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

[Miska]

Hi Alex,

 

If you had a 24bit/24.xMHz parallel PCM stream (upsampled from Redbook with a quality filter) and were going to DSM it to a 6 bit/7 level format, what order of modulator would you choose?

(There is a reason for my question, but I'll not post that here just yet.)

 

I would stick to the same fifth or seventh order modulators I am using now, but just change one parameter to output longer words. I would suggest to go straight to 8-bit.

[Miska]

Sorry - you're interested in what I think of what? In the main I'm not interested in saying what I think, I am interested in giving observations though. One of my favourite sayings comes from Dirty Harry 'Opinions are like a**holes, everybody's got one'. if you'd like my observations then by all means ask.

 

L&V paper is full of just those opinions, look at chapter 4 "concluding remarks", it contains just opinions without anything to back those up. It is full of words that specifically indicate that it is just one big attacking opinion.

 

There are some tricks that can be done that he was clearly not using. Obviously he has been using traditional modulator design models.

 

I won't post my modulator maths, because it would reveal my design I wish to keep secret. I guess your goal is just trying to force me to tell how I'm doing things which I won't. But I have posted both digital domain analysis of my designs with various test signals as well as measurements of real world DACs.

 

If that's not enough, keep on waving L&V paper as much as you like, I don't care. I will continue my work on modulators as well as PCM noise shapers. I'm happy when I can improve obtained performance.

 

For example 6-bit (65-level) SDM DAC has 35 dB lower noise, but otherwise exactly the same behavior when properly implemented. When we turn that same 65-level DAC into very simple DSD DAC we get - drumroll - the same 35 dB noise figure improvement:

 

With a bit of DSD specific modification it can become for example like this:

 

And this is straight from the D/A conversion stage before any analog filter stages.

[opus101]

L&V paper is full of just those opinions, look at chapter 4 "concluding remarks", it contains just opinions without anything to back those up. It is full of words that specifically indicate that it is just one big attacking opinion.

 

By all means cite specific examples.

[Miska]

By all means cite specific examples.

 

"misguided desire to store the data in DSD format"

 

I could myself say "misguided desire to store data in PCM format that has been created as result of lossy conversion from D-S ADC just to be converted back again for D-S DAC playback".

 

"The high levels of ultrasonic noise and spuriae produced by an inadequately-filtered 1-bit sigma-delta converter"

 

I could as well say "The high levels of ultrasonic noise and spuriae produced by inadequately filtered R2R converter".

 

There are many completely hilarious parts:

"©

Finally, consider 8-bit, four-times-oversampled PCM with

noise shaping. This is also a data rate one-half that of DSD and

double that of CD, with a sampling rate of 4 × 44,100 =

176,400 Hz. It can achieve a noise floor 120 dB below full

scale up to 20 kHz, using 96 dB of noise shaping, and a total

noise power of –19 dBFS. Its frequency response would be

flat to 80 kHz."

 

Now this means that this DAC would output around 112 times higher level of ultrasonic noise than a DSD64 DAC designed according to SACD specification. Because you have minimal bandwidth between 80 kHz and 88.2 kHz that is Nyquist frequency of this sampling rate and due to images of this spectrum repeating at every multiple of 176.4 kHz in the output spectrum. So right above 88.2 kHz the same noise repeats with inverse spectrum.

 

Here L&V conveniently forget strong and fully correlated image frequencies at multiples of sampling rate that are generated by a PCM DAC.

 

Plus, as result of any brick-wall filtering this PCM would have extremely poor temporal resolution - a topic completely ignored by the paper.

[Miska]

Just as note, I have used ninth order PCM noise shaper to push 176.4/16 PCM give roughly the same SNR performance as DSD64. However, due to limited bandwidth of 176.4k PCM, the time domain performance is not as good.

 

Bitrates are the same though.

[Paul R]

Hey Miska-

 

It seems fairly obvious this guy does not have a clue and is just trying to pick a fight. I would suggest ignoring him from now on. Sooner or later he will either learn or come to the attention of TPTB. In either case, I doubt seriously anyone will learn from him, since he doesn't seem to have the knowledge, skill, or experience to back up his own assertions. Nor does he even post interesting opinions. Any mouse ID too. (*sigh*)

 

I do so love the ignore list!

 

-Paul

[wdw]

Anyone heard or seen this new Berkeley DAC? Their web page still shows it as "coming". Seems to be a rather slow start for a new product release. A release that was first bought forward a few months ago.

WDW

[ted_b]

Yes, they were having an issue with casework manufacturing but are back on track, according to a dear reviewer friend who is on the early list.

[wdw]

Yes, they were having an issue with casework manufacturing but are back on track, according to a dear reviewer friend who is on the early list.

