XXHighEnd

I hope this is OK to ask this here?

I am going to audition a NOS DAC which will hopefully improve on my SHEK D2 in my system. Would anyone like to comment on the 'quality' of the design and components.

In particular Peter, would one expect it to work best with XXHighEnd with the valve re-clocking option turned on or off.


NOS (Non oversampling –zero digital domain filtering) design

Based on 1x TDA1543 – one of the most analogue sounding DAC chips ever directly connected to CS8412 SP/DIF receiver

Active I/U conversion via a special transistor with a shortest signal path directly connected to the tube grid - unlike other TDA1543 based DACs on the market our DAC is excellent at reproduction of low level musical information

ECC88/6N6P tube output stage

Tube based re-clocking option (can be switched off by a jumper)

Shunt voltage regulators everywhere (much smoother and more analogue sounding)

We only use fine selected audio components such as Allen Bradley resistors, Philips BC, Nichicon Muse, Panasonic Pureism, Kendeil capacitors, Telefunken diodes, PIO output capacitors

The PCB has no mask on the bottom side and all conductive traces are covered with gold and a special varnish for a for a more open and detailed sound

Mains input wiring as well as SPDIF and analogue out is wired in pure silver

Silver plated RCA connectors

This was something like my diy dac in 2000, terribly old school.

I also have the ifi nano idsd for headphone purposes but connected it the other day to my main rig. Big surprise: excellent sound and fun. You can trick this one into pseudo NOS by feeding it a high sampling rate from xxhe like 384khz. Then the xx arc pred filter dominates over the internal one and that is audible in a positive sense.

This dac is dirt cheap and does the works for music lovers, portable too if you have other plans. Only nuisance is that quality listening needs to be done over the battery, but its life is sufficient for more than one evening of listening.

No, it does not live up to the phasure dac, but what do you expect really...

Regards, Coen

P.s. Very good sound is also attainable on the mac with audirvana+(Upsampling)

Thanks Coen  :smile:

Oops ... I wanted to start out the previous post with this :

No Chris, no problem at all with asking about other DACs. The only thing what might happen is that my personal comments seem not valid because Phasure has such a product herself. This latter is obviously true, but the former sure is not (well, I hope !!).

I need to purchase a lot of stuff myself and I guess I am reading more reviews and spec sheets etc. than many of you together. I look through my special glasses and "verdict". Mostly this is about consistency (I mentioned the word in my previous post) and how you can tell that people BS to begin with (be commercial) or lie or don't know. Some times it is a "sport" and not easy to not make mistakes.

What, by all means, should NOT come from this, is that people stop proposing things (like DACs) like you just did, Chris - because it could perceivedbly be "dumn" from the proposer, or something. Not so at all of course.
So what I just did is genuinely sprouted my thinking when I read such a thing and this includes that it tells nothing about the sound (not in this case). Maybe one thing about the sound : since the 1543 is 16 bits only, it can not do (take) any upsampling (formally needs one additional bit per (2x) upsampling step). So might you go for the Arc Prediction filtering in XXHighEnd - it's a no-go.

Peter

It took me 1 minute to find it.
Quite some more to find out who or what the seller is. And THAT alone would make me the most suspicious.

http://www.clar(removethis)tes.co.uk/about
Edit : I added the "(removethis)" so Google won't index this. Meaning : It is fine that someone (me) publishes an objective judgement but it is less nice when quotes from persons can be associated with those persons - especially when the general idea is - or could become a negative one. See and follow the (by now new) posts below.
That is your seller. Hum.

Maybe they got hold of a pallet of DACs but ... I don't know.

Now tell me - If you would be a respectful UK company, would you sell on ebay Hong Kong ?
http://www.ebay.com.hk/itm/SW1X-Audio-Design-DAC-01-NOS-TDA1543-ECC88-6N6P-Tube-Note-300B-2A3-amp-/181910924043?hash=item2a5abc6b0b:g:~6EAAOSwo6lWKhT6

Now what's crucial here :
This is brandnew (from the Ebay UK page), right ?
Of course it is brand new. But as the text is old, see the picture below for what's in there. Indeed that's from my old Telefunken TV they probably tore down for this.
Or maybe it was the 2nd WW Philips radio - I forgot.

Still it can sound good.

