What DACs are inside Naim CD Players?

There is always the danger, when trying to analyse something great, of failing to understand the unspecified love, care, attention to detail and effort that turns a bunch of components into that particular "something great"....  I didn't query or specify the particular breed of chicken that laid the eggs used to bake our wedding cake!

I'm not entirely sure what this info will add to the Naim listening experience, but until somebody can come along and be a little more specific, I think you can safely assume that every physical attribute lurking within all of Naim's players will accurately be encompassed below.......

Have a great Saturday, and enjoy the music!

Mike

Thanks Mike.   I had it coming and was expecting something.  I am just wyling away this Saturday night and was just curious.  Sometimes I have to know how stuff works.  I used to work on F1-11Cs and knew how their avionics worked and these products of Naim have me intrigued.  I am the happy owner of a NAIT XS 2.  The CD player I am using is nothing flash.  A Marantz CD-85 from 1991.  I am looking to upgrade one day and am enjoying the current setup very much.  In fact it might just do my wife and I fine.  I am just one of those guys when four years old had to pull clocks apart to see how they work.  Just be glad I am not your surgeon

I happen to remember that Naim DAC uses the same DAC chip as CD555, which is Burr Brown PCM1704K. This chip is highly regarded and kind of legendary DAC chip used by many High End manufacturers. 

Quote from Naim DAC white paper:

"Two mono, true multi-bit Burr-Brown PCM1704K DAC chips are used in the Naim DAC – the same type as in all Naim high-end CD players including the CD555. The PCM1704K is a precision, 24-bit digital-to-analogue converter with exceptional dynamic performance whose ultra-low distortion and excellent low-level signal performance make it the ideal choice. Precision lasertrimming at the factory minimises differential linearity and gain errors, and the PCM1704K incorporates a BiCMOS sign-magnitude architecture that eliminates glitches and other nonlinearities around the zero crossing point. It is capable of running with an input data clock of 25MHz, which allows sample rates of over 700kHz."

Based on the text, I guess that at least CDX2 uses 1704K also. 

But as written earlier, DAC chip is only small part of the DAC and even smaller part of a whole CDP. 

Good ones

Gigantor posted:

Thanks Mike.   I had it coming and was expecting something.  I am just wyling away this Saturday night and was just curious.  Sometimes I have to know how stuff works.  I used to work on F1-11Cs and knew how their avionics worked and these products of Naim have me intrigued.  I am the happy owner of a NAIT XS 2.  The CD player I am using is nothing flash.  A Marantz CD-85 from 1991.  I am looking to upgrade one day and am enjoying the current setup very much.  In fact it might just do my wife and I fine.  I am just one of those guys when four years old had to pull clocks apart to see how they work.  Just be glad I am not your surgeon

You seem a great candidate to apply the scientific method.

Borrow a Naim cd player from a local dealer and have a listen  If your wife enjoys music and has a good ear, let her do the blind listening tests and see what she prefers.  That's how we do things in my home.  Enjoy!

DAC families would likely make more sense - and even within those, there are selection criteria that might make two identically-numbered parts quite difference in terms of performance.

I have a fondness for the Wolfson DAC used in the CD5i-2 (and in other manufacturer's players); some absolutely love the Philips TDA1541A used in the first Naim CD players (I have one in a synthesizer of mine, oddly enough, and it is outstanding in that application). The original CDS had a double-crown version, though frankly that tells you more about the manufacturing consistency across the range than the performance.

In general, most of the upper-level Naim players of a recent generation use TI/Burr-Brown DACs, some with Analog Devices SHARC DSP filtering, others with the Pacific Microsonics-branded Motorola 56302 DSPs (as PMD200). All in all, though, this is really window dressing - I've heard some good implementations using the Analog Devices DACs as well as Cirrus Logic's CS series.

DAC IC really has very little to do with performance, in the end, and is more about features and implementation. Some people do quite well with an FPGA, a few transistors, and some passive support parts.

Gigantor posted:

I am still trying to get my head around over-sampling and noise levels in dB. 

How much enlightenment did you glean from that knowledge sandwich?

Hi Gigantor, DAC families does make sense as described above. Although Naim use in their higher end DAC the Texas Instrument PCM1704K converter chips (Burr Brown is no more) albeit they are getting long in the tooth now and is expensive to produce for the performance obtained.. Technology has advanced here, and TI don't recommend this chip for new designs. Now Naim tend to off board the over sampling, filtering and other DSP. In the CDPs (that support HDCD) they use Pacific Microsonics licencesd DSP ASIC chips. In the streamers and DACs they forsake the ASIC and use FPGA type SHARC DSP chips.. This gives more customisable control. Arguably it is the DSP implementation that makes a significant contribution to the overall sound.