 

High expectations from this great design team

[Allan F]

I do so love the ignore list!

 

+1

 

That's where trolls belong. What took you so long?

[wdw]

Anyone heard this one yet?

 

Cheers,

 

WDW

[The Computer Audiophile]

Anyone heard this one yet?

 

Cheers,

 

WDW

Haven't heard it yet. I believe it will be used by The Audio Salon at the Newport show in less than one week.

[wdw]

Haven't heard it yet. I believe it will be used by The Audio Salon at the Newport show in less than one week.

 

Hello Chris,

I had thought that you would have been offered one for review by now. Are you scheduled to have one in for review in the future?

Regards,

Warren

[The Computer Audiophile]

Hello Chris,

I had thought that you would have been offered one for review by now. Are you scheduled to have one in for review in the future?

Regards,

Warren

Robert Harley and I will receive the first two units off the production line :~)

[wdw]

Robert Harley and I will receive the first two units off the production line :~)

Great...keep us posted!

[Kelly]

Robert Harley and I will receive the first two units off the production line :~)

 

You keep good company!

[Sam Lord]

He keeps good company!

Fify

[The Computer Audiophile]

Just received the first real photo I've seen of the new Reference Series. Maier Shadi of The Audio Salon in Los Angeles is showing it at The Show Newport. He shot this photo for me of the DAC warming up for the show.

 

 

 

[Paul R]

Just received the first real photo I've seen of the new Reference Series. Maier Shadi of The Audio Salon in Los Angeles is showing it at The Show Newport. He shot this photo for me of the DAC warming up for the show.

 

 

 

SWEEET! I like the look of of it, very retro-futuristic. Look alike no USB built in and that means probably no DSD. But who knows?

 

Paul

[The Computer Audiophile]

SWEEET! I like the look of of it, very retro-futuristic. Look alike no USB built in and that means probably no DSD. But who knows?

 

Paul

Correct, no USB. Berkeley Audio Design believes DACs can only be optimized for PCM or DSD, but not both. Thus, only PCM is supported. Berkeley recommends converting DSD to PCM offline with software included with the DAC.

[ted_b]

SWEEET! I like the look of of it, very retro-futuristic. Look alike no USB built in and that means probably no DSD. But who knows?

 

Paul

 

No DSD. Don't you remember the fuss that was made when they first intro'd the DAC spec, and told everyone to downsample/convert their DSD to 24/176k offline? It makes sense when you are not doing DSD (why ask the DAC to to that heavy lifting) but one press release was vaguely worded as if that is how they are DSD-capable. No big deal.

 

I am anxious for our fearless leader to review this beast.

 

Edit: I see Chris beat me to it.

[Paul R]

No DSD. Don't you remember the fuss that was made when they first intro'd the DAC spec, and told everyone to downsample/convert their DSD to 24/176k offline? It makes sense when you are not doing DSD (why ask the DAC to to that heavy lifting) but one press release was vaguely worded as if that is how they are DSD-capable. No big deal.

 

I am anxious for our fearless leader to review this beast.

 

Edit: I see Chris beat me to it.

 

Now that you mention it, I do recall that brouhaha... It's a shame, I was saving for a Berkeley Alpha DAC, but since I hit DSD lately, my thinking has been changing around a bit. Shame though, a Berkeley quality DAC for DSD would likely be utterly awe inspiring...

 

-Paul

[ted_b]

I would think PCM may well be awe inspiring...but for this coin I am disappointed USB requires another investment.

 

By the way, this will make the new google database we are working on, since the DSD one seems to be no longer about unique product releases. The Berkeley Reference will find a founders location on the new "Database of stuff NOT supporting DSD". Very small list of new products.

[Paul R]

I would think PCM may well be awe inspiring...but for this coin I am disappointed USB requires another investment.

 

By the way, this will make the new google database we are working on, since the DSD one seems to be no longer about unique product releases. The Berkeley Reference will find a founders location on the new "Database of stuff NOT supporting DSD". Very small list of new products.

 

Sure, PCM can be awe inspiring, but that same kind of awe inspiring performance appears to happen a lot more frequently with DSD, and on far less capable gear.

 

I am beginning to be convinced that the file format (PCM vs DSD) and how it is processed has more to do with this than anything else. So a very good DAC doing very good DSD processing would, if that theory holds any water, blow away even extraordinary PCM only DAC.

 

Obviously, the Berkeley people disagree, and God Bless 'Em - they put out some wonderful sounding DACs.

 

-Paul

[mindnoise]

Kinda reminds me of an old Sony DAT!

 

Just received the first real photo I've seen of the new Reference Series. Maier Shadi of The Audio Salon in Los Angeles is showing it at The Show Newport. He shot this photo for me of the DAC warming up for the show.