Analog Reseach Technologies (ART) Legato: First impressions

[Part-Time Audiophile]

Everyone has one of those moments, I think, when they just do something and almost immediately wonder WTF they were thinking. I wish I could say that this is unfamiliar to me, or at least, less unfamiliar. Oh well. I guess impulse control isn't really one of my strong suits. Next time on the wheel, maybe that'll be the lesson I'll be here to learn.

 

In this latest bit of needless excess that is my life as a stereophile (this is opposed to an audiophile, who's after great sound -- I'm after that, yes, but I also have a fetish for nicely made gear), I purchased another USB to S/PDIF converter. The Analog Reseach Technologies (ART) Legato showed up on Friday, in the middle of lovely sleet shower, brought by those wonderful rain-sleet-or-snow guys at the USPS. Nothing like putting that to the test.

 

Anyway, I'm still not totally sure why I picked this up. I already have an (RCA-based) M2Tech HiFace that is connecting my loaner Berkeley Alpha to my G5, and doing a bang-up job of it at that. Truly. And the kicker is that the HiFace supports hi-res (like the Alpha) and the Legato is strictly redbook only.

 

I've been following the Legato for a while now, chatting periodically with Pat, reading his posts on this forum, and I guess I must have been very impressed with Gordon Rankin's unequivocal recommendation of him. Witchery, all 'round.

 

The Legato, unlike the HiFace, is based on two things. One is that asynchronous USB code, licensed from Rankin's Wavelength company. Like Ayre did with their truly wonderful QB-9 DAC, my personal top contender as a replacement DAC until I ran foul of the Alpha. Second is his hand-picked clock, a particular piece of kit that Pat claims is the best available. Together, I was convinced, would be a piece of kit that would virtually eliminate (or at least push past the threshold of problematic) the jitter bugaboo that virtually everyone spending any time reading up about computer/digital audio is deathly afraid of. Pat ices the cake with his U-Byte BNC coax S/PDIF cable that he has all manner of charts, diagrams and test results about that show all manner of technical superiority over the implementation and design. Pat is very much the evangelist about this part of the design, so don't take my hand-waving as anything other than what it is; I was sold on the fact that the Legato had BNC, my HiFace didn't, and I wanted to go adapterless into the Alpha.

 

A note about that cable. I bought it as part of the unit, and he sold it for $50 off because of it. I mention it here because that damn thing is 5m long (I really ought to have read more before I bought the unit)! A 16' digital cable for $150 is a steal by audiophile standards, so what's not to like? Well, all manner of badness happens in long cables (reflection, skin effects, interference, just to name a few) that simply aren't pertinent to shorter cables, where things like resistance, capacitance and inductance are minimized due to their negligible lengths (my theory on why shorter cables sound better, BTW). So, a 16' cable? Wha ... ??? Isn't that a problem?!? My emails expressing this concern met with a polite smack to the back of the head (via email), indicating that if I was truly curious, I'd have to read up on it on his site and the various forums. LOL. Touché. Pat says that more info about the cable can be found here:

 

http://analogresearch-technology.net/ubyte.html

 

Anyway, I ordered this thing and almost immediately had cold feet. What was I thinking? I was interested in the Alpha because of high-resolution audio! Wouldn't getting the Legato pretty much eliminate that as an option -- DUH??? So, I wrote him to cancel and simply refund my money. To my chagrin, he'd already shipped it -- but he said, quote:

 

"We always take things back if the customer is not happy."

 

Ok, no problem, I thought. Pat said I have 30 days to get it back to him if I don't like it, so what's the harm?

 

For various reasons, my unit is lacking the final faceplate, so I can't really comment on the fit and finish of the finalized product. My unit is a silver, sealed box. Metal on 4 sides, with a plastic front which'll be swapped out when the final plate arrives. Four black plastic footers, two black plastic bands circling the casing -- one around the front face plate, one around the back. A BNC connector. An IEC connector. And a Type B USB connector. And that's all, folks. No diodes. No LEDs. No switches. No dials. Nada. The quintessential black (okay, silver) box. Just plug it in and it works. Right?

 

Well, yes. It popped right up in the Sound applet in System Preferences. The cable came with a BNC-to-RCA adapter that I was too stupid to figure out how to remove without 5 minutes of fiddling, but then, I'm an idiot.

 

I plugged everything in, turned on the system, and ... magic.

 

Seriously? Um, no. If you're reading this, you've probably already had the misfortune of reading a previous post of mine, and for that experience, I apologize. As you'll no doubt recall, I'm hardly one to start hyperbolizing, and you'll be happy to note that I won't be starting now.

 

That said, the Legato is very good. Very good indeed.

