HOLO Audio Spring DAC - R2R DSD512

[pkane2001]

3 hours ago, Miska said:

 

Few spurious tones you see in direct USB plot wouldn't be even visible in SU-1 plot, because they would be buried in the noise floor. But those are not jitter. For jitter, can you zoom-in +-3 kHz around the main lobe and increase resolution?

 

What input format did you use?

 

Here are DSD512 figures for both.

 

Spring 1 USB:

 

Spring 2 USB:

 

I think my Spring 1 is with the older USB input module type.

 

 

Interesting. The Spring 1 measurement I posted was using PCM format. Looks like Spring is happier receiving DSD256. Here are the charts for comparison (lower traces are using Spring USB input, upper using SU-1). Not clear to me that SU-1 is helping at all here

 

 

 

[Miska]

1 hour ago, ferenc said:

Would it work with an iMac Pro or latest gen 15 lnch MacBook Pro? 

 

I'm not sure, I think Apple added the new security chip to some of the latest generation computers preventing booting up anything else than macOS and Windows. Because they have bootloader signing keys only for Apple and Microsoft. But other than that it probably works if it gets as far as loading the EFI bootloader.

 

[Miska]

1 hour ago, pkane2001 said:

 

Interesting. The Spring 1 measurement I posted was using PCM format. Looks like Spring is happier receiving DSD256. Here are the charts for comparison (lower traces are using Spring USB input, upper using SU-1). Not clear to me that SU-1 is helping at all here

 

 

 

 

Yes, that's typical for many DACs... I guess yours have the newer USB input module?

 

DSD256 gives lowest audio band noise floor. DSD512 gives flattest wideband noise floor (> 100 kHz). But I didn't have DSD256 results saved for Spring 1 so I posted DSD512 results instead.

 

But certainly at least I wouldn't complain about the internal USB interface... It is among the best ones I've seen.

 

[pkane2001]

5 minutes ago, Miska said:

 

Yes, that's typical for many DACs... I guess yours have the newer USB input module?

 

DSD256 gives lowest audio band noise floor. DSD512 gives flattest wideband noise floor (> 100 kHz). But I didn't have DSD256 results saved for Spring 1 so I posted DSD512 results instead.

 

But certainly at least I wouldn't complain about the internal USB interface... It is among the best ones I've seen.

 

 

How can I tell if my USB module is older or newer? My Spring is a couple of years old.

[Miska]

7 minutes ago, pkane2001 said:

How can I tell if my USB module is older or newer? My Spring is a couple of years old.

 

I'm not sure, @Bimmer100?

 

My Spring2 says it's "Gen2". While Spring1 says it's "xCORE USB Audio 2.0". I'm not sure if this part was changed at all in the middle.

 

[matthias]

25 minutes ago, pkane2001 said:

 

How can I tell if my USB module is older or newer? My Spring is a couple of years old.

 

The newer one should be this:

https://wildism-audio-hk.myshopify.com/blogs/news/holo-audio-xmos-xu208-spring-dsd-r2r-dac

 

Matt

[pkane2001]

40 minutes ago, matthias said:

 

The newer one should be this:

https://wildism-audio-hk.myshopify.com/blogs/news/holo-audio-xmos-xu208-spring-dsd-r2r-dac

 

Matt

 

Thanks, I think I may have the older one based on this, since I bought it in 2017. I also see the updated board available from Tim:

 

https://kitsunehifi.com/product/holoaudioxu208usbmodule/

 

But based on what I measured, I don't see the need to upgrade 

[ferenc]

I finally got my Spring 2 work at DSD512 from my iMac Pro and SMS200 Neo as NAA. It works only in native DSD did not work in DOP. The 18 cores of the iMac Pro needs to be used in multicore modes in HQP preferences, otherwise it is stuttering a lot. In this mode poly-sync works very well. This way the Spring 2 sounds quite a bit smoother, fuller that the SU-1 in DSD256 and in DOP. Actually like a 38 cm/s half inch master tape, it has a stability like the best of the tapes. 

 

As the iMac Pro has a lot of CPU power not fully used even for poly-sync for DSD512, I think the 18 cores would be enough to use for DSD1024, but it does not work with the the SMS200 yet, I tried and it was something very distorted and like a slow motion. Anyway I hope a firmware update will bring DSD1024 to SMS200 soon. Otherwise I need to find another NAA configuration.

[ferenc]

I wonder if using 1.536 kHz sampling with a Mac OS X 10.14.4.x only in 16 bit  as 24 bit gives a distorted sound, is a problem of Core Audio in the Mac OS X or a problem of the Holo firmware?

 

Would be nice to find it out. 

[ferenc]

7 hours ago, ferenc said:

I wonder if using 1.536 kHz sampling with a Mac OS X 10.14.4.x only in 16 bit  as 24 bit gives a distorted sound, is a problem of Core Audio in the Mac OS X or a problem of the Holo firmware?

 

Would be nice to find it out. 

 

I tried the 1.536 kHz 16 bit with HQplayer. It works. However, if I try higher than 16, 17 bit for example, I got the same distortion as with Core Audio, does not matter with what kind of algorithm and noise shaping. 

 

It can mean few things: 

 

- HQplayer is using the same Core Audio (or AV Foundation) built-in up conversion, I do not think so. 

- There is a general problem with the USB connection at this bitrate

- There is a problem with the Holo firmware or the hardware itself. 

[matthias]

41 minutes ago, ferenc said:

 

I tried the 1.536 kHz 16 bit with HQplayer. It works. However, if I try higher than 16, 17 bit for example, I got the same distortion as with Core Audio, does not matter with what kind of algorithm and noise shaping. 

 

It can mean few things: 

 

- HQplayer is using the same Core Audio (or AV Foundation) built-in up conversion, I do not think so. 

