Submitted by Bob Stern on Sun, 11/22/2009 - 03:22
Dan Rubin asked me to write this up and said he’d supplement it:
Dan recently had the opportunity to borrow a Sonic Studio Amarra 4 Firewire D/A & A/D converter for six days. I auditioned it only as a Firewire DAC (ignoring its A/D and DSP functionality) in my system for parts of two days. Dan also borrowed an Ayre QB-9 USB DAC ($2500) for 2 1/2 days during this period. Both of us own Bryston BDA-1 DAC’s ($2000), so Bryston is our frame of reference. We also had the opportunity to compare a Metric Halo ULN-2 ($1700).
The Sonic Studio Amarra 4 has just been reduced in price to $5200 according to this thread:
http://www.computeraudiophile.com/content/ULN-8-shootouts-Next#comment-3...
The Sonic Studio Amarra 4 is the same hardware as the Metric Halo ULN-8 praised in this forum by Barry Diamant, but with different software features. I would like to confine the present thread to a discussion of the hardware. I will start a separate thread on the differences between the software/firmware of the Sonic Studio and Metric Halo products.
Please do not confuse the Amarra 4 hardware described here with the Amarra playback software. I tried the latter only briefly, and that is the subject of another thread.
I listen only to classical music and opera. Dan listens to jazz, rock and classical.
Dan and I both concluded that the Amarra 4 is the best DAC we’ve heard in our respective home systems. (But neither of us has heard the Weiss and Berkeley DAC’s at home, not to mention the upcoming Antelope Zodiac ...) The outstanding characteristic of the Amarra 4 compared with the ULN-2 and Bryston was extreme purity and clarity. The latter two DAC’s were hashy/fuzzy and congested in comparison. The Ayre had much of the purity and clarity of the Amarra and is discussed below.
Early in my listening I decided that the ULN-2 had none of the special purity of the Amarra 4, and that it had no marked advantage over the Bryston, so I soon dropped the ULN-2 from the comparison and concentrated on the Amarra 4, Ayre and Bryston.
The purity, clarity and delicacy of the Amarra 4 made the timbres of different instruments more colorful, beautiful and engaging. Subtle details of phrasing and tone production were more apparent and captivating. Interestingly, this was true on both smooth, modern recordings (e.g., Mahler 10th by Litton on Delos) and bright 1980’s recordings (Mahler 6th by Bernstein on DG).
Piano was especially impressive on the Amarra 4: pure, solid, and subtle. Attacked bass notes are amazing, and you can hear subtle details like the resonance of the strings and the release of the felt. (Download from 2L a free high-res sample of the first movement of Beethoven’s last piano sonata -- an incredible recording and interesting interpretation.)
Initially I felt the Amarra 4 had a bit of mid-treble brightness that was annoying with massed violins and on dramatic peaks of male operatic voices. However, the Amarra 4 was so consistent in its purity and freedom from hash or fuzz that I suspect that the Amarra 4 probably was accurate, and that I was hearing the brightness of my Thiel 3.7 loudspeakers and room (which has a degree of slap echo). However, without auditioning the Amarra 4 in other systems and auditioning other equally expensive DAC’s in my system, I cannot dismiss the possibility that this brightness is a characteristic of the Amarra 4.
The Ayre was very impressive for less than half the price of the Amarra 4, but I only had a few hours with it. In the midrange it equalled the Amarra 4 in purity and subtlety, which is a huge accomplishment. However, the Ayre lacked the wonderful solid bass of the Amarra. (John Atkinson measured 60 & 120 Hz jitter components in the Ayre, so perhaps that’s a factor in its less solid bass.) I’m guessing that the purity of both DAC’s is a consequence of their async Firewire (Amarra 4) and async USB (Ayre) designs that allow them to use stable, fixed-frequency internal clocks that do not have to track the clock rate of the source.
