Rockna Wavelight - Holo Audio Spring2 KTE listening evaluation

Discussion in 'Digital: DACs, USB converters, decrapifiers' started by atomicbob, May 13, 2021.

  1. atomicbob

    atomicbob dScope Yoda

    Pyrate BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    18,653
    Trophy Points:
    113
    Location:
    On planet
    Rockna Wavelight - Holo Audio Spring2 KTE listening evaluation

    All evaluation performed with the following setup conditions
    1) Dante Digital Data Delivery System to DAC AES inputs
    2) 88 KHz sample rate to DAC AES inputs
    3) JRiver SOX SRC used to homogenize source material to 88 KHz sample rate
    4) NOS filter used on each DAC
    5) DAC outputs level matched to within 0.031 dB
    6) listening levels 70 dBSPL average with 80 dBSPL peak at headphones
    7) Genelex Gold Lion 300B tubes used in Eddie Current Studio B
    8) Studio B adjusted for 0 dB gain input to output
    9) levels adjusted with Goldpoint SA2-47 stepped attenuator
    10) listening room ambient SPL level observed at 26 to 30 dBA; 38 to 42 dBC

    setup interconnection diagram
    20210513 D16R - LiveClock - FS725 - Wavelight - Spring2 KTE - Studio B - listening setup.png
    Note: ABX is inserted between the DAC outputs and the Goldpoint SA2-47

    ABX listening station
    DSC_0625_cropped_small.jpg

    Calibration
    Prism Sound dScope III used to adjust levels
    signal generator set for 0dBFS 1KHz to AES output at 88KHz

    ABX level matching setup
    IMG_2863_small.jpg

    ABX level match results
    Rockna Wavelight level match
    20210513 0dBFS level match EC Studio B - Rockna Wavelight.png

    Holo Audio Spring2 KTE level match
    20210513 0dBFS level match EC Studio B - Holo Audio Spring2 KTE.png

    Matched to 0.031 dB in Left channel
    Matched to 0.024 dB in Right channel

    Source material used:
    Brad Mehldau - It's Allright with Me
    Canadian Brass - Purcell Trumpet Tune
    Gary Stroutsos - Forgotten Seas
    Ray Bryant Trio - All Blues
    Brad Mehldau Trio - It's Alright with Me - Brad Mehldau Trio
    Michael Gettel - Summer Rain
    Blood, Sweat & Tears - Blues - Part II
    The Beatles - Blackbird
    John Hammond - Get Behind the Mule
    Steve Treseler Group - Foxy Trot
    Taj Mahal - Mind Your Own Business

    Blind ABX testing used to confirm ability to discriminate differences between DACs
    The results are shown below
    IMG_2866_small.jpg

    During the first run (training) challenge step 5 was not correctly discriminated.
    The two successive runs were perfectly discriminated.

    Commentary
    At this extraordinary level of performance Wavelight and Spring2 KTE sound more similar than different. Please keep in mind the differences described below are subtle but differentiable as demonstrated in ABX testing above.

    Differences observed
    1) Leading edge of kick drum exhibits greater impact with Spring2 KTE
    2) Spring2 KTE upright bass presents with greater richness
    3) Wavelight has enhanced clarity in the 100 to 300 Hz range on dense passages
    4) 300 to 3000 Hz the DACs sound fairly even in presentation
    5) Above 3 KHz Wavelight presentation is brighter
    6) Saxes and brass have more bite on Wavelight. Some may prefer this enhanced bite. I was in the studio with the Steve Treseler Group and heard Steve's tenor sax along with Ingrid Jensen's trumpet. To my ears the Spring2 KTE renders the recording more closely to what I heard in the studio (Some of my mics were being used on the project.)

    Conclusion
    These are two very fine DACs. I have a slight personal preference for the Spring2 KTE and it will not be displaced by the Wavelight.
     
    • Epic Epic x 60
    • Like Like x 30
    • Miss Information Miss Information x 1
    • List
  2. earnmyturns

    earnmyturns Smartest friend

    Pyrate
    Joined:
    Sep 25, 2016
    Likes Received:
    3,218
    Trophy Points:
    113
    Location:
    Palo Alto
    Home Page:
    Thank you for your as always educational comparison. I've been upsampling all 44.1 to 88.2 in Roon ("smooth" aka slow rolloff linear phase filter) for my Spring 2 KTE as I find this a tad more defined than 44.1 NOS. As you note, leading edges and double bass are especially convincing. My question: Wavelight has built-in filters, does any of those change the comparison vs SOX SRC+NOS?
    PS: Your comments on "Foxy Trot" remind me to re-listen to that whole wonderful album, I wish I had been there.
     
