DSD comparison to PCM.

[LL21]

I can contact both Lynn and Rene Jaeger (Berkley designer) and see if they will post here. Lynn likely will, Rene', not so sure, but I will prod him tomorrow.

Bud

Please do! Would be interesting to read more from Lynn...the article posted earlier was interesting reading...i read it 3x over a couple days to try to get thru it properly.

 

[opus111]

I have read someone else (Mike Lavigne?) once post something to the effect that SET amps are surprisingly linear at low levels...its just that expecting them to produce linearity at higher levels of output is not what they were designed for...thus one must DEFINITELY look at system matching to get the best out of SETS for what they do.

That's interesting - I haven't knowingly heard SETs either, I'm just surmising that this is what they're good at, given they are simple designs in terms of the number of active elements.

I believe Lynn is suggesting Class A amplification (zero feedback) in general follows along similar basic principals which he champions regarding low levels of disortion, no feedback loops which have their challenges, no switching from Class A to B...but again, he points out that unless you want to drive a furnace...Class A DOES have to switch at some point unless we remain at lower levels of output.

Seems to me that feedback vs no feedback is where the discussions have the tendency to get rather heated. My take on feedback is that its presence is correlated with poorer low level linearity, not that using feedback causes the problems. After all, the 'no feedback' brigade often (not always, depending on how hard-line they are) make a distinction between 'loop feedback' and 'local feedback' which to me is an entirely bogus one. If I have one overriding design principle (and I don't but for the sake of argument if I did) then it would be 'keep loops tight' - so the bigger the loop enclosed in the feedback the worse the problems are likely to be.

 

[LL21]

That's interesting - I haven't knowingly heard SETs either, I'm just surmising that this is what they're good at, given they are simple designs in terms of the number of active elements.

I have not heard Lamm, for example, and i do not know how they measure at low levels...but i certainly have not ever been given the impression by any review nor any comment that they are 'colored', distorted or are in any way a personal coloration-type of amp. On the contrary, i have had the impression they are natural, tonally pure, yes...and if anything, highly linear in their delivery. Just to make sure i get it right, aren't the ML2/ML3 SET amps?

Edit - sorry - do not mean to derail this thread which is DSD to PCM...which in fact i really enjoy reading about.

 

[microstrip]

(...) I have read someone else (Mike Lavigne?) once post something to the effect that SET amps are surprisingly linear at low levels...its just that expecting them to produce linearity at higher levels of output is not what they were designed for...thus one must DEFINITELY look at system matching to get the best out of SETS for what they do. (...)

Although most people explain the SET typical sound by the increasing distortion as power increases, I can not accept this explanation so easily for good designs. Many years ago I played with a Cary SE 805 - an amplifier that created a lot of debate, as it was a SET with real power. I listened to them with speakers having 86 dB/w and 101 dB/w at equivalent sound levels - not very loud, as I usually do not appreciate rock concert levels. The 15 dB difference in efficiency means that for equivalent loudness the amplifier was delivering around 30 times more power in one case thnn in the other. But irrespective of speaker the main characteristics of the typical sound of the Cary 805 were always the same. But if you look at measurements taken at .1, 1, 10 or 30W they are completely different. Perhaps it is a matter for another thread, but the consistence of sound type in amplifiers, even when they are used with speakers with very different sensitivities always puzzled me.

 

[Orb]

yeah Micro,
most comments relating to SET is not on the very best engineering design and implementation of said products.
Look how many we see about tube amps in general (not just SET) even though McIntosh (easiest example) can provide performance comparable to SS.
Also quite a few SET amps can also provide ultralinear mode.
But we are digressing guys

Cheers
Orb

 

[TBone]

Over our build of several dac platforms, we find that the digital section is not the ultimate arbiter of the final sound. If the implementation is so so, most times the ESS dac will win in a casual listening, it is simply more impressive upfront (like a good new world vs old world wine kinda way). Only when the implementation is good enough, then the difference is obvious. It also takes some training to discern the difference as most people would prefer the ESS sound as it ticks most audiophile checkboxes ie. more spacious, airy, smooth, detailed etc etc. But do they hear the phrasing and inflection from the throat box, the sticks on drum skins, the hammers and bow on strings and all the glorious resonance created that color each performances? Can they hear each individual notes, esp. piano starts and ebb without blurring into one? It took me sometime to get there and to me, this is the biggest difference between high end and the rest and no, you can't get there just with a super DAC chip.

