Why is the nDAC so cheap?

quote:

Originally posted by pcstockton:
ok,ok,ok..... just a second. To address the OP's Thread title..... $3500 is "cheap"?

Get some perspective. I am guessing .001% of the world can afford that and far fewer would spend it on a DAC.

Hi

I understand the O/P's statement from my own perspective.

I have been in the Industry since Compact Disc was launced and have heard and sold many prestigious brands of CD players and DAC's.

One time we were asked to show round a girl who worked at the accounts dept of Radfords in Bristol.She was visiting her parents in London and popped in to look round Windsor on the way.
She had no interest in Hi-Fi whatsoever and I played her a Linn Karik with and without a Numerik D/A which cost £1500.00 twenty years ago.She "got it" 2 straight away as the musical timing was superior as well as being a little more transparent.
What shocked her though was that the £7000.00 machine next to the Linn which was on loan and highly regarded in it's day made Eric Clapton's foot tap out of time with his guitar.
I was impressed as she had no prior knowlege or "baggage" about reputation or price and saw the Linn as good value at half of the price of the competitor.

We have only just received our demonstration DAC and I did the first demonstration last week.

I am not easily impressed but I was stunned by the performance I heard.The demonstration was of CDX2/2 with and without the Naim DAC.

Vastly superior musical timing and phrasing with gobsmacking transparency which made the CDX2 on i't own sound very confused,thick and muddy.The discs chosen were quite random as we had only finished redecorations most of the discs were still packed.We used some Live Tommy Emmanuel,Erica Badou and a clarinet quintet.

I have very rarely heard such an elevation in performance in comparing Hi-Fi equipment and even the best DAC's I have ever heard and sold have made a difference that can only be described as minimal compared to the demonstration last week.In real terms the Naim is VASTLY superior to the Numerik and also VASTLY cheaper as £1500.00 twenty years ago is going to be alot more that £2000.00 today.

As an analogy imagine four "£9000.00" cars.(In the UK this is going to be the "cheap" end of the market and would be a very small engined basic runabout of 1000cc capacity).
After driving all four and maybe having a small preference for one you are next asked to drive a 200BHP Golf GTI.All well and good and you should be suitably impressed by the build,space,refinement and performance.Enormous fun,but of course the Golf is £25,000.
But what if the Golf wasn't £25,000 and was "12,000 instead?.

I know many people would never, ever be able to spend even £9000.00 on a new car but I would call Brand New Golf GTI at £12,000 cheap at the price and a "Bargain".

Regards

Colin Macey

quote:

Originally posted by Rockingdoc:
Well, I have said the same for many years, playing my TV sound through a good stereo system, but I'm afraid that the arrival of HD and Blu-Ray has changed things for me. Having tried it, I now believe that a centre channel is neccessary as a minimum. In fact, I have just gone for a complete 5.1 package, not for dramatic sound effects, but because it makes general viewing far less fatiguing than listening to the sound converted to stereo and played through speakers either side of the screen.
I have moved my music system to the other end of the room, although I'm sure the presence of the surround speakers will be polluting the music with their passive movements when the hifi is on. Didn't Linn used to claim that a telephone in the room could destroy the tunes?

You don't need a centre speaker with side speakers like these - trust me



PS. manky table now replaced with sexy glass designer one Really should update the photo - perhaps when I've tidied up all the cabling when the htpc is finished (waiting for some fan controllers to get the quiet fans in there inaudible

quote:

Originally posted by Andy S:
PS. Downloaded some 44.1/16 and 96/24 stuff yesterday. Have a difficulty telling the two apart, but could tell the difference - just.

Tried the two samples USB vs optical through the HTPC and couldn't tell any difference. I'm coming to the conclusion the DAC is pretty much source independant for me. Now some possible thoughts come to mind:

- Placebo effect (I'm expecting the sources to sound the same so they do)
- There is no difference between sources (my current thought)
- My system has so little resolving power to make all differences negligible
- My hearing is so bad to not notice any.

I'm kinda discounting the last two options as changing the cables (from a £4 RCA->DIN cheap job to Naim SNAIC) did make a difference - but not in the way I expected it to (which leads me to discount the Placebo effect too). The difference was immediate and noticeable within the first few seconds of the song.

