What makes some digital sources sound more analog ?

To make cd's sound like analogue devices....one needs a fat purse, fatter the better ! :-)

Manek

Sound like analogue – can be achieved without breaking a bank. There are a number or sources/solutions in the sub 100k INR which do this.
Sound like the real thing – May need to break a bank or DIY

Well, my question was intended on the technical sides of it.
Why some digital gear sounds more analog ?
There should be some macro level parameters which would be the deciding factors I suppose. Jitter is one of them...that I have read. What else ?

I would give real time example.
A couple of days back I had a lot of time to kill so I thought of trying something 'useless'. I have a entry level Pioneer DVDP (DV-535) which is tucked away in some corner of my house and is pulled out only rarely to play some DVD movies for the family. This time I wanted to use it as a transport with my DAC and see what happens .

The results were as expected...the combo sounded very 'digital'. Yes, there were lot details, nice dynamic sound, good speed and rhythm but the tones had sibilance in some areas, the sound was edgy and the overall presentation was not at all relaxed. One would get fatigued within no time.
I was happy that my investment in a good standalone transport (TEAC VRDS) was a wise decision.

Out of curiosity of I connected the Pioneer DVDP as a standalone CDP. I was expecting something horrible this time. But I was taken aback when I heard a smooth fatigue-free very analog like sound out of this DVDP when used this way . The reason I used the phrase 'analog like' is because music flowed in naturally, very unforced and fluid. The tones were good and real. Ok, the details were less and so were the overall dynamics. Micro details were not even close. But that hardly mattered.
My DAC is a Reimyo and I think it is safe to assume that it would be way better than the cheapo DAC inside the DVDP in every way but then things turned out to be much more complex. What is it that made the Pioneer to sound more natural than the Pioneer + Reimyo combo ?

Things like these are no way related to pocket or bank breaking theory. It would be good to know whats going on in such cases .

Abhi,
I assume this is the same DVP you picked from me.
You noticed a natural unforced sound but lacking in dynamics, detail, micro/macro detail. Your answer is right there
It is sacrificing all the finer details and giving you a fluid-unforced sound. I guess when you do hi-end your transport and Dac has to match perfectly. Or rather complement each other. I’d rather not go too technical here because I do not know much.
Btw: I remember this very same dvd player used as a transport by a reviewer on six moons many moons back
I had picked this up many years back at a “princely sum” of 10k

Hi Vinny,
Yes, it is the same DVDP I picked up from you.
As for details...I would not say less detail = more natural or more detail = more analytical/digital---Not at all.

I have heard sources with poor detail and dynamics and sounding very digital, grainy and irritating.

I have also heard very detailed sound which was not at all analytical or forced but very natural and easy on ears...this had happened when I had inserted the MFA (Music first audio) passive preamp into my system sometimes back. The sound was unimaginably detailed yet very easy on ears and should I say fluid.

I do understand that some equipments are designed to sound pleasant by chopping of the highs and rounding off the edges but I dont think the Pioneer is a sophisticated enough machine which will have such 'designers' voicing . It just sounded plain and natural.

My point is, there must be something wrong happening after the bits are read from the disc and by the time the bits reach the DAC. That is making all the difference.

square_wave schrieb:

May need to break a bank or DIY

Haha just as an aside - Square did you say break and bank or do it yourself! Hmm you are one brave one I say to break a bank yourself, haha.

Sorry bad joke

Hi abhi,

Maybe your expectations have changed. You recognise that harmony and integrity in music is more important for the enjoyment of it.

Not all equipment, even real expensive ones, manage to give you that. Not all companies even strive to preserve this in music. So, price is no guarantee.

Maybe, you're just discovering that.

Regards,
Viren

abhi.pani schrieb:

What is it that made the Pioneer to sound more natural than the Pioneer + Reimyo combo ?

maybe it is just system synergy ?
i think you should buy a Reimyo transport and link them with Harmonic cables

Arj schrieb:

maybe it is just system synergy ?