The very latest streamer, the 272, uses a newer  DAC converter chip that can support PCM and DSD natively, however I am not sure whether even here Naim stick with the PCM i2S interface from the SHARC FPGA DSP chip as with the other DAC/streamers to decode DSD.

Simon

I forgot to say, the Naim CD1, CD2, CD3 and CDI players use the Philips TDA1541A and variants as the converter chip... another legendary device..

The founder of Naim said this many years ago, people think the DAC chip makes the biggest difference, but in his experience is was the least important part of the CD player, and more important was power supply, analog stage, CD mechanism, digital filter, analog filters,  and many other issues, other than the DAC chip itself.

It does make a lot of sense when you think of the fact that Naim have always made their all CD players sound excellent, regardless of which DAC chip they use.

Hope that answers the question.

Thank you everyone for your posts.  I am learning quite a lot and am amazed how CD players ever got off the ground with all their inherent issues.  I am still confused no matter how many times I read that over-sampling can extend the dynamic range of the original 16 bit word that is encoded on the CD.

It was pointed out that Naim use a 24 bit DAC.  Where do they get the extra bits from for the DAC?  My CD player when I purchased it back in 1991 advertised that it has 4 times over-sampling and thus is a 20 bit machine.  Back then I did not understand what they meant and just assumed it was better than a standard 16 bit word.  I have thus learned that over-sampling increases dynamic range and pushes the aliasing frequencies away allowing for simpler analogue filters to cut off frequencies above 20kHz.  I read from this that it is a pretty standard thing to do and recommended by, I think Phillips at the time.

I have also read and will have to re-read about jitter noise being added to help alleviate the issues of noise at low dynamic range levels.  I hope I have conveyed this meaning correctly.  So to make a CD listenable to, you have to add noise which is white noise.  I can see the vinyl brigade getting up on their soap boxes now.  It is all so interesting and confusing at the same time.

The full length articles can be found at SoundonSound "All about Digital Audio".  Six pages in all.  If there are any inaccuracies then they are all mine in trying to interpret what was written in this article and others 36 hours later.

I did read the Naim white paper on power supplies and found that most interesting and one review on the Naim CD5 XS.  There is quite a selection process in selecting for the right DAC.  Evidently some DACS maybe better than others electrically though can suffer from (I think the term was) thermal shock/stress.  So on paper they are better though once in an operating environment that show up the short comings.  So best is always not better.  As also has been pointed out the use of analogue stages and DSP.  I have yet to read the article on DSP and the article on error correction.  As I said before.  With all their foibles I am surprised CDs and CD Players ever got off the ground.  It was pointed out that the mass production of solid state chips made it an economical proposition over time.

Further reading highlighted that the current range of Naim CD players date from 2009.  There are two schools of thought here.  One they are showing their age and I believe that the the DACS may be no longer available and if not broken.

I hope I did not come across as too analytical.  This has been an enjoyable process with a fair amount of head scratching.  A pretty tough sandwich to chew.

Still researching,

Warm regards,

Paul.

Hi Paul, lots of terms... I can point to a few good books if you are genuinely interested... Wikis only take you so far.

But jitter is never added... it is the enemy of discreet sample streams. However dither can be added. Dither is noise that is added to the samples to reduce the effect of quantisation errors. These are the sampling errors between discrete samples (think of a stair case of sample values) of a continuous signal such as analogue audio.. This is more noticeable with 16 bit as opposed to 24 bit.

Analogmusic, there are many contributors to the sq of a converter, but the DAC chip is a key contributor.. and Naim have always appeared to use highly respected and renown devices... and use higher tolerance versions of the DACs whether they be TI or Philips for higher DACs/CDPs. You can see also the decoupling and clocking of the DAC chips defines how well the converter performs as well. Much written on the web on this on how different manufacturers have achieved this for a given converter.

Simon

G'day Simon,

Thank you for correcting me.  Yes dither is the term.  I noticed my error later today though did not get around it making the amendment.  To be honest I had to check my source first and we had visitors this afternoon and it never happened.

I find it truly staggering at the persistence of earlier engineers with all the tweaking that is required to get CDs off the ground.  I now sit back with amazement and more pleasure and understanding to appreciate all the technology involved.  I place it in a similar boat to those old sea going clocks of yesteryear with all their inaccuracies and tweaks to keep them on time.

For the moment Simon I think I will pass on the books.  Thank you for the offer.  I think I will sit back and spin a few disks and enjoy what we have.

Thank you very much.

Paul.