Andy,

You will find any DAC source dependent.

The difference between 16/44 and 24/96 can be modest if the mastering has been done correctly. Music on CD at 16/44 that is mastered and produced properly sounds fantastic, the fact that it seldom is done properly is an entirely different matter of course.

I can't get excited about samples of great mastering in Hi Res, because I think there is little chance of the owners of mainstream labels providing it. A few samples of the usual lobby muzak or occasional jazz/classic piece in Hi Res for demonstrating what is possible, doesn't make for much of a music collection.

Joe

quote:

Originally posted by Joe Bibb:
You will find any DAC source dependent.

You may for any DAC, I haven't yet for the Naim DAC (for the same source material ).

Of course, it could be that all my sources are equally rubbish and I'll see the light once I've been shown the error of my ways. It's not looking good for the source makes a difference brigade so far, but we'll see - I'm open minded enough to try it

quote:

James

Would the quality/accuracy of the data leaving the IC be dependent on the quality of the power supply feeding pin 14, given the use of a decoupling and bypass cap.

This circuit was used in a mid priced CD player. Presumably a high end player would use an expensive/better method of providing current. Perhaps this would have an effect on sound quality

As Andy says, it depends on the context of where its being used. The way they've done it is quite normal. There are usually 4 gates per package so the 4th is being used elsewhere and its an easy way of minimising the number of devices used. The caps - pretty standard close to the supply pin decoupling and yes a noisy suuply here will affect the output. On a high end player / dedicated transport more care would be taken (seperate regulated 'clean' supply to the ouput driver etc. They have gone to the trouble of fitting a pulse transformer to isolate the ground of the player from the DAC it will be connected to which is good. I'm sure moderation may come soon so perhaps PFM might be a better place

James

quote:

Originally posted by james n:

quote:

James

Would the quality/accuracy of the data leaving the IC be dependent on the quality of the power supply feeding pin 14, given the use of a decoupling and bypass cap.

This circuit was used in a mid priced CD player. Presumably a high end player would use an expensive/better method of providing current. Perhaps this would have an effect on sound quality

As Andy says, it depends on the context of where its being used. The way they've done it is quite normal. There are usually 4 gates per package so the 4th is being used elsewhere and its an easy way of minimising the number of devices used. The caps - pretty standard close to the supply pin decoupling and yes a noisy suuply here will affect the output. On a high end player / dedicated transport more care would be taken (seperate regulated 'clean' supply to the ouput driver etc. They have gone to the trouble of fitting a pulse transformer to isolate the ground of the player from the DAC it will be connected to which is good. I'm sure moderation may come soon so perhaps PFM might be a better place

James

James

It was a rhetorical question. I’m not asking for advice. I know very little about electronic, but I know the importance of regulating power supplies. In fact it’s probably true that the performance of Naim equipment is proportional to the number of local regulators used.

I’m trying to make a point to the people who believe, wrongly, that all CD player transports digital outputs are equal.

The circuit is the digital output of a CD player. The S/PDIF data enters the circuit at the bottom left. The only actual output is the Coax.

The circuit is identical to that used in the Rotel 855 Richard tested with the Naim DAC. Apart from addition of the 22N cap.
It seems logical that one reason he found the Meridian superior to the Rotel was due to local regulation.

quote:

Originally posted by fatcat:
I’m trying to make a point to the people who believe, wrongly, that all CD player transports digital outputs are equal.

Has anyone said that on this thread I'm not aware of it - and I'm not saying that's the case either. What I am saying is that the technology in use in the nDAC ensures that it doesn't matter how good the output circuit is (as long as it's functional) it will completely reclock the data once received. As SPDIF is lossless (despite other people on this thread claiming otherwise) the logical conclusion is that the choice of transport should have no influence at all (assuming it is capable of reading the disc correctly) as to the sound coming out of the back of the DAC.

Given the tests I've done so far, to my ears, that is what I'm finding.

You could look at this another way. Naim have a brand "sound" - and they protect that sound by ensuring it's very difficult to tap into the replay chain to alter it. You don't hear of many people running 3rd party pre's into Naim power amps for example and they haven't put SPDIF output on their CD players so that the output can't be "improved" upon (it also locks people in to an upgrade path too - cynical - me ).