    Last edited: May 13, 2021
  3. skem

    skem Friend

    Pyrate
    Joined:
    Nov 1, 2017
    Likes Received:
    1,911
    Trophy Points:
    93
    Location:
    Charles River
    In other words, the DACs are very different.
     
  4. mitochondrium

    mitochondrium Friend

    Pyrate
    Joined:
    Sep 1, 2017
    Likes Received:
    1,115
    Trophy Points:
    93
    Location:
    A Cell
    When the May (rain) comes all of this shall be washed away(?)
     
  5. totlsota

    totlsota New

    Joined:
    Sep 1, 2019
    Likes Received:
    30
    Trophy Points:
    13
    Location:
    London, United Kingdom
    That is a fine write-up. I have a Studio B arriving next week. It is already here in the UK needing customs and such. Of course, I am very much looking forward to it.

    I was recently looking for a second DAC to use for my office setup and I got a second-hand S2KTE. I've been enjoying the S2 in my current main setup (soon to be office setup). I was looking for either a used Wavelight or S2KTE. The S2 came up and answered that. You've minimally made me feel like there is probably not a million miles between them, thus shoving down my FOMO. Thanks for that.

    I was wondering how relatively great your upstream clocks, focusrite, etc are. I was considering a move from [SOTM streamer > Singxer SU1 > I2S > DAC] to a [Pi2AES running Roon > I2S > DAC] . Everyone here seems to be a big fan of Pi2AES here. I'd not seen a chain like yours and was wondering how heartily you recommend a path like yours?
     
  6. spoony

    spoony Spooky

    Pyrate
    Joined:
    Sep 28, 2015
    Likes Received:
    651
    Trophy Points:
    93
    Now THIS IS audio science review.
     
  7. Baten

    Baten Friend

    Pyrate
    Joined:
    Mar 18, 2018
    Likes Received:
    1,130
    Trophy Points:
    93
    Location:
    EU
    Audio actual expertise with science on top review.
     
  8. Hands

    Hands Overzealous Auto Flusher - Measurbator

    Staff Member Pyrate MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    12,285
    Trophy Points:
    113
    Location:
    Colorado
    Home Page:
    @atomicbob Time and sanity permitting (a tall order, I know!), I would be curious if you could test again with both set to OS mode and without prior software upsampling.

    Alternatively, run both over USB, use software upsampling at 8X for the Spring in NOS mode, and the Wavelight in LIN OS mode without software upsampling.

    I don't think the Wavelight's NOS mode operates like true NOS. And that you found the Wavelight brighter in the treble, even with both at 2X software upsampling, suggests to me this could have been a clearly audible factor at play.
     
  9. atomicbob

    atomicbob dScope Yoda

    Pyrate BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    18,653
    Trophy Points:
    113
    Location:
    On planet
    General comment on ABX testing. It takes significant time and concentration to setup, calibrate, control environment variables and perform such a test. This report represents a week of evenings for setup, cal and familiarization. Actual testing took place during quiet ambient times over many hours. One doesn't simply knock out a proper level matched, blind ABX test.

    I haven't spent sufficient time listening to other filters in the Wavelight yet to comment.

    Steve Treseler Group recorded in several studios in the PNW; Studio X and Robert Lang Studios. I had Martech MSS-10 preamps, schoeps cmc641, AEA R92 ribbon, Elux 251 and MD431 mics deployed in the sessions. Reed Ruddy was engineering.

    Ability to differentiate required a very quiet environment to minimize auditory masking. Had the room ambient noise been much higher than the stated 38 to 42 dBC discrimination might have failed. See these posts for details on my room ambient:
    Ambient levels and noise floor discussion
    Ambient levels and noise floor discussion

    The KTE May was here and I was able to compare with KTE Spring2.
    System Synergy - Special Sound
    In the end, I preferred the KTE Spring2.

    I use Dante in my audio / acoustic lab as it simplifies audio distribution and comparison evaluations. There are four rooms with Dante AES drops. Atomic Clock disciplining effects are described here

    See first comment about time required. My next task with the Rockna is to finish measurements. With the further complication of having my semi-retirement derailed my spare time has been seriously squeezed, at least until the end of May. So I may be able to accommodate, but also it may not happen.
     
    • Like Like x 10
    • Epic Epic x 1
    • List
    Last edited: May 14, 2021
  10. Merrick

    Merrick A lidless ear

    Pyrate
    Joined:
    Jan 6, 2016
    Likes Received:
    12,569
    Trophy Points:
    113
    Location:
    Portland, OR
    @atomicbob Is there a benefit to resampling all content to 88.2, or is it that the Dante doesn’t change sampling rates on the fly? I also use JRiver so I could enable SoX if there is a tangible benefit to resampling all content to 88.2.
     
