More improvements in DirectStream firmware


#21
Elk said
gordon said There is a natural immediacy to the attack in timbre . . .
What do you mean by this?

Timbre is tone color (in contrast to pitch and loudness). What is attack in tone color?

I think it goes further than that. Without an envelope just "tone" might be a bit less musical and expressive.

#22

You are conflating an ADSR envelope with timbre. Timbre is the color of sound. The ADSR envelope defines changes in amplitude. An attack transient is not an “attack in timbre;” it is instead the attack or beginning of a note.

Now that we know you are talking about the amplitude change at the beginning of a note, do you want the edge to be sharper and more defined? In synthesis, this is accomplished by shortening the time of the attack portion of the ADSR envelope; that is, making the first line in your graphic more vertical.

The typical brickwall filter used in PCM reproduction results in pre-ringing of the transient. This actually changes the shape of the transient from what it actually is. Many like this as it adds energy; it precedes the actual attack and effectively makes the amplitude rise quicker.

A more sophisticated presentation of an ADSR may assist. This representation is a closer analog to a musical note played by a woodwind, strings, brass or organ (the amplitude of a piano’s or guitar’s sustain keeps decreasing with time as no new energy is added to the system; there is only the initial strike or pluck):


#23

Interesting. This appears to be yet another of those subjective arguments where people attempt to segregate the elements of a term and can only be correct in very specific contexts. I usually identify timbre as what makes a particular musical sound different from another, even when they have the same pitch and loudness. To me, and I am far from being an expert, the envelope [ADSR] is a critical attribute of timbre. It would be like saying that yeast must stand on it’s own and not be included in discussing bread making because it could, in another context, be considered a unique property.

We are talking about the same thing so let’s not get lost in abstract definitions.

The timing/acceleration rather than just the force of the amplitude can generate the excitement of the “attack” timing+ speed will give me all the immediacy I am looking for.


#24
Elk said You are conflating an ADSR envelope with timbre. Timbre is the color of sound.
...and it pronounced tam-ber [like amber].

#25

We blame it all on the French and their hammers and bells. They add to the confusion with timbre as non-audio postage stamps.


#26
gordon said I usually identify timbre as what makes a particular musical sound different from another, even when they have the same pitch and loudness.
Yes, as I initially explained above.
To me, and I am far from being an expert, the envelope [ADSR] is a critical attribute of timbre.
Except ADSR is loudness, which you just wrote above is separate and distinct from timbre. :) Thus, the confusion created by your terminology which conflated the two.

While different things, both timbre and the amplitude envelope are critical attributes of a musical note. There are a number of additional elements, but these are the most important.

The timing/acceleration rather than just the force of the amplitude can generate the excitement of the "attack" timing+ speed will give me all the immediacy I am looking for.
This was precisely my question (" . . . do you want the edge to be sharper and more defined? In synthesis, this is accomplished by shortening the time of the attack portion of the ADSR envelope; that is, making the first line in your graphic more vertical.) So the answer is, "yes." :)

Unfortunately accomplishing this distorts the sound, making it less “real.” As I explain above, this is the sound of a brickwall filter. As also already mentioned, some like this edgy, energized presentation of sound; it adds excitement. Others find it fatiguing.

For me, one of the defining positive characteristics of the DS is its timing accuracy - as I described early on in our discussion of the DS. This precision in timing recreates the sound of how each note develops. The ear is very sensitive to attack transients.


#27
gordon said It would be like saying that yeast must stand on it's own and not be included in discussing bread making because it could, in another context, be considered a unique property.
I'll try to make this analogy work. Wish me luck.

Leaving aside there are unleavened breads, a simple bread is comprised of flour, water, yeast. The finished loaf is a musical note.

The choice of flour determines the flavor (leaving aside yeast has a flavor). It is the equivalent of timbre, tone color. Rye v. bleached white. The tone color of a 'cello v. a penny whistle.

Yeast controls the rise in amplitude; it is the ADSR envelope. If we are discussing rise, we talk about yeast - amount, type, how much to feed it, temperature, time, etc.; i.e., the amplitude envelope (ADSR).

It is a weak analogy, but perhaps helpful. I will defer to Paul if he desires to push this further as he is both a make of artisanal breads, and has a good deal of experience with synthesis. A unique combination. :)


#28

ELK- “For me, one of the defining positive characteristics of the DS is its timing accuracy – as I described early on in our discussion of the DS. This precision in timing recreates the sound of how each note develops. The ear is very sensitive to attack transients.”

Exactly and thanks for the clarification.