The Ayre had some character that I had trouble pinning down, especially since I heard it only for a few hours. One was a relaxed, gentle quality that sometimes seemed unnatural, as if a bit of liveliness or dynamic contrast was sacrificed.
The other Ayre characteristic was a treble that peculiarly seemed to display different character at different times. On Nielsen sym #4 (Schonwandt, 3rd mvmt), high massed violins had realistic body with the Ayre, while the Amarra 4 was thin. Sometimes the Ayre treble seemed sweeter than the Amarra 4, and other times it had a bit of lower treble glare, even in different parts of the same recording. My preference between Measure and Listen depended on the recording, but this is a rear panel DIP switch that’s not suitable for regular use.
Please do not interpret my praise of the Ayre and Amarra 4 as a condemnation of Bryston. Dan and I both consider our Bryston’s much superior to the Lavry DA-10, Benchmark, and Bel Canto DAC-3 previously owned by one or the other of us. I had considered the Bryston well above those other DAC’s in purity and clarity until auditioning the Amarra 4 and Ayre QB-9.
High quality DAC’s that I have not heard in my home include the Weiss DAC2 and Berkeley Alpha, not to mention the unreleased Zodiac DAC advertised on the Antelope website.
Footnote re extracting digital audio from computer:
Before comparing the Amarra 4 with the Bryston, I wanted to give the Bryston the benefit of the best possible digital input. In my brief auditioning, the best was the Amarra 4 functioning as a Firewire to AES digital interface connected via a Mogami digital cable to the Bryston XLR digital input, so I used that in all the comparisons with the Amarra 4 and Ayre DAC’s. The other digital inputs that fell short were a Wireworld Toslink cable direct from my Mac, an Empirical Audio USB Off Ramp via Stereovox BNC cable to the Bryston, the ULN-2 functioning as a Firewire to AES digital interface via a Supra digital XLR cable, and a Levinson 37 CD transport via the Stereovox BNC cable.
My system: 2.5 GHz Macbook Pro (4 GB RAM, internal HD), Bryston DAC, Parasound JC-1 amps, Thiel 3.7 loudspeakers.
Normally I use a Placette passive attenuator, but in the reported comparisons I either replaced it with Dan’s Ayre K5xeMP preamp, or else connected the DAC under test directly to the JC-1 power amps using Nordost Heimdahl XLR cables. In the latter case, volume control was digital, via either iTunes or the digital faders in the Console software of the Amarra 4.
Mac Mini > Metric Halo LIO-8 > Parasound JC-1 > Thiel 3.7
I own an QB9 and have similar impressions. I love the way it handles classical music (body). But sometimes I miss bass as well. Interesting that you mentioned "different character at different times". I have this impression as well. Not only with treble. In the beginning I thougth it was the burning in process (the manual says: up to 500 hrs) but I still hear more or less subtle changes. I thought it might be a power issue at my home but then again it might not be.
Looking forward to your comments on Amarra Software.
Thanks
Claudius
Amarra 1.2, I-Tunes, Mac-Mini (4GB) SSD, Transparent Cables USB, Ayre QB-9, Octave HP-500SE, all PCs Lessloss Signature, HMS Sestetto, ADAM Pencil MK2 (active)
Lessloss blackbodies (3)
Very useful comparison! I too have the Bryston dac and was wondering how it compared to the Ayre. Other reviews seem to go one way or the other without too much detail.
to hear your impressions, based on what you've heard so far (albeit realizing that the DACs are no longer inhouse for you). At a little over $5k the Model 4 is quickly becoming a real serious choice (and that doesn't include the other functions)!! :)
Ted
"We're all bozos on this bus"....F.T.
Is it supplied with a wall wart? It appears that it would be easy enough to power it with an SLA battery power supply, such as the Kingrex SLAP, or the new one from Red Wine Audio.
I'd guess this might make a big difference.
db
I have a battery supply from Red Wine shortly for the Model 4 so I'll make sure to write up some feedback on the impact. can't say if relative impact will be the same with the ULN2 or not, but perhaps directionally.
