Review and Measurements of Holo Audio Cyan DSD DAC

[gvl]

I have this Cyan DSD and also both Spring 1 and Spring 2 DACs and running at DSD256 they sound about as neutral as the RME ADI-2 running at DSD256, with less characteristic sound than ESS Sabre which always seems to have it's own immediately recognizable characteristic flavor

This community doesn't believe in sound flavors. I thought I did but no longer sure I do, or at least not to the extent I originally thought.

 

[bennetng]

Too young, too simple, sometimes naive.

He was criticizing some audiophiles.

 

[Miska]

This community doesn't believe in sound flavors. I thought I did but no longer sure I do, or at least not to the extent I originally thought.

I never thought that as thing matter of belief.

My approach for selection has been to first perform sanity check by measurements and then do selection based on listening. Not much different from buying a car or motorcycle. One can check the specs, but when you start driving one you notice how it behaves and no specs can describe that. I don't select a car or bike based on power and torque curves either... Although I usually use software to improve my car too, on my previous car which was Mini I had SuperChips ECU tuning to bring quite a bit of extra power and torque (it improved fuel economy by 0.5 l/100km too). Now I need to look up a good tuning for my current Volvo too. Yeah, both measurable and noticeable difference as well.

 

[pkane]

I never thought that as thing matter of belief.

It is a matter of belief that all things matter for audio. You spend a lot of time fixing and making things appear better in measurements. I'm all for that. I'm an engineer and I also use your software (mostly for DSP). But, as far as I can see it is only a matter of belief that these things really need to be fixed. A lot more objective evidence is needed to demonstrate that there are audible effects to a -100dB spuria at 1MHz that need to be corrected.

And I've certainly mucked around with reprogramming ECUs on my cars. But I tell you, a carwash and new coat of wax increases torque and HP by at least 10%!

 

[gvl]

I never thought that as thing matter of belief.

My approach for selection has been to first perform sanity check by measurements and then do selection based on listening. Not much different from buying a car or motorcycle.

Do you have any theories based on the data you collected as to why ESS chips have that characteristic sound that's attributed to them?

 

[Miska]

You spend a lot of time fixing and making things appear better in measurements.

That is part of the picture. But OTOH, I have spent even more time thinking and listening different filter designs which all provide very good measured performance. Same goes for modulator designs too. Talking about objective numbers, especially on forums like this, is much easier than talking about listening experiences.

And once again, I want to combine this with other important things like digital room correction and headphone processing, etc. It all goes along nicely in the same DSP engine. Once you run one of these, you can as well have the other.

A lot more objective evidence is needed to demonstrate that there are audible effects to a -100dB spuria at 1MHz that need to be corrected.

I've been mostly talking about correlated images around 352.8/384k at level around -60 dB. It is not about just broadly spuria, it is specifically about correlated spuria.

Given that there are audible differences between different scenarios, and certain measured differences as well between the same scenarios. Given all the options I have some picture how different things correlate. There are of course many things one can measure.

As I earlier said, there's some amount of gap between "I can measure things you can't hear" and "you can't measure what I can hear" camps. And I firmly stand right in the middle. Sometimes feel like being ripped apart.

 

[Miska]

Do you have any theories based on the data you collected as to why ESS chips have that characteristic sound that's attributed to them?

Not very good one. But it to some extent seems to be about modulator design, but not entirely. To other extent about how the DSP + conversion stage behaves together. It is not all that different from what ESS has said either.

 

[amirm]

Sometimes feel like being ripped apart.

Ex CTO of Intel once told me: when you stand in the middle, you get hit by both sides. That is what was happening in the last forum I started so decided this one would not be in the middle. That said, I also don't want us to be too orthodox.

BTW, took me a bit to realize you are the author of HQPlayer. Do you mine putting that in your signature for the benefit of others? It is good to know where someone comes from. HQPlayer shows a lot of knowledge on signal processing side and that is a good thing to identify to membership.

 

[Miska]

BTW, took me a bit to realize you are the author of HQPlayer. Do you mine putting that in your signature for the benefit of others? It is good to know where someone comes from. HQPlayer shows a lot of knowledge on signal processing side and that is a good thing to identify to membership.

