Best USB to SPDIF converter

[wisnon]

Very cool. Not much Stahl-Tek talk around here.

 

But you can buy a Auralic Vega for that price or a E20 and pocket the change!

[rbmyers]

I have tried all the convertes from your list (minus ART anad BADA) and in my opinion Audiophilleo 2 with the Pure Power battery PSU option betters them all. It is a big step forward compared to the regular AP1/2 (which is USB powered).

 

dCS is OK, but only make sense in an all dCS setup where you can use the clock link feature and by soing so, squeeze some extra performance out of it. Soulution, apart from beeing hugely dependent on the quality of USB power, is way overpriced for what it is. Has a cool case though.

 

The other converter which can be a contender to the current king-of-the-hill title is M2Tech EVO, but ONLY after connecting the optional EVO Clock and EVO battery power supply. Without those, is is only avg.

 

 

Elberoth, you say that the AP best the OR-5. Could you give a little more detail about the differences between them. If they are close in performance, the AP seems the way to go, given the price difference

[Elberoth]

Elberoth, you say that the AP best the OR-5. Could you give a little more detail about the differences between them. If they are close in performance, the AP seems the way to go, given the price difference

 

They are a bit difficoult to compare, as they sound very different to each other.

 

ORT5 is a bit dark sounding. You will not get the vast sound stage that AP1/2 will throw, you will miss some spatial clues and decay trails. On the positive side, ORT5 has the best texture out of all converters I tried, and was the only converter that fully matched my Scarlatti CD/SACD transport in that regard.

 

AP has superior resolution and better microdynamics than the ORT5. If you voice your system around the AP, ORT5 will most likely sound a bit muted.

 

That being said, in some systems AP may be too much of a good thing. Some people reported that AP sounded a bit bright in their systems. I belive this is may be computer dependent (since AP is USB powered, it relies heavily on the quality of USB power, so cards like SOtM USB with their own ultra low noise voltage regulators can help here).

 

Both converters can be improved by adding a decicated PSU.

 

A friend of mine - Marcin of JPLAY fame - who borrowed ORT5 from me, reported what he called 10% improvement in SQ by substituting the supplied switch-mode wall wart PSU with his KingRex linear PSU. Aparently, Empirical even makes their own battery PSU, called Monolitic, but it adds another $1k to the hefty price of ORT5.

 

AP recently introduced what they call a PurePower battery PSU. This is a $449 option for both AP1 and AP2. I recently got one, and highy recommend it. AP+PP sounds smoother, has better texture and even better soundstaging that the plain vanilla AP.

 

AP+PurePower can now rival the M2Tech 3-box solution sound wise, but it is cheaper, much easier to use (no manual frequency change necessary), comes in a one box less and brings much less cables.

[Elberoth]

I just got a Stahl~Tek A.B.C. (MSRP $3,500): http://www.stahltek.com/abc.htm; this converter is amazing! Digital never sounded so good in my system and my DAC (Musical Paradise D1) probably reached its full dynamic range that was only on paper in the past! As a matter of fact, all other components in my system (amp and speakers) are possibly revealing their full capabilities as well. I currently also have a Bel Canto uLink (MSRP $675) and Stahl~Tek easily blew it out of the water; well, considering the huge price difference between the two, this should be expected. Previously I owned a Musical Fidelity V-Link (first version). I highly recommend Stahl~Tek if you have the funds - it will transform your digital system! However, since I didn't have the chance to compare Stahl~Tek to other high end converters like Off Ramp, Sonicweld, and HiFace EVO stack, please take my recommendation with a grain of salt.

 

Thanks for sharing, Looks really nice inside - battery PSU, quality digital transformers. Like the BADA Alpha USB, they probaby use ADuM family digital isolators (those chips combine high speed CMOS and monolithic transformer technologies) to isolate the 'dirty' USB side from clocks and SPDIF out.

 

One thing that makes me wondering is the absence of the clocks. I can only see an ordinary looking, single (common ?) clock here.

[barrows]

Yeah Adam. You are seeing the same thing I am. Looks like an XMOS based converter with isolation chips between the XMOS/computer side and the clock/SPDIF side with a simple battery supply. I only saw one oscillator as well, perhaps Stahltek prefers to use a single reference frequency and a digital clock synthesizer? Not sure one should be asked to pay $3.5K for this though... perhaps it has a really fancy chassis?

[Elberoth]

True - price is on the high side.

 

If the clock they use is 14.112MHz they can do simple math as 14.112MHz / 294 = 48kHz and 14.112MHz / 320 = 44.1kHz.

