The Transition to True All-Digital Audio - AVS Forum
Forum Jump: 
Reply
 
Thread Tools
post #1 of 98 Old 03-07-2011, 04:53 PM - Thread Starter
 
HiFiFun's Avatar
 
Join Date: Nov 2008
Location: USA
Posts: 2,209
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 17
In the beginning there was all-analog
In the middle there was an unhappy marriage
In the end there was all-digital

If you look at the history of both audio and video, audio just now entering stage three. Video entered stage three about 15 years ago. With the video transition virtually complete, consumers enjoy high resolution, noise and artifact free, crystal clear, involving picture quality. Display prices and weight have dropped and quality is quite high for both flat panels and front projection technologies.

Analog audio reached its peak before the introduction of the first transistor circa 1960. Since then it’s been a very rough transition period for music lovers. In fact for several decades analog audio still had better sound quality than digital, in spite of the Phillips claim in 1983 that CD format was “perfect sound forever”. Digital finally began to show its merits in the 1990s and begin to supersede analog playback.

However many sonic anomalies remained in spite of the refinements in both digital and analog processing. Early digital was bright and constricted. Manufactures responded with a warmer but less resolving sound quality. Between these extremes, niches were created for “high-end” cables, a dubious solution made popular for several decades.

Today the most common technique is to convert digital audio streams to analog for processing and then several stages of amplification. The problem is low-level analog signal are easily corruptible by higher levels of digital interference.
The obvious solution is to not convert the digital signal to analog. This implies the use of an all-digital processing chain and power amplifier.
However the traditional audio mindset that heavier, bigger and more expensive is always better.

Analog Class D Amplifiers
The most common type is the European based ICE analog class D analog amplifier, which found acceptance in subwoofer amplifiers and then more recently in a few expensive receivers.
Because of the technology available at the (~1990) time, the digital signal was converted to analog. This allowed the delicate mico-volt analog signal to easily become contaminated when input to noisy pulse-width-modulated Class D power amplifier. This also assumes that traditional D/A converters are transparent. Trust me they are not.
These pulsed receivers with traditional D/A converters are at a distinct disadvantage regardless of cost. The analog signal is corrupted during analog conversion and then at every stage thereafter. Then it is amplified 30db! Is this madness or what. Pioneer claims it took their designers two years to make respectable.

Digital Class D Amplifiers
In 2011 the venerable analog conversion is finally avoided, with all-digital amplifiers containing built-in pcm-to-pwm converters and all-digital feedback The best example is the eight channels, 120 watt, HDMI 1.4 Samsung 700 receiver using Pulsus Korean technology. The great efficiency and reduced part count lowers the weight and price, beside offers superior sonics. What more could a consumer ask for?
http://www.pulsus.co.kr/english/products/pro_02_05.html

Texas Instruments has also developed an all-digital two channel reference design which features low distortion:
http://www.ti.com/ww/en/analog/tas5630/index.shtml

These advances in sonic fidelity and realism are possible since any possible extraneous contamination only occurs after the typical 30db of amplification, in the final output stage.
With minimal legacy analog processing, the receivers back panel real-estate can be drastically reduced. In fact receivers could be 1/4 the current size with no loss in performance. In these all-digital amplifiers, the volume control can change the amplifier gain without distortion.

The benefits of all-digital receivers when combined with other new breakthrough technologies such as speakers incorporating high-gauss neodymium magnets are both compelling and already in place. Companies such as Bose and Apple are already making a fortune. The bigger, heavier, more expensive enthusiast believer will be last to make the change.

A Word of Caution
Class D amplifiers are inherently noisy. It takes expensive gear to measure and properly suppress the interference. However by keeping the signal in digital format until the final amplifier output stage, the high frequency noise is easily filtered out.
Two obstacles preventing progress are identified: consumers who have made large investments in traditional gear and the industry including manufactures and advertiser supported reviewers

Lessons Learned
I’ve been following this simple rule with excellent success in three systems: avoid mixing analog and digital processing. It’s like mixing oil and water. Don’t do it, as you will never be satisfied with the long-term sound quality, as the interference is everywhere.
HiFiFun is offline  
Sponsored Links
Advertisement
 
post #2 of 98 Old 03-07-2011, 05:15 PM
bfm
Senior Member
 
bfm's Avatar
 
Join Date: Jul 2005
Location: tampa
Posts: 349
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
I'm confused. Why should amplification be so hard to comprehend for me!!!

I thought source mattered. So when you posted your thread title I thought you were referring to the material we use to store our toonz. Happy Fat Tuesday!
bfm is offline  
post #3 of 98 Old 03-12-2011, 07:56 AM - Thread Starter
 
HiFiFun's Avatar
 
Join Date: Nov 2008
Location: USA
Posts: 2,209
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 17
Imagine the advantages of true ALL-digital receivers:

The Samsung class D receiver "eliminates from a traditional signal path all the electronics of a DAC as well as the active analog gain stages of a preamplifier and power amplifier. It does this by converting the PCM signal from a digital source directly into a pulse-width modulation (PWM) signal that turns the output transistors on and off. That’s it—no digital filter, no DACs, no multiple stages of analog amplification, no interconnects, no jacks, no analog volume control, no preamp. The conversion from the digital domain to the analog domain occurs as a by-product of the switching output stage and its analog filter. This is as direct a signal path as one could envision."