 

While I've only had it a day now, the unit isn't broken in (if such is even required) and the cable is brand-spanking new (again, saying this presumes that cable break-in is also a real phenomena, but I'm putting that aside for now). So, draw from that what you may. But what leaped out at me was how clean the bass was. Note that this is precisely what I learned the Alpha did above and beyond the PS Audio -- and with the Legato, it was more so.

 

Let me unpack that a bit so you have some understanding why this is so critical to me. I have a big room, if oddly shaped. It's 13' across the listening wall, and 40' long. In fact, the section of the room, which is a basement, that I sit in is the top-left bar of a capital letter "T". Let's call this non-ideal and move on. The challenge for me has never been one of bass boominess or lack of articulation -- bass was simply absent. I thought my speakers had "gone bad" or something, or perhaps the amp was limping along on half it's fuses or something. I bought a couple of Stereophile test-tone CDs and learned two things. One, bass performance drops radically at ~50Hz. Two, this is perfectly normal for human hearing (oops). Okay, so maybe bass was something other than what I had always assumed (we learn something every day, no?), which was probably artificially boosted bass, à la my home theater with it's monster dual-12" cone subwoofer (that is some serious bass). I then got over it. I was simply not going to load a room that size with the passive 10" woofers on those Totems, regardless.

 

Until I got the Alpha in here, and suddenly, I had some bass. Not HT bass, but there was some punch. Crisp. Clean. Delineated. Natural. Coherent. All that good stuff that you read about.

 

And then I added the Legato. And everything tightened up more. I think (again, a/b tests have yet to be done, but honestly, I'm having trouble caring, this sound is addictive).

 

In short, I'm loving this solution.

 

Before I sign off, a comment about the HiFace. It's great. And for $150, you can't shake a stick at it. But, so far, I think the Legato is better. I'll spend some more time with it to do more a/b testing. Promise.

 

Another side note about the lack of hi-res support in the Legato. Currently, I have ten files at 176400Hz, twenty-two at 96000Hz, twenty three at 88200Hz, thrity at 48000Hz. 85 high res files. By contrast, I have 3,251 at 44100Hz. I'm guessing that this ratio is similar to many of you. No doubt that hi-res is coming, but for now, my collection is 98% redbook, not including my vinyl collection (~100 albums), so not sure that I'll notice the lack -- for now, at least. While not ideal, all 85 hi-res files play back via the Legato (down sampled and dithered by iTunes, I'm sure), and all sound very good. And those redbook files? Yeah. Well, I think it's safe to say my system has never sounded this good. More listening is required, yes, and some a/b testing to boot. But in the mean time, I'm having a ball.

 

Great job, Pat!

 

[jnrich]

I did the same thing. I impulsively bought a Legato just before Christmas. I didn't really need it, but I wanted to try asynchronous code for a small financial outlay. I have been using it ever since, feeding bits into my Lavry DA10. I can't say I was shocked by its performance, but I sure am happy with it. I finally have a usb solution that removes the mist and grain that I heard from the previous solutions I have tried. I finally have a solution that sounds as good as a transport feeding my DAC. Red Book has never sounded so good for me.

 

Not exactly audiophile bling, but I can say that the Legato does what it is advertised to do. I was also leery of the long digital cable, but my ears don't seem to think that it's detrimental. And I agree that even hi-res (though downsampled) sounds mighty good. So maybe it's not my "final solution," but I'm sure pleased with what it has been doing in my system over the last month. For me, it's computer audio taken to the next level.

 

Pat's the man!

 

 

 

[kamil]

Scot, nice review! Have you tried the U-byte on the hiface? How much of the sound difference was from the Legato and how much from the cable? Thanks in advance!

 

[Part-Time Audiophile]

I haven't swapped cables, no. My HiFace is RCA, the Legato is BNC -- so the cables won't swap without an adapter on each end. Didn't seem fair! LOL.

 

FWIW, I was running the HiFace with a DH Labs digital cable ~6m long, so the lengths were approximately the same.

 

Sonically, I think the primary difference is due to the clocking. Just a guess, though. But note that the character of the Alpha never changed -- it just got better at what it already did well. Which makes me think that jitter is what is being addressed, and that the Legato handles it better than the HiFace.

 

As always, YMMV.

 

[kana813]

Scott-

 

I guess I was right.

 

As it breaks in, the sound will become more addictive.

 

I made the mistake of lending the prototype Legato Pat sent me to a friend. He refuses to give it back until Pat sends him a production unit.

 

Since 97% of your files are 44.1khz, I think you made a good call.

 

Dan

 

[Part-Time Audiophile]

Dan, yes, you called it. Still need to do some head-to-heads, but I'm running signal through it 24/7 in order to facilitate things. We'll see where things stand on Monday.

