Rockna Wavedream DAC: Awesomesauce

The WD looks rather plain to me so I’d call it “audio ecstasy” instead.

 

You have reached a stupid high level, friend. This is an hidden gem that sounds meh on even pretty good systems. One of the tracks I tune speakers with.

 

I would disagree that it sounds meh on other systems, but I love that entire album, its a progressive metal masterpiece. but I would say it sounds f'ing spectacular with the Wavedream Sig, in a way I haven't heard before.

The songs I spoke about in the impressions

 

I would disagree with your disagreement. That track was one of definitive wow moments in my experience. So powerful that I'm still without an amp and taking a forced break from audio yet I don't feel any rush to resume, I know what's waiting, the holy grail or precision meets musicality.

The whole Tool album has an exceptional synergy with the WD sig, not so much on other DACs. I would expect other albums would sound better with other TOTL dacs.

I can't figure out why the NET has the impact it has, if you think you're impressed by the WD, get your hands on the NET for a demo, the spacial precision is increased by a notch or two.

 

I don't have a Rockna. But I do have a May KTE, a very resolving amplifier, and very transparent speakers (Dyn Confidence C2 Platinums). So I've kinda been picking up on what others are hearing on these higher end DACs and trying them in my space.

I've long used CCT as a reference track. I can definitely see the other fella's point about it being "meh" on even pretty good systems. That is NOT a knock on the track. It just means that the awesomesauce it can deliver is only truly realized (IMO of course) on very resolving setups which are properly set in the room. I feel like my rig is that way, and I hadn't yet listened to it on the May KTE. It did not disappoint. What a trip, pun intended.

So now I'm listening to Elegy. And I am a huge fan of metal of all stripes. But this is GD nuts. Holy chaotic harmony, Batman. This is awesome! Great track for "so this one track I use - there's kind of a lot going on!"

Great suggestions!

 

I've had the Wavedream for about four weeks now and so wanted to share some impressions and thoughts. First a BIG THANK YOU for the impressions on this thread from @purr1n, @Xecuter, @famish99, @animus, @SoupRKnowva which were super useful in getting me here. Useful enough, in fact, that I want to argue against the saw that when audiophiles reach level 14 and gain double wielding they should graduate from SBAF and embark on their own heady voyages. Basically SBAF is unique in having members who are not deaf, are able to describe critically what they are hearing, and can compare against known referents in an accurate manner, and this is an even more important service for the top-shelf gear that you can't just churn through. So keep those high-end impressions coming!

Part 0 - Torq-esque blueballs drivel (skip to the end)

My preferred sound is quite dry, neutral bordering on thin/bright (slope up preferred to slope down), super fast, delicate, but with plenty of low-level detail, immediacy and microdynamic variation. Think Donald Fagen's The Nightfly if it was ADD rather than DDD.

For amp and transducer I have Aficionado (reputedly the driest of the recent EC amps and with plenty of mid-high energy) and the Omega 3i's (which are ultra-fast and forward and live in the midrange). My DAC for the last 5 years has been Yggdrasil, first A1 then A2, which after endless tweaking I have got to perform about as well as I think is possible. Nonetheless I still had a few issues with it:

• Despite heroic efforts to mitigate, it was always a bit sibilant
  
• Mid tonality was not completely convincing. Listening to Wire Pink Flag the guitar tones sound quite off, with a slight bubble-gum edge, almost like the bad old ESS Sabre though nowhere near as egregious.
  
• I liked the forwardness but sometimes it was a bit much in my chain. Something like Weyes Blood Titanic Rising is grand and nicely produced, but the mastering is a bit marginal and it got pretty relentless on the Yggdrasil... sometimes I wanted it to calm the F down a bit.

So, I was looking for something that would fix these points while also delivering moar plankton and microdynamics. The Wavedream sounded like the ticket.

Reading @animus impressions, given I wanted something a little calmer, and prefer lean and dry to warm and sweet, I figured the Edition fitted the bill better than the Signature. And as I am SET 4 Lyfe I have no need for balanced. This landed me on the Edition SE. Conveniently, this is also the cheapest Wavedream which, due to poor life choices (job that doesn't involve putting things in people's mouths) is the only one where I can actually justify ("justify") the outlay.

So... what do I think?

 

Part 1 - Listening impressions

Chain for these impressions was

PI2AES -AES-> Wavedream Edition SE -> Aficionado -> Omega 3i + Rythmik FM8

On the WD: linear filter, local clock, dither off.