Peter

Good detective work Peter  :thankyou:

I only stopped from putting the name on here as I didn't want to be seen as advertising someone else's products.

It has all the hallmarks of a re-badged Chinese product, but the guy lives local to me so I will go and have a listen.

Indeed I might just ask for a tour around the English factory. :grazy:

Here is a bit of correspondence I had with the seller. Does this man sound like he knows what he is talking about or is it more BS?

Me

"As my music source I am using computer software that does all the upsampling before it sends out the data. It is capable of outputting data at 32bit/768Khz.

In terms of ultimate SQ how will your NOS DAC be handling a 32bit/768Khz signal?

Please keep the response simple as I am taking advice from the people who generate the software and am not skilled at all this technical stuff."

 
Seller

"I can see where you coming from. A resolution is only one aspect of a good sound. Higher resolution does not guarantee a good sound anyway, when there are so many other factors just deteriorate the sound beyond comprehension. Many years ago I also thought that the higher the resolution the better it is, until I realised that the potential of 16 bit Red Book (CD) format was far from being fulfilled. Wait until you hear a properly designed 16 bit NOS R2R DAC that can resolve as good or better than a 32 bit Delta Sigma DAC.

Speaking of resolution, the highest resolution of an mass produced R2R DAC chip known to me is 24 bit (which requires digital filtering) and a NOS capable R2R DAC chip is 20 bit.

R2R means resistor network based chip architecture, the most accurate and most expensive DAC chip to design and to produce. Only the R2R chips can provide a true resolution in its original sense. If you have a R2R DAC chip with 16 bits resolution and low noise capacitors around, most likely you will come close to the specified resolution.

On the other hand, all modern 24 and 32 bit DAC chips of Delta Sigma type are <1 bit noise shaping by design, i.e. produce noise and that clean up the signal with a demodulator, which supposed to be 24/32bit. These modern chips do not really resolve the claimed accuracy and are quite forgiving with low quality components and all of them require filtering by design.

On top of that, what is the benefit of upsampling to 32 bit if the maximum resolution of digital SP/DIF signal is 20/24 bit and the recording itself is 16 bit only?

The 32 bit upsampled signal is re-formated to 24 bit SP/DIF, where 12 bits are lost. When a signal arrives at a SP/DIF receiver chip of a DAC, the last 8 bits of 24 bit get truncated to the resolution of the DAC chip e.g. 16 bit. It is like taking a low resolution picture and zooming it in, then adding additional bits to make the curse looking squares appear smoother and then zooming it out again. In other words, there is no value added with upsampling. In my expreience, the upsampled signal cannot and does not sound better relative to the original 16 bit.

I hope that everything I wrote makes sense"



 

I played for some time with the cheapest philips CD player of early 90s which I bought for 5 euro at the second hand store. This one had the TDA1543 and I managed t lift some processor pins to disable the digital filter and put it in NOS (1x) mode.

Though it was NOT the last word in resolution the CD player was a lot of fun to listen to and had great tone. Actually it was generally accepted that the TDA1543 has a maximum of 12 bits resolution which is more than enough for todays over compressed pop albums, but not for high end listening and certainly not enough for through-dac-digital volume control.

I was surprised to learn that the Doede Douma Dac (DDD dac) is still alive and kicking today. Here resolution is enhanced by averaging many (multiples of eight) TDA1543 dacs. In a classic setup with analog volume control and tubed amplifiers this DAC probably sounds very nice since real 16 bits of resolution will surpass the capabilities of the other components and its remaining distortion is likely pleasant to the ear. Please note that the tube is needed to conceal or ameliorate the rough distortions that NOS (1X OS) exhibits.

Yet technology moved on in 20+ years and really I think there is no comparison with my IFI nano iDSD, which can sound like a true high resolution DAC and is super versatile and super cheap. It supports all current formats without external conversion like you would have to do with a 16/44.1k NOS DAC. It really is able to let you familiarize with the potential of PC/computer based audio.

Of course there are similar offerings as the nano iDSD to consider and as a starter in this field I think this amount of money will be well spend to jump on the learning curve and decide later if this suffices or that you wish for something else.

regards, Coen

Hi Coen,


I know you are not saying that the resolution becomes higher of it. But for understandings :

What happens with that is that the stack of same D/A chips get more linear because they-are-not-so inherently (actually they are not so much the same ;)).

