New Berkeley DAC? This article implies as much but ... ?

[Hiro]

Everything could be much easier if there was a music format which was, say, 5 bits and 2.8 MHz, of course, file size would be pretty large!

 

The output of such a Delta-Sigma modulator would be comparable to quad-speed DSD bandwidth wise.

[elcorso]

Hi Guys - I received one last response from Berkeley Audio Design. It addressed some items directly.

 

 

 

Berkeley Audio Design:

 

DSD versus PCM part 2.

 

 

 

This looks to me more as a "Yellow Paper" that a White Paper.

 

Berkeley Audio Designs want to confuse readers even more?

 

As I said before DSD & PCM can survive together, but each one on their side.

 

If they don't want to built a DSD DAC, OK. But please, don't try to force anybody to conversions, isn't the best way for SQ...!

 

Roch

[edorr]

Yep, my ExaSound E28 review said as much (multichannel in one dedicated box is very nice) and now I own one.

 

Thanks for the feedback on clocks. I take you comments to mean that although every DAC (in this example) is made to a price point, and creating logical upgrade paths for such things as external power supplies might make sense as a field upgrade capability......but cutting costs on a clock and later adding via cabling is not necessarily a good upgrade path (unless synching to other dacs, a practice not common amongst consumers). Makes sense, especially on hi-end cost no object models like this Berkeley.

 

A lot of people agree on this, but dCS obviously did not get the memo.

[Elberoth]

The fact that Antelope sells any of their more expensive ones to the pro market is indeed a wonder of the world, then. ?

 

It is a good clock. And the magic word 'Atomic' does wonders in terms of marketing and sales

[Elberoth]

It is much better to send the clock signal over a few mm of well engineered, impedance controlled PCB trace, than it is to send it over a cable, through multiple jacks and plugs, none of which can be "perfect" from an impedance standpoint.

 

Don't forget about the PLL, that still has to be present on any DAC with WClk input. It also adds considerable amount of jitter.

[Elberoth]

A lot of people agree on this, but dCS obviously did not get the memo.

 

dCS started using external clocks on their pro-audio oriented gear, and then added this option to their audiophile range.

 

I was puzzled to see the external clock option on the Vivaldi line. But then again - maybe they thought that the compatibility with other models was more important. Afterall - not all Vivaldi buyers get the whole stack all at once. Many upgrade component by component.

[barrows]

A lot of people agree on this, but dCS obviously did not get the memo.

 

or, as I thought m post already suggested, dCS likes selling their customers very expensive additional boxes with "better" oscillators inside, or perhaps even more accurately, likes their customers' money.

Why did dCS not put the better oscillator in the DAC itself from the start?

[edorr]

or, as I thought m post already suggested, dCS likes selling their customers very expensive additional boxes with "better" oscillators inside, or perhaps even more accurately, likes their customers' money.

Why did dCS not put the better oscillator in the DAC itself from the start?

 

As you said; why deliver a superior, cheaper architecture when you can add an expensive shiny box to the lineup?

[Miska]

or, as I thought m post already suggested, dCS likes selling their customers very expensive additional boxes with "better" oscillators inside, or perhaps even more accurately, likes their customers' money.

Why did dCS not put the better oscillator in the DAC itself from the start?

 

What I like about in dCS is that they actually design their own DACs. There are not that many such companies. Meitner, Playback Designs and MSB comes to mind. Others?

[jhwalker]

What I like about in dCS is that they actually design their own DACs. There are not that many such companies. Meitner, Playback Designs and MSB comes to mind. Others?

 

Chord?

[Miska]

Chord?

 

Oh, definitely, I'm already ashamed that I forgot it from the list.

[barrows]

What I like about in dCS is that they actually design their own DACs. There are not that many such companies. Meitner, Playback Designs and MSB comes to mind. Others?

 

Yup, Chord, TotalDAC… I am not sure what Light Harmonic does for the R2R DAC in the Da Vinci, it may be discrete as well...

[Miska]

I am not sure what Light Harmonic does for the R2R DAC in the Da Vinci, it may be discrete as well...

 

It was tested on HiFi-News April 2013 issue and seems to be 4x ladder. Article doesn't say directly, but I'd understand that it is discrete DAC.

