More improvements in DirectStream firmware


@Ted: do you expect that even more improvements can be made in the DirectStream firmware in the near future?

Maybe a direct DSD (without upsampling) path would be possible or even better jitter handling using multiple buffers?

I still think single DSD sounds a bit 'soft'. Not sure this is the fault of the DS however...



I do hear, every once in a while, this idea that DSD sounds soft. I am curious. Soft compared to what? Live music or another DAC? In my experience DSD sounds more like the sound of live music while many PCM based systems seem to sound a little edgy. Do we get to the point where this slight edge becomes the norm? Or am I missing something?

I am not being critical or judgmental at all. I am just curious. In my system, the last thing I would want is for the sound to be “edgier” or “livelier”.


My .02

I do find DSD [with the last few versions] to be a tad less "exciting or perhaps the word “immediate” is more applicable, than our original Beta starting point. The attributes of 1.21 trump this with amazing "air, stage and listen-ability so definitely there are no complaints and the market reaction speaks for itself.

I realize that subjectivity comes into play here so I would NOT get too excited as there are so many other system and room variables involved that there may never be one perfect version for anyone. As I have mentioned in the more recent Beta discussions, every upgrade has been a genuine enhancement of the overall SQ and enjoyment factor. For me it is still the “timbre” and in particular the “attack” of the timbre that would rock my cookies a bit more. Not hardness or harshness or edgier, just a more poignant attack timing.

I am extremely happy with the sound to date but the greedy little piggy inside me just craves a teeny bit more.party_gifmusic-078_gifhappy-132_gif


DSD flows over my ears like warm buttermilk. If that is “soft” then I like it!


No more firmware changes, ever! :)

Okay, that’s asking too much, but wait quite a while, okay?




I agree with Gordon that the DSD sound could have a little more attack. In my system the sound could also be a little more ‘alive’.

I am still hoping this can be improved with optimized FPGA firmware or even a DirectStream Signature / MKII.


@Ted: do you have any thoughts on this?


What many of you are hearing with PCM is likely the pre-ringing of the typical PCM filter preceding the beginning of the transient attack.

This is an artifact of the format, just as soundstage is an artifact of stereo recording.

My bet is Ted can easily introduce pre-ringing into his filters, but would be loath to do so as he is a phase coherency advocate.

adriaan said @Ted: do you have any thoughts on this?
I've heard live recordings and the DSD feeds of them - double rate DSD is very close to the live performances. In my experience PCM recordings aren't.

I do think that many PCM recordings today take advantage of the preringing from the brickwall filters to add extra energy or excitement to the recordings - i.e. the mastering engineers (consciously or not) master to present the music with the extra perceived energy that’s really an artifact of the technology. In life we don’t get “warnings” of an upcoming transient, so getting that preringing let’s our brains know about something interesting/important is coming. That can translate into more excitement, but it can also translate into fatigue since it’s not what the ears were designed to deal with on a long term basis.

As Elk mentions I want to preserve the shape of transients and pick my filters appropriately - BUT the preringing of PCM is what gets the closest to what the PCM mastering engineer was hearing.


I agree with hearing the flatness of DSD when used to listening to PCM - especially with the latest software. I also agree that DSD sounds more Tape like (which, IMHO is my goal rather than “live” sounding). I’m enjoying my PCM recordings so much more lately, and my DSD files less.

I think I need to do a week of DSD only, and see how I feel about PCM after being away from it for a while. I’m betting it will sound edgy and fatiguing, and maybe even a little artificial.


Even better, listen exclusively to unamplified acoustic music for a bit.

All of the artifacts of recording are thus brought to the fore: the artificially magnified attack, the creation of a soundstage and instrument separation, the heightened emphasis on ambiance, the lack of the enveloping nature of bass voices and instruments.

Live acoustic sound does not possess the snap and sizzle of attacks presented by a recording. It exhibits an entirely different type of energy. It is perhaps the equivalent of the artificial lighting and highlights of film.

adriaan said I agree with Gordon that the DSD sound could have a little more attack. In my system the sound could also be a little more 'alive'.

I am still hoping this can be improved with optimized FPGA firmware or even a DirectStream Signature / MKII.

Generally the latest firmware is great, but “more attack or slam” is exactly what I think would make it even better. This of course may be system depended, but this is area where it would be interesting to test multiple versions and views to hear different approaches.


My bad – I wanted to thank Elk for his post about unamplified music and accidentally hit the button under jka’s post (I do not want more “slam” in new firmware). Is there a way to cancel after hitting the ‘Thank’ button?


I thanked to see if it would also “un-thank”. Seems to me that it used to, perhaps the old forum or another.

Slam is not what I might describe my wishes for. Perhaps “immediacy” might be a less misleading or more gentle term.

There is a natural immediacy to the attack in timbre whether amplified or not and certainly varies by the type of instrument or intention of the musician. This, I feel, as part of the synergy between the player and the listener similar to how tone, loudness and spacial timing changes in conversation add emphasis to the emotional expression of the speaker.

I’m not complaining, just expressing my wish list.

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?

Elk said Even better, listen exclusively to unamplified acoustic music for a bit.

I’d love some recommendations - I’m always looking for some new music.

I might need to find some DSD acoustic music to listen to for a week straight.


I always evaluate my system using acoustic jazz or classical. Unless I were the band’s equipment tech or recording engineer rock might sound great, but how do I know how accurate it is? Could have tizzy treble or wooly bass added and still be “rockin’” :)

I use Miles Davis “Sketches of Spain” (the MoFi SACD is amazing) and the Vegh Quartet Beethoven pieces lately as a tool. And I do use some electric rock and blues material, but it’s material I recorded of bands I was in in the 'eighties recorded by me in my then garage apartment. I know how that should sound. I also use piano recordings to gauge a system’s tone and dynamic natures as I have spent a lot of time with pianos both at the bench and in the room. Lately I’ve been using solo pieces in the Horowitz Original Jacket Collection, especially the mono recordings as these are very well-recorded with great presence to the piano sound and significant dynamic swing. And also any of the three volumes of “Earl Hines Plays Duke Ellington” on Master Jazz; when you can really feel the swing and that piano sounds real, I’m really on the right course.

Lonely Raven said
Elk said Even better, listen exclusively to unamplified acoustic music for a bit.
I'd love some recommendations - I'm always looking for some new music.
I'm very sorry; I meant live unamplified music.

Audiophiles often confuse the sound of various forms of reproduction with reality. And then obsess over aspects of reproduction which exist only as artifacts of reproducing music, such as soundstage. Recorded music always sounds vastly different than actual acoustic musicians performing live.

But lonson’s suggestions on what to listen to and how when evaluating a system are superb.


I agree Elk, unamplified instruments in the air, that really is the way to build a sense of “real” sounds and textures. I think Mr. Raven has had a goodly amount of experience along these lines. :)