Main: Mac Mini (SSD/Bolder PSU/Amarra) -> Amarra Model 4 DAC (w/ RWA Black Lightning battery power; Antelope Trinity+10M) -> Cary CAD-211 FE Monoblocks -> eFicion F300 Speakers (ASI Liveline ICs/speaker cables)
Headphones: MBP17 (SSD) -> PS Audio PWD -> Woo Audio WES (maxxed) -> Stax Omega II (mk1, mk2), RWA Isabella -> HD800 (ALO recabled) / Audeze LCD-2
Nice review. It actually gives a good idea how Metric Halo products perform compared to other highly reviewed DACs.
However, I don't quite understand why you counted out the ULN-2 so quickly. By writing " the ULN2. . . had no marked advantage over the Bryston" , you implied that it is as good. Thus wouldn't it be worth auditioning since it costs less than the Bryston?" I would think for computer use that it would outperform the Bryston since it uses an Asynchronous interface which the Bryston doesn't.
Also, seems like the Ayre comes close to the Amarra for less than half the money.
Regards,
Larry
Not sure if anyone caught this, but you stated the ULN2 doesn't have the same purity as the other DACs, but then you stated that the purity you heard with the async Ayre or the firewire A4 might be attributed to their interfaces. But the ULN 2 was via firewire too, right? I am a bit fixated on the Firewire interface and the ULN2 right now, which is why I mention this. Maybe I need to rethink the ULN2 as my DAC.
Can you share more about your room? Is it dedicated? What are the dimensions of the room and where are the speakers placed? And do you have room treatment?
Thanks for taking the time to share your experiences with these DACs! All of them are on my mind.
Bryan
Dedicated 240V balanced power, Torus RM20-BAL. Mac Mini/Ayre QB-9. LSA Group Signature integrated. Eminent Tech LFT8B speakers. Real Trap and GIK bass traps.
Bryan,
What was stated was:
"Early in my listening I decided that the ULN-2 had none of the special purity of the Amarra 4,"
And neither should it considering that the Amarra 4 costs 3-4 times as much money. :)
clay
I also noticed he stated that the ULN2 is basically equal to the Bryston and so it was dropped. This is a bit of a rush job with 4 DACs auditioned in a short time frame. I'm not sure what to think of the audition impressions.
I'd like to know more about the room and acoustics.
Bryan
Dedicated 240V balanced power, Torus RM20-BAL. Mac Mini/Ayre QB-9. LSA Group Signature integrated. Eminent Tech LFT8B speakers. Real Trap and GIK bass traps.
Great comparison. I look forward to your further posts.
If it sounds better to you then it is better...
As to whether the inferiority of the ULN-2 disproves the potential value of async Firewire or async USB, I think it merely demonstrates that the implementation details are at least as important as the top-level architecture. It makes sense that the 3X more expensive ULN-8 would employ a more stable clock and more extensive decoupling of noise in signal, power supply and ground lines than the ULN-2.
The reason I dropped the ULN-2 from the comparison is that I am not a professional audio reviewer trying to review every product under the sun. I was trying to determine whether to replace my Bryston with a better DAC. Since the ULN-2 did not seem like a step up from Bryston, and since I had the Amarra 4 and Ayre only a limited time, I wanted to focus on products I had a serious prospect of buying.
I did anticipate that the CA community would appreciate feedback on ULN-2 versus Bryston, but I did not have enough time. Dan still has the ULN-2, so it's possible he will be able to offer some comparisons, but both of us are short of time now because of work.
ULN-2 and Amarra 4 both came with the same model of external power supply, and I did not try any other: Fairway VE50-2400.
http://www.fairway.com.tw/English/En_ProductShow.asp?ArticleID=222
http://www.fairway.com.tw/English/En_CoProfile.asp?Action=Profile
Mac Mini > Metric Halo LIO-8 > Parasound JC-1 > Thiel 3.7
Thanks to both Bob and Dan for reporting on their efforts.