How this is useful to a better sound or distortion of if you like "resolution" ?
Well, see next post (that severeness can't be overcome by any means), because I don't see it.
Yes, more current output ... (easier I/V and IIRC the DDD does this passiveley).

Regards,
Peter

My initial impression :

If I write him an email he could buy one of my DACs.

Read :
His response seems to be genuine.

Try to see this in the context of the so many persons/companies who put up large stories about why 96KHz is sufficient (while of course their DAC can't do more). This includes very well known and respected "names". Still it is "BS" so to speak, to not offer more. Thus, what those persons / companies should do instead, is explicitly telling about the negatives of e.g. 192KHz; why it would be devistating. THEN I'd buy the 96KHz storie(s).
But of course this is difficult, so you don't see that happening (but see next post as well).

---

Chris, it is all rather complicated. So something to really learn again :

We both know that you and me are from the BD-Design Forum era where I started out with XXHighEnd. Say this was over a year earlier than this forum emerged (2007). I started to implement upsampling;
I tried many 3rd means of it myself (like Rabbit stuff from Foobar) and in the end - as I recall it - ended up with something which was at least not worse than doing nothing (so just NOS). But really helpful ?
No.
At that stage, my stories would have been - and were if you ask me - the same as your Seller just put up to you.
But hey, this was 9 or 10 years ago and a few things changed;

I created Arc Prediction which was actually my own promise to the NOS1 being in deveplopment because I refused to make a ringing filter (that would kill the NOS virtues) and against all odds I even succeeded with that. From there more started to do this explicitly, and while Coen mentioned Audirvana, that would indeed be one ho got the idea (Damien (Demian ?) from Audirvana even implemented the "always even" upsampling for me).
Much much more has been going on and the whole idea of in-DAC doing things (or do nothing) started to shift towards the PC. Ask "Miska" from HQPlayer and see how actually (only) two guys on this globe understand the principles (he thinks the very same as I do, although he is the DSD guy).

What am I saying ?
Well, that it is VERY EASY to miss out on this all because one forgets to read a couple of forum posts and a similar couple of years (or is not into that at all).

Back to your seller ...
BS ?
Not as I read it from that text.
Old fashioned (if I am allowed to see it like that) ... sure.

Tell him that the digital filter (in or outside of the DAC) makes the sound. And not his Telefunken GPS-towers in there.
And how he sells D/A converters while he should be doing energy stuff (IIRC - I already forgot).

Still it can sound good.
Nah ... it can't. LOL.

Look below. I coincidentally needed that this morning for something else. It is a 10KHz test signal and genuine NOS.
If one likes that, one likes distortion. It is from this topic/post :
World's best measuring NOS DAC : Phasure NOS1 (http://www.phasure.com/index.php?topic=642.msg6497#msg6497) (it is a large topic so let it rattle until it downloaded all the pictures - it is Reply #130 in there).
The first picture (the orange one - the same as you see below) is a NOS DAC without filtering (remember, filtering = upsampling !)
The second picture looks quite 99% the same, but it is from a DAC which is now used as NOS and thus no filtering again.
The third picture is that same latter DAC and but now 4x Upsampling with a nice filter is applied (I don't know any more what filter that was at that time).
All you need to do is envision that the first picture (also shown below) can not be improved upon because the DAC is not capable of a higher sampling rate and/or does not supply the necessary number of bits to do that.

You know what ?
Squeeze a 20Hz tone through your system. What do you hear ?
Well, if it is 40Hz capable, you will hear 40Hz. That is a similar distortion as you see below but easy to proove because you can hear it so easily.
(with the notice that 20Hz should be inaudible to you)
Disclaimer before I'm aimed at : People owning the Orelo MKII speaker of course will hear nothing ineed.

Yup.

Peter

Help ...

I am laughing myself, now I see this picture pop up in this post.
My story is nice and valid, but please notice that the "harmonics" you see are a bit outside of the audio band. That happens with a 10KHz base tone ... (which I thus coincidentally ran into this morning).

Anyway, still the idea is clear hopefully.

Peter

Edit / PS : But you can compare the distortion figures (the 10% you see) because they are calculated from the in-band response only). For that, dive into the link I gave, where you see that the same signal can end up to be 0.004% but which can not be justified by these plots (so you need to believe me).

