Discussion in 'Digital: DACs, USB converters, decrapifiers' started by Torq, Nov 7, 2016.
I'm definitely IN for this loaner
I would be in for the loaner as this Dac interests me
I'm more interested in seeing more data points for this and the metrum pavane, and some of the other metrum line and other worthy dacs. I love the schiit mb dacs, but they should not be the end all. More choices please .
I'm not sure if the problem is so much a lack of choices but that $5k+ DACs like the Pavane are in cost-prohibitive territory for most people.
The Spring is a much easier pill to swallow and given the price and initial impressions I hope it disquiets DAC value proposition much in the manner the Yggdrasil did.
And so it begins... if I see enough interest in my stuff, I might be dumb and get a Holo.
I just spoke with Tim Conner @bimmer100 at Kitsune Hifi and placed an order. The unit should be coming in around mid December. Very much looking forward to this. Thanks for your impressions guys and pointing this out. We need more R2R type DACs.
In the meantime, I've invited Tim to feel free to chat with us and answer any questions we might have on the DAC. He runs a small business and is restricted to what he can say on HF. Yeah, so he's approved here. I like the fact that Tim has exhibited a lot of class: mostly lurked, done a proper introduction - making clear that he is an MOT, and never pushing his stuff. These are traits of a quality person.
Sounds like the i2s is the best input for the Spring?
I have a decent HDMI coming to try the Singxer via i2s vs the Lynx via AES before the Spring has to go to the next person.
@bimmer100 (Tim) is super awesome to work when ordering stuff, even if I can ask a lot of question or be a bit forward at times.
I would be curious to see if more people can hear differences between fuses in the Spring and get their subjective thoughts. I'm at the stage I can start worrying about insane things like that and cables.
BTW, there are a lot of good technical details about the Spring on the product page and a few forums. I don't always understand it, but it seems a lot of thought was put into this. Or it's Jedi hand waving!
I can also confirm the technical measurements on the DAC are superb. It maxed out my ADC in almost every possible way. Sure, you get the usual high-frequency droop at redbook, and THD/IMD might be a bit high (on my incredibly modest measurement rig - common on NOS DACs) when you start pushing high level, high treble frequencies, but...really, subjectively and objectively, this is a NOS DAC with very few compromises.
I'd definitely be interested in the loaner tour for this. Been looking for an R2R DAC to compliment my main source (vinyl).
I thought so.
The difference is there, it's not huge, but it was the best result for me.
AES performance out of the SU-1 was also very good, though I felt the Aries and RedNet 3 were both just slightly ahead there ... and there's not much point in using the AES output of the Singxer into the Spring when I2S works better than all other options on the table.
One important caveat with I2S ... it was originally intended as an short/direct internal connection and as a result it is something you want to minimize connection lengths for. Shorter the better. I used a 0.3M cable with my testing.
Which I2S cable do you use? Just a generic HDMI or something a little more fancy? Cheers
Trigger pulled. Estimated delivery mid-December. Thanks Tim!
For the audition it was a 0.3m Wireworld Starlight 5.2 HDMI cable.
I really didn't spend any time playing with different cables, beyond that necessary to try the different inputs, during this audition, since I wanted to evaluate the different inputs themselves first and see if/what/where things like the JCAT isolator were useful for.
My experiences with I2S in general, which has been in an internal context for the most part, but does include the old Audio Alchemy stuff, I strongly suspect that length, or lack thereof, will be the primary consideration in cable choice.
Awesome intel. Thanks
An important note for those of you that haven't used I2S connections before ...
In general it is not implemented in a manner which makes it safe to make/break the I2S connections with the hardware powered on. So please make sure you power down the DAC and the DDC you're connecting it to before you plug-in or un-plug the I2S connection (be it via an HDMI cable or some other scheme).
Random thought on I2S, would it be possible to go directly from the I2S pins on a Raspberry Pi out to the Holo? Would that sound awful because of the clocks on the Pi, or would the Holo reclock them?
Maybe there's an I2S reclocker hat one could use for the Pi if the Holo won't reclock. That would eliminate the need for a USB->I2S converter.
@Scott Kramer you've played around with I2S a bit, right? Is this something feasible?
You could go from the I2S output on a Raspberry Pi to an I2S compatible DAC like the Spring DAC. That is assuming that you, or someone you know, is comfortable soldering in the necessary optional pin-header to get access to that output. The lack of audio-frequency-specific clocks will result in higher jitter than is desirable, yes.
It's important to understand that I2S is intended as an internal digital interconnect. So, you're generally going to find that it's a bigger pain in the arse, with much stricter implementation constraints, than other interfaces. The original design intent was to communicate between the data-discriminator in a CD-player's read circuits and the built-in DAC chip(s) in said CD-player. And in such cases, the clock is common/shared within all circuits in that unit.
I don't know, for certain, if the I2S input on the Holo Audio unit buffers and re-clocks that input, but I'm inclined to doubt it, as that's really not how I2S is generally implemented. It CAN be done, but it isn't usually - though that may change as it becomes more widely offered as an input option.
The HiFi Berry Digi+ Pro AND DAC+ Pro do have audio-optimized clocks, and I2S output (again, requiring soldering to access) which would improve I2S jitter performance over a raw Raspberry Pi connection. This is probably the best approach.
Beyond that, while it's another story, for another time, there's a little project I'm working on that'll combine NAI/AOIP capabilities with AES, S/PDIF (Coax and TOSLINK) and USB output - and for which I began adding I2S support this week. Some other discussions, including with @Marvey, need to occur before I am willing to talk more about that, however.
Hmm, seems like it would be a bit of a pain to get it set up but in theory it could give you a cleaner digital path than USB->Singxer->I2S->Holo unless the presence of the Digi+ Pro as a reclocker mitigates whatever advantage you would have had by bypassing USB.
I have a question regarding I2S vs AES/EBU.
I know that I2S input is generally superior to AES/EBU in most DAC's that offer an I2S input, not just Holo Audio Spring. So, if I were to use a DDC that can serve as both USB-to-I2S converter and USB-to-AES/EBU converter (such as Singxer SU-1 or Empirical Audio Off-Ramp), it would be more beneficial to use I2S.
However, in either case, it doesn't change the fact that you would still have to use USB somewhere in the audio chain, since these DDC's must receive a USB input signal from the computer. The thing is, as recently demonstrated by "USB-less" DDC's like Focusrite RedNet D16 or Lynx AES16e, bypassing USB is generally superior to most USB DDC's.
So my question is, would using these USB DDC's with the superior I2S outputs still outperform these "USB-less" DDC's with AES/EBU outputs?
And, better yet, is there a "USB-less" DDC which completely bypasses USB that also has an I2S input (in the form of HDMI)?
Trusted Friends only. Putting a list together for now, DAC goes out later when @Marvey is done.
Separate names with a comma.