Holo Audio - Spring DAC - Level 3 - "Kitsune Tuned Edition" - Impressions & Reviews

Shipping information received! Now let's see how long it take for my L3 to make it from HK to Philly. Cheers

 

Could someone compare Holo Spring vs AGD Nos7 (or Nos11) or Master7 (or Master11) ?

 

I was fortunate to be able to try the Holo Spring Kitsune level 3 loaner. Overall it is a good dac subject to certain conditions, but not quite as good as Gungnir Multibit. It needs 3-4 days to warm up. A high quality source like rednet is necessary to get any soundstage depth (to my ears, though I don't have much success hearing headphone soundstage depth in general). The main sonic differences I heard vs Gungnir Multibit were a more forward sound, less sub-bass, slightly less smooth, and slightly less resolution (mainly micro-dynamics).

Longer chronological version of events:

I received the unit and then went out of town for two days, leaving it on to warm up during that time. My chain has had some big upgrades recently and is now rednet d16(spdif)->Gungnir Multibit(xlr)->jensen isomax(rca)->445->utopia. I first inserted the holo spring in the chain directly replacing Gungnir Multibit without changing any cabling.

It sucked. Vocals sounded rough and forced compared to the more smooth effortless presentation I'm used to. Resolution was clearly inferior. Basically it sounded like everthing I had gained when I got the 445 was lost. I was about to box it up right then, but I figured I have an SU-1 so I should at least try the I2S input.

I got a 1-foot HDMI cable two days later, so the holo spring has been on for 4 days. I connected it though the SU-1 I2S, and surprisingly it was much better. It didn't sound forced, the roughness was nearly gone, and resolution was close enough to Gungnir Multibit that it was hard to tell a difference. The main issue was there was basically no depth to the soundstage. Now I've never heard what I'd consider to be good soundstage depth from any headphone system, but at least with Gungnir Multibit I get a little bit of 3D/layering effect. With the holo spring on i2s everything was in a 2D plane.

Then I switched back to the rednet and it sounded even better. There was actually some depth. So I guess it was the fact that it had been on for 4 days that made it sound better and not the i2s. It is still a more forward sound than Gungnir Multibit, think first row vs 5-10th row. It makes vocals more forward in the mix, which is not necessarily bad, but they still do not quite have the silky smooth sound I am used to with Gungnir Multibit. It was hard to tell the difference in resolution, but after listening for awhile I felt like I was still missing some of the microdynamics that I had gained with the 445. So while much closer than at first, overall I'd still have to say its not as good as Gungnir Multibit.

I pretty much only used it in NOS mode as everyone says that's the only mode that's any good. I did briefly switch it to OS, and the main difference I noticed is the soundstage was squished in towards the center. I've never heard a NOS dac before, and based on this one I have to say I don't get the whole NOS=nonfatiguing thing. If anything it seemed like the holo spring would be more likely to get fatiguing than Gungnir Multibit. It comes off as slightly bright due to the relative lack of sub-bass and slightly sharper sound.

 

You must have had the $100+ fuse in upside down, there is no way the basic fuse could sound better! ;-)

 

I find my fuses work and sound better when I wrap them in foil first. I think it keeps cosmic radiation from contaminating the internal electronics of these high tech devices...

 

Could well be the case. The overall magnitude of the differences I heard vs Gungnir Multibit is not large.

Interestingly, I had to unplug my Gungnir Multibit for a while today due to power issues. When I listened after it had only been back on for an hour or so, I felt like it had some of the same issues I had with the holo spring - lack of stage depth, roughness in vocals. I wonder if the holo spring needs more than 4 days to fully stabilize.

 

I should be in bed now, but I'll give a quick chime to back you up (just to prove you are not crazy ) - The stock fuse did sound better in at least one off my applications. Off the top of my head, had that with an Audio Analogue amplifier once.

With power related tweaks I personally tend to avoid (those which makes a significant difference may cause issues with my housing/fire insurance policies, and power cables, fuses, power boxes/filters tend to be too neurotic or just gives me more problems to solve), but I am fortunate in the sense that I live in a relatively new building without wiring/dirty power line issues.

 

Very interesting discussion going on here! I was content with my early run Vinshine Audio DAC-R2R-REF (0.02% resistors) but reading all of this hooplah on the Holo Spring Level 3 made me jump on a pre-order earlier this week. I'm expecting to hear great things from the Holo Spring L3. My setup is for stereo so I'm interested to see how my experience compares to what I'm reading here.

 

Could it be the mythical 500 hours of burn in time myth, that I see being thrown around at various places, as the cause for the Spring DAC not lining up with Torq's impressions?
Especially with all the silver used in the the DAC, maybe a few weeks of burn in will change the sound signature for the better?

 

Isn't it more stuff getting up to temperature, and the accuracy being somewhat dependant on that, rather than magic?

 

The Holo uses LVDS signals to transmit the I2S data, so the original spec is irrelevant. You also need to get the LVDS / PSaudio spec version of the Pink Faun card.

 

The I2S spec is concerned with more than just wire line characteristics; the reduced susceptibility to common mode noise that differential signaling provides doesn't change that.

I don't think I "need" to do that at all.

It would do me absolutely no good. I don't run Windows, I don't use a PC or other computer as a directly connected source (outside of occasional testing, but that's never my actual source when listening for pleasure) and the computers I do use, outside of my lab/workshop don't have card slots.

...

Interesting first post; hope you didn't sign-up just to "make corrections" that are based on faulty assumptions.

 

Tim would tell you that most of the burn-in time is for the Jensen caps.

I'm not a big believer in burn-in for the most part (outside of transducers and tubes, both of which have readily observable, if minor, changes over time), I find it to be, mostly, a too-convenient excuse that probably results more in conditioning of the listener than it does the gear - usually over a long enough period that either return periods have expired or the user stops caring.

Caps can settle to a certain degree with use, but if it was that big a factor I'd expect manufacturers to have big racks of them cycling under load to address that. And they still operate within spec during this period.

This is more likely what happens for "burn in" with DACs, at least in my opinion.

And this is more than likely what's actually going on. Particularly as can't say I noticed any particular difference going from the loaner DAC to my own new copy. Nor can I say I've observed any obvious change in the performance of my unit beyond the initial warm-up period.

 

I've been looking at/listening to various DACS for ages now, trying to sort out what is what, and what I'm looking for sound wise. I was/am really interested by the Aqua Acoustics models, but even the La Voce (their "entry" level offering) will be 3 grand USD + import fees. The spring DAC would be 1700 USD if I went for the base model. Might as well save some money, especially if the level 1 sounds good.

I know this thread is about the Level 3 model, but if anyone has experience with the Level 1 model and could chime in with how it sounds, especially in comparison to the Level 3, that would be awesome.

 

@ufospls2 I cannot offer you any advice on the Holo Level 1, but I do want to comment on the AQUA DAC. I am going to re-post a lengthy impression in the La Scala thread so I don't sully up this Holo Spring thread. You can see the post here.