This has been my experience with digital. I remember years ago when I was auditioning two fine CDPs; 1 had a high grade MultiBit design, the other a DeltaSigma design. Both were priced similarly. I fully expected to prefer the MB design based on theory, but that turned out to be not the case in actual practice. Same goes for discrete circuits vs op-amp in CD players, Like certain DACs, I've heard pros & cons. Implementation, on an overall basis, remains the key.

The biggest problem I have with those reviews is the idea that the sonic ideal (to strive for) is embodied by SET amps. They may be euphonic, but few think they are anything even approaching a "straight wire with gain". Dismissing all solid state equipment (as he does in the longer article on DACs) is a poor way to start.

Agreed. Although his original premise peaked my interesting, his obvious biases & generalizations lost me early. I also didn't find it particularly well written.

tb1

 

[jkeny]

SHHHHH let's not start that fight over here too please. Three wars is enough! Thanks for the welcome guys!

Yes BudP - a very warm welcome to a much more tolerant, open-minded group of people here.
Feel free to be worry-free with what you post (in terms of what are considered controversial topics on other fora)

 

[garylkoh]

Welcome, Bud. Thanks for coming in

I'll see you and Rene later. Hope you have time to listen to some PCM vs DSD

 

[jkeny]

I might be wrong (BudP will correct me) in my recollection but I think Lynn O references a SS amplifier as one of the best he has heard - Gary Pimms SS Tabor amplifier http://www.pimmlabs.com/web/solid_state.htm.

Interesting distortion measurements which track what others are saying about SETs - great low distortion at low level outputs - sharp increase in distortion above about 7 watts.
BudP has also heard this amp, AFAIK, & concurs with all those who have heard it as one of the best (again, I hope my memory is serving me well here, Bud - apologies if not)

 

[Mike Lavigne]

Always enjoy reading your posts Opus. I have read someone else (Mike Lavigne?) once post something to the effect that SET amps are surprisingly linear at low levels...its just that expecting them to produce linearity at higher levels of output is not what they were designed to do.

i did say something like that. but i was not referring to SET's in general at all. what i was talking about was a Mercury Vapor tube rectified, 45 tube based, 1 watt SET built by Found Music. and then later the same thing from a 2a3 tubed, Mercury Vapor tube rectified SET also built by Found Music.

what these 2 amps did i've not heard from any other amp in the same way. and that was an amazing window into the inner detail of the music. the 'speed inside the notes' and the sense of nothing between you and the music was astonishing.

unfortunately this view of heaven came along with considerable real world limitiations to drive speakers.

i'm not a techie and don't know about how tubes work or circuit design. but i did get a sense to what tubes can do potentially that solid state cannot in quite the same way.

that approach is not one i choose to use as i go forward. but i do get what those particular tubes can do.

 

[MarinJim]

Oh that's what makes WBF a little different, we're on the whole more into playing than fighting

+1. And our motto is, "Too many chefs do not spoil the broth!".

 

[LL21]

i did say something like that. but i was not referring to SET's in general at all. what i was talking about was a Mercury Vapor tube rectified, 45 tube based, 1 watt SET built by Found Music. and then later the same thing from a 2a3 tubed, Mercury Vapor tube rectified SET also built by Found Music.

what these 2 amps did i've not heard from any other amp in the same way. and that was an amazing window into the inner detail of the music. the 'speed inside the notes' and the sense of nothing between you and the music was astonishing.

unfortunately this view of heaven came along with considerable real world limitiations to drive speakers.

i'm not a techie and don't know about how tubes work or circuit design. but i did get a sense to what tubes can do potentially that solid state cannot in quite the same way.

that approach is not one i choose to use as i go forward. but i do get what those particular tubes can do.