From the $6000 NAD all-digital stereo amplifier.
http://www.avguide.com/review/nad-m2...lifier-tas-198

Summary: HDMI 1.4, 7.1 120w channels for $300. Obviously a no brainier decision to buy as nothing else compares.

Realize that hardly anyone here will recommend this superior solution, as it allows consumers to spend their disposable income elsewhere. I took my savings and bought a $1300 MSRP Mitsubishi 4000 Digital Light front projector, which used to cost $10k!
HiFiFun is offline  
post #4 of 98 Old 03-12-2011, 08:20 AM
AVS Addicted Member
 
MichaelJHuman's Avatar
 
Join Date: Nov 2006
Location: Colorado Springs, CO
Posts: 18,955
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 239 Post(s)
Liked: 118
Nothing wrong with all digital, if they can get good results. A few engineers have been skeptical about all digital because feedback was hard to implement (compared to analog class D.)

"But this one goes up to 11"
MichaelJHuman is offline  
post #5 of 98 Old 03-12-2011, 09:34 AM
AVS Addicted Member
 
JOHNnDENVER's Avatar
 
Join Date: Dec 2005
Posts: 13,377
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 12
I like this AVR, I almost purchased it twice now. The video handling isn't robust enough for me I don't think anyways. Why" Becasue I still run a lot of analog video source.
JOHNnDENVER is offline  
post #6 of 98 Old 03-12-2011, 10:19 AM - Thread Starter
 
HiFiFun's Avatar
 
Join Date: Nov 2008
Location: USA
Posts: 2,209
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 17
Reviewer Robert Harley explains the theory. It is remarkably how many unnecessary electronic stages and parts can be eliminated. Why didn't we do this years ago?

"This difference might not seem that great at first glance, but consider the signal path of a conventional digital-playback chain driving a switching power amplifier. In your CD player, data read from the disc go through a digital filter and are converted to analog with a DAC; the DAC’s current output is converted to a voltage with a current-to-voltage converter; the signal is low-pass filtered and then amplified/buffered in the CD player’s analog-output stage. This analog output signal travels down interconnects to a preamplifier with its several stages of amplification, volume control, and output buffer. The preamp’s output then travels down another pair of interconnects to the power amplifier, which typically employs an input stage, a driver stage, and the switching output stage. In addition to the D/A conversion, that’s typically six or seven active amplification stages before the signal gets to the power amplifier’s output stage.

To reiterate the contrast with the Nad M2 (and the Samsung 700), PCM data are converted by DSP into the pulse-width modulation signal that drives the output transistors. That’s it. There are no analog gain stages between the PCM data and your loudspeakers. The signal stays in the digital domain until the switching output stage, which, by its nature, acts as a digital-to-analog converter in concert with the output filter. The volume is adjusted in DSP."

Nad choose to take the legacy S/PDIF bus route by omitting HDMI, yet still charging an-arm-and-a-leg. However this allowed traditional high-end reviewers to take the technology seriously (let the advertising dollars flow). They too are in awe of the technology, both in theory and in listening. The difference being I've moved-on to recommend eminently superior multi-channel solutions, while they still refuse to take HDMI seriously. Hello?
HiFiFun is offline  
post #7 of 98 Old 03-12-2011, 11:27 AM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 18,375
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 981 Post(s)
Liked: 547
While class D amps on paper look ideal, in practice, designing good class D amps is exceptionally challenging. Design requires more expertise then typical analog designer usually has. Circuit-board design is also complex. Here is a simplified overview of the design and what to watch for: http://www.eetimes.com/design/audio-...mplifiers-work

As you see, some of the major problems are balancing distortion against reliability of the system (switch time/gap). Next one is having a super stable power supply as any variations there directly translates into output error unlike traditional amplifier designers which are pretty resilient that way. This is why amps like Mark Levinson NO 53 are so enormous in size: http://www.marklevinson.com/ProductDetails.aspx?prdid=1



Filter design becomes quite tricky too. The speaker and wires now become part of your filter network. This is not a big deal when you are talking about an amp mated to a specific sub. But for a general purpose amplifier, you can't rely on that.

PWM frequency is also an issue. Ideally you want a frequency that is 10 to 20X higher than the highest frequency you want to reproduce. But higher frequency designs are even more challenging. The above Mark Levinson uses a clever multi-phase interleaving solution that has an effective frequency of 2 Mhz. Compare that to ~300KH frequency of typical ICE amp uses in mass market consumer products. (See a bit more here: http://www.whatsbestforum.com/showth...ll=1#post39815).

Integrating the DAC for the most part, is frowned upon. I have to run so no time to explain the details now. But as intuitive as that might sound, you are better off using a conventional DAC and then amplifying that using class D amp topology. Remember, class D does not mean digital! It was just the next letter in the alphabet.

The typical answer to all things distorted in amps is to use feedback. And lots of it in case of class D amps. I won't get into the argument relative to merits of feedback here . But suffice it to say, an amp topology needs to be good and performant without resorting to ton of feedback.

Net, net, it is easy to get efficiency of class D amps. Getting the full fidelity though, is much, much harder. So that is the reason the whole world hasn't gone there .

