Article: Berkeley Audio Design Alpha USB Review

[Vangelis]

Chris, thank you for sharing your experience with the Alpha USB. <br />

<br />

I'm under the impression that Berkeley Audio design the Alpha USB around Macs and they recommend using Pure Music right in their owners manual. If the Alpha USB was dialed in / voiced around Pure Music did this perform near or better than Windows with J Rivers?<br />

<br />

Did you try the Alpha USB into a DAC8 and if so was it a large improvement?<br />

<br />

Thanks Again<br />

Dave

[agentsmith]

Very cool, I would be very interested on your findings.

[easternlethal]

One would expect betters result using Alpha USB than without (just Debussy) because of more isolation + separate clocks.<br />

<br />

Using the USB, between the alpha dac and dcs the difference in technologies I can surmise are: dsp filters + mechanical isolation (dcs) vs electrical isolation + separate clocks (BA). Again one would expect BA to come out on top because it has a superior clocking solution, which I think the best way to address audio transfer, and dsps are a big no-no.<br />

<br />

It would be great to compare the results against this hypothesis. If dcs comes out on top we might have to hand it to them for superior implementation or the better quality components.

[Mark Powell]

I have both a Debussy and a Cambridge Dacmagic. The Dacmagic was purchased as a 'starter' DAC when I decided to have a go at computer audio. I quickly purchased a Musical Fidelity V-Link II for three reasons. Async, galvanic isolation via Toslink from it to the Dacmagic, overcome the DACs 48K USB limitation.<br />

<br />

Though a 'budget' device the V-Link greatly improved the sound quality of the Dacmagic.<br />

<br />

Then I purchased the Debussy.<br />

<br />

Since Chris' review of this Berkeley converter, I have, out of pure curiosity, been feeding the Debussy from the V-Link.<br />

<br />

I hear no difference at all. The V-Link most certainly does not degrade the SQ, but I do not think it improves it either.<br />

<br />

These USB to S/PDIF and/or AES converters are pretty simple devices. I have read a couple of 'multiple comparison' reviews, and the reviewers have found they are all pretty good and found little difference between any of them. The improvement, if any, has only occured when the USB implemenrtation of the following DAC has been poor, which is the case with the Dacmagic.<br />

<br />

So that is my experience with the V-Link to the Debussy - no change. I would be most interested in your experience should you get to try the Berkeley converter.

[Mark Powell]

pawel8's results will be very interesting.<br />

<br />

But I am curious about your statements. Note I am not criticising them. The difference in technology between *all* dCS DACS and any others on the market is fundamental and total.

[easternlethal]

sorry - I forgot another big difference. custom dac chip (dcs) vs. delta-sigma chip(BA). <br />

<br />

hmm... <br />

<br />

@mark powell - reputedly. because dcs makes their own chip. but the technology is not fundamentally different (they still use clocks, FPGA, circuit boards etc...) only the design and layout is different.

[Mark Powell]

They don't use any DAC chips at all, custom or otherwise. It is not delta sigma, its not R2R, its not single bit, its not bitstream. They use about forty 'DAC processes' which they have put into FPGAs. In what they call a 'ring', and look at the outputs in a random sequence, which they then combine.<br />

<br />

Sort of. Their DAC is unique, nothing else even remotely like it.<br />

<br />

But the do uae clocks, circuit boards, and electricity

[manisandher]

<em>"Am now comparing a Phasure NOS1 (usb) dac and hope to come up with some findings to report."</em><br />

<br />

I look forward to hearing your thoughts.<br />

<br />

Judging by the snippets of info that PeterSt is giving us, the next version of XXHE looks to be revolutionary, so it may be better to wait a short while until this is released.<br />

<br />

These are exciting times in computer audio!<br />

<br />

Mani.

[agentsmith]

Thanks for your informed analysis. <br />

<br />

I sort of assumed that a Firewire or USB to SPDIF converter should not be rocket science, and properly designed devices should not cost an arm and a leg to produce, and should all sound pretty much the same.<br />

<br />

And it would be interesting to listen to elaborate devices such as the Alpha, Weiss and Sonicweld. Obviously Chris seems to be convinced that the Alpha is a significant step up from budget converters. <br />

<br />

From your signature I am quite curious on your choice of Debussy matched with entry naim pre and Mac power amp. As an ex Naimee (282/200/Hicap/nDAC) I am very interested in how you find the new ND5XS as a source, VS the PC-Debussy setup. Are you also using the ND5XS with the Debussy DAC? or just analogue out from the ND5XS?