[4est]

Yeah Adam. You are seeing the same thing I am. Looks like an XMOS based converter with isolation chips between the XMOS/computer side and the clock/SPDIF side with a simple battery supply. I only saw one oscillator as well, perhaps Stahltek prefers to use a single reference frequency and a digital clock synthesizer? Not sure one should be asked to pay $3.5K for this though... perhaps it has a really fancy chassis?

 

Agreed-including the silicone attaching the LED indicators!

 

It would be my guess that the clocks are beneath silicone blob near the XMOS chip. The one across the isolation bridge is most likely the HDMI receiver clock. Xtal oscillators are very sensitive to temperature and vibration as well as P supply. The blob of goo may be just to act as a stabilizing agent. I know in my TP diy DAC the on board(now unused) Crystek clock was very sensitive to temperature. If you blew on it it would lose lock.

[barrows]

Agreed-including the silicone attaching the LED indicators!

 

It would be my guess that the clocks are beneath silicone blob near the XMOS chip. The one across the isolation bridge is most likely the HDMI receiver clock. Xtal oscillators are very sensitive to temperature and vibration as well as P supply. The blob of goo may be just to act as a stabilizing agent. I know in my TP diy DAC the on board(now unused) Crystek clock was very sensitive to temperature. If you blew on it it would lose lock.

 

4est: I doubt the clocks are on the XMOS side of the isolation chips, as this would be the "wrong" way to do it. Typically, USB-SPDIF interfaces isolated in this fashion put the clocks (and their ground and power supply) on the isolated side, and then send the clock signal back through the isolators to the XMOS chip. Then they re-align the I2S lines with "clean" clock signal on the SPDIF side. This is how Berkeley, Weiss USB, Ayre, and Wavelength all isolate the critical clocks from the "bad" ground plane of the computer and XMOS processor. The USB receiver does require a clock of its own (non-critical), that might be under the silicone. Silicone is great for damping parts on PCBs, Empirical damps its oscillators (CCHD on the "TurboClock" versions). I really suspect Stahltek is using digital frequency sysnthesis from the single oscillator on the isolated side. While doing it this way generally results in a little more jitter than fixed frequency oscillators, it may make it possible for them to manipulate the jitter spectrum for a euphonically pleasing presentation by designing their algorthm appropriately.

[4est]

Yes, that makes sense Barrows. I was thinking about it backwards. Not the first time!

 

There is an additional pad for a second clock near the other. They probably made provisions for both one or two within the design, choosing one in the end for reasons akin to your suggestions.

[tubesound]

Thanks for sharing, Looks really nice inside - battery PSU, quality digital transformers. Like the BADA Alpha USB, they probaby use ADuM family digital isolators (those chips combine high speed CMOS and monolithic transformer technologies) to isolate the 'dirty' USB side from clocks and SPDIF out.

 

I was scratching my head about why they charge so much for this converter. The case/parts certainly didn't add up. Stahl~Tek is a small Texas based company. R&D and small volume of sales probably all played a role. Ultimately, I believe they have high faith in the sound quality of this converter.

 

I don't know much about the technical stuff. The original V-Link used only one clock but worked pretty well too (maybe people can argue this is a not true asynchronous application). I don't believe the ADuM USB isolation chips support USB 2.0 High Speed 480Mb/s yet while Stahl~Tek's specifies A.B.C. can output up to 24/192. I'm using an iFi iUSB in front of A.B.C. and I got a bit smoother sound with this addition.

[dallasjustice]

They are a bit difficoult to compare, as they sound very different to each other.

 

ORT5 is a bit dark sounding. You will not get the vast sound stage that AP1/2 will throw, you will miss some spatial clues and decay trails. On the positive side, ORT5 has the best texture out of all converters I tried, and was the only converter that fully matched my Scarlatti CD/SACD transport in that regard.

 

AP has superior resolution and better microdynamics than the ORT5. If you voice your system around the AP, ORT5 will most likely sound a bit muted.

 

That being said, in some systems AP may be too much of a good thing. Some people reported that AP sounded a bit bright in their systems. I belive this is may be computer dependent (since AP is USB powered, it relies heavily on the quality of USB power, so cards like SOtM USB with their own ultra low noise voltage regulators can help here).

 

Both converters can be improved by adding a decicated PSU.

 

A friend of mine - Marcin of JPLAY fame - who borrowed ORT5 from me, reported what he called 10% improvement in SQ by substituting the supplied switch-mode wall wart PSU with his KingRex linear PSU. Aparently, Empirical even makes their own battery PSU, called Monolitic, but it adds another $1k to the hefty price of ORT5.