Amir
Retired Technology Insider
Founder, Madrona Digital
"Insist on Quality Engineering"
amirm is online now  
post #8 of 98 Old 03-12-2011, 12:53 PM
AVS Special Member
 
M Code's Avatar
 
Join Date: Mar 2003
Location: Joshua Tree, CA
Posts: 9,967
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 152 Post(s)
Liked: 133
Quote:
Originally Posted by amirm View Post

While class D amps on paper look ideal, in practice, designing good class D amps is exceptionally challenging. Design requires more expertise then typical analog designer usually has. Circuit-board design is also complex. Here is a simplified overview of the design and what to watch for: http://www.eetimes.com/design/audio-...mplifiers-work

As you see, some of the major problems are balancing distortion against reliability of the system (switch time/gap). Next one is having a super stable power supply as any variations there directly translates into output error unlike traditional amplifier designers which are pretty resilient that way. This is why amps like Mark Levinson NO 53 are so enormous in size: http://www.marklevinson.com/ProductDetails.aspx?prdid=1



Filter design becomes quite tricky too. The speaker and wires now become part of your filter network. This is not a big deal when you are talking about an amp mated to a specific sub. But for a general purpose amplifier, you can't rely on that.

PWM frequency is also an issue. Ideally you want a frequency that is 10 to 20X higher than the highest frequency you want to reproduce. But higher frequency designs are even more challenging. The above Mark Levinson uses a clever multi-phase interleaving solution that has an effective frequency of 2 Mhz. Compare that to ~300KH frequency of typical ICE amp uses in mass market consumer products. (See a bit more here: http://www.whatsbestforum.com/showth...ll=1#post39815).

Integrating the DAC for the most part, is frowned upon. I have to run so no time to explain the details now. But as intuitive as that might sound, you are better off using a conventional DAC and then amplifying that using class D amp topology. Remember, class D does not mean digital! It was just the next letter in the alphabet.

The typical answer to all things distorted in amps is to use feedback. And lots of it in case of class D amps. I won't get into the argument relative to merits of feedback here . But suffice it to say, an amp topology needs to be good and performant without resorting to ton of feedback.

Net, net, it is easy to get efficiency of class D amps. Getting the full fidelity though, is much, much harder. So that is the reason the whole world hasn't gone there .


There are some great sounding Class D solutions..
Their refinements include an analog front end and ultilize afeedback system that reduces power supply dead time..

A couple of points of reference are Intersil (D2) Ice Power (B&O) and Zetex..

Just my $0.02...
M Code is offline  
post #9 of 98 Old 03-12-2011, 02:38 PM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 18,375
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 981 Post(s)
Liked: 547
Quote:
Originally Posted by M Code View Post

There are some great sounding Class D solutions.

Clearly there are. I gave one example in the Mark Levinson above. That is a 125 pound mono amp btw!

At the same time, there are also a lot of not so good sounding class D amps. I have heard B&O Ice amps driven by clean linear power supply but yet be noisy, have poor high-frequency response, and generally not too pleasant. To be sure, they put out a ton of power so if you just listen to dynamics, they sound fine. But at detail level, they might now.

Note also that per my post, class D amps (outside of the extreme design of the Mark Levinson) can be highly load dependent. So a good sounding system can sound worse with a different speaker and cabling So that makes it hard to generalize their performance.

Amir
Retired Technology Insider
Founder, Madrona Digital
"Insist on Quality Engineering"
amirm is online now  
post #10 of 98 Old 03-14-2011, 05:51 PM - Thread Starter
 
HiFiFun's Avatar
 
Join Date: Nov 2008
Location: USA
Posts: 2,209
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 17
So where do we start to cleanup with an ultra expensive analog amplifier in an all-digital thread?

Lets start with the review in HomeTheaterReview.com that Harman/Levinson recommends. Its titled:

Mark Levinson No 53 Hybrid Digital Power Amplifier Reviewed
.
.
.
The first comment by read karbunkle corrects the inaccuracies:
"It may be semantics, but as far as I can see, this is NOT a digital amplifier. It is a "Class D" amplifier, judging from the review and considering ML's reputation it'll probably be a bloody good one. But digital, no. On the Mark Levinson site there's no mention of it being a digital amp either.

To my knowledge, there are only three truly digital amps: A sharp from 10 years ago or so, the Tact/Lyngdorf Audio Millenium and the recently introduced NAD M2.
The difference; True digital amplifiers will take a digital input and convert the PCM stream directly into a PWM stream which switches the output devices. No need for pre-amp, power amp or D/A converter. These amps are all these rolled into one and all are also Class D amps.

The difference between Class D and Digital amplifiers will become more important over time, so the two shouldn't be confused or be used interchangeable. Nice though the review is, Andrew Robinson would do well to use either term appropriately."
------
The reviewer then response:
Karbunkle,
"You are correct, the No 53 is not a digital amp, it's a hybrid amp using bits of digital amp technology and classic amplifier technology together to create its unique albeit non-digital sound."
-----

So what is an ALL-analog amp doing this "true All-Digital Thread"? This is exactly what makes AVS a good read is the irony: the poster is guilty of what he accuses others of: only a novice would assume the D in Class D stands for digital. Therefore its a red-herring.

Amir, you have a unique path being a former Microsoft vice president turned salesman with a track record for picking the lesser technology. What do you think of Blu-ray 3D? i luv it!

