Denafrips Ares technical measurements

Discussion in 'Source Measurements' started by atomicbob, Jul 23, 2017.

  1. Armaegis

    Armaegis Friend

    Friend BWC
    Joined:
    Sep 27, 2015
    Likes Received:
    3,680
    Dislikes Received:
    2
    Trophy Points:
    113
    Location:
    Winnipeg
    @atomicbob
    I think at some point people might appreciate a post or video detailing just how much work goes into taking these measurements. It's not to toot your own horn, but I think the average person here just doesn't grok what goes into it and imagines it's something akin to hooking up a couple wires and pushing "start" on the magical measurement machine.
     
    MrTeaRex and Elnrik like this.
  2. johnjen

    johnjen Doesn’t want to be here but keeps posting anyways

    Friend
    Joined:
    Sep 29, 2015
    Likes Received:
    613
    Dislikes Received:
    15
    Trophy Points:
    93
    Location:
    Well Grounded
    I 2nd that request.

    However it too takes time etc, time that could be used to measure etc.

    A dilemma of sorts.

    JJ
     
  3. Dino

    Dino Friend

    Friend
    Joined:
    Jun 21, 2016
    Likes Received:
    3,508
    Dislikes Received:
    2
    Trophy Points:
    113
    Everyone I know (including me) finds documentation to be a much more unpleasant (and longer) task than actually doing the job. Maybe @atomicbob is different.

    (Yeah, it would be very interesting, if @atomicbob feels like doing that.)
     
  4. atomicbob

    atomicbob dScope Yoda

    Friend BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    10,188
    Dislikes Received:
    3
    Trophy Points:
    113
    Location:
    On planet
    Somewhere in one of the forums (maybe the predecessor changstar) I did a brief outline of the major tasks from setup to teardown. A video would be wonderful for the viewer but it is a lot of work for the producer. I've worked videos from scene capture through post production. A lot of work in itself. Maybe screen captures, pictures, and a description of what each represents, but even that could easily take 6 to 8 hrs. It is like a miniature version of a studio setup prior to session start or venue setup prior to sound check, through actual showtime and then teardown / documentation / post production.

    Very astute JJ. Quite the balancing act between documentation and actual measurements given time constraints on every day. Even organizing the data collected takes time before it is ready to post.

    Dino, you are a man after my own heart. The documentation is easily the most difficult, especially data quality assurance, error checking and cross checking labels to avoid misrepresentation. Tonight, rather than vetting today's measurements I'm having a glass of good red and listening to Muddy Waters via A/B of dac1541 and Gungnir MB into Liquid Crimson and HD800. Tomorrow I really must wrap up and send the dac1541 on to @Hands who has been waiting patiently for me to quit fiddling around with measurements (and a vacation in between.)
     
    Aleatorius, landroni, Dino and 2 others like this.
  5. gixxerwimp

    gixxerwimp Professional tricycle rider

    Friend
    Joined:
    Sep 30, 2015
    Likes Received:
    2,356
    Dislikes Received:
    13
    Trophy Points:
    113
    Location:
    small island claimed by China
    Maybe just one camera with a wide FOV that you then speed up to something like 10x and then throw onto Youtube. This would give people an idea of the amount of effort involved, and hopefully wouldn't require too much work on your end.
     
  6. Elnrik

    Elnrik Super Friendly

    Friend
    Joined:
    Jan 10, 2017
    Likes Received:
    7,604
    Dislikes Received:
    31
    Trophy Points:
    113
    Location:
    Denver CO
    Home Page:
    I could just follow him around with a video camera for a few days. It gets less creepy after 6 hours or so. Just ask my Ex. Actually, don't.

    I'd assume 1 hour setup, 20 minutes per pic he attached to this thread, and a lot of booze. So that's about 9 hours and a hangover. You have my appreciation sir!

    Back on topic, I'm not sure how long this thing should take to settle in. Left it on since I got it a few days ago, and haven't noticed much (if any) change. @atomicbob did you notice a change after initial power on vs after?
     
  7. atomicbob

    atomicbob dScope Yoda

    Friend BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    10,188
    Dislikes Received:
    3
    Trophy Points:
    113
    Location:
    On planet
    You aren't far off. Setup is in a shared space so it is similar to a temp Sound Reinforcement gig in a venue except I know the space and locations for power sources.