This is a topic for another thread & another heated discussion!

abhi.pani schrieb:

Hello Buddies, In the last few months I have been playing a lot around "Source". Fortunately or unfortunately I could also try some analog sources in my setup. A Nakamichi DR-10 Tape Deck and a Technics TT. In the digital domain I could try a number of Transports with my DAC. The "thing" that I have realized is when the source is "right" everything falls into place. Music comes naturally. You dont stress yourself to find "musicality" in your setup. Cables, Power conditioning, Isolation etc all seems very secondary and not really a hindrance between you and music. Not that they are unimportant just that you think about them a lot less.

alright! yet another step forward in your life as an audiophile.

abhi.pani schrieb:

But it is true that some digital sources do sound more analog then others...why ?

complicated subject! Sivat has a lot of experience in this field & I'm hoping that he'll jump in & offer some thoughts (but I have my doubts ).

abhi.pani schrieb:

1. Jitter: I have only read about them. But frankly I do not know the exact sonic impact that a higher jitter transport has over a lower jitter transport. But would love to know more..as in what exactly does a low jitter transport translate into sonically ? And what are the ways to reduce them.

Jitter is, in the digital world, the uncertainty in the clock edge. The analog equivalent is called phase noise. There's much written on this subject on the web & the amount of available reading is limited only by your willingness to read the volumes written!
There are several sources of jitter some of them being an imprecise clock generated using a large variation clock xtal. Another source of jitter is poor slew rate in output buffers - when the buffer output cannot transition from logic 1 --> logic 0 or vice-versa & the signal "drags its feet" the ensuing electronics transitions at variable times creating an uncertainty in transition edge. Jitter can also be created when amplitude noise (say, noise on the power supply) converts itself phase noise (they call this AM to PM conversion). This happens because as the power supply varies, the threshold voltage at which logic transitions from logic 0/1 --> 1/0 varies (the transition voltage is a function of power supply, often a strong function) thereby creating clock edge uncertainty. Similarly, jitter can be created due to ground bounce i.e. ground not being a rock-solid 0V but having noise riding on it. This time around the logic circuit might not transition at the designed value but might need a higher input voltage. Again, the end effect is jitter on the signal. There are more sources but this gives you an idea that IMPLEMENTATION of a CDP/transport is (nearly) everything!
Jitter causes the DAC to make errors - a logic 1 is misinterpretted as a logic 0 & vice-versa because the data input to the DAC & the corresponding clock signal are no longer lined up as the designer intended. Distortion increases due to these errors & they manifest themselves as (generally) "digititis". Almost every audiophile has his/her favourite description for this - "when will digital get soul?", fatiguing, hardness, too much shimmer in the cymbals, etc, etc, etc.......

abhi.pani schrieb:

2. Oversampling: In general non-oversampling DACs are considered to be more analog sounding in nature. But I would not really want to believe that oversampling DACs cant sound equally "right".

That's right! OS DACs can sound correct. There are a few in the commercial market such as Wadia (some on this forum do not think so but each of us is entitled to his opinion ), Esoteric, TEAC, EMM Labs, Muse, Accuphase, Meridian, etc, etc.
In non-OS DACs the DAC reconstruction filter naturally rolls off the signal. I believe that the designers do not make any attempt to transition the filter into the stop-band in the 2.05KHz that is officially assigned. The roll-off is much more gradual & that seems to make a lot of difference - no in-band distortion. Others use vacuum tubes to the job since tubes have a limited bandwidth due to their physical construction.
The key to OS DACs I believe is the OS algorithm - the algorithm to fill in the blanks when the incoming 44.1KHz signal is over/up-sampled. Boothroyd Stuart Meridian has built a whole company on this philiosophy & they have several patented OS algorithms.
When OS is done wrong the reproduced music seems to have too much air around instruments & artists & in general a very artificial feel to the played-back music - something you'd hear & say that it was too good to be true (using your experience of going to live concerts).

abhi.pani schrieb:

3. Error Correction: Again I have only read about it that transports which are very well built have fewer reading errors so they undergo fewer error correction hence preserving the integrity of the information.