Bob says:
"I am not a professional audio reviewer trying to review every product under the sun."
It's a pretty thankless job, despite the thanks usually proferred, the 'thankless' part coming in the form of 'why not this?, what about that?' from audiophiles who would have others do their listening for them. No offense intended to any who do that, just please recognize that this is almost universally experienced when anyone posts a review. Just ask Chris. :)
It should perhaps be noted than even professional reviewers avoid the comparison game for the most part.
Re the dropping of the ULN-2, given that it was NOT a noticeable step up from the Bryston, it wouldn't have gotten more time were I in your shoes either. :)
It should be noted that the ULN-2 is NOT a stereo version of the ULN-8. I"m not surprised at extra purity over the 2. I would have been surprised by a change in character, however.
It bears stating again - as Async interfaces have gone from unheard-of to overly-hyped in less than one year - Async interfaces, whether Firewire or USB, are NOT a panacea. They can vary significantly due to implementation. At the cheaper levels (certainly less than $500 DACs), the interface is NOT the most significant consideration.
Let's not forget folks, there's an analog section in a DAC as well, and it shares the same PS as the Digital section. Lots can go wrong, which is why some DACs sound much better than others, no matter what Ashley James (and his engineer) says! ;)
Thanks again, Bob AND Dan.
Cheers,
Clay
Bob, I understand why you dropped the 2 from the audition. Thanks for clarifying that. At least you were able to add it in to the mix. If I was you I would have wanted to see how it compared to the Bryston and Ayre for sure, wondering if it bettered a couple of the well known $2K+ Dacs. You're a lucky guy to have been able to do the comparison.
So what are you going to do now that you've done the comparison?
And I'm curious why you don't reply to my query about your room.
Thanks,
Bryan
Dedicated 240V balanced power, Torus RM20-BAL. Mac Mini/Ayre QB-9. LSA Group Signature integrated. Eminent Tech LFT8B speakers. Real Trap and GIK bass traps.
(deleted)
Mac Mini > Metric Halo LIO-8 > Parasound JC-1 > Thiel 3.7
Bob, thanks for the reply to my question about the room. But the post was deleted.
The reason I ask about the room is sometimes people have really horrible setups where the speakers are located in very bad locations, and there is no room treatment. The first be the worst of the two. If I'm interested in someones post about products I'm interested in (3 of the DACs in your post, in this case) I'd like to know if they have a bad setup or not. Not that I"m going to make a buying decision based on a post, but I would make a trial/demo decision based on the weight of a series of posts.
It sounds like you are listening near field with the speakers on the long wall and first reflection points won't be at play, which is nice. I had a room that was like that and near field worked great. Nice clean sound.
If you are looking for high WAF reflection absorbers check out the GIK Art Panels. Any wife can't refuse them.
Bryan
Dedicated 240V balanced power, Torus RM20-BAL. Mac Mini/Ayre QB-9. LSA Group Signature integrated. Eminent Tech LFT8B speakers. Real Trap and GIK bass traps.
Hi silverlight,
It will be interesting to hear your take on the battery supply.
I just completed testing one (not the Red Wine) with my ULN-8 but will wait for your feedback before posting what I heard.
(I'm also have a linear supply on hand, that I'll be testing soon.)
Best regards,
Barry
www.soundkeeperrecordings.com
www.barrydiamentaudio.com
Hi Barry - good to hear from you. will probably give you a shout on this as I'm curious as to your experience. I'm guessing that the degree of improvement or change in sound by adding battery power will be so heavily dependent upon one's own power setup that this one will be a bit tough to generalize. While I have done a lot in terms of trying to stabilize and clean my power, I'm in a pre-war NYC apartment and very limited control over the power supply not to mention pollution levels prior to my fuse box (which b/c of my bldng's requirements I can't mess with; so limited myself to passive filtering on the mains in and have several large shakti stones up and down the fuse box and conditioning on the audio circuit, few other things here and there on the audio and non-audio circuits). So I wouldn't be surprised at all if the scope for improvement is much better in my type of situation than it is with someone in a more sparse suburban environment with dedicated high quality circuits / subpanel, etc. In any event, looking forward to hearing both before and after's (the Model 4 shows up hopefully this week).