The hardest part was looking for this amplifiers oil-rich dictator price. Hopefully the reviewer got this right:
"However, it was the No 53's price that took me by surprise at $50,000 for the pair."

http://hometheaterreview.com/mark-le...fier-reviewed/
HiFiFun is offline  
post #11 of 98 Old 03-14-2011, 06:23 PM - Thread Starter
 
HiFiFun's Avatar
 
Join Date: Nov 2008
Location: USA
Posts: 2,209
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 17
So what would it cost to equip a home theater with seven channels of amplification:

Samsung All-digital 1.4 HDMI receiver with seven 120watts $300

Mark Levinson NO 53 $25,000 x 7 = $175,000..
Add in the AV controller and cables for $25,000

Total $200,000

Let the gentle reader make there own judgment behind the motivation for these posts.
My goal is to raise the awareness of state-of-the-art ALL digital Class D receivers to the average Joe consumer. I'm not trying to make a buck. It's just that no one in the industry is stepping forward to explain the benefits in cost, weight, power consumption, reliability and price.

Most likely, the industries stumbling block is they can't figure out they can add value. For the record I have never cared for analog class D amplifiers because the Radio Frequency Interference (RFI) seriously degrades ALL the analog stages in the surrounding gear. Again don't mix digital with analog, especially analog processing with class D amplifiers.

Over the past 12 years Pulsus Corporation has integrated a single system-on-chip solution for multi-channel AV systems such as 5.1 channels, 6.1 channels or 7.1 channels. Combine this extremely high level of integration with high-power efficiency and reduced parts count. The Samsung receiver wins my award as the most significant product since the Sony PS3. (Remember Sony fired the designer for that design!) Anyway too much irony make me choke-up as people do the craziest things to make a buck.

Technically the Pulsus PS9830B is 8 channels and 3 additional channels PCM to PWM modulator with full 8 channel sample rate converter and pre-amplifier functions. It features 24-bit audio digital-to-digital converters and over-sampling digital filters, a sample rate converter,an audio DSP that functions as an on-chip pre-amplifier with equalizer, volume control, bass management and automatic gain limiting functions.

The sample rate converter is allowing for input of up to 96kHz, 8 channel and 192kHz, 2 channel. It automatically detects the sampling frequency and converts input sample rate to internal 192kHz sample rate. It makes easy for user to handle this chip.

The PS9830B accepts industry-standard audio data formats with 16- to 24-bit audio data. Sampling rates of up to 192 kHz are supported.
HiFiFun is offline  
post #12 of 98 Old 03-18-2011, 08:55 AM - Thread Starter
 
HiFiFun's Avatar
 
Join Date: Nov 2008
Location: USA
Posts: 2,209
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 17
One critically important area largely ignored in high-end audio is system level power conditioning, grounding and shielding techniques. The emphasis here, for the first time, will be placed upon audible superior system integration strategies. However such blanket guidance is only possible when transitioning to an all-digital system.

First lets discuss the gerbil-on-a-wheel issues in the current mixed signal (analog/digital) marketplace: Since traditional advertiser supported reviewers focus on selling an individual component, typical system optimizations and set-up strategies are seldom discussed (except for them foolishly expensive tweaks, like $300 cables). This is partially true because every system is unique in location, components, cables, imperfect connections, defective components and quality of A/C power. Further, the more components, the more complex (and non-ideal) the system grounding becomes, with an increased risk of audible anomalies being introduced. Typically H/W integration issues are bound to occur, yet there is no way to test, identify and correct; other than to listen; which is quite impractical.

Worse at the component level, each designer implements their own philosophical grounding schemes internally and makes assumptions about how it will interface to other components. Since there is no perfect ground, every analog circuit stage is susceptible to noise and interference besides generating its own distortion. Grounding techniques are also frequency dependent. A 60Hz star ground is ineffective for digital square waves/radio frequencies.

The traditional wisdom has been that expensive, separate (analog) components sound better. This philosophy is partially true as manufactures cut corners where the consumer is not likely to notice (this is why marketing departments sell an image and not a product). However this is worst case for noise and grounding issues to develop. No wonder going the analog separates route has lost popularity with the increasing levels of digitally generated interference.

One of the industries response to this unpleasant sound quality is, it must be caused by simple frequency response deviations. So add-in huge amounts of tone-control Digital Signal Processsing (DSP) which, in reality, only exasperates the harsh digital interference coupling into the analog stages. But it did raised profits and add "value". So what is music sounds like crap? its compressed, resampled, mp3 anyway!

Traditional sixty hertz sine wave power supplies are far from the ideal way to deliver power to audio power amplifiers as they are slow to respond. As everyone knows, they are big, heavy and expensive. The faster the high current diodes switch, the more EMI they generate. Electrolytic capacitors exhibit the worst behavior of all capacitor types. Capacitors all have their own resonance frequency above which they worsen the problem they are supposed to solve.
At the most basic level, all A/C signals (analog and digital) generate an electro-magnetic field. There are all sorts of parasitic coupling effects occurring between adjacent analog and digital circuitry.

Traditional analog amplifiers require negative feedback. Negative feedback is analogous to a cat trying to catch its tail. Analog circuit designers view it as a necessary evil, with radio interference mucking-up the theory.
The FCC only requires a components external emissions to be measured. There are no standards for equipments susceptibility to external emissions. The official solution is if there is a problem, then move it.
But what if the problem is at a low level, enough to degrade performance but not be overtly noticeable? This is what consumers are hearing and why there are many disagreements. Usually there are multiple issues occurring simultaneously.
I've only touch upon this analog can-of-worms.