[Hiro]

Chord?

 

The only DAC from the bunch I could ever afford...

[Kevin M.]

 

The rumor is that Berkeley's software is an included copy of JRiver. However, JRIver currently only does DSD-to-PCM offline into 24/352.8K.

Ted I believe it's at 64/352 and then downconverts to 24/176 or whatever format you chose. From the Wiki,"

By default, DSD will be converted to PCM for playback.

The DSD to PCM conversion process converts from 1-bit DSD to 64-bit PCM at 1/8th of the sample rate. The total amount of data from this conversion grows by 8x, so the process is effectively lossless / perfect.

Once you have PCM, it will be 64bit @ 352.8 kHz for DSD, and 64bit @ 705.6 kHz for DSD 2x.

It is rare for hardware to support these high sample rates, so downsampling is required. The option 'Greater than 192kHz' in DSP Studio > Output Format is what is used. There is no way to configure DSD 1x and DSD 2x independently."

[dallasjustice]

My local dealer has one of their new dual DACs. It has 2 separate DACs for PCM and DSD. I believe it's all discrete.

 

It was tested on HiFi-News April 2013 issue and seems to be 4x ladder. Article doesn't say directly, but I'd understand that it is discrete DAC.

[Sam Lord]

What I like about in dCS is that they actually design their own DACs. There are not that many such companies. Meitner, Playback Designs and MSB comes to mind. Others?

 

Mola Mola, though not available just yet.

[Miska]

Mola Mola, though not available just yet.

 

Yeh, and that's a 1-bit DAC...

[Superdad]

My local dealer has one of [Light Harmonic's] new dual DACs. It has 2 separate DACs for PCM and DSD. I believe it's all discrete.

 

I have searched the web quite a bit and nothing I have ever found con confirm that the Da Vinci DACs are a discrete resistor design--nor does the company ever claim that. I am betting that that they are actually an NOS PCM1704, or maybe one of the other T.I. chips that allows for turning off of its built-in filters (though that would then not be an R2R design). If anyone can find a photo or quote to prove otherwise, I would be interested.

[Miska]

I have searched the web quite a bit and nothing I have ever found con confirm that the Da Vinci DACs are a discrete resistor design--nor does the company ever claim that.

 

Same here. HiFi-News only mentioned that it's an array of four DACs with two as complementary pair and two time-delayed (staggered). I couldn't find any detailed enough photos either.

 

I am betting that that they are actually an NOS PCM1704, or maybe one of the other T.I. chips that allows for turning off of its built-in filters (though that would then not be an R2R design). If anyone can find a photo or quote to prove otherwise, I would be interested.

 

I originally thought that it would have been PCM1704, but after checking out the HiFi-News review and looking at the measurements I'm not sure. Seems to be linear within +-0.5 dB down to around -93 dB which is spot-on PCM1704 (-90 dB in datasheet) in staggered configuration (+3 dB). Also the complementary pair is just like PCM1700/PCM63P/PCM1704 are constructed.

 

Maybe Larry could enlighten us on this topic?

[Adyc]

There is a big fuss in another forum on a new dac called Trinity from Germany. It uses 16 PCM 1704 in time staggered configuration and claims there is no need for any digital filter. It retails EUR 45k. Judging this DAC interior, I do not understand how they can charge that much money.

 

Anyway, what is so special about time staggered configuration? If it is so great, why not more people using it?

[CKehoe]

Interesting article here:

 

pcm dsd

 

Any of the engineer types care to comment or enlighten us as to the article's insights / mistakes?

[opus101]

Very interesting question - so much so that I signed up in order to have a stab at answering it.

 

Firstly Lynn Olson's a very interesting and valuable contributor to the field of quality audio design - particularly in relation to speakers and valve amps. Almost everything I've read from him - even if I've disagreed with it - has been worthwhile and stimulating. Very generally speaking I agree with him in this article, though I'd like to highlight a few oversights. Rather than write a single long post, I'll take his numbered paragraphs one at a time in individual posts.

 

1.

