Digital edge, a thing of the past?

My CD source sounds stunning and there certainly isnt any digital edge. It is also has the most detail retrieval of any source I have heard, analogue or digital. Mind you to get this kind of performance from CD does not come cheaply

 

OK, I'm going to shock you Neil. Vinyl does some things better than even the Eclipse. I'm not saying it is a better format full stop, and contrary to Michael's comments, CD will remain my main source of music. But when it comes to realistic high frequencies, I have yet to hear a CD player that does it.

The upper harmonics of many instruments are going to be artificially truncated by any form of Nyquist filter. Whether you think this has an audible affect at such high frequencies is the can of worms that ain't worth opening.

 

But you see all of the event as defined by the sampling frequency. If you choose the sampling frequency correctly then you see all the event that you are capable of seeing. Similarly for CD. A bandwidth of 20kHz is available and you get everything within that bandwidth at a resolution and channel separation way better than vinyl.

Bringing up the old canard of what's between the samples is really just a demonstration of ignorance, it's like little old ladies not wanting to remove a plug from a mains socket lest the electricity escapes and runs around the room biting their ankles.

Paul

 

So Paul,

you reckon a 16 bit signal is every bit as good as the analogue master then?

Isn't that at odds with the opinions of most experts in this field who consider 20bit depth to be the minimum requirement for accurate reproduction of complex waveforms?

You seem to be sidetracked by the frequency issue, 44.1 is fine for the audible frequency range in theory. It's what the filter does above it that's the problem.

 

merlin,

I am sure you are right, I have not really heard that many all singing and dancing vinyl setups to make an accurate assessment. However, I am sure you concur with me in my view that the eclipse has some of the best treble and is one of the most detailed disc spinners you can lay your hands on.

 

Yep Totally

It gives you everything that's on the disc in as natural and lifelike fashion as I've heard (with the possible exception of the dCS, but that don't boogie) for a Red Book player.

It's just that to these ears, CD has never sounded totally convincing on certain things, and it took a reunion with vinyl to understand why. Swing and roundabouts, really, horses for courses, etc etc....

 

Well, today i had the fortunate experience of listening to the following vinyl & digital fronted setup (I'll leave it to you guys to decide if it's worthy or not as I'm not a vinyl afficianado .] :

Transrotor Fat Bob S.
Transrotor reference phone stage.
Linn Uni-Disk 1.1 CDP.
Ayre Cx-7 CDP.
Ayre k-5x Pre Amp.
Ayre v-5x Power Amp.
B&W 704 speakers.

All-in-all, not bad :MILD:. We listened to a "QSound" encoded pressing of Madonna's 'Immaculate Collection'. Quite frankly - f*ck me. Really scary stuff . Good scary though .

But........we were listening to some 180g pressing of Elvis :MILD: and I hadn't spotted that it was on vinyl and thought it was the Linn Uni-Disc (which we had been listening too previously). Granted, it sounded great but the leap from a half-decent digital end to a well set up vinyl front end was not IMHO totally jaw dropping.

Granted, different music etc. but my point is that switching between the two I wouldn't be able to tell you which was vinyl and which was digital !

I would comment that the vinyl was certainly 'warmer' and instruments such as Double Bass sounded that touch more convincing but overall there wasn't much in it. background noise hardly even came into it .

 

Not to my ears, or to the ears of those I respect! In my experience analogue high frequencies tend to sound 'real' and natural, low sampling rate digital like red book CD sounds 'false' and etched like there is actual information missing. If it sounds wrong then it is wrong, no science is going to convince me otherwise.

To go back to my daft train analogy, my feeling is that with the low bandwidth digital sampling of just twice the required frequency there is a chance of blinking at the same rate that the coaches of the train pass, therefore seeing the wrong information of a stationary train.

There has been plenty of evidence that humans are aware of high frequency signals way higher than 20Khz even though we can not accurately hear it, hence last years trend for super-tweeters (which are the worlds most stupid invention when combined with the curtailed and distorted top end of red book CD). A good record deck can easily produce output at 60-70Khz, so this may have something to do with the analogue mediums superiority at reproducing high frequencies.