Because of these largely analog issues the music loving consumer has spent vast sums of money constantly upgrading with no long-term listening satisfaction end in sight. Living with a mixed signal system (digital and analog) system is like being in purgatory or limbo. Forgive me of my sins and get me outta here!

How about some new technology solutions?

The Superior ALL-Digital Solution
Enter the ALL-digital processing age, first with high frequency, quasi-resonant zero-current/zero-voltage switching power supplies. There are many designs which greatly reduce EMI emissions, increase efficiency and lower weight and size by increasing the soft-mode switching up into the megahertz range. For example, the traditional power supply weight can be reduced from 20 to 2lbs, a 10x reduction. The size can easily be reduced by a factor of four. Now the receiver case can be made corresponding smaller, without even taking into account the more efficient output stage and its much smaller heatsink.

ALL digital amplifiers require no feedback. Is that simple or what?

ALL digital systems are more relaxed about grounding and interference. Who cares if a bit get in? Its ignored! Expensive perfectionist analog designs are no longer required. However precise digital clock timing is required, but is that is inexpensive.
For example I largely use the Samsung HDMI 1.4 chips which are very clean. It is well known that Samsung LCD displays offer extremely clear and natural picture, especially for 2011. HDMI is for 24 bit digital audio too. So why is traditional high-end audio stuck on inferior 1970's S/PDIF digital?

Sorry for rambling, but protecting claimed -120db microvolt analog signals from 50 volt digital interference is an expensive, crazy, losing battle. It must stop. The superior technology is here today and its dirt-cheap.

Now that analog processing has been dumped (well almost), my next post will discuss my simplified and higher performance all-digital system grounding strategies. Remember, the goal (and the humorous irony) is to make an all-digital system sound like perfectly-pure analog.
HiFiFun is offline  
post #13 of 98 Old 03-18-2011, 11:28 AM
AVS Special Member
 
M Code's Avatar
 
Join Date: Mar 2003
Location: Joshua Tree, CA
Posts: 9,967
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 152 Post(s)
Liked: 133
Quote:
Originally Posted by HiFiFun View Post

One critically important area largely ignored in high-end audio is system level power conditioning, grounding and shielding techniques. The emphasis here, for the first time, will be placed upon audible superior system integration strategies. However such blanket guidance is only possible when transitioning to an all-digital system.

First lets discuss the gerbil-on-a-wheel issues in the current mixed signal (analog/digital) marketplace: Since traditional advertiser supported reviewers focus on selling an individual component, typical system optimizations and set-up strategies are seldom discussed (except for them foolishly expensive tweaks, like $300 cables). This is partially true because every system is unique in location, components, cables, imperfect connections, defective components and quality of A/C power. Further, the more components, the more complex (and non-ideal) the system grounding becomes, with an increased risk of audible anomalies being introduced. Typically H/W integration issues are bound to occur, yet there is no way to test, identify and correct; other than to listen; which is quite impractical.

Worse at the component level, each designer implements their own philosophical grounding schemes internally and makes assumptions about how it will interface to other components. Since there is no perfect ground, every analog circuit stage is susceptible to noise and interference besides generating its own distortion. Grounding techniques are also frequency dependent. A 60Hz star ground is ineffective for digital square waves/radio frequencies.

The traditional wisdom has been that expensive, separate (analog) components sound better. This philosophy is partially true as manufactures cut corners where the consumer is not likely to notice (this is why marketing departments sell an image and not a product). However this is worst case for noise and grounding issues to develop. No wonder going the analog separates route has lost popularity with the increasing levels of digitally generated interference.

One of the industries response to this unpleasant sound quality is, it must be caused by simple frequency response deviations. So add-in huge amounts of tone-control Digital Signal Processsing (DSP) which, in reality, only exasperates the harsh digital interference coupling into the analog stages. But it did raised profits and add "value". So what is music sounds like crap? its compressed, resampled, mp3 anyway!

Traditional sixty hertz sine wave power supplies are far from the ideal way to deliver power to audio power amplifiers as they are slow to respond. As everyone knows, they are big, heavy and expensive. The faster the high current diodes switch, the more EMI they generate. Electrolytic capacitors exhibit the worst behavior of all capacitor types. Capacitors all have their own resonance frequency above which they worsen the problem they are supposed to solve.
At the most basic level, all A/C signals (analog and digital) generate an electro-magnetic field. There are all sorts of parasitic coupling effects occurring between adjacent analog and digital circuitry.

Traditional analog amplifiers require negative feedback. Negative feedback is analogous to a cat trying to catch its tail. Analog circuit designers view it as a necessary evil, with radio interference mucking-up the theory.
The FCC only requires a components external emissions to be measured. There are no standards for equipments susceptibility to external emissions. The official solution is if there is a problem, then move it.
But what if the problem is at a low level, enough to degrade performance but not be overtly noticeable? This is what consumers are hearing and why there are many disagreements. Usually there are multiple issues occurring simultaneously.
I've only touch upon this analog can-of-worms.

Because of these largely analog issues the music loving consumer has spent vast sums of money constantly upgrading with no long-term listening satisfaction end in sight. Living with a mixed signal system (digital and analog) system is like being in purgatory or limbo. Forgive me of my sins and get me outta here!

How about some new technology solutions?