That's the one! Thanks for the clarification, Mike. I have always been intrigued by your observations and hope to hear some of these Found Music SETs some day.

 

[jkeny]

+1. And our motto is, "Too many chefs do not spoil the broth!".

+ 1000 Exactly & each chef gets a hearing & feels included - its just what a forum should be - it's the beacon for what a forum can be if moderated well!
Huge congrats to the owners & mods in achieving this!!

 

[LynnOlson]

Hi Bud, thanks for the invite to the forum!

I am astounded at how closely the PFO articles have been parsed and analyzed! I am not an expert on the TDA series converters; the ones I heard back in the late Nineties left an OK-but-not-great impression. I suspect I'd probably think otherwise if I heard what Thorsten has done; much respect for his designs.

I should mention there are certain constraints on how I write for PFO. I've been friends with David (the editor) since the early Nineties, when I wrote the first Ongaku review in the US. It was living with the Ongaku and the Reichert Silver 300B that inspired me to design my own amplifiers - first, the Amity (named after my daughter), then the Aurora (which was a failure), and the Karna (named after my sweetie). Anyway, I've written for David, Glass Audio (Ed Dell), and was Tech Editor for Vacuum Tube Valley (Charlie Kittleson) for a couple of years. It's a gigantic hassle being an editor and publisher of a magazine, and I am very sympathetic to all the pressures they experience.

PFO does not want to follow the example of the mainstream magazines and on-line sites. We specifically avoid what we call "hatchet-in-the-head" and "damn with faint praise" reviews. If it's not good enough, or just isn't to our tastes, it goes to another reviewer or back to the manufacturer. The Invicta review was difficult to write, on PCM, I prefer my old-school PCM1704 DAC with passive I/V conversion and a non-feedback 6DJ8 analog stage. On 64fs DSD, the Invicta pulls even; I wasn't able to get 128fs DSD to work with Pure Music. Pure Music would convert my 128fs DSD files to 176.4/24 PCM, which definitely did not sound as good as 64fs DSD, so there was a very real conversion loss.

The difficult of writing the review was the reason for writing the longer "Mountains and Fog" article. My system sounds so different than most audiophile systems I don't know if what I write has any validity for PFO readers. The Karna amplifiers sound *nothing* like most commercially available SET amplifiers - and they don't sound like commercially available transistor amplifiers either. They are very analytical and intensely musically vivid at the same time; with most commercially available equipment, you get one or the other, but not both. The analytical aspect is convenient for debugging sources - single-part changes are easily audible, usually in the first few seconds. The intensely musical aspect is useful for hearing when something goes missing - which seems to be a problem for a lot of delta-sigma converters.

I had cheerfully assumed that the latest generation of delta-sigma DACs, so glowingly reviewed on Computer Audiophile and elsewhere, would sound absolutely fantastic on my system. They didn't. I'd twiddle around with various settings in Pure Music, try various USB 2.0 and USB -> S/PDIF gizmos, but it never sounded as good as what I already owned. The glowing reviews did *not* match what I was hearing - in any way. That's what prompted me to take a closer look at the reviewer's systems - which were very different than mine. Not better, not worse, but aimed at different tastes.

The Monarchy N24 tops out at 88.2/24 and 96/24, so I was naturally curious about 176.4/24, 192/24, and DSD content. I'm reluctant to splash out $6000 or more for the Playback Designs DAC or single-box unit, since I have only about 10 SACD's, and was not at all sure I'd like it better than the Burr-Brown DAC & modified Denon 2900 transport I already have.

Having met the Resonessence engineers at the 2012 RMAF show, I was curious to hear what the Sabre 9018 might sound like. I'm not a fan of opamps, particularly for I/V conversion, but I figured the Resonessence people probably know a lot about the converter/analog interface, since they know the internal architecture of the chip.

 

[garylkoh]

Hi Lynn, welcome to the forum. I sent the note to Bud asking if you would join us on WBF. He in turn sent me your thread "Beyond the Ariel". The first few pages were fascinating, and I'm going to have to take a week off work just to read the 8,000 posts on that thread!