Tonality. Coming from the Yggdrasil it's not dramatically different. The overall tonal balance is very similar. The first thing that stands out is how much more articulate and well-defined the Wavedream is in the upper bass -- the Yggdrasil is very squidgy by comparison.

The second thing is how much smoother and more refined it is in the treble. I don't actually find it the quantity of treble to be much different from Yggdrasil, but the way it presents it is much less challenging: there is no grit, no digitus, and if you hear sibilance, it's because the producer skipped the de-esser -- and this is speaking as someone who is ultra sensitive to treble nasties. As for the mids, they are pretty similar to Yggdrasil but without that slight artificial tinge. Those hard to reproduce distorted guitar tones are on point.

This very even tonality gives excellent timbral reproduction. Where it really comes out, I think, is in hearing all the different recording techniques. Is this 60s tape slammed to f**k, refined 70s tape through a Neve, or big-desk 90s SSL? And not just as an effect, hey I can hear all the recording techniques, but something that really imbues the sound from the word go, you can instantly hear it and be transported into that era. With Yggdrasil this is there, but you have to listen for it and know to listen for it.

Staging. This was a big surprise for me. Not that I wasn't expecting it to be good, more that I wasn't expecting to care. I generally don't give a shit about staging, I listen near-field in a triangle of around 1.5m and rarely bother to sit even close to the sweet spot. The Wavedream is arresting enough that I actually want to do so.

Yggdrasil presented a stage sitting on a fairly flat plane that stopped precisely at the two speakers. Instruments had a small amount of front-back depth but the stage itself was not very deep. The pWavedream stage extends some distance outside the speakers left-right and is significantly deeper, maybe about half as deep as it is wide, and in a ovoid shape. Within the stage, instruments have significant depth of their own, but relative to Yggdrasil what is most noticeable is how stable and well-delineated they are. With closed eyes the projected space is really convincing.

Dynamics. I'm not going to say anything about macrodynamics because I don't really care that much about them or pay much attention beyond the fact that I hate gear that is overly dynamic or slamming (eg MJ2). Suffice to say that the WD Edition is not that.

The best thing about the microdynamics of this DAC is something that's been mentioned before, its ability to distinguish fine gradations of volume near full-scale. This is really nice to hear. The loudest guitar strums on Joni Mitchell Amelia normally blow out to white, but here retain all their inner texture. In other regards, the microdynamics are a step up from the Yggdrasil, but not in a whole different realm.

There is a point here which is that Yggdrasil is fussy as all f**k and really forces you to get your source side in order to stop it sounding average. With PI2AES+linear power, plus an uber power conditioner, plus careful thought about grounding, Yggdrasil reveals itself as a much better performer in many aspects than it appears at first blush, and this pulls it much closer to the Wavedream (which seems to be much less fussy about source). However Wavedream pulls away again with further optimisation and I'll touch on this below.

Technicalities. For blackground, clarity, transient response there is again, to my ears, not much to choose between Yggdrasil with really clean power and the Wavedream. @kukur9 touched on this in the LIM thread, how good power cleans up the background on the Yggdrasil and I totally agree. However, separation is much better on the WD just because of the great staging.

There is one time-domain feature which is noticeably different, which is to do with sustain. With the WD there is a sort of confidence and correctness to how sounds evolve and decay past the initial transient which is missing with the Yggdrasil. In part this is to do with bass handling, but not entirely. I have tried the Yggdrasil through Gigawatt power conditioners with big capacitor banks and while they totally screwed the sound in plenty of other ways they also imparted some of this same feeling of correctness. So I think it could be power-supply related.

Overall presentation. The one thing living with the WD has revealed is how on edge I was the whole time listening to Yggdrasil. With Yggdrasil there is a collection of very good technicalities overshadowed by the constant fear that it's going to crap the bed. This could be because it suddenly hits sibilance or digitus (fret noise/tambourines) or just that things get complex and start to fall apart. Often this was worse with sausaged mastering, which tallies with the lesser ability to distinguish near-full scale dynamics.

The WD is frankly imperturbable. You can throw anything at it and it doesn't get upset. As a torture test I made a "mix" which was just lots of different kinds of music playing simultaneously on top of each other, and they were all just perfectly rendered each in its own space, Meshuggah and Jeff Mills and Vaughan Williams, which is kind of amazing.