The Superior ALL-Digital Solution
Enter the ALL-digital processing age, first with high frequency, quasi-resonant zero-current/zero-voltage switching power supplies. There are many designs which greatly reduce EMI emissions, increase efficiency and lower weight and size by increasing the soft-mode switching up into the megahertz range. For example, the traditional power supply weight can be reduced from 20 to 2lbs, a 10x reduction. The size can easily be reduced by a factor of four. Now the receiver case can be made corresponding smaller, without even taking into account the more efficient output stage and its much smaller heatsink.

ALL digital amplifiers require no feedback. Is that simple or what?

ALL digital systems are more relaxed about grounding and interference. Who cares if a bit get in? Its ignored! Expensive perfectionist analog designs are no longer required. However precise digital clock timing is required, but is that is inexpensive.
For example I largely use the Samsung HDMI 1.4 chips which are very clean. It is well known that Samsung LCD displays offer extremely clear and natural picture, especially for 2011. HDMI is for 24 bit digital audio too. So why is traditional high-end audio stuck on inferior 1970's S/PDIF digital?

Sorry for rambling, but protecting claimed -120db microvolt analog signals from 50 volt digital interference is an expensive, crazy, losing battle. It must stop. The superior technology is here today and its dirt-cheap.

Now that analog processing has been dumped (well almost), my next post will discuss my simplified and higher performance all-digital system grounding strategies. Remember, the goal (and the humorous irony) is to make an all-digital system sound like perfectly-pure analog.

Do you work for Samsung or their advertising agency??

Are you aware that Harman/Kardon actually built & delivered about 7 years ago a line of Digital Path AVRs..
They used the Intersil (D2) solution which utilized a Digital In > Digital Out with the only conversion before the loudspeaker output..
In the last few years, the Class D solutions have been refined significantly and sonically can sound very good..

Just my $0.02...
M Code is offline  
post #14 of 98 Old 03-21-2011, 06:03 PM
AVS Addicted Member
 
noah katz's Avatar
 
Join Date: Apr 1999
Location: Mountain View, CA USA
Posts: 20,777
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 398 Post(s)
Liked: 183
Quote:
Originally Posted by HiFiFun View Post



Digital Class D Amplifiers
In 2011 the venerable analog conversion is finally avoided, with all-digital amplifiers containing built-in pcm-to-pwm converters and all-digital feedback The best example is the eight channels, 120 watt, HDMI 1.4 Samsung 700 receiver using Pulsus Korean technology. The great efficiency and reduced part count lowers the weight and price, beside offers superior sonics. What more could a consumer ask for?
http://www.pulsus.co.kr/english/products/pro_02_05.html

Texas Instruments has also developed an all-digital two channel reference design which features low distortion:
http://www.ti.com/ww/en/analog/tas5630/index.shtml

These advances in sonic fidelity and realism ...

I'm all for all-digital, but on what do you base your statements of sonic superiority?

I can't find reviews of the Samsung receiver, and don't know of any products using the TI reference design.

Can you point to any reviews

Noah
noah katz is online now  
post #15 of 98 Old 03-21-2011, 06:24 PM
AVS Addicted Member
 
penngray's Avatar
 
Join Date: Feb 2006
Location: Florida
Posts: 26,779
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 1 Post(s)
Liked: 30
Interesting stuff Im still waiting for the product that keeps everything in the digital domain, allowing me to create my XOs for my active designs (already in the digital domain) and keeps it digital right on pass amplification. At some point it has to convert to analog though so that each speaker driver can accept the signal. Speaker drivers ONLY accept analog signals the last I check.

It is not "open-minded" to reject knowledge - Bob Lee
penngray is offline  
post #16 of 98 Old 03-21-2011, 08:08 PM
AVS Addicted Member
 
MichaelJHuman's Avatar
 
Join Date: Nov 2006
Location: Colorado Springs, CO
Posts: 18,955
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 239 Post(s)
Liked: 118
Quote:
Originally Posted by penngray View Post
Interesting stuff Im still waiting for the product that keeps everything in the digital domain, allowing me to create my XOs for my active designs (already in the digital domain) and keeps it digital right on pass amplification. At some point it has to convert to analog though so that each speaker driver can accept the signal. Speaker drivers ONLY accept analog signals the last I check.
Of course, if you believe some people, a class D amp digital inputs, using PCM to PWM conversion is a power DAC. The implication being that it's digital up to the final filter, I guess (if you want to call PWM digital, or maybe it's just digital up to the conversion to PWM.)


"But this one goes up to 11"
MichaelJHuman is offline  
post #17 of 98 Old 03-22-2011, 02:20 PM
AVS Addicted Member
 
noah katz's Avatar
 
Join Date: Apr 1999
Location: Mountain View, CA USA
Posts: 20,777
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 398 Post(s)
Liked: 183
Quote:
Originally Posted by penngray View Post

At some point it has to convert to analog though so that each speaker driver can accept the signal.

Depending on the freq, that may require nothing more than a series inductor.

In fact, I think that's what class D amps do now.

Noah
noah katz is online now  
post #18 of 98 Old 03-22-2011, 02:25 PM
AVS Addicted Member
 
MichaelJHuman's Avatar
 
Join Date: Nov 2006
Location: Colorado Springs, CO
Posts: 18,955
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 239 Post(s)
Liked: 118
Quote:
Originally Posted by noah katz View Post

Depending on the freq, that may require nothing more than a series inductor.

In fact, I think that's what class D amps do now.

