Discussion in 'Digital: DACs, USB converters, decrapifiers' started by Torq, Nov 7, 2016.
What AES cable were you using?
Just the AES out that comes attached to the Lynx L22 Sync Cable set. I haven't done any modding or anything to make an AES out only cable.
Ah yes, I see.
Just a comment in passing…
There be gains to be realized by using 'better' AES cables.
Perhaps enough to significantly sway your results, at least if the results I'm getting are any indication.
I was also splitting hairs when I was trying to figure out which was the best sounding interface between the Yggdrasil & Mutec MC3+USB, it was between the Glass Toslink and AES cable. They sound really extremely close performance wise, I am very sure that to most untrained or inexperienced ears, they could not tell any difference at all because the overall frequency response and sound signature seems to be exactly the same.
That was when(after hours and hours of listening) I figured that you must pay especially close attention to how your DAC performs on human voices(unplugged music or live jazz or leonard cohen) on these to determine a winner. The better interface will deliver a better "emotional" or "authentic" connection for human singing. You can hear more clearly, the very tiny detail of straining/wavering/very slightly ever changing pitch of singer's vocal cord on the better interface. Also how the overall attack profile of each musical transient is. The better interface will provide a better defined envelope(either better decay or strengh & speed of initial attack or more noticeable sustain).
Also you must ensure all your equipments in your chain are properly warmed up(>4hours or more) and freshly rebooted(including powercycling of dacs, ddc and etc) before you evaluate this kind of hair splitting things.
That's my own observations and this might work or not work for you.
@MisterRogers has thrown a wrinkle in my reality by pointing me at some excellent DSD recordings by Stockfisch.
Here's a free DSD128(5.6MHz) music:
Does the Holo do DSD in NOS mode, or only in OS DSD mode? Also does it max out at DSD 256?
Yes, DSD in NOS mode.
It does DSD in NOS or OS. It maxes at DSD 512. Not sure if it does 512 in NOS mode or OS or both. The highest I've tried in 256 because I can't find files at 512.
I don't think any 512 files currently exist for commercial sale, none that I'm aware of anyway. But you can set up hqplayer to upsample to 512 in the meantime.
For the sake of testing while messing with the built-in USB (which is better than I expected), I upsampled to 8x DSD in JRiver and it worked. Screen says DSD512 as well.
Hi Guys. Been following thread awhile. I bought a Spring L 3 about 3wks ago. Also using the Singxer SU-1, Sonore Microrendu, Uptone LPS-1, Intona and utilizing the I2S input. Using the same generic USB that came with the SU-1 and the short generic HDMI cable Tim sells for $10.
I'm upgrading from an iFi Micro DSD. I can easily agree with all the reviews here. Huge detail and resolution. WIde soundstage and quite deep. Bass is tight and deep, midrange lush and fantastic upper freq. resolution.
Also agree, that it's a bit too analytical for 'dirty' music, Industrial/Rap for example. I too prefer to listen to these in OS PCM mode. Other than that, all NOS. Upsampling from HQP.
Tim's been an awesome person to deal with, and has helped 'educate' me thru this whole new game to me.
I can easily recommend Tim, and the Spring.
Finally got around to posting my Spring impressions. Includes direct, A/B comparisons against the Metrum Menuet.
See here, post #2 in the thread that @Torq kindly reserved for me.
Sorry I wrote so much. Too late now!
Very nice write-up!
And it definitely increases my desire to revisit the Metrum Pavane in a back-to-back evaluation with the Spring DAC (so far, I've gone on memory alone). My Yggdrasil isn't going anywhere, but even without proper warm-up, I've been enjoying the Holo Audio unit for the better part of the day.
I am thinking that, when I kick off "round 2" of my "Life after Yggdrasil?" quest, it'll begin with an Yggdrasil/Pavane/Spring DAC evaluation ...
That'll have to wait until the new year though ... there is much music to listen to (promised myself a complete tour of Leonard Cohen's works, in cased I managed to miss any ... [not so far] before I got back to that), and there is much vinyl still to acquire in my respective ATT100 lists ...
@Torq and @Hands
You guys are skilled swordsmen!
Good stuff @Hands! Very useful information, that makes me even more eager for my Spring L3 to show up soon. Cheers
Thank you Hands for you impressions. Can't wait for some level headed impressions of lvl 1 & 2. 1000$ price difference is quite steep.
I want to be very clear and reiterate my point that non-oversampling DACs can be very hit-and-miss with different individuals. Most will prefer oversampling DACs. Some will feel indifferent or be OK with either. A small minority will have a preference for non-oversampling. What does this mean? Don't go in blind, and don't buy out of hype or a desire to blow money on a shiny new toy. Experiment with a variety of other non-oversampling DACs and participate in loaners if and when you can before you jump into a pricey DAC like the Spring or Menuet/Pavane.
If I had to guess, even the base level Spring should sound pretty darn good. Should make the Musette feel very uncomfortable at its price point as well as provide a strong NOS alternative to the Gungnir Multibit. I could be wrong, having only heard the Level 3.
BTW, I would not say that DACs like the Spring or Pavane are necessarily better than the Yggdrasil. They're more like non-oversampling alternatives that are on the same playing field, so to speak. Be wary of anyone going around claiming the Spring is some DAC that has ascended from the heavens to defeat all other DACs. It's not that simple, and claims that like should make you pause.
Great writeup, @Hands (and of course, @Torq). Very interesting observations. In my limited experience, I have found just the opposite regarding the 2D/3D of NOS vs OS. I lived with a AD1865N-K filterless NOS DAC for quite a while, and then brought in the Yggdrasil. Soundstage was more holographic with the Yggdrasil, and Yggdrasil's own soundstage seems to become more holographic as I increase the sampling rate - e.g. x4 with 176.4kHz out of Audirvana further improves the holographic effect. This, however, introduces iZotope for sample rate conversion. But thus far, I've found it preferable.
Nevertheless I'm very interested to compare Yggdrasil again to a NOS implementation, in this loaner tour. For the longest time, I had convinced myself that NOS simply had to be more natural sounding. That all went out the door when I finally critically compared to the top notch multibit Yggdrasil implementation. Time for round 2.
Which AD1864 DAC to be specific? And are you sure you meant 1864 and not 1865? There are a lot of bad NOS DACs on the market...And a lot of NOS DACs have really compressed, claustrophobic stages, if not a host of other major issues.
To clarify, my point was more that while OS can have a more precise sounding stage and more depth to/between layers, thus leading to a more holographic stage, all the elements to me with OS sound like they are 2D. Think something like the Paper Mario games. The world is 3D, but all the characters are 2D. It's hard to replicate that same holographic nature with NOS (Spring and Pavane are about as close as you can get from NOS) but have 3D sounding individual elements. When I used to play guitar (no, I wasn't good, and, no, I didn't play for long), NOS so far has better captured the sense of 3D body a good acoustic guitar is capable of. OS flattens that all down into a 2D plane to me.
On a side note, I noticed the Metrum Acoustics Facebook page posted something about discrete R2R DACs. They aren't fans. Maybe they got wind of the Spring? LOL (To be fair, and as I understand it, their critiques are valid but don't take into account some of the design choices in the Spring that are supposedly meant to address this. I'm not an engineer and can't verify the Spring is really mitigating the weakness of discrete ladder DACs.)
Separate names with a comma.