Perhaps they haven't built a high end DAC until now because they couldn't guarantee the front end put on it and that would alter the sound coming out of the back. The technology in use in the nDAC should mean you can put "any old source" onto it and you will "Naimify" it. Why would you put a 2 channel DAC with so many inputs on otherwise?

Just my brain doing overdrive...

quote:

It was a rhetorical question. I’m not asking for advice. I know very little about electronic, but I know the importance of regulating power supplies. In fact it’s probably true that the performance of Naim equipment is proportional to the number of local regulators used.

Ah ok. Looking at it that way, there is far more going on before that point to affect things before the final output.

A quick read of the Hi-Fi news review of the nDAC's technical measurements showed as state of the art Jitter reduction performance of just 15psec. So ...

Either Andy is right (and i'm not disputing he isn't we all perceive things differently)

There are differences heard between different sources (as a lot of users have found)

There is a subtle interaction between the source and the nDAC which affects the resultant sound (see my earlier pontifications)

Work calls...

James

quote:

Originally posted by Andy S:
You don't need a centre speaker with side speakers like these - trust me

I'm just envious. The DBL is my all time favourite speaker.
Don't the enormous magnets bend your picture, or does that only happen with cathode-ray tubes?

quote:

Originally posted by Rockingdoc:
I'm just envious. The DBL is my all time favourite speaker.
Don't the enormous magnets bend your picture, or does that only happen with cathode-ray tubes?

Had those nearly 11 years now. Knew I HAD to have a pair one day after I heard one of my first Naim dems at the Bristol hi-fi show in the mid '90s.

The bending only happens on CRTs, but they never did when they were either side of my old TV. To put some scale on that picture - the TV is a 50" The speakers are walnut as is the stand. The stand was new when that photo was taken - it's now dulling to a nice orange colour (a couple of years down the line).

I have to say, they are immense speakers and I'd only ever be parted from them if I desperately needed the money. Once you have DBLs, you don't think about changing them - only putting better electronics in front of them and, IMHO, they have been transformed with the nDAC into something special. My partner who spent 13 years in the audio industry as a singer and dealer of studio equipment - she has heard seriously high end stuff and is used to listening to master tapes - thought it sounded so good we spent the whole night rediscovering our music collection.

quote:

Originally posted by Andy S:
What I am saying is that the technology in use in the nDAC ensures that it doesn't matter how good the output circuit is (as long as it's functional) it will completely reclock the data once received. As SPDIF is lossless (despite other people on this thread claiming otherwise) the logical conclusion is that the choice of transport should have no influence at all (assuming it is capable of reading the disc correctly) as to the sound coming out of the back of the DAC.

Given the tests I've done so far, to my ears, that is what I'm finding.

So if the CDP has a crazy noisey power supply that is transmitted in various ways through the SPDIF cable to the DAC, does the re-clocking remove all unwanted PS noise, ground loop noise, RF picked up along the way, microphony (SP????) and other what not?

I get the rejection of jitter and all that, but is it unreasonable with all of the other possible factors that there could be slight SQ differences.

-patrick

Patrick,

from what I deduced from a discussion with the digital engineers at Naim, this is key, and can only really be dealt with at source (i.e. minimising noise as much as possible in the transport/headunit/streamer what-have-you), otherwise the noise just get modulated into the digital signal and then demodulated at the other end.

quote:

Originally posted by Richard Dane:
Patrick,

from what I deduced from a discussion with the digital engineers at Naim, this is key, and can only really be dealt with at source (i.e. minimising noise as much as possible in the transport/headunit/streamer what-have-you), otherwise the noise just get modulated into the digital signal and then demodulated at the other end.

Richard,

When did you talk to the engineers at Naim? Was it specifically how the nDAC works or how SPDIF in general works? The nDAC is a new design that shouldn't have these issues through DESIGN. Older designs would have had these issues.

Richard,

Thats what I am gathering.... It is very nice to know that the Naim DAC does narrow the gap between sources and can provide great pleasure even through an inexpensive source.

Also it is good to know it can be improved on as desired.

In a way the Naim DAC could be said to be source independent in that it provides great replay and wonderful VFM (NOT cheap), from almost any source. This need not entail that all sources sound identical.