    Approximately:
    2 hrs - setup, warm-up and calibration checks, verify noise floors. DUT is on one shielded distribution circuit. All SMPS powered devices are on a separate shielded circuit.
    0.5 hr - generate folders with template filenames, rename the files (bulk rename utility used), print appropriate data collection checklist Configure periodic backups to server.
    1 hr - setup / teardown camera, tripod & hot lights for photo documentation, then pick the one with best lighting, reduce size in editor for forum limits.
    4 ~ 6 hrs for each superset of 25+ measurements per configuration (spdif, ASIO, WDM, Balanced outputs, SE outputs, Filter modes, etc.) at 44.1 KHz 24 bit.
    This includes final measurement setup adjustments if any, run the measurement, assess quality of measurement, screen capture results, save to correct folder / filename.
    Repeat 4 ~ 6 hrs for each configuration to be measured. If spanning multiple days, repeat warm-up, cal checks, noise floor verification each day.
    2 ~ 8 hrs (depending on quantity of measurements) organizing and vetting data for forum presentation.
    1 ~ 2 hrs agonizing over any comments and annotations to be made.
    1 ~ 2 hrs posting data to forum.
    No libations until the end of each day. Too much opportunity for mistakes. But I appreciate the sentiment.

    Additionally, I do this as part of my corporate responsibilities, so when corporate fills my queue, it takes precedence over my discretionary measurement time.

    denefrips requested I perform 200 hr burn-in before measurements and listening evaluation. I did not listen other than to verify audio was being produced before the 200 hrs was completed. Yggdrasil did make significant , measurable and audible changes after 168 hrs so it is plausible, though I think 72 hrs is sufficient for most devices, first power on. There-after 1 to 4 hrs for thermal stabilization.
     
    Last edited: Aug 30, 2017
  8. johnjen

    johnjen Doesn’t want to be here but keeps posting anyways

    Friend
    Joined:
    Sep 29, 2015
    Likes Received:
    613
    Dislikes Received:
    15
    Trophy Points:
    93
    Location:
    Well Grounded
    The one aspect that can all to easily be overlooked is the degree of focus and attention that must be maintained by the operator, continually, in an ongoing basis for each and every one of those 4-6 hr measurement sessions. Not to mention verification of the setup and remembering to document the notes and info that pops up during these sessions.

    It is a demanding mental challenge to remember ALL of the pertinent details, over and over again, with out slipping up.

    These sorts of data generation sessions are definitely not a set and forget, just mash down the button and walk away until its all done kind of thing.

    JJ
     
    Brad Tombaugh, atomicbob and Dino like this.
  9. atomicbob

    atomicbob dScope Yoda

    Friend BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    10,188
    Dislikes Received:
    3
    Trophy Points:
    113
    Location:
    On planet
    Thank-you for pointing that out jj. It is a demanding task requiring extreme, sustained focus.

    Earlybirds viewing the postings may have noted I removed a 20 Hz -3 dBFS square wave response for balanced output because I forgot to Alt+PrtScr before CTL+V to irfanview and then saved the the screen capture of the 0 dBFS version. which had been the measurement preceeding. Very tired when I went into the posting cycle. Saw the mistake just after saving the uploads.
     
    Elnrik and Dino like this.
  10. Grahad2

    Grahad2 Red eyes from too much anime

    Friend
    Joined:
    Jan 26, 2017
    Likes Received:
    1,218
    Dislikes Received:
    2
    Trophy Points:
    113
    Location:
    Singapore
    Might I suggest Greenshot? This is one of the essentials I install on every machine now since it's light and yet pretty useful.
     
    atomicbob and gixxerwimp like this.
  11. mitochondrium

    mitochondrium Friend

    Friend
    Joined:
    Sep 1, 2017
    Likes Received:
    450
    Dislikes Received:
    0
    Trophy Points:
    63
    Location:
    A Cell
    @atomicbob

    Just want to thank you for the work you put into these measurements. It is very nice to have unbiased test results on gear.
     
  12. atomicbob

    atomicbob dScope Yoda

    Friend BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    10,188
    Dislikes Received:
    3
    Trophy Points:
    113
    Location:
    On planet
    Denafrips Ares data updated to include measurements with USB input, Schiit Eitr DDC and some extra information. See first post and posts 5 through 10. Updated post 1 highlights concerning correlation of measurement data to @Torq listening impressions.

    Begin new data here
     
    Last edited: Nov 14, 2017
    Joshvar, Dino, jowls and 4 others like this.
  13. Reggy

    Reggy Rando

    Executed
    Joined:
    Jun 12, 2018
    Likes Received:
    4
    Dislikes Received:
    17
    Trophy Points:
    8
    Location:
    Niagara-on-the-Lake, Ontario, Canada
    Hi atomicbob,

    I really enjoyed your data analysis you provided on the Ares, it was quite thorough. I do have one question that may or not be possible to answer, would you happen to know the variation of the output impedance on the DAC for the range of frequencies. This is important as I am in the hunt for a preamp that will properly match the impedance ratios. I have an active studio monitor with an input impedance of 10 Kohms.

    Thanks
     
  14. atomicbob

    atomicbob dScope Yoda

    Friend BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    10,188
    Dislikes Received:
    3
    Trophy Points:
    113
    Location:
    On planet
    I did not measure that parameter nor do I have access to the unit, which was on loan at the time. This is not a usual attribute to measure. However, the measurement input impedance was 100K and frequency response on the Ares was typically flat.
     