Hopefully Rene joins in here as well because I know nothing about DSD vs PCM.

 

[NorthStar]

-- Hi Lynn, and welcome to WBF! :b

Bob

______________

Lynn, just a simple question: The Burrr-Brown PCM-1704K DAC; in quad differential balanced mode implementation (four mono DACs per channel)?

Also, do you like the Pacific Microsonics HDCD digital filter/decoder?

 

[LynnOlson]

Yo don't need to get lost in the "Beyond the Ariel" thread. Just go to the last few pages, which describe the overall design.

The new speaker is mostly done, and a good friend of mine has the working prototypes in Dallas. I spend a week there last Spring fine-tuning and measuring them - and yes, they sound very good. Measurements, although preliminary, are pretty good: a true Theile/Small efficiency of 99 dB/metre/watt, frequency response +/- 1.5 dB, and impulse response that decays in less than 0.5 mSec. Comparable to the Ariels, but 7~8 dB more efficient - which is *very* audible as increased headroom and greater resolution. We only auditioned them on a flea-power 1 watt SET amplifier (not of my design), but measured peaks of 103 dB at the listening position.

I've been at a loss for a name; if I can't think of anything better, I might just name them after myself - the "LTO" loudspeaker.

 

[TBone]

PFO does not want to follow the example of the mainstream magazines and on-line sites. We specifically avoid what we call "hatchet-in-the-head" and "damn with faint praise" reviews. If it's not good enough, or just isn't to our tastes, it goes to another reviewer or back to the manufacturer.

Well, personally, I rather decipher faint praise than hype.

(BTW, I was not referencing your particular contribution)

tb1

 

[LynnOlson]

The other surprising thing I discovered was the *big* difference between upsampling algorithms; some sound like crap - well, actually, most do - but some sound very good indeed. I guess this shouldn't be surprising; we see the same thing in video; upsampling 480i to 1080p varies enormously in quality. Pixel-doubling (which is similar to the oldest oversampling algorithms) looks really terrible, and straight-line interpolation looks blurry (and the audio equivalent isn't much better). There's now enough processing power available to make intelligent guesses about the shape of the waveform as it rolls along.

I suspect this is the proprietary part that vendors are not telling us about. The dithering and filtering is probably dynamic; that way, it measures well, keeping the meter-readers satisfied, and can follow signal content to optimize dither, filtering, and noise-shaping on the fly.

 

[Orb]

Hi Lynn and welcome
Quick question regarding the Invicta; sorry if I missed it in the review but does the Invicta have the option of enabling a slow roll-off filter for PCM, and if so how do you feel that compared to the traits you associate with delta-sigma DACs?
Just curious if the issue is more to do with fast roll-off filters and the limited bandwidth of CD.
From a few measurements I have seen the slow roll-off seems to behave closer to NOS in regards to frequency domain and also has better time domain/ripple behaviour than "default" type filters found within the delta sigma DAC chip software.

Just to add to your own thoughts.
Interesting note regarding the disappointment of Pure Music transcoding from DSD to PCM.
Paul Miller (Hifi-News) noted the same even with the SOTA dCS latest 4-box solution where it is better to keep them as is (native to PCM or DSD) and also within same multiplier for upsampling so synchronous.
Looking back in the discussions a few pages, one of Bruno Putzey's papers mention that for like-for-like transcoding 32bit and 352.8kHz conversion to PCM is required, and in other discussions another consideration is processing capability meaning this would normally require a capable DAW rather than say a consumer DAC.
From Bruno's paper-article

When Philips engineers were building DSD editing tools the design criterion was to fi nd a PCM format that would not detract from the sonic capabilities of DSD.
It was found that it was possible to convert a DSD signal to 352.8kHz/32 bit and back without incurring any audible quality loss, as long as good care was taken with the filtering and remodulation stages.

Edit:
I appreciate the Invicta is not a delta sigma DAC in the traditional sense, but I do notice you still preferred PCM1704.
Cheers
Orb