[easternlethal]

I didn't think using an FPGA to implement a dac was that unique. Xilink, Entegra have done it as well as diy'ers in the past (even incorporating PLL and DSPs).<br />

<br />

But I dcs have a patent so I suppose something must be different. Question is.. what? and why would changing the topology of a board (something that can be done inside a single chip) improve anything? would like to look into this somemore..

[PeterSt]

<cite>One would expect betters result using Alpha USB than without (just Debussy) because of more isolation + separate clocks.</cite><br />

<br />

Sometimes I'm asked what could be a good converter between the PC and the NOS1. Of course it must be a USB converter.<br />

I'm always having a hard time in explaning that something like the NOS1 is its own converter; it won't get any better by stuffing another converter in between (and next ask for SPDIF input - no wait, i2s, that's better).<br />

<br />

This (see quote abobe) is similar;<br />

<br />

So, we have this PC (etc.) and we need to go to something like the Alpha. Fine. Everybody learned to use a Lynx for that. Works. Should work well either. Clocks are in-PC, and something could improve there (use another brand of soundcard, apply some SOtM stuff, whatever).<br />

<br />

Now the USB-SPDIF converters come about. Okay, they may have AES connections; soundcards can have that too. No difference there.<br />

<br />

OF COURSE the converter must be async. That is better for jitter, they say. Good.<br />

So there we go, anync to that other box. Can be the Alpha USB converter. There we come - low jitter as they say.<br />

Then what ?<br />

<br />

Clocks are in that box now. Yes, they can be better than what was there in-PC. But now we have to travel over SPDIF again.<br />

True, reclocking makes a difference for sure. Even (and as for sure) when SPDIF is after that. But that doesn't take much. A few $20 oscillators will do the job. Make them $40 and the job might be better. Still there's devistating SPDIF behind it.<br />

<br />

What the async USB does in front of it, is a bit beyond me. But maybe I am missing a few things. Okay, it allows for the reclocking more easily (I think).<br />

<br />

But maybe someone can tell me how this can -theoretically !- be the best converter (sound wise of course) while converters exist with i2s output, which at least gives the theoretical reason to outbetter any SPDIF means, and then most probably already with $3 oscillators. Not that this few $ makes a difference for the sales price ...<br />

Or do we now say that the i2s connection will be too long, and SPDIF will always be better ?<br />

<br />

Assuming I did not make too many mistakes, I hope that people see how someone like me easily thinks that something else is going on, and that moving the USB (async) connection into the Alpha itself will always be a better solution, sound wise.<br />

If Berkely say they can not do that, then, well, I don't believe that.<br />

And it has few to do with a complete overhaul of the D/A section itself (what was suggested as the means to bring better sound).<br />

<br />

Peter (just trying to explain how I seemingly "blatantly" came to my so nicely in here quoted expression in my own forum)<br />

<br />

<br />

[easternlethal]

ha ha.. that is too cynical. <br />

<br />

This is a temporary solution for BA. I bet anything they will produce an Alpha III in the future incorporating USB, that will cost more than Alpha DAC II + USB. They will not be able to not produce non USB dacs. Even Weiss (longtime firewire proponent) has started incorporating USB into their 202. Even Phasure! Is this an 'honest' strategy? Well... it does benefit existing users.<br />

<br />

Anyway I don't think it's whether it's inside or out that is the issue. I think it's whether USB is implemented properly. If so then a USB converter would not do much. And I guessed dcs would benefit from the BA USB because I didn't think dcs implemented USB as well (their efforts were spent on the FPGA). speaking of which... doesn't NOS1 also use FPGAs?

[Mark Powell]

I still have the V-Link inserted because in previous trials of things I have found that on making a change it is sometimes difficult to hear a difference, if any. But if I leave the change for a week and then revert to as it was the difference, if any, seems to become more obvious.<br />

<br />

Will remove the V-Link today, but I don't expect a change. These converters are usually pretty transparent, as of course they should be. The original Dacmagic is known for its poor USB input (which of course I did not know about when I bought it) so the V-Link was a good purchase.<br />

<br />

Naim. I have always been keen on them, since they had factory open days in the mid 1980's. I live in the UK, only 15 miles from their factory, so I went. But I have only had the Naim stuff for a few months. I use the ND5XS with the Debussy, but the laptop and JRiver are my main source. The Naim preamp is used because I have other sources, not relevant to my 'signature', such as a turntable, an FM tuner, and a budget AV receiver. Using the HT bypass on the preamp lets me use the McIntosh for everything, including regular two channel TV broadcasts, without spoiling the 'purity' of my 'hi fi' setup.<br />