I think we say the same.
Bit linearity is enhanced by paralleling (or Dynamic Element Matching if you want to keep the current small) from 11 to 15-16 bits, but even perfectly linear you still have to deal with the gross distortions that NOS 1x (16/44.1k in) implies.

By NOS, both Peter and I mean that the DAC does not filter digitally but the PC does that for you. Now you can apply a filter to the bare music data that sounds and measures well and feed it to the DAC WITHOUT additional in-DAC processing.
Please note that the bitrate to the DAC needs to be high to overcome the distortions and that's the job of the digital filter. As a bonus extra signal bits can be calculated (ie 24 bits) to mimic the signal closer.

So: signal processing in the PC and the DAC adding NOTHING to it at a HIGH BITRATE (i.e. 192KHz or higher). Iow preferably you need a very bit-linear NOS DAC that works on these high frequencies. Unfortunately these do not exist with an old fashioned SPDIF interface (nor do the sources) and you need USB or something else. Note that the IFI doc works up to 384kHz (not completely NOS) and the NOS1a up to 768KHz.

nuf said,
regards, Coen

Ok that makes sense, but the job of the firewire in my system is to carry the signal from XXHE to the Fireface 400 which is an external sound card and is much better than the internal computer sound card.

If I fed a good NOS DAC with a USB signal from the computer I seemed to have omitted the Fireface 400, so what is being used as a sound card  :dntknw:

Eh?

Usb is not a signal, it is a cable system to transfer data on a short distance and a simple way to connect devices to your pc. Only the initiated know the electrical and pc stuff that makes this possible. All pcs have usb ports, the digital music signal is transferred via the usb system to the Dac which also has an Usb port. There are different flavours of usb operation, i guess all modern DAC use the same mode today, anyway all will work. Several years ago people called these dacs 'Usb dacs'.

Regards, Coen

P.s. A sound card is a big DAC albeit not/never Non Over Sampling, whats your point really?

Coen, thanks for baring with me, perhaps I have got this wrong all along :)

When I started with computer music, first option was to let the computer play using its own sound card (sounded rubbish).

Then XXHE came along and the advice I took in (perhaps wrongly) was that the internal sound card would hold back the sound quality and to switch to an external sound card attached by a firewire. This was and still is the Fireface 400 which is used solely as a sound card and the internal DAC is not used as this is not very good.

Signal from the Fireface 400 is taken via a single digital cable to the DAC.

This is how it is currently wired and plays very nicely. My assumption therefore was that somewhere in the system a quality sound card was required. Is this all nonsense?

Ok!

the rme fireface itself looks like a nice but littlebit outdated device to function as a quality two channel audio dac itself but you are not using it that way. Apparently the rme 'soundcard' as you call it has an good working spdif output interface for your quality dac. I see nothing fundamentally wrong with this setup, yet moderner solutions are available. I don't think a 'better' soundcard will improve on this merely because it is "better'. The attention given to the spdif output is usually not clear.

There are these specialized usb to spdif converter devices that can do the spdif trick of the rme in a smaller package and have been developed with audiophiles in mind. As a matter of fact the ifi idsd nano has a specialised spdif output and you can connect your external dac to it. No clue though if this kind of connection enhances your sound, it might or might not.

I may sound like an ifi commercial by now but have no ties whatsoever. I just think its design, engineering and sound are great,especially for the price.

Like i said your DACs spdif interface has its limitations wrt the possibilities of software based optimisation in Xxhe. You need both the higher sampling rates and increased bit numbers and linearity to fully benefit from it (and preferably no in dac sampling like on delta sigma style dacs). So if you really want to move forward this is something to consider.


Regards, Coen

P.s. Obviously you do not need any 'soundcard' when using the usb to spdif device. Just plug it in one of the free usb ports on the pc (and then experiment which port sounds best :grin:).

P.s.2 does your Dell on board soundcard have a SPDIF output? Not all of them do. Maybe it will be worth trying since it is closer to the pc's hardware and xx is likely to have more impact, on the negative side noise on the spdif line is bound to be higher but not neccecarly so.

Thanks again Coen

So, if I wanted to progress in a good direction I should use a USB cable to connect to a DAC with USB input such as the iFi.  The Fireface 400 can then be sold on E bay as no soundcard is needed with this configuration.  :smile:

The USB DAC will have phono connections to connect to audio amplifier.