  11. msommers

    msommers High on Epipens

    Pyrate
    Joined:
    Nov 6, 2016
    Likes Received:
    2,750
    Trophy Points:
    113
    Location:
    Calgary, Alberta
    Home Page:
    Did we ever get a Morpheus in your hands @atomicbob ?
     
  12. earnmyturns

    earnmyturns Smartest friend

    Pyrate
    Joined:
    Sep 25, 2016
    Likes Received:
    3,218
    Trophy Points:
    113
    Location:
    Palo Alto
    Home Page:
    If @atomicbob feels like evaluating the Morpheus, I could lend him mine as it sits next to an Yggdrasil, maybe a bit more grit and less romanticism wouldn't be a bad idea for a while ;) My opinions are not worth much compared to his expertise, but since I also have a Spring 2 KTE, on another system with very similar amps and headphones, I'd say that Spring 2 KTE bests Morpheus in every department I care about. My (from memory) take: Spring 1 KTE < Onyx < Morpheus < Spring 2 KTE.
     
    Last edited: May 15, 2021
  13. atomicbob

    atomicbob dScope Yoda

    Pyrate BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    18,653
    Trophy Points:
    113
    Location:
    On planet
    Yes.

    Here is 20 Hz square wave response for Wavelight with Lin filter selected at 44 KHz
    20210515 SigGen Spring2  20 Hz square 2000mVpp 200nS div 5MHz filter AES 44K SE - Lin.png

    Here is 20 Hz square wave response for Wavelight with Lin filter selected at 88 KHz
    20210515 SigGen Spring2  20 Hz square 2000mVpp 200nS div 5MHz filter AES 88K SE -Lin.png
    Note how transition time and ringing both have shorter durations at 88 KHz than 44 KHz.
    Nothing is free though. Upsampling will also have some filter ringing but in the grand scheme of engineering and listening tradeoffs I prefer to upsample to 88 KHz and experience better DAC recover filter performance.

    Morpheus evaluated here
     
  14. penguins

    penguins Friend, formerly known as fp627

    Pyrate
    Joined:
    Sep 26, 2018
    Likes Received:
    3,451
    Trophy Points:
    113
    Location:
    SouthernCA
    For the tiny speckle that this is worth after atomicbob's ABX test, I would completely agree that the WL (loaner) and Spring 2 (have had a level2 for a few weeks now) are more similar than different compared to many other DACs but that many a "good listener" would be able to tell them apart most of the time assuming the "setup was good", etc.
     
  15. BarnBurner

    BarnBurner Acquaintance

    Joined:
    May 24, 2021
    Likes Received:
    53
    Trophy Points:
    18
    Location:
    New England, USA
    Could someone explain to this noob what the handwritten test results mean, with the 3 runs?
     
  16. noisyscott

    noisyscott New

    Joined:
    Apr 13, 2021
    Likes Received:
    31
    Trophy Points:
    13
    Location:
    Austin TX
    It shows that AtomicBob was damn-near-perfect when choosing whether the randomly chosen "X" was A or B.

    ABX test:
    1. Listen to A
    2. Listen to B
    3. Testing device presents user with X for evaluation
    4. User chooses whether it is A or B
    5. Score the result
    6. Wash rinse repeat until you achieve a confidence level you want or you go insane
     
  17. BarnBurner

    BarnBurner Acquaintance

    Joined:
    May 24, 2021
    Likes Received:
    53
    Trophy Points:
    18
    Location:
    New England, USA
    Thanks Noisy, now I get it.
     
  18. etherealsound

    etherealsound New

    Joined:
    May 30, 2018
    Likes Received:
    17
    Trophy Points:
    3
    Location:
    Madison, WI
    Really impressive work you've put in here. It's not too often you see people who take this much care and preparation into comparing any gear. I am curious though, could you perhaps give just a bit of insight as to why you preferred the Spring to the May?
     
  19. atomicbob

    atomicbob dScope Yoda

    Pyrate BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    18,653
    Trophy Points:
    113
    Location:
    On planet
    My perception of the audible differences between KTE May and KTE Spring2 was less important than a practical consideration of single chassis vs. dual chassis in my lab. If this was going to be the only DAC in my listening station I might have chosen the KTE May. From the picture above you see I am often running two or more DACs.
     
  20. msommers

    msommers High on Epipens

    Pyrate
    Joined:
    Nov 6, 2016
    Likes Received:
    2,750
    Trophy Points:
    113
    Location:
    Calgary, Alberta
    Home Page:
    Sounds like I need to get my hands on a Spring2. Or is the equivalent the May L2 now?
     

Share This Page