In general, read-back lasers these days are very good & even the cheapest DVD player can read a disk back w/ practically zero errors.
There are several manuf, most notably TEAC, that have designed & built the sturdiest of transports where all the mechanically moving parts are damped & geared minimize read errors.
In modern HDD based transports, CDs are read & re-read many, many times until zero errors are obtained. This ensures the best (minimal digititis) digital playback.
However, whenever there are errors & error correction is done, different algorithms used have different effect on the sound.

abhi.pani schrieb:

4. Deliberate Coloration: Some CDPs are deliberately voiced to sound warm using tubes for coloration (Unison Research for example) but thats hopeless. We are not discussing them.

Besides the company you cited, AudioMeca, Cary, Lector, Raysonic fall into this category.
There are some of us who believe that tubes do not belong to the digital playback realm & there are others who do. To each, his own, I suppose.

abhi.pani schrieb:

There could be a lot of other factors, please do bring them forward. I firmly believe that CDs can sound "right" and can give a good dose of what a good analog source can deliver. Lets start by discussing what it takes to achieve it !!

The other interface that is of importance is that DAC output interface. The DAC o/p is usually a current & there are ways to convert it to digital such as simply using a resistor, I2S interface, using small-signal transformers. I believe that Sivat implemented each one of them in turn in 2007-2008 & concluded that I2S was the best interface. Maybe he'll join this thread & share (or point us to the existing thread where this was already discussed)??

Yet another important aspect is parts selection - there is merit in selecting Blackgate caps over XYZ brand, a particular audio-grade power xformer over an AC xformer, one DAC IC over another, etc, etc. However, mere parts selection will not overcome a bad implementation of the digital playback machine. The implementation is key.

Still another aspect is attention to the power supplies - making sure that the analog & digital PSUs do not talk to each other, making sure that they are low impedance, making sure that they can supply the RMS & peak currents with headroom to spare, making sure that the ripple is minimized, wire routing of the power supplies so that all the work done to make them wold-class is not un-done by a hodge-podge of wire routing, making sure that ground is a really low impedance point so that it sinks all the nasties like it's supposed to.

Some food for thought for you......

Arj schrieb:

abhi.pani schrieb: What is it that made the Pioneer to sound more natural than the Pioneer + Reimyo combo ? i think you should buy a Reimyo transport and link them with Harmonic cables

I'll agree ...or may be a pi-tracer

One more point to consider I think without malice or offence to anyone.....

The designers knowledge of music, its structures and his ability to marry his knowledge of music with electtronics. If he plays an instrument, well even better....that would help in making the device sound closer to the real thing.

A musically trained ear would help I feel. That old acronym "PRAT" would take a new meaning.

Manek

Manek schrieb:

One more point to consider I think without malice or offence to anyone..... The designers knowledge of music, its structures and his ability to marry his knowledge of music with electtronics. If he plays an instrument, well even better....that would help in making the device sound closer to the real thing. A musically trained ear would help I feel. That old acronym "PRAT" would take a new meaning. Manek

Agree with you here.
Reflects an article by the Zanden designer which I happen to read recently. According to him there are unknown phenomenon due to various implementations/choices which cannot be measured accurately or rather “the measurement does not reflect how it will sound”. A musician’s ear is required here to maintain a balance.

Square

I haven't read that article, could you send me the link ?

I really would like to read it.

Manek

Many of you agree that cables make a difference in a equipment right ? ....well, if you do not agree, then please do not read further

Cable is "part" of the "interface" between two components. Just as important as cables are - the other components involved in the interface between two equipment is also very very critical.

