Burl B2 Bomber DAC Review: Holy Sh**, This is Good

Discussion in 'Digital: DACs, USB converters, decrapifiers' started by purr1n, Jul 14, 2020.

  1. Gungrave

    Gungrave Rando

    Joined:
    Jul 12, 2020
    Likes Received:
    19
    Trophy Points:
    3
    Location:
    miami
    interesting, I find the lack of USB troublesome if I were to go this route, oh wait there is coax :O Eitr saves the day ?!?! future upgrade to dante perhaps ! will there be any impressions using the Dante card ?
     
  2. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Friend BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    82,575
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    SUBJECTIVE IMPRESSIONS - FINALLY

    I've had a hard time really thinking about how to describe the sound of this DAC hence why I've taken to so long. If I had to summarize things in a few sentences, it would be this:

    The Burl B2 Bomber DAC is something that I simply listen to. I don't get distracted by wow this or wow that. It's not that it doesn't do certain things well as all good DACs do or have its own flavor. It's just that goes about its business in a nonchalant way. This isn't Zhang Zimou were the audience thinks "wow, artsy fartsy, look at all those colors and synchronized movements - he's so amazing", this is Steven Spielberg where we are engrossed in the story.
    That being said, it think it's important for me to describe its sound characteristics, lest you do decide to buy this DAC and end up disappointed because it doesn't meet your personal preferences. Most of my listening so far was been with EC 45 Studio (with Monolith interstages and the custom OPTs used in last batch of the Studio Jr.) with a variety of headphones: HD650, Grado HF 3 (TTVJ Deluxe pads), Elex, Verum One, etc. The Burl B2 Bomber was fed AES from the Pi2AES.

    The B2 Bomber uses the old AKM4399 chip. The chip is from the prior generation of AKM before their Velvet Sound stuff. Some of you may be familiar with the OG Gungnir DS or even the Modi 2 which uses a lower line chip from AKM. In general, I feel this generation of AKM chips sounded more analog or R2R of whatever among most of its delta-sigma peers (except the AD1855) and the latest Velvet Sound stuff. The OG Gungnir DS was competent not the best. Long ago before Schiit came out with the multi-bit stuff, I asked Jason at Schiit if he could make the DS better. He simply replied that he couldn't - given the budget constraints and that most of their eggs were going into the multi-bit DAC basket. The Gungnir DS already had slam in spades. However it was inelegant and couldn't do the nuanced thing. Also the micro-stuff wasn't quite up to par and the soundstage was small. The highs were easy on the ears though in terms of digititus, despite having good attack in the upper registers were some DACs seem to intentionally hold back just to be a bit more forgiving. Yup, the Burl will slice and dice with its guitars, trumpet blares, and snare shots. If you are the type of person who cringes too easily at these things, look elsewhere.

    The overall tonality of the B2 Bomber is slightly warm. By slightly, I mean quite a bit less warm than something like the Bifrost 2. There is a smidgen of outward warmth (give the DAC's output stage a few minutes to warm up) in terms of upper bass to lower mids. And then there is quite a bit of Moto Ultralite internal warmth that runs consistently throughout the audio band. Call it line transformer warmth, or old school console warmth if you will. The warmth doesn't impact it's clarity or blackground. We are not talking about warmpoo murky or organic. We are talking about muscular warmth. Blackground is better than Gungnir Multibit A2 (which in turn is blacker than Yggdrasil A2). But still not as quite black as the Bifrost A2 or Modius. It's closer to the latter though. Bass texture is available in spades, equaling the best R2R DACs, surpassing the Convert-2. Midrange textures are just like how we remember it from the prior generation AKM DACs, not blended into baby food like with most AKM Velvet Sound DAC implementations of today (I'm sure the audio designers will figure out the Velvet DACs, eventually). The tone density is solid. Not overly thick and dense like with the typical NOS R2R designs, but solid yet textured.

    The sub and low bass slam isn't as exaggerated like on the Convert-2 and Schiit Multibit DACs; however, it's hardly weak - let's say it seems proper. If there is any emphasis, maybe some midbass punch, but even then it isn't pervasive at all volume levels. It seems that the louder the music, the more it has punch. Which then goes to what @Psalmanazar mentioned a few posts back: the sound varies on how hard that output stage is it. More precisely, there seems to be a built-in approach to saturation effect (so far I'm seeing measurements that may indicate this) akin to the old-tape decks and transformer. This is probably why one of the first observations I sent to @Clemmaster (this is his DAC, so kudos to him for supporting SBAF) was this as a vintage sort of sound from the days were I would monitor the tape deck levels while making mix tapes from record players.

    The one thing that it doesn't quite do at the highest level is resolving the lowest level information. The Burl guys did an amazing job extracting what they could with a single AKM4399, but that chip is showing its age in this respect. In reality it's surprisingly not that far behind and way more resolving of plankton and microdynamics than any other AKM4399 DAC I've heard (I have not heard some of those crazy Japanese specials that stacked 22 of these chips). Also, you love the immense air and top octave of the new Velvet Sound DACs, the B2 Bomber may not do it for you. Personally, these things are not a deal breaker since the most of the music that I listen to isn't recorded and produced well enough. Besides, the point of this DAC seems to be just STFU and enjoy.
     
