World's first NOS 24/384 filterless DAC

[GerardA]

Peter,
Just for curiosity,
I thought you did not use resistors in the signal path?
But if you do then a good quality volume control is easily made by adding a potmeter after the resistor between signal and ground.
The wiper you connect to signal and this way there are no bad contacts or bad resistors in the signal path.
Maybe this is what you already tried but I thought I'd ask.

BTW. If you postpone delivery a bit more maybe I'll have some more funds to buy one! 
BTW2, What is the way to order and what will it cost?

[PeterSt]

Hey Gerard,

Thank you for your tips, but this is what I mainly have been trying to do, and it is indeed the one way which doesn't degrade the measurements so to say. However, this means -sadly- degrades when compared to "not in there at all" afterall, because there is no way to separate this from a direct output. So, maybe you can imagine, but with a parallel resistor you "shortcut" everything, and no uninfluenced direct path can be there now. So, compared to "direct" all is okay, but direct itself already isn't okay anymore (and *is* easily measurable).

But the main culprit is -to my findings- about how an analogue volume works in the first place, compared to a digital volume; As you may recall from a long time ago I ever put this on the forum as a kind of quest "what is the real difference", and so far still nobody came up with the answer. At that time, it was theory only, and for me only the audible difference with something like a pre plus volume, and nothing at all. Now, however, with all down to the real merits and one pot doing it etc., it is audible as h*ll. Only at attenuating, but hey, it is just about that. So, now I know that theory is real practice, and any analogue volume is not preferrable.
Maybe you can guess what it is about ?

So yes, there is *one* resistor in the signal path (per channel of course) and it is there to terminate / cover for the capacitance of cabling and it (a.o.) determines the impedance (which is 33 Ohm and of which I think I talked about could increase to 100, excactly for the volume reason). And during that process it was found overlooked to the sense of it needs to be a "best" one. So this is what I talked about in my post from today as the one(s) arriving next week and which should theoretically improve on THD figures. For sound I'll have to see.

Between now and "then" there should be a webshop popping up where all should go automatically, but this isn't 100% arranged for yet (didn't I plan that for last January ? ). Price wil be 2.900 euro, ex VAT and shipping for the normal 2 channel version and includes the interface to the PC.

Again, thanks for the advices; really appreciated !
Peter

[Telstar]

I will be short and I wont report on what we did in sequence - it doesn't matter in the end.

First of all, thanks again and again to Peter, Ciska and Paul for the hospitality and the great time - I was never unconfortable with them. OK, this includes Roy

IMO the NOS1 keeps the promise of providing transparent and uncolored sound, and it will be extremely enjoyable provided that what is fed to is good enough. This means decent recordings and XXHE run from a ramdisk in Vista (so far no other options work).

-A poor recording will show all the issues where they are as they are not "corrected" by filters or other means. Really poor recordings are unlistenable, and it cannot be otherwise if we are not adding or correcting anything. But so-so ones can be enjoyed or re-discovered (i.e. REM - Everybody hurts in the original release, way better than the remaster).
-The Software played from a SSD will also contaminate the sound in a way that has been described by the three crazies (me, Peter and Roy) as "aggressive" and "messy" and it took us just a few seconds to want to go back to the ramdisk/vista.

About "highres" recordings, i'm waiting for Peter to open a dedicated thread.
Instead, i'll mention the really highres stuff - the DXD material which is UNFILTERED. Peter had 4 tracks provided by 2NL (i think), I chose a sonata of Beethoven and we played it. Strangely Peter never played it... it showed the true potential of the DAC, but it also showed a bug in the software that played it at HALF SPEED :D While there were too many bass notes and too few highs (which is weird for a Beethoven piece), there was also an absolute blackness, room-filling harmonics of 20th+ order, and complete lack of fatigue that 80db seemed like 60. Really a pity for the halved HF, cause they could have been the first high piano tones that would satisfy me, apart from the real thing.

Comparing to the very old version of the DAC that I auditioned 1 year and half ago, the progresses are huge and are all in the right direction. I was finally able to perceive the emotion of the music at a more intimate level, which means chillings and/or goosebumps.
I cannot point ANY of the small defects perceived sometimes to the DAC itself. There is a bit of dryness and that's the horns. We changed the amplifiers and there came a sonnolent smoothness similar to many Pass Labs amps, that i dont like at all. Speed speed speed.

Now I can patiently wait for the 8-channels version with a working software XO. It seems it will take between 6 and 12 months, while the 2 channel version will be ready soon.

The future is bright (but uncoloured).

[AUDIODIDAKT]

-A poor recording will show all the issues where they are as they are not "corrected" by filters or other means. Really poor recordings are unlistenable, and it cannot be otherwise if we are not adding or correcting anything. But so-so ones can be enjoyed or re-discovered (i.e. REM - Everybody hurts in the original release, way better than the remaster).

This is for me the MAIN problem, it can never be good (not anoying, enjoyable) without some kind of filtering.
Not to blaim this on Peter's DAC, but it is about those bad recordings the industry provides us with !!
almost all HiRez albums, 20-24bit remasters are "touched" in a cerain way during mastering. (bad filters etc), its destoyed "forever"
And ofcourse the usual not so good recordings because of recording equip.