EDIT: forgot to mention, I have only one potential caveat w battery power. some DAC's or electronics are said to perform their best after 24 hrs of being left on to reach circuit stability. so battery power that necessitates leaving equipment off could have an effect. This can be handled by switching over to AC power and back onto battery but just a thought.
Best,
Geoff
Main: Mac Mini (SSD/Bolder PSU/Amarra) -> Amarra Model 4 DAC (w/ RWA Black Lightning battery power; Antelope Trinity+10M) -> Cary CAD-211 FE Monoblocks -> eFicion F300 Speakers (ASI Liveline ICs/speaker cables)
Headphones: MBP17 (SSD) -> PS Audio PWD -> Woo Audio WES (maxxed) -> Stax Omega II (mk1, mk2), RWA Isabella -> HD800 (ALO recabled) / Audeze LCD-2
Hi Geoff,
You are correct of course regarding power being different in various locations.
I'm running dedicated lines and a Monster power conditioner.
I can say that some aspects of using the SLA supply were immediately obvious and others perhaps less so. Then again, I did have two immediate reactions.
I'll hold on specifics of the experience though, until you've had a chance to get a good listen. Which DAC(s) will you be trying it with?
Best regards,
Barry
www.soundkeeperrecordings.com
www.barrydiamentaudio.com
Barry says:
"I'm running dedicated lines..."
I (think I) learned something new today, from the unsolicited response to a similar comment posted on a forum dedicated to a supplier of ac line conditioning gear that I use.
"There is no such thing as a dedicated circuit in the sense of noise free isolation. All circuits are connected in parallel at the panel and protected with a circuit breaker. Noise travels up and down each circuit all day long by finding the path of least resistance. In a 120 volt installation the path of least resistance happens to be the neutral and earth conductors. Filtering the non-Audio/Video circuits using one of our active models lower noise in the exact location where it lives and breathes. Lowering noise on the neutral and earth can offer a dramatic improvement in system dynamics and soundstaging."
I've just begun using the equipment from this manufacturer, and I can report a huge improvement in sound quality. Like Geoff, I live in a NYC apartment (although NOT pre-war, in my case), and have no ability to implement things such as the so-called 'dedicated' circuit, in which no other devices are on the circuit other than the audio gear.
clay
"I can say that some aspects of using the SLA supply were immediately obvious"
Can I guess before Geoff responds? :)
Lower noise and lessened dynamics? Or is that too obvious?
No need to respond until Geoff does, but I knew he wouldn't be swayed by my guess.
Clay
Hi Clay,
Regarding dedicated lines, I'm not so sure about what that person said.
There was a larger difference made a larger difference with the same power conditioner I'm using now, before I had the dedicated lines. After installing the lines, the difference went from "night and day" to "I think it is better".
I'm sure much depends on how the lines are installed, etc.
Or maybe the electricity where I am just coincidentally got appreciably better right at the same time I had the dedicated lines installed. ?
I guess I can test that by switching to one of the non-dedicated outlets.
So much to test (this, hardware, software, etc., etc.), so little time. ;-}
Best regards,
Barry
www.soundkeeperrecordings.com
www.barrydiamentaudio.com
Hi Clay,
There was something else, particularly noticeable I believe, with the ULN-8 (and I would imagine, with the Sonic boxes that share the same design).