    Last edited: Jul 16, 2020
  3. Gungrave

    Gungrave Rando

    Joined:
    Jul 12, 2020
    Likes Received:
    19
    Trophy Points:
    3
    Location:
    miami
    How close is it to balanced modius in resolution ? Did you run it through an eitr mostly ? Will you be trying out the dante path? Awesome impressions btw thank you for your service !!
     
  4. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Friend BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    82,575
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Forgot to mention and just updated the above post. I used pi2AES as the source using AES. I will let @Clemmaster comment on the Dante as I have no desire to use it or make my life more complex. @Clemmaster already wants it back ASAP, so I want to spend the rest of my limited time on the technical measurements.
     
  5. Psalmanazar

    Psalmanazar Most improved member; A+

    Friend Slaytanic Cliff Clavin
    Joined:
    Sep 27, 2015
    Likes Received:
    5,066
    Trophy Points:
    113
    This has the good pro gear analog warmth but also the class a warmth. The discrete opamps are pretty much just distorting in a cool way though so it’s not going to be perfectly accurate to what’s being fed. It doesn’t look like they’re socketed like API style class ab opamps, where you can just shove in a 990 if you want to perceive a more linear response. You might be able to get it hot rodded with the new ones by Burl themselves but it will sound completely different. They certainly could have switchable ones in the rack mount for more options given that they offer so many options for what op amps and transformers (or none) you want in the modules.

    Burl made newer opamps for their new DAs meant for mixing and mastering. Who knows if their colored or not. They’re only in the Mothership. The bomber uses the original old op amps, the BOPA1. Here is the newer module https://burlaudio.com/bda4m
     
    Last edited: Jul 15, 2020
  6. Vtory

    Vtory Audiophile™

    Friend MZR
    Joined:
    Sep 12, 2016
    Likes Received:
    9,970
    Trophy Points:
    113
    Location:
    East Coast
    Wondering -- if k4 hears this dac, will he drastically update his inconvenient truth thread the other way?
     
  7. Vtory

    Vtory Audiophile™

    Friend MZR
    Joined:
    Sep 12, 2016
    Likes Received:
    9,970
    Trophy Points:
    113
    Location:
    East Coast
  8. k4rstar

    k4rstar Britney fan club president

    Friend BWC
    Joined:
    Jun 11, 2016
    Likes Received:
    6,191
    Trophy Points:
    113
    Location:
    Toronto, Canada
    no

    it's not for me, but I have no doubt based on what Marv has said so far that I would prefer it greatly to the Solaris. I don't need a rack mounted space heater with XLR outputs that require a line transformer to use with any of my gear though.
     
  9. Psalmanazar

    Psalmanazar Most improved member; A+

    Friend Slaytanic Cliff Clavin
    Joined:
    Sep 27, 2015
    Likes Received:
    5,066
    Trophy Points:
    113
    Agreed. Vivid but a bit washed out hiding detail. Convert 2 would be color processed and more vivid in an artificial way like a modern Hollywood Blu-ray that’s sharp but with way too much denoising. Lavry would be Excalibur with all the soft hazy lens filming blood spurts on grainy film stock. Prism FireWire/USB DACs are Gladiator.

    Tape? Good tape is Conan the Barbarian. The gold standard.

    The best modern workhorse pro dacs (Lynx, MOTU, DAD) are the brutal shower knife fight from Eastern Promises. Think of how much Hollywood would’ve fucked that up. That’s the music industry. Good thing it was done by Cronenberg and Suschitzky.
     
    Last edited: Jul 15, 2020
  10. Gungrave

    Gungrave Rando

    Joined:
    Jul 12, 2020
    Likes Received:
    19
    Trophy Points:
    3
    Location:
    miami
    please elaborate ?
     
  11. Gungrave

    Gungrave Rando

    Joined:
    Jul 12, 2020
    Likes Received:
    19
    Trophy Points:
    3
    Location:
    miami
    @Clemmaster which input would you suggest for the burl b2 bomber, if ethernet is the key, what would be the best source ? Ex: rednet pcie card ?
     
  12. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Friend BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    82,575
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    The Burl, being a piece of pro gear, only has balanced outputs. You are sort of SOL if you have only SE inputs on your amp or preamp. The best way to convert from balanced to singled ended this: https://www.superbestaudiofriends.o...le-ended-converter-xlr-to-rca-bal-to-se.9506/ Cheater XLR to SE cables that tie - to ground can work, but can result in higher distortion at best or magic smoke at worst.

    As an aside, the Burl's outputs don't have an issue driving the 600-ohm studio grade transformers such as the CMLI-600/600C. These generally sound better than the CMLI-15/15B2, but really hefty output stages that can drive 600-ohms are needed to get the best from them.
     