Hence it would be common-sense to assume that all products take good consideration of this aspect...but alas, this is what many commercially available product screw up (Simply because it is easy make fool of people with some fancy numbers.... )

The reason why i'm saying is

-> I do not believe a "0.00%" distrotion level compared to "0.00000%" distotion at the digital secion of your DAC is going to matter much....infact i do not believe it will make any sonic different at all (I'm betting my reputation on this comment )

-> One thing that always attracted me towards tubes were NOT really thier "sonic" qualities by nature ( i feel they just sound similar to solid-state, if implemented correctly), but -in the ability of tube design to provide a better "interface" between two equipment. The end-product just feels far more at ease driving the equipment on the other side.

The above is based on my limited experience. However, one thing i'm very convinced is that - A simple circuit (like PCM1704) with a cost-no-bar output stage will be as good as a DAC can get.

I feel the over-sampling and artificial predictive algorithm only color the sound and affects the overall harmonics of the sound....

I think products like Reimyo are a good example of this ..

sivat schrieb:

I think products like Reimyo are a good example of this ..

ahaa...because I am not buying a pie-tracer, you are asking me to ditch the Reimyo ..

sivat schrieb:

Many of you agree that cables make a difference in a equipment right ? ....well, if you do not agree, then please do not read further Cable is "part" of the "interface" between two components.

While I am a out and out cable believer, I am yet come across a cable which can make an analog source (TT for example) sound digital. Thats the reason I started believing that cables are a not the solution to fix the anomalies of a digital source.

only way of making an analogue into digital is by putting an ADC in the line.. no cable to do that:) however Bad it is

Manek schrieb:

Square I haven't read that article, could you send me the link ? I really would like to read it. Manek

Here's the link.
http://www.audioasylum.com/scripts/d.pl?audio/faq/joes-tubes.html

abhi.pani schrieb:

sivat schrieb: I think products like Reimyo are a good example of this .. ahaa...because I am not buying a pie-tracer, you are asking me to ditch the Reimyo .. :P

I think you mis-understood.

Reimyo is a good example of a simple digital circitury with an excellent output stage

abhi.pani schrieb:

sivat schrieb: Many of you agree that cables make a difference in a equipment right ? ....well, if you do not agree, then please do not read further Cable is "part" of the "interface" between two components. While I am a out and out cable believer, I am yet come across a cable which can make an analog source (TT for example) sound digital. Thats the reason I started believing that cables are a not the solution to fix the anomalies of a digital source.

I'm not saying Cable is a solution. Please read more carefully

sivat schrieb:

abhi.pani schrieb: sivat schrieb: I think products like Reimyo are a good example of this .. ahaa...because I am not buying a pie-tracer, you are asking me to ditch the Reimyo .. :P I think you mis-understood. Reimyo is a good example of a simple digital circitury with an excellent output stage

Oh...ok...I was almost about to buy the pie-tracer, because I love the Reimyo .

On a more serious note, what I understand from your post, you seem to believe in non-oversampling and its likes for the source. But AFAIK, PCM-1704 is also an oversampling DAC (please correct me here)!!!

sivat schrieb:

abhi.pani schrieb: sivat schrieb: Many of you agree that cables make a difference in a equipment right ? ....well, if you do not agree, then please do not read further Cable is "part" of the "interface" between two components. While I am a out and out cable believer, I am yet come across a cable which can make an analog source (TT for example) sound digital. Thats the reason I started believing that cables are a not the solution to fix the anomalies of a digital source. I'm not saying Cable is a solution. Please read more carefully

Ya, I know, I was only sharing my experiences regarding cables.

Yes.Most PCM1704 implementation involves the use of DF1704 digital filter. You can control the level of ditial filtering you would like to do with this chip.

I have experimented with various settings. You could try and see if you hear any difference from exterme levels of oversampling.....atleast i did not.

The other problem is the that most of the modern chips do not directly work with a native 16 bit I2S signal. So you have no choice but to oversample to atleast 18 or 24 bit !!