There is also the mellowness that people have mentioned. There are I think a few aspects to this. One is the fine gradation of near full-scale dynamics, which turns blunt full-scale impacts into something more nuanced (less ouch). Another is the super smooth top end. Another is the imperturbability, it's sort of like the mindfulness DAC, it induces the sounds to arise in your speakers but can do so without needing to get involved. In the end it might come down to a world's tiniest violin kind of problem, which is that it makes things sound so good that it seems slightly unbelievable that it's a realistic depiction.

The best I could come up with to describe the mellowness is this: imagine you went to an all night party, got home, stumbled around, dozed fitfully for an hour or two, woke up in streaming sunlight, and put some tunes on while dust motes danced in the air before your eyes. Everything sounds a bit special, you have some new insights into sound, but there is a pleasant mellowness to everything. Well, it's kind of like that.

 

Part 2 - Tweaking

After getting my head around the WD sound, and running an apples-to-apples chain against the Yggdrasil, I spent the last couple of weeks dinking with the various options to see if the sound can be stepped up even further. Answer: yes!

DITHER
This does nothing really on 16 bit sources. On 24 bit source, it adds a little extra depth and roundness to the staging, and does nothing else I could hear. Sometimes I wondered if it made the top end slightly harsh, but it's edge of perceptibility. Not worth thinking too hard about this

INPUTS
I only tried AES and I2S from the PI2AES -- though S/PDIF would probably also be worth a shot. With AES, I felt the WD is on a par with Yggdrasil for transient response and speed of stops and starts. With I2S it puts clear ground between the two, with the WD much more precise. This is particularly noticeable in bass -- AES sounds a lot looser. No changes in other technicalities though -- stage, plankton, detail and microdynamics seem pretty similar.

Initially I thought I2S sounded a bit etched by comparison, but then I realised that this is because I had both AES and I2S connections from PI2AES to WD plugged in at the same time. I think this does something ground loop-y, which always seems to introduce sharpness and hash in the treble. Unplugging the other input solved the problem.

Cables seem to matter here. Cheapo Belkin was bested by a Neotech HDMI from Audio Sensibility (still reasonably cheapo at $90CAD). I find these comparos hard but it seemed like there was more tonal colour and less sense of strain. Not a great difference but a good last 3% thing.

I2S just works from the PI2AES without installing any jumpers. Plug and play. Rockna advised to use the second HDMI connector, though not really sure why. Hit the RPT button on the remote to change the I2S +/- convention (swap channels and inverts polarity) until it sounds correct.

CLOCK
We have "stream" and "local", which are described as follows: "stream mode will track the input frequency constantly while local mode will adjust the lock only from time to time, using a few stages of memory buffers." From the PI2AES they sound quite different. Local has a slight gloss to it and is a bit rounder and bloomier. Stream is more immediate, up-front, transients are better resolved, you get some extra plankton, and a welcome tiny bit of edge and texture to the sound (local is so smooth, I feel it glosses over some detail). "Stream" is an easy call.

FILTER
Didn't do more than 5 seconds on anything but linear, because why would you? The upsampling filter on the WD is actually pretty good. I don't really like the megaburrito filter on the Yggdrasil, it screws with the treble transients too much. On the WD the filter keeps them pretty intact.

That being said, you can do better than the internal filter. Upsampling 4x (most I can do from the PI2AES) with a very naive 64M tap rectangular windowed sinc removes a layer of greyness and plasticity, fixes a slight digital tinge to treble transients, and steps up transient response, speed and articulation (much better bass articulation). I prefer to leave the filter set to linear when upsampling, to avoid collapsing the soundstage.

Conclusions
Taken together these last three steps really elevate the WD from good to great in technicalities, and actually go a long way towards counteracting the mellowness: rather than mellow I would now describe it as delicate, still lacking that last bit of energy if you are so inclined but pretty much ideal for my sensibilities.

 

superb assessment.

 

@Garns Great review! It took me a while to find a good upgrade from my Yggdrasil A2 and I think the WD was a solid step up from it. Everything the WD did was just better than the Yggdrasil. I always felt there was something better but knew it came at a hefty price upgrade. While I don't have the WD anymore, I do think it's one of the better DACs out there and an easy endgame for someone who wants a no fuss DAC where you don't feel like you're losing out on anything, whether it be details, emotion, timbre, etc.

 

Hello, why you are stating that the linear filter is not messing up with the upsampling? To my understanding what you are describing should be achieved (in theory) with the ''Nos'' filter, as it should not upsample at all.