Is this about the best configuration to go for, or is there any other approach you would recommend (other than Peters DAC which is a bit too expensive for me at present).

Nope, looks like a fine plan. There a a lot of usb dacs in any price segment, so you'll have to do some homework on the internet. The positive side of this is the fierce competition leading to better base quality for the dollar all the time.

I'd advice to look for a high as possible samplerate capability, like 384khz or more. Personally i do not value the DSD capability since it adds little if anything to sq and its catalog is allways also available in normal format (pcm), but it may be a nice to have feature.

The other thing is the pc. Some xx settings resulting in good sound can be really tough on the processor. Even my 4core machine is not really up to the task of delivering the bits on time consistently. Today you can make it really complicated with pc setup but if you renew make shure to step into the i7 range for the audio pc.

Regards, Coen

Good man Coen. I feel like I am getting somewhere now.

I have been reading reviews on the iFi range and its all very favourable.

You mentioned the nano kit which is also the cheapest. I guess that moving up to the micro range is better still. I don't need headphone facilities and reading some of the specs gives me a headache  :wacko:.

Ignoring cost for the time being, what would be the best USB DAC to fit my needs.

I also see there are USB cable power supplies, mains power supplies, and audio perfect USB enhancers, so clearly an upgrade path is opened.

I have been smiling a lot, reading the last posts ...

Apparantly time has gone so fast that us "modern youngsters" have forgotten (or never have known) how this world was constructed.
And I am talking maybe 5 years ago ...

A FireFace is a sound card because ... it just is.
It is by the grace (is that Dutch ?) of it containing D/A converters.

Something like a FireFace was used back in the days (now think 10-11 years ago) because, well, uhm ... I came up with it.
:thankyou:
It extended the SPDIF connection from the PC to more close to the DAC. Well, sort of;
Think like the Firewire connection being regarded (!!) harmless and extendable to xx(x) feet, so your SPDIF connection (which can't take that length) now always being close to the DAC. Small SPDIF cable - done.

Notice that USB devices did not exist at that time.

Also notice that the FireFace these days also exists with USB connection and things get more vague.

Chris is a bit more correct than Coen in thinking that a PC needed a sound card and the Fireface (but many more of such devices) was just that. So remember, supposed you don't have analogue out on the MoBo (and that sure did not need to be there), where is the soundcard ?
It is not there. And so a FireFace sure was (and is).

Now USB DACs start to emerge and things get more vague;
What was all new is that now a PC did not need a sound card any more !
And to the letter this is true. And that the USB DAC exhibits as a sound device - not different from how a real sound card (or FireFace alike) would do it, is even more obfuscating (mind you, it was also for me back at the time).

The point is that all got more transparant and which is exactly the reason why we don't see through it these days.
For example, when you have a USB DAC, it can connect directly to the PC, or to a FireFace with USB output (not sure whether they can do that these days, but I sure have an alike which can) or to an "Interface" which serves the purpose of actually connecting legacy DACs (SPDIF-in only), but which generally are versatile enough to have e.g. USB-in and also USB-out and ...
well, it does not matter, because all connects everywhere.

N.b.: It does matter because the PC can only control the first-connected device for buffer sizes etc. (that is, as far as I know).

The sort of sad thing is that if one misses only the past five years, he doesn't understand one bit (pun) of it any more.

For me too it isn't always clear yet;
It was only last week that someone wanted a NOS1a but with SPDIF-In. I right away said "sure !, possible !".
Then I started thinking ...
The SPDIF-In is "on" the USB Interface. Sure it will work.
But will it work without USB connection to a PC ?
haha, I don't think so.
And I am mot even sure.

Peter

Peter

Thanks for restoring my sanity and making me feel very old at the same time  ;)

I knew I had put my system together with advice from you and Bert because with electronics and computers I need expert guidance. :friends:

So today I wake up in 2015 and must readjust to a Hi Fi world that has shifted on its axis.

All is not lost as a pristine Fireface 400 seems to be fetching reasonable money on E Bay and will help finance a change to a USB DAC.

I'm not sure I even dare to ask this question but both USB 2 and USB 3 exist. How does this factor in to the decisions?

