HOLO Audio Spring DAC - R2R DSD512

[guymrob]

Thanks for the reply. I was asking about the analogue filter. If it was designed as a NOS DAC normally it would be a filterless DAC or have a relatively sharp filter. From your reply it seems to have an analogue filter with a gentle slope (probably designed for upsampling in the DAC or in the computer), that works well with high res files or upsampling in the computer. This brings another issue in my case, 176.4, 192, 352.8, 384 would have to go through USB, and if the USB implementation in the DAC isn't good enough that is another can of worms.

 

@axel_69, may I know what issue you encountered using the built-in USB input? The built-in USB (with XMOS U8) board is made by Holo Spring (Jeff Zhu), you can see the name and the Holo Spring logo printed on the USB PCB board. This is NOT an off selves or OEM USB input board where you can just find anywhere else.

 

Jeff Zhu spent a lot of his effort optimising the USB driver (Thesycon and Windows signed driver!) plus latest firmware so that it can work reliably on Linux and Windows.

[axle_69]

I didn't say I had. Trying to make a decision between the Spring and another DAC. It seems some people feel the need to put something like the Singxer SU-1 DDC in the middle. In the case of the other DAC it would be computer without upsampling > Toslink up to 96 > upsampling in the other DAC, whereas in the case of the Spring it would have to be upsampling in the computer > USB > ? > Spring. The Spring may take the lead on the DAC itself, the other DAC will take the lead on the upsampling algorithm (probably even compared with upsampling in the computer), but what happens in the middle is important as well. I could upsample 44.1 to 88.2 and use Toslink with the Spring but my guess is that the analogue low pass has a gentle slope and a higher sampling rate will be better. So it ends up being a question about upsampling algorithms and digital connections as well. How does the USB input (with output directly from the computer) compares with the others?

[Middy]

http://www.superbestaudiofriends.org/index.php?threads/holo-audio-spring-dac-level-3-kitsune-tuned-edition-impressions-reviews.3172/

 

Been enjoying reading about this Dac.

 

This is a nice comparison with the Yaggy if you want to read it?

 

Good luck

Dave

 

Sent from my SM-G900F using Computer Audiophile mobile app

[guymrob]

I didn't say I had. Trying to make a decision between the Spring and another DAC. It seems some people feel the need to put something like the Singxer SU-1 DDC in the middle. In the case of the other DAC it would be computer without upsampling > Toslink up to 96 > upsampling in the other DAC, whereas in the case of the Spring it would have to be upsampling in the computer > USB > ? > Spring. The Spring may take the lead on the DAC itself, the other DAC will take the lead on the upsampling algorithm (probably even compared with upsampling in the computer), but what happens in the middle is important as well. I could upsample 44.1 to 88.2 and use Toslink with the Spring but my guess is that the analogue low pass has a gentle slope and a higher sampling rate will be better. So it ends up being a question about upsampling algorithms and digital connections as well. How does the USB input (with output directly from the computer) compares with the others?

 

Thanks for your reply, first let me put things in a simpler way. There are people who use Singxer SU-1 reported some improvement in sound quality, for me I did hear a minor different; I feel the background is a little 'darker' but that doesn't mean the built-in USB is inferior. In fact, the difference is so small it doesn't warrant me to upgrade to Singxer SU-1 at all!

 

Now, I don't just listen to USB input but I also have a CD player that hooked up to the coaxial digital input too. When I compared 44.1k from the CD and one from a PC, I can't tell the difference between both of them!

 

Regarding on the different type of over sampling and digital filter, whether it is hardware or software, the hardware over sampling and digital filter will never sound the same from different manufacturers; because either they use a different type of chip or the over sampling and digital filter have a different algorithm. When you buy them, you have to like the signature of the sound!

 

For software up sampling and digital filters, all these can be customised and tailored to your liking, the flexibility of software is simply out of this world.

 

In order to do that, the DAC must have a provision to bypass the hardware over sampling and digital filter, so software up sampling and digital filter will really shine. The Holo Spring give user a lot of flexibility not found in most high end DACs. By doing that, people who are 'purists' who just want to listen directly without any form of over sampling can do so (better impulse response) or can use the hardware or software up sampling to suit their likings.

[GUTB]

http://www.superbestaudiofriends.org/index.php?threads/holo-audio-spring-dac-level-3-kitsune-tuned-edition-impressions-reviews.3172/

 

Been enjoying reading about this Dac.

 

This is a nice comparison with the Yaggy if you want to read it[emoji4]

 

Good luck

Dave

 

Sent from my SM-G900F using Computer Audiophile mobile app

 

SBAF is a cult / insular ego bottom feeder site. The general audio knowledge of CA is significantly higher. Marvey, the owner of the site, also reviewed the Spring and found it to be slightly inferior....to the Gunjir Multibit.

 

 

Sent from my iPad using Computer Audiophile

[barrows]

Thanks @ted_b, what I like is I can simply bypass the built-in over-sampling and digital filter and use a software base like HQP or the latest Roon 1.3 with up-sampling DSP to tailor the sound I like.