With the filters on (minimal, hybrid or linear) you are upsampling twice, on player and then last mile on WD.

 

That's correct, and this I think sounds better than just upsampling 4x in software. If I could give it a 16x fs input I agree there would be no point in having the filters on. But just 4x fs isn't enough to retain good soundstage and transients. (There are also practical considerations, even if 16x fs input were possible. The necessary files would chew up a lot of space, and I'd be shooting for a 256M tap upsampling filter which needs 16/32Gb RAM to run.)

 

Ok, all clear now
What do you think about NOS filter?
I'm trying it lately more extensively on my WD and it sounds just correct. Probably the most natural of the filters... without emphasizing or excelling in any aspect.
Still I'm not sure if I prefer it to the linear filter.
I'm not oversampling via player.

 

Question for wavedream users: @animus @SoupRKnowva

Animus in your review you mentioned that firmware 4.55 makes it sound mellow and 4.8 much more slammy. Is there that big of a difference in the firmware? This is concerning because what happens if you update the firmware after dialing in the sound you want, the whole system would sound different. What happens if one prefers the sound of the older firmware?

 

They stick to that firmware and stop thinking about it (like I do with FW 1.07 on the Wavelight).

 

I never compared firmware, I updated it to the newest when I received it and moved on.

 

So are you never going to upgrade your firmware again? Sound could change for better or worse, it's a gamble

 

I would like to upgrade my current DAC to either Wavedream Signature SE or Chord Dave. Hence I just tried Wavedream / Wavelight / Hugo TT2 yesterday. Chord distributor in my country does not have Dave unit for demo so I had to try TT2 instead. I took Utopia and Ether 2 with me, but Ether 2 does not have good synergy with Leben CS600X Amp ( connected with both Rockna ) and TT2. It's totally lean and dry, especially with Leben. So my impression is based on Focal Utopia only. I stopped by at Rockna distributor first then Chord distributor. At Rockna distributor, I started with Wavedream Edition, Wavelight, then back to Wavedream before finished.

- Rockna Setup : Spotify on iPad connected via WIFI -> Lan Interface -> Wavedream Edition/Wavelight via Coax -> Leben CS600X EL34 mode via RCA -> Utopia
- Chord Hugo TT2 Setup : Spotify on iPhone via Airplay -> Chord Streamers -> Chord Hugo TT2 -> Utopia

- Wavedream : Starting with MJ's Billie Jean, first thing I noticed is the drum and bass impact at intro. It's tense and slammy. It's the way I like, but it's too much. I tried other songs. I found it's quite good with Pop, Acoustics ( like James Taylor : Fire and Rain , Earl Klugh : Wishful Thinking ) , and R&B. Wavedream has lush, full sound, and layers I've never found in any DAC before. Only issues is the amount of slammy.

- Wavelight : With the same tracks I tested with Wavedream, the instrument is less slammy and more neutral, but lose lush and full sound on Wavedream. Funny thing is I heard singers breath-in ( before they sing ) almost every time, which is also annoying. One good side I found on Wavelight is that it sounds sweet with acoustics music.

- Wavedream again : The amount of slammy decreased this time. I'm not sure whether Wavedream reached its warm up time or I already used to its sound. It's still quite too much for my taste though. I also noticed the singer breath-in sound, but not obviously like Wavelight.

- Audiobyte The Hydra Stack : Audiobyte is owned by Rockna as a second tier brand. The distributor recommended this setup. He said it has the most neutral sound among Rockna setup. I did not have time to try it.

- Chord Hugo TT2 : Before I tried Hugo TT2, I hope its sound is bad so I could focus on Wavedream. However, TT2 has the most neutral, smooth, relax sound with details. Until now, Billie Jean via TT2 is the best version I've ever heard. Only negative is that it lacks the bass and drum impact for my taste.

That's all for my impression. I'm sorry that I could not explain in terms of audiophile. I'm just an amateur who love music. Now for the decision, I still have to study more.

1. If I already tried Dave, it could be a lot easier. If you tried both Wavedream and Dave before, please share your experiences.

2. I just realized that I set Wavedream volume at maximum ( 0 dB ) , which likely equals to 6.6 V per spec. I adjusted the volume at amp instead when listening. I'm not sure whether the volume on Wavedream could be a cause of slam. If I understand correctly, normal DAC voltage should be at 3 - 4 V. Single Ended. Please advice.