Right you are Peter, I also paradigm shifted along with the developments from SPDIF in the time of the CD player to a Firewire 'high res' DAC and to an USB one in the PC era. God knows what's next (Ethernet?).
Also even tapped the I2S from the source when it was en vogue in the diy world (90s). That idea seems to be echoed again in a recent commercial implementation (Pink Faun) or was the first non-USB NOS like that?

As for USB2 vs USB3, the data rates are well within spec yet there seems to be a preference for USB3 on the PC side. I've to catch up here since I don't know if it will matter on the Dac side. I would not consider this to be very important on the basis of reason alone.

regards, Coen 

Guys,

Re the USB2 vs USB3 ...
I don't think USB3 DAC's exist. Partly this will be because it is not necessary (USB2 is fast enough), partly this will be because someone has to make the drivers and it probably is too difficult for any party to start such a project.
But I don't follow much regarding this, so maybe I am wrong.

Chris, what Coen responded to, is that *everything* matters and thus also the USB3 interface of a PC sounds different than the USB2 interface. Or a separate interface card - again sounds different. The stories are too long, I'm afraid.
Indeed the trend seems to be that UDB3 sounds better than USB2. But mind you, the DAC is always USB2 so no USB3 speed is utilized/used anyway.

... And then to think that the above "info" actually already is from 3-4 years or so ago. At this moment we all take it for granted ...

Regards,
Peter (who also must be precautious not to get rusty)

THE ifi has an USB3 interface receptor plug. Don't know if it really supports usb3, since this is irrelevant for the data transfer. since USB is backwards compatible it'll work with any USB cable.

Regards, Coen

I did go over to visit Dr. Slawa Roschkow at SW1X yesterday and was VERY VERY impressed.

He is a Hi Fi fanatic and an electronics genius just beginning to market some of the stuff he has developed (start up company) He certainly has no affiliation to any Chinese company.

He has three systems set up in three dedicated rooms, all of which were wonderful.

Everything was valve based 300Bs, 2A3s and the best system used open baffles speakers with a full compliment of field coil drivers. Silver wiring was everywhere including mains cables.

If this is old technology, then I love it.

I am not trying to promote this man, but in hindsight we gave him a bit of a hard time on here and I just wanted to restore the balance.

The biggest compliment I can pay Showa is that I have not met a person of such qualities since I met a young Peter Qvortrup many many years ago, and he went on to develop the Audionote brand.

Hi there Chris,

To hopefully set things straight :

I don't think anyone really has given anyone else a hard time. But might you judge the "old stuff" as giving someone a hard time ... I'm afraid that's the wrong thing to think. Because remember what I said a couple of times : it still can sound good.
That it cannot meet today's technology (if it were about the filtering) is just a truth and if the other party deems it necessary to counteract that with a "but that is useless anyway" then all what counts is our ears.
And that is just what you did.

I like to explicitly add that it was me who removed (or obfuscated) the link to Mr Roschkow's website as it was also me who deliberately did not post that link (but I did link to the eBay page). You probably haven't noticed, but I obfuscated that link because it was you yourself who -in my view !- was too negative ("too" in my book means : unjustified - without really knowing).

Last thing - and that is to show how complicated communication can be :
What is not in this topic is your kind of "quest" (put elsewhere) to Upsample/Filter etc. and everything what XXHighEnd could do for you; a kind of prerequisite. This, obviously, should droop from at least my responses. The TDA1543 just can not do that and with that should never be a good choice for you.

BUT (!!!)

Chris, you could be too theoretical here; following some pack perhaps. Look :


If that is you real stihl, then please adhere *that*.
What did I say ?
N-O-S
But now the real thing, meaning no upsampling/filtering etc., exactly as PQ adheres that (btw, to my recollection this was long after Audio Note existed, but alas).

My cautious conclusion :
You don't want any 24/768000 blahblah input on a DAC. You want 16/44.1 - period. It is what your idea is about how digital audio should work, because it *is* a way of thinking (seriously).
That I proceeded on that with the same ideas and especially the idea that it could be improved, well ... that costs extra $$ (apparently).


Of course you are. And you know what ? I personally can not think of a reason why not to. But to be very honest, I don't see a single difference with *any* of our customers because we all - no one excluded - look honestly for the very best (ok, against our $).
I saw that Mr Roschkov signed up for the forum and I did not know what to expect, nor why that happened in the first place. Possibly he wanted to counteract some of what has been said in here. Well, not necessary (I hope I just showed that). But, perfectly allowed. So Slawa, please do ! Warm welcome !