 

[Part-Time Audiophile]

Okay, pulled out the HagUSB and the HiFace again for a little back-to-back testing. I ran the HagUSB with a Belkin USB and a DH Labs RCA coax digital cable (6m) with an RCA to BNC adapter into the Alpha. The Legato was also run with a Belkin USB, but I used the ART U-Byte BNC coax (5m) cable with that.

 

First up, Hagerman.

 

For $129, this unit performs rather well. It's pretty much head-to-head with the Legato for features in that it supports only Redbook and DVD-A (44.1 and 48, respectively) over USB 1.1. It has a little light to let you know its on, which is nice -- and something that both the HiFace and the Legato do without. There is also a mini-jack for headphones (!) which I didn't bother using (another time). The box is plastic and is so light it feels like it's empty. Setup was simple, and was quickly recognized by my Mac. Sound was very nice -- the Alpha is a great DAC. Just a note: for those of you that didn't know, there is a new version that sports a USB to XLR interface for a whopping $149. I don't know of a cheaper USB-to-XLR interface out there, and if I didn't have other options, I'd strongly consider it.

 

When switching to the Legato, I noticed very little different, but both things that did leap out were bass-related. First, the bass with the Hag (and the HiFace) was deeper than the bass played back through the Legato. Second, the bass was muddier on the HagUSB, and my preliminary experience stands -- the Legato tightened this up. But as to the deepness ... I'm just guessing here, but I think the Legato+UByte combo isn't broken in while the Hag+DH Labs most definitely is. Why do I say this? Because when I switched from the HagUSB to the HiFace, the bass quality didn't change noticeably -- but the HiFace did play louder. As if it's enjoying a 3dB gain, or something. Dunno what that's about. Level setting pretty much eliminated the audible improvements that louder playback created (louder is better, BTW), but the overall performance wasn't noticeably different from that of the HagUSB. It might have been, but I certainly wouldn't swear to it.

 

So, there you have it. The Legato still isn't broken in after 72 hours. ;-)

 

[Laurenta]

There is something very weird in your description. Having a 'gain difference' (as much as an impression of 3dB) for three similar products simply doing in pure digital the USB to S/PDIF translation job is simply impossible...

 

I don't know what's wrong in the set-up. But level output (at least your auditive impression) should be very similar if not exactly the same. Otherwise this would mean the HiFace is playing with the bits somehow ;-)

 

[Part-Time Audiophile]

I can't explain it. But the HiFace plays louder. Damndest thing.

 

[Audio_ELF]

Is the increased volume measured or your perception?

 

[Part-Time Audiophile]

I didn't whip out the SPL meter to verify -- it was a subjective experience.

 

[Wavelength]

Gang,

 

Simple... The Legato registers as a 16 bit device. Any 24 bit device will be louder as the 16 bit data is shifted left 8 bits when playing 16 bit material through a 24 bit interface.

 

Thanks

Gordon

 

[ar-t]

My guess would be if anything needs a lot of "break-in" time, it would be the cable.

 

I know, you would need another BNC-RCA adapter to know for sure. Or one in the opposite direction.

 

[Marcin_gps]

Hi,

 

Does Legato support ASIO? If not by default, did anyone try ASIO4ALL?

 

[idiot_savant]

sample alignment doesn't work like this...

 

and if it did, it would be 48dB louder...

 

your friendly neighbourhood idiot

 

[ar-t]

A few folks have asked me about a competing unit, that claims it is probably the lowest jitter available. Part of that comes from the fact that we do not spec our jitter. But, using the same method they do (I believe), ours is about 1/5 the amount.

 

If one were to closely read our website, we do mention that we believe that we have the lowest jitter clock on the market. However, we do not make that an absolute claim, as it is possible someone will make a better one than we do. (I have no idea how someone can read it, and not come to that conclusion. Someone obviously has.)

 

The main reason we do not spec it is because it is just a number, which really does not mean anything by itself. The frequency content, and whether it is data-correlated, or Gaussian, are more important that some superfluous number. (Ours is totally Gaussian in nature.)

 

Hope this answers some of the issues that several of you have contacted us about.

 

At this point, we have no intention of starting a spec war, but rest assured that we will not allow our brand name to be sullied, by any of our competitors.

 

Units are slowly shipping. I am contacting everyone who is in the queue. Some of you may need to check your mail, or you will lose your place in the queue.

 

[Part-Time Audiophile]

I opened this thread a couple of months ago, and at that time, I was a happy if slightly disbelieving customer. Gordon Rankin had recommended ART, and for whatever reason, that was enough for me. Our own Clay recommended the uber-USB cable approach (ahem) and I chose the Synergistic Research Tesla Tricon from that bunch.

 

In the intervening time, I have written about what a great combo this is and how my computer audio listening experience has been transformed.