3. Other setting I applied was Hybrid filters + Dither On. I also swapped between filters and dither on/off, but could not catch the difference between them. Thank you.

 

DAC makers tend to manipulate output voltage to higher levels to create more positive first sonic impressions. Matching loudness by ears is hard, very hard. The best practice is to match listening volume by a device. Other than that, a sensible way to get a hopefully reliable impression is to audit a DAC repeatedly, ideally along with another DAC with a similar output voltage.

 

Finally got to hear one of these WD signature SE over winter break. It is by far the best modern DAC ive heard so far and while i still think there's room for improvement, but for now this is the new top hurdle. Hard to believe that a little over a year ago i could not imagine anything 'more than marginally better' over Yggdrasil A1.

We warmed it up for about 4 hours before playing with it, Linear filter, fed Coax SPDIF from my Arcam CDT, AVC autoformer volume between Dac and Amps. Listened to mostly on headphones HD580, 600 via EC 2a3 amp and the Germanium Amp. I really wish i got to play with this DAC more on speakers because i think that's where it might be stronger, but thatll be another day. WD does not appear to be sensitive to source compared to other dacs. Yggdrasil requires a AES source, Holo needs HQP and a 1.4536mhz OS from a gaming PC, vintage Dac needs vintage SPDIF. Rockna WL and WD work perfectly great out of their RCA output and yggyA1 has Cukt SE outputs. Basically no contest here, Rockna WL WD digital input stage and USBwipes the floor with Everything. They got it right and that the fact that you dont need to worry about source and decrapification helps justify its price point. This IMO is why WL is still a relevant dac in any conversation.

Wavelight and Yggdrasil, which i've spent quite a bit of time with, is my reference to mainstreet these days. WL to me now is like a YggyA1 with many of its shortcomings fixed (mid bass hump distorted like HD800S, extension lacks, stage proportion narrow on Yggdrasil). WD is in a higher tier - super smooth* and more downsloping FR than WL. WD has more bass heft than WL from a FR balance point of view (WD slams and WL slaps more). WL is flatter sounding (a perfectly flat FR curve in my mind sounds a little bright) with those R2R peaks just like Yggdrasil, bf2, even HD650 etc. Color tone wise, i think WD is more neutral and pale coolness and not as hot as WL (WL and yggyA1 is like a HD800 and WD like a HD580). Tone color is very important to me and i strongly prefer the colorful vibrancy of vintage R2R like the 2.3 dac. I don't know what it is about modern DACs missing this. I do prefer a little bit of warmth (color tone warmer not merely more bass), but it's not excessive. It's far less than Sonic Frontiers or DNA kind of sound.

WL sounds more like it is putting in effort and WD is quite relaxed (and lower energy). WD has nice engaging mids and lower slam that kind of pop subtly which helps it be more engaging and WL emphasizes up higher in its slap transient. I still felt WD restrains the music unlike my 2.3 dac, the decay particularly 'the Air' still feels dampened and truncated to get that smooth sound or texture. The 2.3 excels at sounding fully effortless and free, that air goes on for insane depth especially compared to width that is the tubed analog stage doing magic. Engagement factor is something that 2.3 dac excels at over anything else ive tried and i thought the WD would get totally rolled on this like the WL, but it did have a nice response to show for. I would like to spend more time to know WD better in this aspect.

In truth i expected WD was going to be a WL..yggyA1 on HGH deep down that same tone line, but it wasnt. It sounds totally different and more normal, which is why it's better in my mind. I was pleasantly surprised that WD was meatier than WL. I would say WD's strength is in raw resolution, best ive heard period. It is wide and deep equally, imaging and placement are exceptional. I think this will show more on speakers than headphones unfortunately i didnt get to try that this time. WD smoothness is a blessing and a curse and i dont know how to feel about this when i encounter it; it's like committing no sin, but then lacks 'character' (someone else used the term 'grit' at the same time. 2.3 has lots of textures, the right air and decay, big transients slam and slap especially right the right tube rolled in, and a very organic analog timbre. 2.3 gets destroyed in stage width which is its weakpoint relative to high end OS Dacs (still better than yggya1 but worst than WL). It's kinda like Yggdrasil where depth is far better than width, but since its still a NOS dac the fact that it has any stage at all makes it likely unique. WD still sounds like a modern dac and has lots of cool tricks to show off. 2.3 is just pure fun while still trading blows in technical; the impossible combination.