  13. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Friend BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    82,575
    Trophy Points:
    113
    Location:
    Padre Island CC TX

    Used Mehta / LA Phil for Copeland this time around (the performance is better than the Telarc / Kunzel, well, because it's Telarc, but otherwise some similarities in style which aren't totally far off yonder like Bernstein's):
    • Tympani hit harder on the Burl, which is expected since Gungnir A2 (GA2) is cooler and only hits hard in low/sub bass. Overall more body on horns on Burl than GA2. GA2 picks up ambient cues, noise, and that last bit of microdynamic shading. GA2 has better localization within the orchestra arrangement whereas Burl its doesn't quite pinpoint the location and separate things out as well.
    • The above are the biggest differences. There are also differences with the following, but they are very very small: highs on GA2 are more focused but also more "spitty". Burl is actually smoother. Lows on GA2 are just a tiny bit murkier. Tone density seems to be on par. GA2 wins on texture.
    • In the end, however much I wanted to like prefer Burl, the Schiit Gungnir A2 wins.
    Rickie Lee Jones: Now the Burl just works beautifully with these kinds of studio recordings. The Burl just nails it.
    • GA2's spittier highs mentioned above become more of a liability. Snares, high-hats, and even finger snaps just sound more right on the Burl, as the GA2 seems to emphasize their harmonics rather the the fundamental or body of the sound. The Burl actually has a stronger attack too. So stuff like this and sibilants and still going to hurt (which is how I think it should be).
    • Similar to above, after getting used to the Burl, after the intro where the bass and keyboards come in around 00:18, the less cohesive nature of the GA2 (between focused highs and murkier lows) becomes very evident.
    • That internal warmth (and sense of body) just seems to be totally in place with studio recordings like this.
    • Burl B2 Bomber wins.
    This usually isn't my style, so I hope this helps.
     
    Last edited: Jul 15, 2020
  14. Vtory

    Vtory Audiophile™

    Friend MZR
    Joined:
    Sep 12, 2016
    Likes Received:
    9,970
    Trophy Points:
    113
    Location:
    East Coast
    Thanks. You wonderfully nailed it. (I kinda like this way because it's easier to imagine what were actually brought to the table.)
     
  15. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Friend BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    82,575
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Measurements now. Output level selector set to -18db (which translates to +22dbU at 0dbFS). 8192 FFT Window size. Let's progress from low level signal to a high level signal - see what's up with the intent of the cook - as opposed to taking measurements for the sake of measurements and the one number that rules them all (there is no such thing).

    -90dbFS 1kHz
    upload_2020-7-15_12-37-53.png

    -60dbFS 1kHz
    upload_2020-7-15_12-40-15.png

    -30dbFS 1kHz
    upload_2020-7-15_12-41-20.png
     
  16. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Friend BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    82,575
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    -20dbFS is where we start to see the intentional tape saturation effect. At its highest, the distortion is -75db below the fundamental, still a low level, on par with may high powered class AB power amps. The distortion starts with the second harmonic higher than the third (a good pattern) and ends up with 2nd and 3rd order being close to even (3rd order just a bit higher).

    -20dbFS 1kHz
    upload_2020-7-15_12-43-54.png

    -15dbFS 1kHz
    upload_2020-7-15_12-52-8.png

    -10dbFS 1kHz
    upload_2020-7-15_12-45-40.png

    -5dbFS
    upload_2020-7-15_12-48-19.png

    0dbFS
    upload_2020-7-15_12-49-11.png
     
  17. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Friend BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    82,575
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    Here's another way to look at it.

    Burl B2 Bomber DAC
    THD+N vs Input (dbFS)
    upload_2020-7-15_12-57-11.png

    For comparison this is how a "normal" DAC with excellent measurements would typically look like (Modius).
    upload_2020-7-15_13-2-7.png
     
  18. gridmaster

    gridmaster Facebook Friend

    Joined:
    Oct 1, 2018
    Likes Received:
    121
    Trophy Points:
    33
    Location:
    United States
    I recently had a discussion about this with some people based on your Bal -> SE thread with the cinemags.
    I was told the best way to do it is actually float the third pin, and don't tie it to ground.
    Which at the end of the day doesn't make sense to me, if it was really that simple, why all the Bal -> SE transformer stuff?

    I feel like I'm missing something important. (sorry for the aside not related to thread topic)
     
  19. purr1n

    purr1n Desire for betterer is endless.

    Staff Member Friend BWC
    Joined:
    Sep 24, 2015
    Likes Received:
    82,575
    Trophy Points:
    113
    Location:
    Padre Island CC TX
    If you float the third pin, half the DAC, one half of the differential signal isn't being used. Will also lose CMRR, the benefits of lower noise and some cancelled noise (AC mains) of balanced. Depending upon the implementation, this could sound better. But it's still using half the signal. In other cases it could sound worse, for example if a DAC design relied on the differential signal to cancel out DAC ladder glitch.
     
  20. Gungrave

    Gungrave Rando

    Joined:
    Jul 12, 2020
    Likes Received:
    19
    Trophy Points:
    3
    Location:
    miami
    .
     
    Last edited: Jul 16, 2020

Share This Page