 

AP recently introduced what they call a PurePower battery PSU. This is a $449 option for both AP1 and AP2. I recently got one, and highy recommend it. AP+PP sounds smoother, has better texture and even better soundstaging that the plain vanilla AP.

 

AP+PurePower can now rival the M2Tech 3-box solution sound wise, but it is cheaper, much easier to use (no manual frequency change necessary), comes in a one box less and brings much less cables.

If you like the AP so much, why selling it?

Audiophilleo AP1 + PurePower - free shipping | DA Converters | AudiogoN - The High-end Audio Community

[Elberoth]

If you like the AP so much, why selling it?

Audiophilleo AP1 + PurePower - free shipping | DA Converters | AudiogoN - The High-end Audio Community

 

Because BADA Alpha USB - which images I posted on the prev page - bested all converters I tried, including AP+PP combo.

 

AP+PP is still a good value though. AP2+PP (which is the BADA equal in terms of features) is $1000 and doesn't need an expensive SPDIF cable, so it is still 2-3x cheaper than BADA option (depending on the digital cable you will buy to go with BADA).

 

AP1 is $400 more than AP2, but has no equal in terms of features among other SPDIF converters. For many, the integrated display or dithered volume control will be worth $400 extra.

[barrows]

I was scratching my head about why they charge so much for this converter. The case/parts certainly didn't add up. Stahl~Tek is a small Texas based company. R&D and small volume of sales probably all played a role. Ultimately, I believe they have high faith in the sound quality of this converter.

 

I don't know much about the technical stuff. The original V-Link used only one clock but worked pretty well too (maybe people can argue this is a not true asynchronous application). I don't believe the ADuM USB isolation chips support USB 2.0 High Speed 480Mb/s yet while Stahl~Tek's specifies A.B.C. can output up to 24/192. I'm using an iFi iUSB in front of A.B.C. and I got a bit smoother sound with this addition.

 

Yes: the ADuM isolators, and most all other isolators, are not "fast" enough to isolate USB class 2 high speed signals. But in the Stahltek, as in the Berkeley converter, the isolators are not isolating a USB data stream, they isolate the I2S lines between the XMOS chip, and the SPDIF output. Current isolation chip technology is fast enought to isolate I2S at any data rate. Interestingly, it has been reported over at diyaudio.com that there is a new high speed isolation chip just coming available which claims to be fast enough to isolate USB class 2 high speed signals: if this is true, isolated asynchronous USB audio interfaces can get better and easier to make.

[Elberoth]

I don't believe the ADuM USB isolation chips support USB 2.0 High Speed 480Mb/s yet while Stahl~Tek's specifies A.B.C. can output up to 24/192.

 

They put ADuM chips on I2S lines, not USB line.

 

UPDATE: Barrows was first this time

[cowboy]

All those converters discussed here are very expensive. Take a look at this one QLS QA01, dirty cheap with good implementations on external power supply and ASIO support.

[astrostar59]

Quest,

 

i had tremendous sound quality with an M2tech Evo and a custom shunt Regulated power supply, Top notch USB Audiophile and BNC, SPDIF cables and a BNC terminated RF attenuator.

 

Hi Lizard_King

Hey, I have the Audio Note DAC 3.1 SE kit arriving next week, and are to hook it up to an EVO plus clock plus outboard power supply built by Cables R Us UK. I used to have an Audio Note DAC 1.1 kit. with a CEC TL51 CDP. I am jumping into PC audio with a Mac Mini (I use Macs day in day out as am a web designer).

 

So I am pleased to read your post. I know the Hag USB in the AN kits was bad, and really was an entry level option only. I think Brian is looking now at a better board. I kinda prefer the ability to buy it outboard, and play around with options. I am hoping the Mac Mini plus EVO stack will equal or better my CEC CDP. It should in theory I think. I was to use Pure Power software, and stuff 8 gig ram in the Mac Mini.

 

I use Audio Note AN-V RCA Coax for the EVO to DAC, and Audio Note RCA AN-V for DAC to pre amp. I use a factory Audio Note M3 pre and a factory Audio Note Conquest monoblock pair with 2 x 300b tubes and 1 x 6SN7 each channel.

 

I note the DAC 3.1 uses 2 x 5687s on the output board, which my M3 does as well, also the same is a 6X5 for the rectifier (which I swop out for a Bendix 5852 as it has more power draw and better bass).

 

I am a bit nervous doing all these mods to my system in one go. I was happy with the DAC 1.1 sound, though it was a tad forward though very organic and smooth, with no treble edge. It's only failing if it had one was it was a bit hissy and hum levels were a tad high. It couldn't be beat at it's price though, and was close to vinyl.