  15. Reggy

    Reggy Rando

    Executed
    Joined:
    Jun 12, 2018
    Likes Received:
    4
    Dislikes Received:
    17
    Trophy Points:
    8
    Location:
    Niagara-on-the-Lake, Ontario, Canada
    OK, so the 100kOHm input impedance you measured was on the input side of your pre-amp or amp directly after the source.

    What I meant was the 1250 Ohm XLR output impedance from quote https://www.vinshineaudio.com/product-page/denafrips-ares-r2r-dac
    measurement may or may not be the highest impedance, as the output impedance at the audible frequency for which the output impedance is highest is often not known or measured. So perhaps the impedance could be much higher at other audible frequencies, so because of this unknown I am in the hunt to pick a pre-amp, most likely active that will have a significant higher input impedance, preferably 20-50 times higher. If I did manage to somehow find the highest output impedance of the DAC, let's say 2000 OHms, then I could simply use the rule of thumb to pick a passive or active pre-amp with input impedance 10x higher, as is the usual rule when matching the components.

    So basically the reason why the output impedance variation of the DAC is so important for me is because the source impedance does not match my active studio monitors (Genelec 8040B's) with input impedance 10 kOhms and I have to correct for that. if I want to keep my DAC.
     
  16. atomicbob

    atomicbob dScope Yoda

    Friend BWC MZR
    Joined:
    Sep 27, 2015
    Likes Received:
    10,188
    Dislikes Received:
    3
    Trophy Points:
    113
    Location:
    On planet
    Here is a diagram of my typical measurement setup (DUT = Device Under Test):
    Audio Analyzer Diagram - Sig Gen DUT Analyzer.png
    The load placed on the output of the DUT, in this case the Ares, is 100K.
    If you look at the Ares balanced Frequency Response at the beginning of this thread you will see how it varied with a 100K load.

    If you are concerned about the Denafrips Ares 1250R output impedance with a 10K load (which is a factor of 8 larger) then here are some alternatives (and much better sounding in my opinion):
    200R RME ADI-2 DAC
    100R Soekris dac1541
    75R Schiit Gungnir MB
     
    Reggy and Elnrik like this.
  17. Reggy

    Reggy Rando

    Executed
    Joined:
    Jun 12, 2018
    Likes Received:
    4
    Dislikes Received:
    17
    Trophy Points:
    8
    Location:
    Niagara-on-the-Lake, Ontario, Canada
    Got it. That makes sense now about where the 100k comes into play here.

    Thanks for the list of alternatives , it may be better for me to go for one that matches the impedance on the first try. I will take a look at your list, will probably aim for 100R Soekris dac1541, the 4.0V RMS, Zout 100 ohm which gives me a factor of 100 seems really reasonable for my setup.

    I am really surprised on the amount of praise on the Ares, although the amount of enthusiasm seems very local yet concentrated wherever you find it on the web. I don't know if some people are tricked somehow into placebo but it seemed like a reasonable investment without putting too much cash down for possible waste if it didn't turn out well.

    I don't see why people would take the risk to purchase the Pontus or Terminator when there are very few reviews on them across the web, and as good as the Ares could sound for its price point in the r2r technology rave I doubt the others would be considered the huge investment when there are so many other competitors at the $2000-5000 price point . What I don't like is the people who did end up praising Denafrip's higher end models all seem to have slippy trigger fingers, and join the bandwagon or have the fear of missing out sensation.

    Anyway, enough of an aside, cheers for the list in case I get stuck.

    Reggy
     
  18. mscott58

    mscott58 Friend

    Friend
    Joined:
    Oct 13, 2015
    Likes Received:
    1,696
    Dislikes Received:
    8
    Trophy Points:
    113
    Location:
    Philly - Yo
  19. Lyander

    Lyander Too sensitive for SBAF

    Friend
    Joined:
    Feb 25, 2017
    Likes Received:
    4,509
    Dislikes Received:
    24
    Trophy Points:
    113
    Location:
    Pasig, Philippines
    Price was a topic of interest earlier. Apparently this is more expensive than the MSRP at Vinshine (official retailer), which offers free shipping globally besides:
    https://superbestaudiofriends.org/index.php?profile-posts/18079/

    They're currently out of stock though, which bites. Then again, unless a sale or great used deal I don't see wisdom in rushing to get a piece of gear.
     
    Last edited: Sep 9, 2019
    msommers likes this.
  20. mscott58

    mscott58 Friend

    Friend
    Joined:
    Oct 13, 2015
    Likes Received:
    1,696
    Dislikes Received:
    8
    Trophy Points:
    113
    Location:
    Philly - Yo
    Sorry about that. I did a quick search of the web and couldn't see a retail price.

    Can't say I'm surprised. Drop can list RPs that are not always grounded in reality. Sometimes it's higher than Amazon's price and you have to wait for months versus getting it in two days...
     
    Lyander likes this.

Share This Page