Tony.

 

I didn't claim that. You reckon that a vinyl cut is every bit as good as an analogue master?

FWIW most CDs mastered from analogue tapes (and we'll require them to sound great too!) have clearly reproduced tape hiss. That's, IMO, empirical evidence that CD is 'good enough to reproduce an analogue master'.

Show me an expert opinion.

20 bits is theoretically better, especially in a recording environment, but often it's technically worse, especially in a marketing driven implementation.

Now you are an expert on what the 'problem' is?

Paul

 

The Nyquist criterion explicitly excludes this situation. The maths works, many implementations suck.

But the information isn't there to reproduce, at least not in general. A tape machine cannot go that high.

For instance http://www.otari.com/products/MTR15_spec.html even at 30ips it's 20kHz at +-1dB and can just reach 28kHz. The multitrack isn't even that good. So where's the extra information coming from?

I've never had a problem with CD since I started using it seriously, which would be as recently as the late 90s. OTOH I've been using a very good player since then (Linn Karik/Numerik). When I heard the Devil's system in Glasgow I thought his CDP showed some 'glare' or 'haze' on some tracks which I associate as 'digitus'. I never get this at home. Even old CDs from the mid 80s which don't sound great still don't really annoy. I wouldn't simply blame the CDP though, his system may be simply playing it like it is.

Paul

 

I like vinyl because I like having access to at least 50 years of recorded music on the cheap, I still enjoy listening to many of the records I've had for 25 years or so, and because I've had a turntable for about 30 years and it would be unusual not to have one. Do there need to be more reasons?

Oh, and records (usually) sound better than CDs, of course, for whatever reason, which is nice...

-- Ian

 

Well I had a tom evans eikos once, and even a humble vinyl sounded better on some things.

Its the ambience and tonal realism that you get. Certainly with things like pipe organs, classical guitars, and drums on rock tracks...black sabbath.

I don't know if its the frequency, I tend to think 20 khz is enough.

I have an audionote dac without digital filter, and that is marketed as the solution to digital. Its not as such, it is a little clearer, but its still identifiable a digital.

What I found interesting, is I think it is in the recording, I have heard some cds with old analogue recorders, and they do sound MUCH more like vinyl.

Francinstien, the cd device, did go a little way to giving a more defined central image due to crosstalk, and rolling off the treble.

 

Sorry to go back on-topic for a sec

Is there a sensible UK source for the AD8065's?

I agree they are fantastic, along with the AD8610's too, although horses for courses the 8610 is better in some apps, but none of the main UK disti's carry them ex-stock, AFAICT.

Also for those that don't read the data sheets ALL of the devices using the XFCB process are limited on operating voltage compared to the CB process used for most earlier chips.

The max recommended for the AD8065 is 24V, 26V for the AD8610 - this is total supply, not +/-. Check your supplies before dropping them straight into an existing player!

Also beware the low noise quoted too, these high speed amps have high noise corners, and hence higher noise at LF than some 'worse' specced devices.

As usual, there's no universal advice, but for a CDP output stage either will be marvellous. Both these devices are really excellent into cap loads, showing little degradation or distortion increase, so are good for driving cables.

The other issue here is so few players implement adequate filtering at the output, industry norm is 3rd order Bessel, and this only gives about 9bit accuracy, you need a well-designed 7th order to achieve anything like 20bit accuracy, assuming an 8x oversampling system.

Andy.

 

I'd agree with this, also for those that have used Naim for years the suspersonic signals will be attenuated anyway, by design.

I doubt the top end response has anything to do with the sound of vinyl, it's much more to do with resolution (bits) and the fact so many digital system implementations suck, as Paul says!

Andy.

 

Andy, Didn't get on with the 8610's, designed for audio as well , yes both of them have a lower voltage range, prosummably due to the much lower input noise values, On the Belcanto I had to drop the voltage regs down from +/-15v to +/-12v, altough I built a rig to destruction test a few, seems they hold up well
The supply of the newer AD stuff, is difficult but not impossible, PM me on this issue.
Changing the op-amps alone will not produce miricles, along with careful restructing of the analoge stages, very good results can be obtained .Tone