 

I am not sure why more folk don't look at the Audio Note Dac kits (or UK factory products) as they are superb DACs. The nearest I have heard to analogue and can listen for hours with no fatigue.

They major on NONE OVERSAMPLING AND NO FILTERING.

The parts quality is also top notch (BlackGates, Elna Cerafines, Tantalum Resistors, Silver wiring, Top quality design, IV Transformers between the digital and analogue sections (most important) and Transformer Coupled Output.

[astrostar59]

One more thing, I have heard the DAC 3.1, and it is mega quiet, and for a tube unit remarkably so. The design follows elements in the awesome and mega expensive Audio Note DAC 5 many hail as the best ever tubed or none tubed DAC.

 

I think for me, the Audio Note brand has a sound I can live with without trying. I am not sure if it is the None Oversampling, or the lack of Filering, or the fact Tubes are used for power regulation and outputs. Possibly all 3, but it floats my boat.

It just sounds more natural. If anyone says a tube system can not have the resolution and bandwidth of a solid state system, they have not heard a good Audio Note set up.

 

The other cool thing I like is the ability to tube roll, and tweak the sound for system synergy. Once you get to a higher level, it is all tweaking here and there, it's not drastic, but can make the sound go to another level.

[4est]

"It just sounds more natural. If anyone says a tube system can not have the resolution and bandwidth of a solid state system, they have not heard a good XXXXX "

 

I have been fortunate to be around a couple of guys that have really pushed the limits with tubes. Two extremes as well. One into horns and vintage Altec, Jensen and Western Electric, the other scratch built into electrostatics.

[wkhanna]

 

I have been fortunate to be around a couple of guys that have really pushed the limits with tubes. Two extremes as well. One into horns and vintage Altec, Jensen and Western Electric, the other scratch built into electrostatics.

 

I can not resist inserting a recent photo I took at our local shop.

 

 

[Mark V.]

Quest,

 

i had tremendous sound quality with an M2tech Evo and a custom shunt Regulated power supply, Top notch USB Audiophile and BNC, SPDIF cables and a BNC terminated RF attenuator.

 

Hi Lizard_King

Hey, I have the Audio Note DAC 3.1 SE kit arriving next week, and are to hook it up to an EVO plus clock plus outboard power supply built by Cables R Us UK. I used to have an Audio Note DAC 1.1 kit. with a CEC TL51 CDP. I am jumping into PC audio with a Mac Mini (I use Macs day in day out as am a web designer).

 

So I am pleased to read your post. I know the Hag USB in the AN kits was bad, and really was an entry level option only. I think Brian is looking now at a better board. I kinda prefer the ability to buy it outboard, and play around with options. I am hoping the Mac Mini plus EVO stack will equal or better my CEC CDP. It should in theory I think. I was to use Pure Power software, and stuff 8 gig ram in the Mac Mini.

 

I use Audio Note AN-V RCA Coax for the EVO to DAC, and Audio Note RCA AN-V for DAC to pre amp. I use a factory Audio Note M3 pre and a factory Audio Note Conquest monoblock pair with 2 x 300b tubes and 1 x 6SN7 each channel.

 

I note the DAC 3.1 uses 2 x 5687s on the output board, which my M3 does as well, also the same is a 6X5 for the rectifier (which I swop out for a Bendix 5852 as it has more power draw and better bass).

[ATTACH=CONFIG]3353[/ATTACH]

 

I am a bit nervous doing all these mods to my system in one go. I was happy with the DAC 1.1 sound, though it was a tad forward though very organic and smooth, with no treble edge. It's only failing if it had one was it was a bit hissy and hum levels were a tad high. It couldn't be beat at it's price though, and was close to vinyl.

 

I am not sure why more folk don't look at the Audio Note Dac kits (or UK factory products) as they are superb DACs. The nearest I have heard to analogue and can listen for hours with no fatigue.

They major on NONE OVERSAMPLING AND NO FILTERING.

The parts quality is also top notch (BlackGates, Elna Cerafines, Tantalum Resistors, Silver wiring, Top quality design, IV Transformers between the digital and analogue sections (most important) and Transformer Coupled Output.

 

I own AN DAC kit 4.1x, I am using MAC Mini with Amarra for playback. I used to have Halide SPDIF Bridge, but since a week I have AP2 + PP. There is A LOT of difference between those 2 USB SPDIF converters! I never have thought there could be so much difference... At this moment I still can't tell if I am happy with the AP2 + PP, as there is quite a bit more energy in the upper mid and highs. (which I don't like at all...) There is also deeper and wider soundstage with better immaging. (more stable) After reading a lot of reviews and knowing from my own experience what the Halide and the AP are doing in my system, I think Off Ramp 5 will be best USB/SPDIF converter for me. But for now I don't have the Off Ramp money...