I hope all is well now.
And Chris, very good of you to be so honest (with your corrections). Appreciated !

Peter

Well, its a very small world on the internet. I am glad that Slawa found himself here as we spoke a bit about my use of your software.

My feeling is that all people that stray from the mainstream will  get criticism at some time - how many times have we read in forums that all cables 'sound' the same. This is more so, if one has never met someone and has to rely on limited information.

You and Slawa have a lot in common, though you have taken divergent paths you are both aiming towards audio Nirvana.

Long may it continue.

And Slawa, I have only just been told you are here so my response above was not an attempt to ingratiate myself with you :)

Greetings to Chris and Peter,

Just wanted say thank you for evaluating my audio design without any prejudice.

I am also feeling honored to be compared with Peter Qvortrup. I have a deep respect for him as a character, business person and for his courage of going against the wind of the audio industry. I know that there are many haters out there but that is fully justified given the success.

I would like to add a couple of words about the design of the SW1X DAC 01. The most overlooked issue that we are addressing in our designs is the impedance mismatch between the components. If the output and input impedances are not matched, the mismatch acts like a filter in the audible frequency spectrum, where in most cases low frequencies are lost. On top of that, the consecutive stages in the audio amplification chain loose the drive and the dynamics irreversibly. The earlier the impedance mismatch in the chain, the worse it is. The most detrimental impedance mismatch in digital sources known to me is between the digital transport and the S/PDIF receiver of a DAC but that is a different topic for now.

The issue of impedance mismatch is critical in 2 places of a DAC design: A) Between the current out of DAC chip after the I/V conversion and B) Between the first amplification output stage and the preamp/power amp input. Most common approach in the TDA1543 DAC implementation is to use a passive shunt resistor as I/V converter. This approach works fine but has some drawbacks: the shunt resistor worsens the impedance mismatch further and it takes energy away, therefore some of the dynamics are irreversibly lost and all subsequent stage are lacking the drive.

In order to avoid the mismatch in impedances, the output impedance needs to be ideally be multiples time lower than the input impedance. However, the current out DAC chips such as TDA1543 has high impedance output by nature but likes to see a low impedance input of the next stage. That is not happening if the next input stage is a valve since valves have high impedance input by nature.
In order to minimise the mismatch, many designs are using multiple TDA1543 DAC chips in parallel in order to lower the output impedance before the I/V stage. The drawback here is that voltage output after the conversion becomes relatively high, actually already high enough to saturate some of the preamp inputs. Therefore most DAC designs stop here and have no output stage. However, that solution causes the problem of impedance mismatch in point B) especially when valves and passive preamps are used in between.

We at SW1X Audio Design, on the other hand addressing the point A) by using a single transistor that does the job of I/V conversion, lowers the output impedance after the I/V stage and preserves the dynamics. Plus this approach allows us to use a simple but elegant class A, zero feedback valve output stage with low output impedance, which addresses impedance mismatch issue in the point B). On top of that the circuit remains elegant and simple and makes the music come alive with incredible analogue smoothness.

Best regards,

Slawa

I just ran into this little machine:

http://www.6moons.com/audioreviews2/metrum2/1.html

No filtering inside, supports up to 384kHz with USB. Price in tthe order of 1k euro/dollar.

This is the kind of DAC that will suit the virtues of XXHighEnd well.

Regards, Coen

Hi Coen,

Real 'poor man's' choice would be Soekris R2R DAC :)

US : http://soekris.com/products/audio-products/dam1021.html
EU : http://soekris.eu/shop/dac_modules_dam1021_en/

(With something better than suggested 'Amanero' USB/I2S receiver)

Rgds,
George

George,

If you want an in effect less than 12 bits DAC, yes.
And if that is what we call "real poor man's" then OK.

Peter

Interesting, Peter, this sounds much like Miska's comments on the Soekris.

I've listened to Soekris and was not impressed, though a friend listened also and liked it better than I did.

Jud, it is simple : with that kind of resistor accuracy, simple math tells this. Not that the accuracy of the here (optimally) used resistors is that bad, but it must be way way better in order to achieve e.g. over 20 bits of resolvement. All sorts of trickery is needed and it is far from easy.