 

So, let me revisit: it's friggin' awesome, and if you haven't tried it, you're missing out. Truly.

 

I returned my dealer demo cable about a month ago and ordered my own 3m cable, despite Nick Bedworth's comments about the evident superiority of short USB cable lengths. Just got it today, and OMG, I am one happy camper -- again.

 

Thanks to Pat for such a great converter. Thanks to Ted for such a great cable (just wish it cost less than the converter [... sigh ...]).

 

I am LOVING this setup ...!

 

I feel like I just got sent to a place in the middle of nowhere with a big black horse and a cherry tree. Now I won't come back, cuz I'm oh so happy ....

 

[ar-t]

You paid more for a cable, than you did for our doo-hickey? The one that took Gordon countless hours to write the code, and a lot of hard work and effort on our part (not to mention expensive parts!) to create?

 

We gotta raise our prices.

 

No, then someone will complain that it doesn't look expensive enough. It doesn't. Wasn't supposed to. Just be affordable, and sound great. Glad you find agreement on that point. Still, thanks for your kind words.

 

[Part-Time Audiophile]

Yeah. I'm not very bright. I'll be the first to say it ....

 

But thanks again, Pat.

 

FWIW, the USB cable issue is real. While it may not translate to other cables, the differences between a USB cable with a separate power leg and one with one integral to the design are very definitely audible. Sadly, the cables that do this are crazy-expensive. All that quantum-this, dialectric-that, are crap IMO -- all you need to do is isolate that power leg and BAM you're good to go.

 

[ar-t]

That is rather interesting, as the power pin is not connected inside our unit.

 

In which case, maybe we should, in our non-existent free time, try to make a USB cable out of some 124 ohm twisted pair. Using the shield for the ground return.

 

The reason I am skeptical about USB cables sounding different, is because one of the data pins has a pull-up resistor, to tell the computer that something is connected. Does wonders for CMRR. Which is to say it is zilch.

 

[kana813]

Pat-

 

Yes, we need a The U-byte © USB cable that cost less than $200. and drives a stake

into the hearts of all these audiophile USB cable maker vampires.

 

Maybe you can get Father Guido Sarducci to bless them. ;-) Dan

 

 

 

[Part-Time Audiophile]

Pat -- I have no idea why these USB cables seem to work. My theory is that the separation of power and signal is a good thing. But WTF do I know? I do know this -- something is going on here and a better USB would be very welcome. Especially if it kills vampires, too.

 

[cfmsp]

 

I agree with you that the power leg should be relatively simple/cost effective to eliminate and should NOT require cryogenic treatments and other esoteric enhancements that the 'expensive' (digital) cable manufacturers utilize.

 

 

 

However, I find it quite 'interesting' that a difference can be detected if Pat is not connecting the power leg. :0 Although, if there remains a difference, something else is going on other than just the power leg elimination, no?

 

Did you test the Legato without the fancy power leg elimination cable?

 

 

As for me, I use Firewire adapters (to 4-pin) to eliminate the power leg on my $9 Gold-X Firewire cables. One can also pull the power pin (in Firewire) or put a tiny piece of duct tape over the power pin. note: these work only with Firewire 400 cables.

 

 

cheers,

clay

 

 

 

[Part-Time Audiophile]

I knew about Pat not connecting the wire -- the issue isn't at the chip (though that is an issue with USB products generally), its the wire itself. As the signal travels down the wire, perhaps having the power and signal wires not isolated from each other is enough to cause "issues", whatever those are.

 

And, yes, I tried a cheap-o nameless, a Belkin Pro, a Wireworld Starlight and the SRT. Each move bumped up "clarity", including tighter bass performance, less "grain", and better treble extension. Believe me, I did not want this to be the case. But my experience is that the SRT cable is the most relaxed USB-based sound I've managed to get out of the Legato.

 

[Part-Time Audiophile]

"That is rather interesting, as the power pin is not connected inside our unit. In which case, maybe we should, in our non-existent free time, try to make a USB cable out of some 124 ohm twisted pair. Using the shield for the ground return. The reason I am skeptical about USB cables sounding different, is because one of the data pins has a pull-up resistor, to tell the computer that something is connected. Does wonders for CMRR. Which is to say it is zilch."

 

If you're going to fiddle (and I can't tell you how awesome that would be), finding some way to completely isolate the power & ground legs from the two data legs would be really nifty. As in, physically separate them all as in multiple, fully shielded strands of cable. Simply eliminating the power leg (and/or ground?) might make the cable unusable with USB systems that require it, but given that its your cable and you'd be offering it (ha!) with and for your Legato, perhaps chopping it out (if possible) entirely might be a good thing and easier than trying to figure out how to implement a switch that would toggle it on for those situations that might need it.

 

Whaddya say, Pat???