I think that's right. Feed the amplified PWM through a filter, and you magically get back audio. I always thought that was pretty cool.

"But this one goes up to 11"
MichaelJHuman is offline  
post #19 of 98 Old 04-10-2011, 08:00 PM - Thread Starter
 
HiFiFun's Avatar
 
Join Date: Nov 2008
Location: USA
Posts: 2,209
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 17
The 60HZ A/C power cycle is still analog and results in the transfer of energy to power our home theater systems. The main issue is the nominal sin wave cannot supply energy very fast. Sixty Hertz is to slow an update rate.

So I use the Furman Power Factor Pro which has a tuned time constant capacitor to monetarily supply enough energy during the sine wave nulls.

I also use a traditional Belkin power PF60 conditioner.

My primary system has a $5 6-Grounded Outlet In-Wall Adapter. The tap is designed to cover only an existing grounding "duplex" (two 3-prong outlets) wall outlet


Grounds can be relative or absolute. Here we use this wall outlet as the relative A/C power ground for the entire system.

The first branch is to connect the Furman Power Factor to this wall outlet.


The second branch connects the Belkin conditioner to the this wall outlet.

The third branch connects directly to the wall.

In my system I connect the class D subwoffer(s) straight to the third branch socket wall.

The Samsung ALL-digital receiver is connected to the Belkin's high power outlet.

The video display, htpc, blu-ray player are connected all connected to the Furman on the second branch. I use a cheater plug on the video display. That is it!

You are free to connect any components to any branch and experiment to see how dramatically the sound and picture quality changes. Its a make or break situation, even for an ALL-digital system. The less components, the higher the chance for achieving the best sound.
I'm amazed that these basics is not common knowledge, but help explains the endless debates, confusion and upgrades, year after year.
HiFiFun is offline  
post #20 of 98 Old 04-10-2011, 09:27 PM
AVS Addicted Member
 
MichaelJHuman's Avatar
 
Join Date: Nov 2006
Location: Colorado Springs, CO
Posts: 18,955
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 239 Post(s)
Liked: 118
I did not think amps relied on the 60hz charge cycle. With capacitors in the power supply, the AC voltage swing is evened out. And it has to be, or your supply voltage would be a lot power.

I assume a similar reasoning applies to switch mode supplies, but I don't know much about them.

I am also curious about subwoffers

"But this one goes up to 11"
MichaelJHuman is offline  
post #21 of 98 Old 04-11-2011, 04:04 PM
AVS Special Member
 
jvillain's Avatar
 
Join Date: Mar 2003
Location: Burnaby, BC
Posts: 2,110
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 10
From my standpoint with the exception of media all analogue still sounds better than all digital. Sit in a control room of an analogue studio some time and listen to what is coming out of the speakers when it is being fed by some thing being captured in the recording area. Analogue from end to end can't be beat.

Digital keeps getting better. But some thing is still lost in the A/D conversion.
jvillain is offline  
post #22 of 98 Old 04-11-2011, 06:19 PM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 18,375
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 981 Post(s)
Liked: 547
Quote:
Originally Posted by HiFiFun View Post
So what is an ALL-analog amp doing this "true All-Digital Thread"?
No 53 is not "ALL-analog." It uses a multi-phase class-D configuration called class-I. For now, just think of it is a very clever way to push up class-D frequency to sharply reduce its high frequency distortion. The output stage transistors fully operate in switching more and not linear as class A/AB,etc would use. That is how it gets its high efficiency.

Where it differs from many other amps is that its power supply is linear, not switching. As a matter of fact, the power supply is shared with the Reference 532. The choice of linear power supply does contribute to better fidelity but does NOT change the configuration of the amplifier.

Not sure what the reviewer had in mind when calling it "hybrid" but it does sound good. Here is a technical paper on its design btw: http://173.203.156.112/Uploads/Files...10_5.17.10.pdf. Don't see a reference to "hybrid" in there either.

Quote:
Amir, you have a unique path being a former Microsoft vice president turned salesman with a track record for picking the lesser technology. What do you think of Blu-ray 3D? i luv it!
Oh don't be mean. I still have a smile on face, when I look at every blu-ray player with technology out of my team in it, together with billions of other A/V, phone and computer products that use the same. Winning the third technical Emmy was a wonderful thing too. Bad pictures of me here: http://jobsblog.com/blog/the-emmy-go...-say-it-again/

So do I walk around with my head down? No.

As for 3-D, don't get me started here.

Quote:
"However, it was the No 53's price that took me by surprise at $50,000 for the pair."
It is expensive. I am too darn cheap to buy one myself . But it does show for the first time that class D can perform, and perform exceptionally well. I recently compared it to two other class D amps (Crown and D-Premier) and it really is in a class by itself. And gorgeous too.

Amir
Retired Technology Insider
Founder, Madrona Digital
"Insist on Quality Engineering"
amirm is online now  
post #23 of 98 Old 04-11-2011, 07:33 PM
AVS Addicted Member
 
MichaelJHuman's Avatar
 
Join Date: Nov 2006
Location: Colorado Springs, CO
Posts: 18,955
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 239 Post(s)
Liked: 118
I have long thought that some people reject digital just because of their own biases with the term digital.

For example, when people say a class D amplifier sounds "digital", which I have seen a few times, doesn't that suggest a bias? Just like people saying CDs sound "digital."

