind up with an audiophile cult classic.

I was reading again your post and i forgot to mention this:

Sampling is not the weakness in the digital audio conversion process. It is the fundamentals of the digital processing signal, which leave us to Nyquist-Shannon theorem, the basis for all digital sampling of analog signals.

In simple terms, the importance of the sampling theorem is that it allows analog signals, such as audio or video, to be reproduced digitally and then converted back to analog again later. One of the key requirements for accurate conversions (sampling) is the necessary Sample Frequency.

We at ST have an interesting topic about all the digital process here. Try look for "Digital Audio Basics" post.

The problem your tech is refering is in imo, the kind/type of filters used in this process. Some of filters used by some manufacturers traditionally sounded horrible.
No wonder, then, that the freqs top end sounded clangy and harsh and people said digital would never catch on. Improvements in filter design, specifically ones that used more gentle slopes, significantly improved the imaging and, consequently, the sound of digital audio.

A more complete solution would be to sample at higher rate, and thus the filters could be smoother and operate way above anything audible. Sounds good in theory, and this may be where the original push for higher sample rates came from.
Today, however, such filters are implemented digitally, and phase errors aren't such a problem. In addition, the type of converters used today multiply the effective sample rate internally. As a result of this oversampling technique, the filter rolloffs and frequencies can be kept well out of the way of the audio. So there's no need for higher sample rates. After all, there's nothing higher than around 20 kHz to record. Is there...?

Of course there is, but that is another issue....

Just quoting a friend at yesterday's dinner :

"We shouldn't let numbers define audio quality than we would let chemical analysis be the arbiter of fine wines."

Of course wines are made by chemical engineers, even though....

A more usefull -*test*-('") would be how well the CD Player plays a known musical signal, such as a Gong cymbal and compare the original sample to the reproduced . A perfect Square wave is never present in music, unless you have digitally clipped a recording, and even then it isn't perfect. If you want something that reproduces a square wave perfectly, get a synthesizer or signal generator.

One can also wonder how well Phillips actually did on getting an accurate Square wave onto the CD in the first place Basically the same process as playback reversed.

Ian

Well put, except wine isn't trying to reproduce wine

or would you buy a performance car based solely on 'measured' specs and numbers. If so 'the' high-end performance car would likely be a Vette.

Ian

Touché!