 

Concerning the AN DAC kit, it's a nice sounding DAC, but I already improved the SQ quite a bit with Duelund coupling caps and AN factory digital board and digital PS. (which is TOTALLY different from the AN kit digital board and it also sounds TOTALLY different)

 

Mark.

[Jud]

I am not sure why more folk don't look at the Audio Note Dac kits (or UK factory products) as they are superb DACs. The nearest I have heard to analogue and can listen for hours with no fatigue.

They major on NONE OVERSAMPLING AND NO FILTERING.

The parts quality is also top notch (BlackGates, Elna Cerafines, Tantalum Resistors, Silver wiring, Top quality design, IV Transformers between the digital and analogue sections (most important) and Transformer Coupled Output.

 

Watch out for marketing language that cannot be technically correct. "No filtering" is an example. Audio Note emphasizes in their marketing that they have no on board analog or digital filtering. However, it is a simple fact that the digital bitstream must be filtered to get music, because the unfiltered bitstream sounds like a very loud and unpleasant wash of static, as anyone who's ever experienced loss of signal lock in a digital rig can tell you. What these DACs do have is a transformer (so not "on board") that acts as an analog "brick wall" low-pass filter for the bitstream. This "brick wall" filter creates the distortions that led most DAC manufacturers to use oversampling in order to avoid them. A very complimentary review I read of one of the Audio Note kits said the distortions should be inaudible because they were 30db down from the music signal. I note in this regard that with mp3 a sound is usually considered "masked" if it is 40db down. Thus the Audio Note DACs are presenting distortion at a level that would typically be considered unacceptable in an mp3.

 

So you may like the sound of these DACs, and avoiding oversampling does leave out distortions that oversampling itself can cause, but recognize that they have their own inaccuracies.

[Mark V.]

Even if the AN Dac's from "theoretical" view hardly can deliver bearable sound, they actually DO deliver in real live!

[Mark V.]

What these DACs do have is a transformer (so not "on board") that acts as an analog "brick wall" low-pass filter for the bitstream.

 

Do you mean the "interstage transformer"? In that case you can hardly talk about a "brick wall" low-pass filter, as the AN DAC's sound wonderful with the Sowter 9762 Interstage Transformer. This Transformer has the following spec: +/- 1,5 db 5 Hz - 100 kHz. I cannot find any other transformer in my DAC (except for the output transformers of the line stage, which have even more extended bandwith).

[4est]

Do you mean the "interstage transformer"? In that case you can hardly talk about a "brick wall" low-pass filter, as the AN DAC's sound wonderful with the Sowter 9762 Interstage Transformer. This Transformer has the following spec: +/- 1,5 db 5 Hz - 100 kHz. I cannot find any other transformer in my DAC (except for the output transformers of the line stage, which have even more extended bandwith).

 

A good transformer is a thing to behold. Most people that poo-poo them most likely have not heard a really good one. I have not heard this Sowter, but a very good transformer is pretty darn transparent. Sadly, they are pretty darn expensive too.

[Jud]

Do you mean the "interstage transformer"? In that case you can hardly talk about a "brick wall" low-pass filter, as the AN DAC's sound wonderful with the Sowter 9762 Interstage Transformer. This Transformer has the following spec: +/- 1,5 db 5 Hz - 100 kHz. I cannot find any other transformer in my DAC (except for the output transformers of the line stage, which have even more extended bandwith).

 

Oh, I said nothing about "bearable." Many people enjoy the sound of these DACs. I was simply pointing out they cannot be accurate, for well understood mathematical and physical reasons. To what extent these inaccuracies are audible (and might even be responsible to some degree for listeners' enjoyment) is a matter for discussion. Certainly these types of DACs (non-oversampling and using a brick-wall filter on 44.1 and 48kHz signals) have much higher measured distortion than is typical of other DACs.

 

Regarding the term "brick-wall," it has absolutely nothing to do with the transformer specs you've cited, nor with the transparency of sound of which the transformer is capable in various applications. The Audio Note DACs typically use excellent parts. Rather, "brick-wall" refers to the steepness of the slope of the analog filter the transformer is used to implement in this DAC. The filter's slope is extremely steep, as confirmed by test measurements in the review I read. That says nothing about the transformer's quality. It is just the way the designer intentionally implemented the low-pass filter, to almost completely stop anything in the frequency band above audibility. Mathematically and physically, this choice of the designer has well known consequences in the audible frequency range, and indeed the review measurements also confirmed this.

 

Edit: And let me once again mention that the review was not at all critical. To the contrary, it praised the sound of the DAC.