"But this one goes up to 11"
MichaelJHuman is offline  
post #24 of 98 Old 04-11-2011, 07:52 PM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 18,375
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 981 Post(s)
Liked: 547
Quote:
Originally Posted by MichaelJHuman View Post
For example, when people say a class D amplifier sounds "digital", which I have seen a few times, doesn't that suggest a bias? Just like people saying CDs sound "digital."
You are right that some people use it to say something doesn't sound fluid. I think it is poor choice of words as we only hear digital audio as a continuous form as opposed to its digital origins.

Amir
Retired Technology Insider
Founder, Madrona Digital
"Insist on Quality Engineering"
amirm is online now  
post #25 of 98 Old 04-11-2011, 09:44 PM
AVS Addicted Member
 
MichaelJHuman's Avatar
 
Join Date: Nov 2006
Location: Colorado Springs, CO
Posts: 18,955
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 239 Post(s)
Liked: 118
I guess, in my mind, a "digital" sound would mean you could hear distortion from inadequate sampling rate.

I remember this paper where they (Meridian?) claimed 44.1 khz was probably not high enough. But it's never been clear to me what a higher sampling rate would sound like - meaning if you could AB the two, then you would have some idea why they sounded different.

"But this one goes up to 11"
MichaelJHuman is offline  
post #26 of 98 Old 04-11-2011, 10:01 PM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 18,375
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 981 Post(s)
Liked: 547
Quote:
Originally Posted by MichaelJHuman View Post

I remember this paper where they (Meridian?) claimed 44.1 khz was probably not high enough. But it's never been clear to me what a higher sampling rate would sound like - meaning if you could AB the two, then you would have some idea why they sounded different.

The paper you are speaking of is one of the best ever written on this topic by Bob Stuart: http://www.meridian-audio.com/w_paper/multips3.pdf

He shows that using noise-shaping and taking account psychoacoustics, we should be able to live with 14 bits and 58 Khz sampling rate.

Turns out that sampling rate and resolution are interdependent. For every doubling of sampling rate, you gain half a bit. So we should always talk about them in pairs.

Given Bob's work, we can see why CD's 16-bit, 44.1 Khz is just below where we like to be.

Amir
Retired Technology Insider
Founder, Madrona Digital
"Insist on Quality Engineering"
amirm is online now  
post #27 of 98 Old 04-11-2011, 10:29 PM
AVS Addicted Member
 
MichaelJHuman's Avatar
 
Join Date: Nov 2006
Location: Colorado Springs, CO
Posts: 18,955
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 239 Post(s)
Liked: 118
Yep, that was the paper. I have read it a few times. Takes time to absorb it all, if concepts like "noise shaping" are not self evident (which they are not, to me...I am sure I don't understand all of it 100%, but the basic concepts are very accessible to anyone with some basic background in digital audio theory.)

"But this one goes up to 11"
MichaelJHuman is offline  
post #28 of 98 Old 04-11-2011, 10:33 PM
AVS Addicted Member
 
amirm's Avatar
 
Join Date: Jan 2002
Location: Washington State
Posts: 18,375
Mentioned: 4 Post(s)
Tagged: 0 Thread(s)
Quoted: 981 Post(s)
Liked: 547
Talk about negative connotations. "noise" anything sounds like it should make things worse. Yet, in digital audio, it is as close to a miracle as one can get! Formats like SACD would not exist without it.

Let me know if it would be useful for me to explain it and I will be happy to do so.

Amir
Retired Technology Insider
Founder, Madrona Digital
"Insist on Quality Engineering"
amirm is online now  
post #29 of 98 Old 04-11-2011, 11:11 PM
Advanced Member
 
xianthax's Avatar
 
Join Date: Oct 2008
Posts: 514
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
I see its time for the bi monthly HiFiFun thread regarding his marketing job at Samsung hyping the HW-C700.

This has already been beat to death:

Its not all digital (its using International Rectifier analog input Class D drivers)

Even if it was, conversion is not the primary source of distortion in a Class D design.

Here is HiFiFun pawning off a review of the NAD M2 as a review of the HW-C700

The biggest advancements in Class D design are coming from improved feedback mechanisms. Some are all digital (Zetex DFAA used in the NAD M2) some are not. Feedback is much easier to implement with an analog input to the Class D driver, but its difficult to correct for everything with an analog path. The advantage of 'All Digital' isn't that you avoid the digital to analog conversion, but rather that you can implement more feedback mechanisms with a DSP monitoring multiple points and correcting for errors.
xianthax is offline  
post #30 of 98 Old 04-11-2011, 11:16 PM
Advanced Member
 
xianthax's Avatar
 
Join Date: Oct 2008
Posts: 514
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 0 Post(s)
Liked: 12
Quote:
Originally Posted by noah katz View Post

Depending on the freq, that may require nothing more than a series inductor.

In fact, I think that's what class D amps do now.

At low power no filter is needed with some topologies, the inductance of the speaker itself is enough to provide the needed filtering. Its likely if you have a modern smart phone there is a filter-less class D amp driving the speaker.

Actually even in home applications the biggest reason to filter the class D output is that large, high speed switching currents being sent down a long speaker wire results in a TON of EMI emissions. It would be impossible for the device to pass FCC part 15 without a filter.

The filter is generally just an LC filter, a series inductor alone isn't enough, you need a cap as well.
xianthax is offline  
Reply Receivers, Amps, and Processors

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off