<br />

The change from the Dacmagic to the Debussy may seem a very large jump. But I don't believe in 'half measures' I chose the Debussy because I believe (many have, and do, disagree) that it is not worth spending 1000s of dollars on a box containing much the same chips as a 500 dollar box. dCS do it very differently.<br />

<br />

Now my 'hi fi' budget is totally blown for a year or two <br />

<br />

Regards

[Mark Powell]

How they do it is on their site, much better than my probably garbled 'explanation'. Clicking their 'sponsor' tag to the left of the CA home page will take you there. <br />

<br />

Cheers

[PeterSt]

<cite>speaking of which... doesn't NOS1 also use FPGAs?</cite><br />

<br />

Every D/A converter does. But the question is what for. Not for D/A conversion (like dCS seems to do).<br />

<br />

<cite>Even Weiss (longtime firewire proponent) has started incorporating USB into their 202.</cite><br />

<br />

Well, try to get anything decent out of DICE(II). For your fun : I ever was in the state of putting my signature under the software development (and specs) I wanted; in the end the (official) party for it, drew back. The DICE chip just can't do what we want nowadays, no matter that seems less than the Pro world requires (it just works there of course).<br />

May it say something : internal latency (which loops through the chip though) can be a couple of samples for normal soundcards and USB (not async); DICE won't get lower than 512. 256 when you are lucky.<br />

But of course this isn't about chips only; it is about what users want. And this doesn't seem to be firewire anymore ...

[easternlethal]

All the site really reveals is the magic's actually in the code (implying it can also used to manage sample rates). Feel free to share if you have anything more.

[One and a half]

These USB to S/PDIF and/or AES converters are pretty simple devices.<br />

<br />

Oh dear, Mark, what are you saying now. Simple is, even a toaster has a lot of though behind it, the materials to cope with the heat, how the crumb tray works, is it easy for the bread to pull out, does the toast heat evenly...you get my drift.<br />

<br />

Simple compared to what, a toothpick? You think it just has a few wires, IC's and that's it, anyone can make one? Sound much the same? Nice that you can refer to comparison reviews without stating them, just waffle my friend.<br />

<br />

No wonder the level of conversation at CA has gone downhill with fluff comments like this. BTW, am still waiting for a reply as to a $20 USB->Optical converter availability at Maplin.<br />

[Mark Powell]

But simple compared to a DAC, a laptop, a television, a car fuel injection system, etc. etc. etc.<br />

<br />

The comparisons are in HiFi News and HiFi World.<br />

<br />

I was mistaken not to come back to you. I also mistook what I was looking at. My apologies.<br />

<br />

What level of self-proclaimed 'erudition' would you like on this site? As soon as someone tries to get away from the mystical 'beliefs' that audio is somehow different from other electronics he gets shouted down. They are just consumer durables, like toasters.

[Elberoth]

Mark,<br />

<br />

I think that was a wise choice. dCS gear can be upgraded almost to no end. The main board is a bunch of FPGAs and DSP chips, that can be reprogrammed to acomodate any future formats, digital filters, interfaces etc. That gives you a certain peace of mind in terms of future proofnes that others produccts lack.<br />

<br />

In fact Debussy main processing board is exactly the same board that is used in Scaratti. Since Scarlatti launch in 2007, they had no problems in reprogramming it to include asymetrical (apodising) digital filters, to accept 24/192 input signals on all inputs, to perform DSD upsampling (on the same board, in Puccini), to add USB interface (in Debussy), and most recently - to accept DSD over USB. All on the same board !<br />

<br />

The old board from Elgar was used for 11 years, and was only discontinued because some chips they were using become unavailable (those chips were not ROHS compliant and manufacturers of those chips, instead of designing a new ROHS compliant version of those chips, discontinued them). That board was also reprogrammed many times - first to accept 24/192 signals on Dual AES, then DSD over FireWire and also asymetric digital filter for 192 signal (that was around 1999, when their upsampler came out; BTW - dCS, not Meridian was the first company to implement asymetrical digital filters, although that fact is very often forgoten, since it was the Meridian who made the whole fuss about this type of filter in 2009 - some 10 years later !).<br />

<br />

Anyway, getting the dCS stuff, you are subscribing to a certain upgrade program, that noone in the industry is offering.