 

Firstly its a bad idea to feed a lot of ultrasonics into a multibit DAC, but perhaps that's an aspect of Lynn's point (he's no fan of S-D DACs in general). Its going to be worse though than the ultrasonics from a DSD or S-D DAC because the multibit architecture (at least as audio DAC chips go) isn't optimized for that. So whilst he's correct that 'BUD' will linearize the bits, that's not the low-hanging fruit of the audible problems with multibit. Rather, the dynamic errors are - glitching, digital-analog crosstalk and settling transients - ISTM that Lynn may well not be aware of this.

 

So adding all the ultrasonic dither is going to 'grey out' the sound of your typical multibit DAC much more so than for a 'low-bit' DAC which are specifically designed for very low glitch. The dominance of dynamic over static errors in multibit DACs explains why NOS DACs have such a following and why many listeners report reduced dynamics from such DAC the more oversampling they switch in.

[dallasjustice]

Opus,

I'm glad you resurfaced here! :-)

 

Michael.

 

 

Very interesting question - so much so that I signed up in order to have a stab at answering it.

 

Firstly Lynn Olson's a very interesting and valuable contributor to the field of quality audio design - particularly in relation to speakers and valve amps. Almost everything I've read from him - even if I've disagreed with it - has been worthwhile and stimulating. Very generally speaking I agree with him in this article, though I'd like to highlight a few oversights. Rather than write a single long post, I'll take his numbered paragraphs one at a time in individual posts.

 

1.

 

Firstly its a bad idea to feed a lot of ultrasonics into a multibit DAC, but perhaps that's an aspect of Lynn's point (he's no fan of S-D DACs in general). Its going to be worse though than the ultrasonics from a DSD or S-D DAC because the multibit architecture (at least as audio DAC chips go) isn't optimized for that. So whilst he's correct that 'BUD' will linearize the bits, that's not the low-hanging fruit of the audible problems with multibit. Rather, the dynamic errors are - glitching, digital-analog crosstalk and settling transients - ISTM that Lynn may well not be aware of this.

 

So adding all the ultrasonic dither is going to 'grey out' the sound of your typical multibit DAC much more so than for a 'low-bit' DAC which are specifically designed for very low glitch. The dominance of dynamic over static errors in multibit DACs explains why NOS DACs have such a following and why many listeners report reduced dynamics from such DAC the more oversampling they switch in.

[opus101]

Thanks for the warm welcome Michael

 

To continue then with Lynn's diatribe, I have a little more to contribute to his 1.

 

The stated 'problem of the 1000000 to 01111111' transition isn't one related to the resistor weights in an R2R DAC so cannot be fixed up with masses of dither. Rather its a problem of timing errors in the switches - with this MSB transition, all the switches controlling the resistors are changing state. The problem comes when the state changing isn't executed perfectly in synchronism. There are then lots of ephemeral intermediate values being generated by the DAC which aren't related to the initial or final code - they're very distant codes, constituting considerable glitch energy. its for this reason that Burr-Brown came up with their 'Co-Linear' architecture which moves this unwanted energy out to be triggered at the -6dB transition where the theory says it should be less of a problem subjectively. I'm not yet convinced of this, but that's by-the-by. Lynn's dither has maximized the glitching in both a normal R2R and a Co-Linear one, because its full-scale, highest frequency (read maximum number of zero crossings).

 

Moving on then to no.2....

 

The noise shaping algorithm is continuously busy no matter the audio input signal because the algorithm creates the noise - the level of 'busy-ness' though does indeed vary with audio signal to some degree. Forcing the loop to have maximum activity is going to make it all the more sensitive to jitter (Bob Adams wrote a letter to the AES journal about this). I'm not knowledgeable enough about S-D to talk about 'instabilities' - I'd rather hope the designers got those right. Evidence is though there's still significant noise modulation probably due to the difficulty of constraining the dither within a feedback loop into the required probability density function (TPDF).

 

If you dissect practically any S-D DAC datasheet (I've done this with AKM and ADI parts) you can't fail to miss the fact that the noise at full scale output goes up when compared to the -60dBfs stimulus - chances are Lynn's full-on dither will have the same effect, reducing the dynamic range significantly. The argument that 'there is dynamic range' to spare doesn't really wash when this effect is taken into account - for example the AKM part I looked at gives around -105dB noise putput at full scale, but -120dB at -60dBfs stimulus. -105dB is under 18bits of dynamic range.