Peter

While we are talking about resolving bits, did you see that the http://www.msbtech.com/products/dac4select.php?Page=dac4home (http://new top of the line MSB) is supposed to resolve 28.5 bits?  By my calculations that is a noise floor of -171dB...wow.

If you read the blurb they reckon they can now hear CD quantization noise.  US$90k+ though!!

Oh ... What MSB spits out, usually is one big lie. I sorted them out before with their $10K Femto Clock.

When I have nothing to do I'll look into it.

Anyway notice that without the noise being as low as your mentioned 171.5dB, it is even impossible. And, the noise being that low already looks quite impossible *and* measuring it (that low) looks "difficult".

Here's the accuracy of the resistors required for 28 bits (R/2R assumed) :
0,000-000-022-351-742 %
:yes:

Btw, with the NOS1a I can show the 23rd bit, but that won't be consistent (read : when that happens it is a bit ( ;)) of coincidence for the DAC or even channel in question).

Peter

Good to hear!  That 28.5bits resolution is an extraordinary claim.  I was wondering how they even measure such a thing...

They have a measurement device with a claimed accuracy of 27.5 bits resolution where no noise shows up from their DAC, perhaps?   ;)

Nearly a year ago now I started this thread and now feel the story of SW1X requires some conclusions.

Looking back at the early postings, we saw vintage technology being applied to Red Book CD playback, and in a modern World of upsampling and oversampling we were a little harsh on the new boy on the block.

I suspect of all the people here I am the only one to have heard the gear and been mightily impressed. But being impressed is not enough and it would be OK for you to be thinking “yes but was it good enough to buy”.

The answer is an emphatic yes.

I am in my sixties, and for a number of years had tired of fiddling with software settings to try and find the elusive audio nirvana. I am no Luddite. Computer based audio is a reality, but how to get a system that sounded like the best analogue, but was plug and play rather than needing a degree in computer sciences. What's wrong with just wanting to sit and enjoy the music without all the software paraphernalia?

The bottom line is I bought the following items from SW1X

1. Solid silver interconnects including USB cable. Many replaced Audionote silver cables
2. Solid silver mains cables
3. Solid silver speaker cables
4. Signature SPDIF to USB converter
5. Signature DAC
6. My own simple triode amp reworked by Slawa of SW1X at a modest cost, to vastly improve its sound quality
7. Open baffle speakers with vintage Alnico magnets on Grundig tweeters, Greencone Saba midrange, and Altec 416 VOTT base drivers. Speakers rewired by Slawa with silver wires, Black Gates and AN Kaisei capacitors etc

I didn't want this to turn out to be a sales pitch for SW1X, but Peter has agreed with me that the early negativity in this thread could do with a bit of balance, and I post this here with his full blessing.

This is quoted all the time. Why use analogue, why not use live instruments in a real acoustic space.

This is so entrenched that even musicians attempt to emulate the so called analogue sound in new recordings, Mark Knopflers latest "Tracker" is an example. Mark admitted this in a release interview.

I recognise this is about a NOS, in a system somewhat emulated from Audionote whose efforts largely priced themselves into the stratosphere. I'm unable to hear the Phasure NOS or your Signature NOS Dac being neither reside in New Zealand at this point in time. I certainly recognise that NOS is a very valid option and top of my list.

Anyway reading your efforts it probably sounds great(whole system) but possibly on the lean/bright side(guessing but I found this with pure silver cables). I certainly agree with just sitting and enjoying music without fiddling with bits and pieces. Unfortunately fiddling can result in a better sound so should never be ruled out. Fun at times.

You could try Xxhighend on a suitable computer(some computer knowledge required). I've compared it to other audio programs(JRiver, Foobar to name 2) and it still comes out on top. My recent comparison was with Daphile(Linux) running sound in Ram. I do fancy a simpler OS but XXhighend easily beat it using latest W10 14393 Ram file.

Quoting a "Linn" phrase nothing ain't more important than your front end: rubbish in rubbish out. Actually formulated from the famous vinyl LP12 era.

Is digital finally making ground now? I hope so.

All we need is more digital HiRes(24/96/192) quality recordings.

Robert

Cynical 61 year old music/Hi-Fi enthusiast who lives in a nice place with sea and sunshine!!!