The fact is i can provide a playlist, with all amazing albums, they while sound absolutly AMAZING, goosebumps etc
But I can also provide a playlist, with some "critical" albums, you will never play them again.

But there are "critical" albums that I DO like to hear, but they just sound bad

I think the NOS1 is very good (The Best !) its a superclean pipeline, but you have to judge the NOS1 on its thru merits (nl being filterless)

So in the end:

If we do not find some mains in software to "repair" those bad albums with smart filtering somehow. (they are destroyed anyway, so for the sake of god, they NEED to be filtered)

Those albums become onbearable, well, for me that is.

You wont play them again.


Roy

[PeterSt]

Thank you guys !

But before I get confused about things ... I know of one album (track) sounding superbad from Telstar (remember, Vista/RAMDisk), and I guess I will be able to find it back. But can you mention a few others you recall ? I mean, I don't know from the top of my head, and I'd like to continue with those in order to see whether they can be improved afterall (or just *will* be after some upcoming OS tweaks ! ).

The context : I sure know them from W7 and W7/RAMDisk, but not from Vista/RAMDisk.

If I had to name one other, it was REM (that specific album; in fact two, one hires one normal), but I guess Telstar was already referring to that (and then particularly the hires).

But thanks, you were good company, quite OK cooks, and nice ears !
haha
Peter

[Telstar]

If we do not find some mains in software to "repair" those bad albums with smart filtering somehow. (they are destroyed anyway, so for the sake of god, they NEED to be filtered)

it's not that difficult do apply filters in software, but you must know what you are doing and what you plan to repair.
Easiest things:
-HF rolloff to protect our ears from badly mastered music with violins, cymbals etc.
-peak extend for very dynamically compressed recordings (already done in xxhe)
-soundstage expansion or compression
-many more that you find commonly as VST plugins. Just check sites such as waves.

Peter is on principle against any manipulation, but in the end letting us do it by means of external software should be acceptable and in *some* cases beneficial.
it is impossible to undo all the damage done during the mastering because we dont know what has been done to each recording.

[Telstar]

But before I get confused about things ... I know of one album (track) sounding superbad from Telstar (remember, Vista/RAMDisk)

David Garrett - La califfa (from the only cd of his that i have).
to compare to a good track of the same, check the best of Ennio Moricone on the same HDD.

[Per]

I cannot point ANY of the small defects perceived sometimes to the DAC itself.

Wow. What a praise for a product aiming for the state of the art. Congratulations, Peter.

Per

[xp9433]

Peter

Have you got any pre/early production models of your NOS DAC out in the field yet for testing?
Any feedback?

Frank

[han]

Peter,

I'm planning to trade in my Weiss Minerva for the Weiss 202 but of course i am very eager to test your DAC too. So i have the same question for you, is it possible to audition the NOS DAC and when are you able to deliver a "commercial" version? 

Regards, Han

[PeterSt]

Frank, No.

Han, All ...

First of all I want to thank you people for not asking again How long, etc., ...
The things to do in this last part of the traject (project) have been crazy, and in fact still are.

What happened since the last "contact" here, is that I had made another prototype, in fact on the advise of Gerard and Leon - that the outside connections didn't seem all that convenient. So, below pictures are from nearly a month ago already, but I didn't want to promise another date *again*, not being able to keep it. Btw, and for fun, the second picture shows the work bench and the number of cut wires (if you want to count them) needed for the last assembly alone (everything into the case). So somehow that number of wire connections must be somewhere. Anyway ...

Long story short, it should really be so that next week the first handful of production units are ready. With paint and everything (not so with the proto below).
I always planned to let the ordering go via a nice webshop (which software and the ERP system behind it I own from my other life), but I don't think we're going to make that in time. So that will be email stuff for the time being (read : I won't hold up things longer because of a nice administrative system).

On the audition thing :
Two days back I wrote a somewhat longer post about this, but was so stupid to reboot my PC before it was really posted, so that got lost; I didn't find the time to write it again, but I will in (very) due time. But in short : it will be paying, but give back and refund when you think it's needed; no questions asked.

Official written specs are half-way, and should be ready soon as well.

What will happen next is that I'll send an email or PM to everybody ever asking for the DAC, starting at the very first (from quite a while back ) whether the interest is still there. First come first serve, but the first 100 units for which we have the parts and assemblies readily available will be sufficient for that. Still the last assembly (into the case) plus testing has to be done as well, and without much other interference that will produce two units per day at first.

The cases are due next tuesday, and as soon as a first has been assembled I'll show you a picture of that. After that happened (so watch that day) ten days later the first 20 units should be ready for shipment, and I think we'll batch up to that first 20 to streamline things a bit, and finish up the administrative system which really is needed. So, within 3 weeks from now the first units should be breaking in at your homes (needing three days for a good show off, and another near 20-25 to really get the job done -> trial time will be 90 days).

Thank you all for your patience and interest,
Peter

[Leo]

For the moment I will just order one!

Leo

[Gerard]

Looooooooking Good!! 

[Gerard]

(if you want to count them)

mmm 75?? 

[PeterSt]

Haha, I just counted too and came to 74. And an even number seems more logic to me.
But wait, whay about P.E. (protective earth) ? (the three wires from the wall outlet).