In doing my own A/Bs I must admit I dont hear any significant differences between sources. BUt I am chalking that up one or a combination of three things.

1) all of my sources may be on about the same level

2) My 102 isn't capable of realizing the subtleties on a noticeable level.

3) My sources are all actually just fine (could be better of course), but good enough that I cannot hear a dramatic improvement through the USB input.

I am guessing that if the USB input is one of the best ways to play, and I cannot hear a difference between my Foobar>ASIO>Transit>Optichord>DAC and a USB stick, then my PC setup must be doing something correctly.

I would like to hear something that makes me re-evaluate my process and gear. But just because I haven't yet does not lead me to believe all sources sound the same.

-Patrick

quote:

Originally posted by pcstockton:
So if the CDP has a crazy noisey power supply that is transmitted in various ways through the SPDIF cable to the DAC, does the re-clocking remove all unwanted PS noise, ground loop noise, RF picked up along the way, microphony (SP????) and other what not?

I get the rejection of jitter and all that, but is it unreasonable with all of the other possible factors that there could be slight SQ differences.

-patrick

The thing you are missing is that in a CD transport->SPDIF->DAC system, the noise is transmitted as JITTER. The worse the design of the transport, the more jitter it will have. Basically, higher jitter translates into worse sound reproduction if it gets added into the DAC clocking circuits (all else being equal, and assuming the player can read the CD accurately).

Remove the jitter and you remove the noise that comes from the transport. Simples.

This is one reason why Naim recommend optical with the nDAC - you remove ANY electrical (apart from back through the mains supply) connection between the source transport and the nDAC avoiding issues like ground loops destabilising the highly sensitive DAC circuitry.

quote:

Originally posted by pcstockton:
I would like to hear something that makes me re-evaluate my process and gear. But just because I haven't yet does not lead me to believe all sources sound the same.

-Patrick

Also, just because a lot of others hear something doesn't mean it is there

Personally, I think your system should be capable of resolving any differences - if not the differences people are perceiving are being over exaggerated by people on this and others threads.

quote:

Originally posted by Richard Dane:
otherwise the noise just get modulated into the digital signal and then demodulated at the other end.

Yup... it gets modulated as JITTER. Remove the jitter and you remove the noise.

Trust me, it isn't rocket science or magic that is going on. Naim have just done a very well thought through design that eliminates many of the problems that previous DACs have had.

Andy, it was going back a while, during the development of the Naim DAC if I recall correctly. I was theorising that based upon the DAC design, then you could pretty much put any digital source (such as a £25 DVD player) on the DAC and it would sound identical to even the highest-end CD transport. The answer was that while this was, on the face of it, a sound theory, in practice there are a number of factors (touched on above) that would still have a marked effect on performance.

AFAIK, the noise modulation I mentioned above occurs whether using optical or coaxial s/pdif. Optical though is recommended for isolating the DAC from particularly noisy sources where this is also transmitted via the electrical conductivity of the coax connection.

I'm sure my powers of adjudging relative performance are pretty good. I'm not much taken in by placebo effect. If I think something is too close to call then I'll give it more time. If it's still too close to call, then it is. It helps nobody to suggest otherwise.

If there's a marked difference then it boils down to which is the more enjoyable, the more involving, the one that gets me closer to the performance. Even then, you can have a significant difference but each can be just as involving, just as enjoyable and get you just as close. Then it comes down to source material and personal preference.

So far, each CD player I have connected up to the nDAC/XPS has resulted in a different performance, for better or worse. The worst was still very good, but the best was obviously something special, that stood out - And not just in the obvious hifi ways (bass, treble, soundstage etc..). The best had that elusive "je ne sais quoi" that involved me, gave me the goosebumps and drew me in to the performance unlike the others.

Goosebumps are one of the most reliable indicators I know of that something special is going on in a hifi system.

Hi,

I'm a bit of a newbie and have just upgraded from NAC282+HiCap to a NAC252. Source is a CDX2+XPS. NAP250 into Sonus Cremona floorstanders.

Expect you're going to say source first, source first !! But I was unsure about my best options: CDS3 or HDX or nDAC. I don't want to irritate my dealer by not having a clearer preference on what I should audition !