So, if I personally find that the “perception” [mine] of the intended immediacy of the note [dictated by type of instrument as well as the manner of playing it] feels a bit lazier than might have been expressed, how is this less “real” Not every exclamatory trumpet blast is a sweet sounding note. Nor does it necessarily get smudged into other instruments at similar volume and frequency.

To me the “longer” attack timing may tune the SQ to perhaps sound less edgy [not the best word] and, although pleasant, may sound more produced than real. I prefer the feel of raw emotion [undistorted] and accuracy of an instrument to tuned smoothness.

If you wish to break it down even further we Canuks sometimes refer to this as the onset and not necessarily synthesized. not sure if it fits your dictionary?

Keep in mind I am far from critical of the SQ of the DS. This thread was about suggestions for future development and obviously will contain subjective preferences.

It could well be that my amp rack is also anxious to receive the new PSA amp with potentially higher slew rate?


#29
gordon said

So, if I personally find that the “perception” [mine] of the intended immediacy of the note [dictated by type of instrument as well as the manner of playing it] feels a bit lazier than might have been expressed, how is this less "real"


If somehow the attack transient is sluggish, reproduced notes would not sound quite real. Given Ted’s design and concern with maintaining phase accuracy I doubt what you are hearing as lacking is any slowness of the DS. It may be you are accustomed to pre-ringing. Wadia was criticized when it first introduced filters many years ago which gave up frequency extension for timing/phase precision. People missed the bit and energy.
If you wish to break it down even further we Canuks sometimes refer to this as the onset and not necessarily synthesized. not sure if it fits your dictionary?
The nomenclature for acoustics is not mine. I wish I could claim credit as it is quite elegant. On the other hand, I do not understand what you mean by "onset" and "synthesized" in this context.

I will note however that live instruments rarely possess the blunt force energy and edginess many crave in sound reproduction.


#30

I just finished A/Bing the Direct Stream with the latest firmware and the Antelope Gold with Power Supply. The DS overall is a superior product in sound quality. The one area the Antelope is better is the impact and slam in the bottom end. As we all know the pace and flow of the music is controlled by the bass. If I knitpicked the DS it could have a little more impact on the bottom. In comparison to the Antelope the highs and midrange are more like real music the Antelope is a touch on the to detailed lean side. The DS warmer with the right highs and midrange. The DS gets the soundstage more right as well with the height and width of the performance correct. I was at a live event and was able to hear Doug McCloud perform live at my local Audio store and few months back. So after hearing him play live with nothing but his voice guitar and wooden platform that he used his foot on to simulate some bottom end I went home and listened to the same songs through the DS and everything sounded right like he sounded live but the bass didn’t have as much impact. I also have the album on LP and notice the same thing. So Ted and Paul if I could have my wish to improve it would be the bass and that would turn a great DAC into a world class reference DAC.


#31

Great comparison Rich. I’m quite interested in what you said about the bass. Over the short evolution of the DS FW I thought one of the areas that showed significant improvement - especially with the last version - was the control, flow, and impact of the bass. If the improvement there would come at the expense of the excellent resolution and spacious quality that we’re getting from 1.2.1 I’m not sure if that’s something I’d trade. But Ted has worked wonders so far. Maybe he’ll find another small algorithm issue to fix.65_gif


#32

Tony22

I agree with you on the last upgrade made a significant imrovement on the bass over previous. I recently also heard the New Berkley Alpha DAC Reference and it like the Antelope had better low end punch/impact. I realize its a $16k product but both of them have more bass like my analog setup. I am being super critical here because overall I really really like the DS. The thread was more improvements to firmware and if I had my wish thats the only area I see improvement needed. I agree with others here including Paul that the top end is just right and more top end woud turn towards the lean and analytical side which in a highly resolving system I do not need or want.


#33

Thx, richfieldhunter!

Could you please eleborate a little on differences between DS and the BADA Ref. outside of the bass reproduction.

Was it a close call?


#34

Quote taken from somewhere within this thread topic

“For me, one of the defining positive characteristics of the DS is its timing accuracy – as I described early on in our discussion of the DS. This precision in timing recreates the sound of how each note develops. The ear is very sensitive to attack transients.”

I have recently been experimenting with JPLAY mini standalone against JRiver my long time preferred player , both players installed on Windows server 2012 R2 in core mode. JPLAY mini is performing a better job than Jriver in respects to the quote.

When running JPlay mini I am getting more bite and slam - the Bass is better presented, there is something about the timing which is making me lean towards JPlay mini. The only thing it’s not as convenient setup as a Jriver player with Jremote.