[Frank Mercurio]

Chris,<br />

Would be interested to know if and when you had the NAD M2 on the list for reviewing. Given that it is a true digital piece of hardware I would be curious to see what Chris thought about it - any room for this product review on your busy schedule? Thanks

[Paul R]

@Peter St. - I am not sure what, but something seems wrong with that logic to me. If that were really the case, you should be able to get Phasures mass produced in China at 1/20th of the current cost, sell them at $995, and see no reduction in build or sound quality.<br />

<br />

I think you will agree that other factors come into play there, right?<br />

<br />

I do know there is a difference in the sound of a V-Link, a Wavelength, and a Highface USB converter fed into several DACs, so it really is not difficult for me to believe the Berkeley sounds different and better, especially since Chris' review strongly suggests that is the case.<br />

<br />

@Mark - I do not know if the difference in sound holds going into a DAC like the dcs. It may well be that as the quality of the DAC increases, the difference a USB converter can make is less and less, probably a case by case thing. It is interesting to hear from you that a V-Link works well with the dcs. V-Links are one of my favorite hunks of gear. They are great to demonstrate to folks jut how much difference USB can make over say, an optical output on a PC. (A lot in some cases!)<br />

<br />

Paul<br />

<br />

<br />

[PeterSt]

<cite>I think you will agree that other factors come into play there, right?</cite><br />

<br />

Hey Paul - I really don't know what you are getting at.<br />

Is it that a mentioned a few "$" a couple of times, and thus you think I implied it should all be way cheaper ?<br />

This was not my message at all. This was mainly two-folded (without saying it all) :<br />

<br />

1. A device like this is made to keep your legacy DACs, be they 16/44.1 only or higher;<br />

2. I refuse to believe that Berkeley was not able to make a USB DAC out of the Alpha for less money in the same period of time.<br />

(and the remainder is to your imagination, but coincidentally I spoke about that remainder as a one-liner only, on my forum)<br />

<br />

There is no way I implied that a device like this should come cheaper. If you won't believe (the merits of) *that* ... if you want I can dig up a forum post where I proposed what an USB-SPDIF converter should cost when made out of the ready components of the NOS1 (which just *is* that converter for half of its part). IIRC it came down to 1500 euros and it wouldn't go for less. This is more than the Berkeley converter in US dollars ...<br />

<br />

Regards,<br />

Peter<br />

<br />

Edit : (but don't read) : http://www.phasure.com/index.php?topic=1506.msg15135#msg15135<br />

It was 1200 euros. But this was not about the USB interface and the ton of input/outputs which can be there with that interface. The price really would be similar to the Berkeley.

[Mark Powell]

Yes. MF stuff is always good, and only when you get to either their huge amps like the Titan or their class A amps do they get high priced. Then they are quite capable of charging you as much as anyone else <br />

<br />

I heard no difference with the V-Link inserted, which to me, is good. A converter should *always* be transparent. It changes the signal dformat, and that is *all* it should do. It is not supposed to be a 'modifier'. It worries me slightly that differences can be heard between them. To me they should only be used when the DAC has a poor USB input (as in the case of the Dacmagic), or does not have a USB input at all.<br />

<br />

Whatever. Thats my opinion. For intesr4st, I just read some reviews in the latest UI magazines.<br />

<br />

HiFi News thinks very highly of the Sonicweld. They tested only that, and did no comparisions, though they have done before. It is priced to make the Berkeley look 'budget'.<br />

<br />

HiFi World compared several, using their best reviewer, Rafael Todes, a musician with the Allegri String Quartet. They compared the Kingrex UC 192, Halide Bridge, MF V-Link 192, SOTM DX, and Audiophilleo I. They used JRiver with JPlay, and the Weiss DAC 202. They thought the Audiophilleo was the best.

[Jud]

<i>I refuse to believe that Berkeley was not able to make a USB DAC out of the Alpha</i><br />

<br />

That's really what it comes down to: whether the separate box results more, or purely, from audio considerations, or more, or purely, from commercial considerations.<br />

<br />

Since both you and the Berkeley folks are far more knowledgeable than I am, and you have come to different conclusions (you, one box, Berkeley two), and I haven't listened to either, I have no current basis on which to make a judgment. But one box or two is a question that interests me. Instinctively I'd tend to favor one box, but as I say I have no real basis on which to judge.

[Mark Powell]

Using two boxes they have had to go from USB to S/PDIF (as that is all their DAC can take) and then from S/PDIF to the 'internal' I2S, thus introducing an unnecessary double conversion. I am not an audio engineer either, but it smacks of 'let us use what we have already got' rather than anything else.<br />

<br />

But of course, taken by itself, it is a perfectly good USB to S/PDIF converter, suitable for any DAC that does not have a USB interface, such as theirs. I suggested earler that we should take it as that, and nothing more.<br />

<br />

PS: I do wonder why people seem to be concerned. Musical Fidelity, for example, make both S/PDIF converters and DACs, just the same as Berkeley. I do not see the same concern about them.