Isn't it showing there already? I put it there almost immediately after creating the account! I have no intentions being any kind of under-cover person!

Thanks!

 

[amirm]

Isn't it showing there already?

I see it now but didn't think it was visible before. Either that, or I am going blind.

 

[vinh]

Because you get around some of the limitations of DAC chips and get to choose your own modulator as well. Since in modern DAC chips and also in this case (all are SDM DACs) the sound largely depends, in addition to the conversion stage itself, on the digital delta-sigma modulator.

Sure, you can do it with some DAC chips as well, like the more recent AKM ones that have DSD Direct mode, and with TI/BB chips fed with DSD.

I have this Cyan DSD and also both Spring 1 and Spring 2 DACs and running at DSD256 they sound about as neutral as the RME ADI-2 running at DSD256, with less characteristic sound than ESS Sabre which always seems to have it's own immediately recognizable characteristic flavor (and doesn't support direct DSD). The on-board AKM chip to convert PCM to DSD has surprisingly poor performance though, I would have kind of expected more from AKM.

Given selling price of these, like Holo and Chord, surprisingly they are not really more expensive than comparable chip-based designs either.

Hi,
I am deciding between Cyan DSD and Spring 2 Level 1. Since you having both cyan dsd and spring 2 . Can i have the review comparing between them ?
Beside, The Cyan DSD now has a new design as they has removed the amp socket. I wonder if they improve/fix some of the performance.

Warm regard,

 

[mikemaverick]

Hi,
I am deciding between Cyan DSD and Spring 2 Level 1. Since you having both cyan dsd and spring 2 . Can i have the review comparing between them ?
Beside, The Cyan DSD now has a new design as they has removed the amp socket. I wonder if they improve/fix some of the performance.

Warm regard,

I would suggest Cyan DSD if you only listen to DSD musics since its DSD performance is pretty close to Spring 2. However, if you have more budget I would suggest that you wait for Holo May as it will perform much better than Cyan and Spring.

 

[vinh]

I would suggest Cyan DSD if you only listen to DSD musics since its DSD performance is pretty close to Spring 2. However, if you have more budget I would suggest that you wait for Holo May as it will perform much better than Cyan and Spring.

I actually listen to PCM most of the time, but PCM come with the amp as well. I only need a DAC. I thought Cyan DSD is closer to Spring 1 but not Spring 2 ?

 

[Veri]

I would suggest Cyan DSD if you only listen to DSD musics since its DSD performance is pretty close to Spring 2. However, if you have more budget I would suggest that you wait for Holo May as it will perform much better than Cyan and Spring.

Much better? The manufacturer said it will be a very smalll update for those that want to pay for the 'best' so shouldn't be a big difference

 

[donniewn]

did you guys listen to this DAC? measuements don't give the whole picture . does it sound good? I want a good native dsd dac

 

[Veri]

did you guys listen to this DAC? measuements don't give the whole picture . does it sound good? I want a good native dsd dac

Schiit used to make a Loki DSD-native DAC. It never took off so it was quickly scrapped. If a DSD-based DAC is what you're looking for, this Cyan is what you want. It'll sound good, no measurements to prove otherwise. It is expensive for what it is, though

 

[donniewn]

I have the Loki , it’s reasonably good , just looking for an improvement. I’m sure the power supply on the Cyan is much better than the USB of the Loki .

 

[Veri]

I have the Loki , it’s reasonably good , just looking for an improvement. I’m sure the power supply on the Cyan is much better than the USB of the Loki .

Wow you must be one of the rare owners of the Loki DSD, that's actually really cool. The Cyan sure looks like a solid performer to me. If you can afford it, I wouldn't think twice.

 

[pcasr]

ESS has a lot of theories why their dac has that characteristic sound that's attributed to them. And most importantly are resarching for them, which is part of science usually.

 

[Billy Budapest]

@amirm

Amir,

Any chance you would be able to review the latest iteration of the Holo Audio May DAC? Would the manufacturer send you a sample to measure? I think it is conceptually one of the most interesting DACs out there and according to some other reviewers, measures very well. I think its noise floor performance was the best Stereophile ever measured a few years ago—22 bits of dynamic range.