So my question, should I go for

(i) an nDAC fed by the CDX2. If so should the XPS power the nDAC?
(ii) trade-in the CDX2 for a CDS3 ?
(iii) trade-in the CDS2 for an HDX ?
(iv) trade in the CDX2, buy an nDAC and a cheap transport +/or lots of USB sticks ?

I guess SQ is my primary criteria.

Be very interested in your votes (more than today's other voting event !

Robert

quote:

Originally posted by Richard Dane:
AFAIK, the noise modulation I mentioned above occurs whether using optical or coaxial s/pdif. Optical though is recommended for isolating the DAC from particularly noisy sources where this is also transmitted via the electrical conductivity of the coax connection.

Well, in an optically isolated system (which you have with an SPDIF design - and in fact between the digital processor and the DAC section), there are only two ways to transmit noise - jitter and bit dropout. We've discounted bit dropout, leaving jitter, which the nDAC should remove. Or have the Naim engineers found a new law of physics

quote:

Goosebumps are one of the most reliable indicators I know of that something special is going on in a hifi system.

Yup... got those the first time I listened to the DAC in my system - haven't had those for a long time.... Keep getting them too

quote:

Originally posted by Seaslater:
(i) an nDAC fed by the CDX2. If so should the XPS power the nDAC?
(ii) trade-in the CDX2 for a CDS3 ?
(iii) trade-in the CDS2 for an HDX ?
(iv) trade in the CDX2, buy an nDAC and a cheap transport +/or lots of USB sticks ?

I guess SQ is my primary criteria.

Hi,

If you've read this thread, you'll be able to predict my answer: (iv)

quote:

Originally posted by Seaslater:
Hi,

I'm a bit of a newbie and have just upgraded from NAC282+HiCap to a NAC252. Source is a CDX2+XPS. NAP250 into Sonus Cremona floorstanders.

Expect you're going to say source first, source first !! But I was unsure about my best options: CDS3 or HDX or nDAC. I don't want to irritate my dealer by not having a clearer preference on what I should audition !

So my question, should I go for

(i) an nDAC fed by the CDX2. If so should the XPS power the nDAC?
(ii) trade-in the CDX2 for a CDS3 ?
(iii) trade-in the CDS2 for an HDX ?
(iv) trade in the CDX2, buy an nDAC and a cheap transport +/or lots of USB sticks ?

I guess SQ is my primary criteria.

Be very interested in your votes (more than today's other voting event !

Robert

Robert

Regarding your option 1. The XPS should go on the Naim DAC, it has no purpose on the CDX2 when the CDX2 (which should be CDX2-2 or a converted CDX2 when it needs to be able to feed the DAC) is feeding the Naim DAC through SPDIF.

The options are wide ranging, do you already have some incling where you want to go to in the future? Stay with your CDs or go towards computer stored music (incl. HiRes)?

Between your options the prices vary, possiblities vary, the future possibilities vary. And it is no shame to try all your options with the Naim dealer. I will be fun at least. He should also be able to guide you towards what suits your future plans best.

-
aleg
-
aleg

quote:

Originally posted by Andy S:
We've discounted bit dropout, leaving jitter, which the nDAC should remove. Or have the Naim engineers found a new law of physics

Not if that jitter arrives at the DAC as an inherent part of the signal, which seems to be what people are referring to in jitter introduced by the source.

quote:

Originally posted by Naijeru:
Not if that jitter arrives at the DAC as an inherent part of the signal, which seems to be what people are referring to in jitter introduced by the source.

Yes, but with the nDAC it can't (or Naim are lying about how their DAC works!!). By reclocking the data with a high precision clock, the nDAC removes the jitter due to anything arriving at or before the SPDIF interface.

From their white paper:

quote:

Naim’s buffer or memory method of jitter removal relies on a simple concept: the audio data is clocked into the memory at the incoming, inconsistently timed rate and is then clocked out of the memory and into the DAC chips using a precise clock.

The key thing about the nDAC is it operates differently to most other DACs that sync their DAC clock to the incoming bitstream - jitter and all.

I'll ask it another way around. If the nDAC is affected by jitter from the source, by what mechanism does it get mixed in with the DAC clock? I've set out my understanding of the system, based on Naims white paper but no one has managed to describe how this jitter would get into the DAC clock. What is that mechanism?