Best regards,
Barry
www.soundkeeperrecordings.com
www.barrydiamentaudio.com
Well, I'll keep this guess to myself, so as not to spoil the punch line, but if I'm correct, it's something you've mentioned before about the ULN-8. :)
I have an ULN-2, with a linear power supply, BUT, what I'm thinking of was also noticeable in my setup, when I installed the custom PS.
clay
PS, I probably shouldn't have even posted my earlier comment about the dedicated circuit, since I have almost no knowledge of AC-related stuff. I only know that I took a chance on something Geoff recommended (due to similar environments) and it has paid off in spades. I'm fairly confident that if I had researched the products like I normally do, I would not have gone down this path - as it seems a little bit 'voodoo' to me - but I am (not literally) shocked by the magnitude of the improvement.
How are the dynamics on the ULN-8? Having not heard the DAC but having read everything I can abt this unit on this board as well as Gearslutz I'm conjuring up a sonic image of a NOS-like DAC with extreme clarity but perhaps lacking some dynamics? Sorta like the difference between going from a passive, auto-former volume control to a top-notch active preamp. Initially, the passive sounds great but then you realize the music is a bit flat and you're willing to give up that nth degree of clarity for a bit of liveliness offered by the active preamp.
______________________
iTunes (iMac) -->Weiss Minerva-->Cary 306 Professional-->Eddie Current Balancing Act -->Balanced SAA Voice HD800s
Hi jky999,
"How are the dynamics on the ULN-8?"
For my ears, dynamics on the ULN-8 are like everything else on the ULN-8:
in comparison to anything I've tried, I have yet to hear a challenger.
Bring on a "contender" known for its outstanding soundstaging and the '8 makes it seem like that contender's soundstage is a bit smaller than the original space (using minimally mic'd recordings where one has stood at the mic position is great for this).
Similarly, the only time I've heard any dynamic limitations is when I've compared the '8 to other DACs. Switching away from the ULN-8 reveals a bit of dynamic compression in the other.
Just my perspective.
Best regards,
Barry
www.soundkeeperrecordings.com
www.barrydiamentaudio.com
I also evaluated these products in my system as well as in Bob’s. My system is Dynaudio Confidence C1 (and briefly the JM Reynaud Duet) driven by an Ayre AX-7e integrated. Front end was Mac Mini/iTunes/Amarra Mini.
I am pretty much in complete agreement with Bob’s review, and I want to thank him for his efforts not only in writing this up but in bringing precision and discipline to our evaluation process.
I was startled by the Model 4 and, hoping to avoid the audiophile clichés, I will just say that Barry Diament’s praise for it seems entirely justified. It has a purity and directness and “rightness” that had me thinking “live microphone feed.” Not everyone’s cup of tea, I’m sure, so be sure to audition for yourself. You might prefer the Ayre…
I found the Ayre both compelling and curious, and I’m hoping that one of you might have some insight about what I heard as “its sound.” My history with Ayre: I own an AX-7e integrated and a K5xeMP preamp. I briefly owned the original CX-7 CDP and did not care for it. I did a home audition of the first Ayre power amp many years ago and, later, the original Ayre preamp. I’m a big fan of Charles Hansen (his Avalon Eclipse speakers are what I replaced my original Qual ESLs with) and the company, but I have mixed feelings about the sound of some of the Ayre gear.
The QB-9 has a very full-throated sound, seemingly a little thicker than it should be in the lower midrange and bass, very smooth, elegant, sophisticated, warm. I thought its bass left a lot to be desired in the areas of clarity and definition, but the Ayre exceeded my expectations in many respects and it’s a joy to use. It has a truly elegant sound. What really puzzled me, though, was the presence of an overall character that I find hard to describe. I might call it a higher noise floor than I am used to, or a background “matter” that contrasted with the complete absence of anything in the background from the Amarra 4. I do not hear this quality from the Ayre integrated but I did hear it in the original Ayre power amp. I have also heard it from Plinius integrateds. I will add that it was not audible with the Reynaud speakers, only with the Thiels and the Dynaudios, and it was not audible, or at least not objectionable, with all program material. It’s a mystery.
Dan