All I do is select copy album or song from within Jriver Ctl-C and then click in JPLAY Mini to obtain focus and the hit the space bar to play selection - yes I have to be at the PC :-(

Suggest try JPLAY mini as a player and see what you think. JPLAY has try before you decide to buy only restriction from the full version is a few silence passages during play back but enough can be heard where you can hear the difference. For me the difference using JPLAY Mini I’m hearing positive characteristics of the DS in its timing accuracy I think choice of player is critical here…

This difference between players is significant and for serous listening I favour JPLAY mini and switching to jriver for convenience play when I’m not located in the listening chair.

Cheers

Johno


#35

Frode

I will qualify my comments by saying the Berkley listening was not at home on my system but instead at a dealer so It was not the same comparison as the Antelope Gold I mentioned above. I do believe the DS held its own fairly well but the Berkley was just a little better across the whole audio spectrum. Before at the same dealer I they had the Berkley Alpha DAC2 and I felt the DS was every bit as good as it was and maybe better in the natural sound of the female voice and the instruments sounding more real on transients and the leading edge of a guitar or upright bass. The Berkley Reference is spectacular on the leading edge and the transients attacks are amazing. The thing that impressed the most is the the impact of the bass and how that makes the whole pace of the music more real.

Jeff


#36

Some of us seem to be on the same page with suggestions for future development.

We also seem to NOT be criticizing what it DOES already, just what MIGHT make it better, subjectively. That is the OT of the thread.

one of my beta comments was that although the bass was Tighter, it seemed also somewhat softer and this seemed to filter up into the MIDs and tempered some of their “natural” bite.

personally, I miss that and suggest that whether this is True to the sound of real instruments is subjective and largely irrelevant given that we are listening to recordings of mixed or dubious realness potential.

“BITE ME”


#37

My two cents:

As a Magnepan devotee (almost 30 years), I’m not a person who is overly concerned with bass reproduction. My thing is that sense of being there (read: midrange). Recently I had a non-audiophile friend comment (unsolicited) on my DS-based system that he could now sense the venue. His girlfriend: “it’s like you’re there!”

In fact the only time I miss deep bass repro is when playing organ music, but I found the cost/benefit of subwoofers not viable. This being said…

What I noticed with the DS:

  • Initial release - great open sound, great midrange -- the boom/sizzle balance of solid state electronics was GREATLY diminished (as much as solid state has improved, I've always been disappointed by the lack of midrange power from this type of equipment -- snares, horns, rimshots never sounded right to my ears)
  • Middle update - a weird edginess was added to female vocals -- actually put my teeth on edge. Thought this was because of something else in my system, which has gone through a lot of recent change, but realized it was the DS because..
  • Latest update - the female vocal edginess almost gone. The sound is really close to my vinyl setup -- my vinyl sound is just a bit smoother compared to DS -- even when playing close miked, direct-to-disk jazz
So, If I were to have a wish granted as far as DS updates, it would be a slightly smoother midrange (no edge). the funny thing is, the DS has the best midrange I've heard from a DAC. All that midrange exposes stuff that would never be an issue but for the fact that the DS brings it to the fore.

I also realize that live sound is not always best. I actually had my hearing slightly damaged years ago from sitting too close (and having my head turned away) to an unmuted jazz trumpet soloist. So, what I seek to assemble in my audio system is live, but not too live :slight_smile:


#38

Jah, I agree; midrange timbre and transparency is the most critical for “that sense of being there.” This also makes sense as our ears are by far the most sensitive to this range of frequencies.

gordon said . . . whether this is True to the sound of real instruments is subjective and largely irrelevant given that we are listening to recordings of mixed or dubious realness potential.
Hmm . . . so "moar real" is the goal, except when it is not?

#39

I think what is intriguing with the DS is its huge potential related to firmware upgrades. To me it seems that competitors upgrade their firmware entirely founded on the need for bug fixes rather than focusing on improved sound quality.


#40
gordon said whether this is True to the sound of real instruments is subjective and largely irrelevant given that we are listening to recordings of mixed or dubious realness potential.
Commercial popular music is so heavily produced that it's hard to evaluate it in terms of "real" sound. Did the producer want more sizzle than what one would hear live, or less? We can't know. This is also true of some acoustic music; for instance, some symphonic recordings from the major labels.

But with acoustic music from companies that employ more minimalist production values, with the goal of coming as close as possible to what one would hear at the performance, I think we are justified in looking for a ‘realistic’ sound (not that any recording is 100% perfect, but some come quite close).