- There is a general problem with the USB connection at this bitrate

- There is a problem with the Holo firmware or the hardware itself. 

 

In my Spring2 thread Miska mentioned that the Spring is not able to settle to 24bit accuracy at 1,5MHz. If I recall correctly he said he would be surprised if the Spring can settle perfectly even at 16bit 1.5MHz.

 

Matt

 

[ferenc]

38 minutes ago, matthias said:

 

In my Spring2 thread Miska mentioned that the Spring is not able to settle to 24bit accuracy at 1,5MHz. If I recall correctly he said he would be surprised if the Spring can settle perfectly even at 16bit 1.5MHz.

 

Matt

 

 

It sounds perfectly ok in 16 bit 1.536 kHz. It could be used in 24 bit or higher only max at 768 kHz.

[matthias]

5 minutes ago, ferenc said:

 

It sounds perfectly ok in 16 bit 1.536 kHz. It could be use in 24 bit or higher only max at 768 kHz.

 

IMO, this makes perfect sense.

1,5MHz is too high to settle at more than 16bit.

The problem is the R2R architecture.

 

Matt

[Miska]

1 hour ago, ferenc said:

I tried the 1.536 kHz 16 bit with HQplayer. It works. However, if I try higher than 16, 17 bit for example, I got the same distortion as with Core Audio, does not matter with what kind of algorithm and noise shaping. 

 

It can mean few things: 

 

- HQplayer is using the same Core Audio (or AV Foundation) built-in up conversion, I do not think so. 

- There is a general problem with the USB connection at this bitrate

- There is a problem with the Holo firmware or the hardware itself. 

 

On macOS you won't be able to get more than 16-bit out at 1.5M, unless someone creates a custom driver. It won't work with the built-in driver, max is 768k with more than 16-bit. However, this is not really a problem!

 

If you use for example the bootable NAA image you can get up to 24-bit at 1.5M.

 

Generally, Spring is linear up to around 20-bit (which is very good performance for R2R). However, running 16-bit at 1.5M with NS5 or NS9 noise-shaper gets you same SNR as you would get with 20-bit! With NS5 up to 70 kHz and with NS9 up to about 100 kHz. At lower rates set bits to 20 and use TPDF/Gauss1 dither (or either noise shaper if you like).

 

[ferenc]

24 minutes ago, matthias said:

 

IMO, this makes perfect sense.

1,5MHz is too high to settle at more than 16bit.

The problem is the R2R architecture.

 

Matt

 

It does not work with 1.411 kHz either, only in 16 bit too. 

So if it is the case than it should not be advertised as 1.536 kHz compatible. 

 

Actually it does not matter a lot, I like the sound at 768 kHz or in DSD256, 512 too. 

[Miska]

Just now, ferenc said:

It does not work with 1.411 kHz either, only in 16 bit too. 

So if it is the case than it should not be advertised as 1.536 kHz compatible. 

 

Actually it does not matter a lot, I like the sound at 768 kHz or in DSD256, 512 too. 

 

Yes, by "1.5M" I mean both 1.4112M and 1.536M...

 

And it does work with more than 16-bit at those rates and of course DSD1024 too. Just not on macOS (without something like NAA to bypass macOS audio stack).

 

[ferenc]

50 minutes ago, Miska said:

 

Yes, by "1.5M" I mean both 1.4112M and 1.536M...

 

And it does work with more than 16-bit at those rates and of course DSD1024 too. Just not on macOS (without something like NAA to bypass macOS audio stack).

 

 

Unfortunately it does not work with SMS200 Neo and NAA either, but it is a very different distortion than than the MacOS distortion. 

[Miska]

1 hour ago, ferenc said:

Unfortunately it does not work with SMS200 Neo and NAA either, but it is a very different distortion than than the MacOS distortion. 

 

It should work fine with the bootable NAA image version 3.5.6.

 

[ferenc]

9 hours ago, Miska said:

 

It should work fine with the bootable NAA image version 3.5.6.

 

 

Do you have an NAA 3.6 ISO for the SMS200 as well? I see only x64 version of it on your site. 

[Miska]

40 minutes ago, ferenc said:

Do you have an NAA 3.6 ISO for the SMS200 as well? I see only x64 version of it on your site. 

 

No, only for x64... I try to periodically update the CuBox-i and RasPi3 images too, but now those are a bit old and don't have this support yet.

SOtM and Sonore maintain their own OS, although the CuBox-i image now works on microRendu too.

 

[ferenc]

2 minutes ago, Miska said:

 

No, only for x64... I try to periodically update the CuBox-i and RasPi3 images too, but now those are a bit old and don't have this support yet.

SOtM and Sonore maintain their own OS, although the CuBox-i image now works on microRendu too.

 

 

Thanks as I thought. 

[tmitchmd]

I have a large collection of SACDs, DVD-As, and Blu-Ray Audio discs but not all that many CDs. Not much on hard disc, I'm afraid.

 

 I purchased the Oppomod 203 from Jaehong with the added i2s output.
The Oppomod 203 works perfectly with the Spring 2 over i2s. I am also streaming Qobuz over ethernet into the 203 which provides the rendering and then outputs the Qobuz bits to the Dac over i2s. No USB connections are used.
I believe it is this combination that allows both transport and  dac to be used to full potential. I have yet to try HQ Player as I am Chromebook based rather than PC.

 

I encouraged Jaehong to pick up as many new Oppo 203s as he could find at the old price and I believe he may still have a few left in stock but all will surely be gone soon.

[kaka]

Is anything known now about the new Holo Audio dac due mid-year?

[drjimwillie]

Has anyone tried or read about the new Singxer SU-6 with their Holo Spring?

[Account Closed]

There is new firmware for the Spring/Spring 2. Please check the Kitsune website.