 

Most DACs in the market has little provision to allow the user to bypass the hardware over sampling and digital filter inside the DACs (especially those use off selves DAC chips) so even having a good software base up sampling like HQP, one is already severely limited by the built-in hardware over sampling and digital filter. This is main reason I brought Holo Spring DAC and I simply love the sound of this DAC!

 

This is incorrect. For instance, "most DACs" use an 8x first stage filter in their oversampling engine. If the DAC in question can receive a 352.8 sample rate (which is 8x red book, 44.1) and if you oversample in software to 352.8 and send that to the DAC, you will be bypassing the first stage digital filter. Additionally, most audible "problems" of a poor oversampling algorithm occur during the first pass of the oversampling process (any artifacts of the next stage(s) of oversampling will be entirely out of the audible bandwidth). You do not need a DAC which can be set to a NOS mode to enjoy the benefits of software based oversampling.

[Middy]

Tiz why I said If you want to read it?

No offence to anyone anywhere. I just try to glean as pertinent informative information I can find, in this post modern western we call the internet.

CA is a nice place... but I thought that write up was balanced and might help some here.

Kind regards

Dave

 

Sent from my SM-G900F using Computer Audiophile mobile app

[guymrob]

This is incorrect. For instance, "most DACs" use an 8x first stage filter in their oversampling engine. If the DAC in question can receive a 352.8 sample rate (which is 8x red book, 44.1) and if you oversample in software to 352.8 and send that to the DAC, you will be bypassing the first stage digital filter. Additionally, most audible "problems" of a poor oversampling algorithm occur during the first pass of the oversampling process (any artifacts of the next stage(s) of oversampling will be entirely out of the audible bandwidth). You do not need a DAC which can be set to a NOS mode to enjoy the benefits of software based oversampling.

 

Most of the off selves DACs have 8x oversampling filters that are consisted of three 2x filters.

 

When you input data at 44.1/48 (single rate), you get all the three filters. When you input data at 88.2/96 (double rate), you get two filters. And when you input data at 176.4/192 you get only one. So, in all cases, final rate is capped at 352.8/384. So you are right to say if you already up sample in the software to 352.8/384, the DAC will auto select to no over sampling since it is already capped in the final stage. However, there are custom design DAC that uses FPGA have the processing power to over sample at higher rate, like 705/768 or even higher in the MHz range even when input sample is 352.8/384, therefore there is still over sampling going on. To add it further, some even convert PCM to DSD internally! It all depends on the processing power given by chip.

 

By using a NOS DAC, you can be 100% sure that there is no hardware over sampling and digital filter come into play, just a gentle LPF in the analog stage. Over sampling and digital filter is then done on the software level, like HQP, giving users more flexibility and better quality.

[Mogos]

For me something like SU-1 is a must for splitting the digital signal. I am using two, no even three DACs to play the music. Every DAC has a different chracter and strong points. They are respectively connected to amplifiers one tube one and one SS one. The third DAC is controling an active subwoofer. I want to exchenge one of the DACs with HOLO Spring.

[barrows]

Most of the off selves DACs have 8x oversampling filters that are consisted of three 2x filters.

 

When you input data at 44.1/48 (single rate), you get all the three filters. When you input data at 88.2/96 (double rate), you get two filters. And when you input data at 176.4/192 you get only one. So, in all cases, final rate is capped at 352.8/384. So you are right to say if you already up sample in the software to 352.8/384, the DAC will auto select to no over sampling since it is already capped in the final stage. However, there are custom design DAC that uses FPGA have the processing power to over sample at higher rate, like 705/768 or even higher in the MHz range even when input sample is 352.8/384, therefore there is still over sampling going on. To add it further, some even convert PCM to DSD internally! It all depends on the processing power given by chip.

 

By using a NOS DAC, you can be 100% sure that there is no hardware over sampling and digital filter come into play, just a gentle LPF in the analog stage. Over sampling and digital filter is then done on the software level, like HQP, giving users more flexibility and better quality.

 

As I mentioned, yes, often there is still more OS beyond 352.8/384. But, going from 352.8/384 to (whatever, often MHz rage) will generally not be audible, as any artifacts produced by this OS step are so far beyond the audible range as to not matter, where the first OS step, from (for example) 44.1 to 352.8 is very audible. So, my point stands that one not need an actual NOS DAC (or setting in their DAC) to experience the difference of software based OS.

[guymrob]

As I mentioned, yes, often there is still more OS beyond 352.8/384. But, going from 352.8/384 to (whatever, often MHz rage) will generally not be audible, as any artifacts produced by this OS step are so far beyond the audible range as to not matter, where the first OS step, from (for example) 44.1 to 352.8 is very audible. So, my point stands that one not need an actual NOS DAC (or setting in their DAC) to experience the difference of software based OS.

 

If 44.1 to 352.8 is very audible in hardware over sampling inside in a DAC then obviously you want to remove this and apply external software based OS. After all, you don't need to have both hardware and software OS. And since people use HQP to do OS because it sound much better than those built-in hardware OS, obviously bypass it will make software OS really shine. Besides, bypassing it actually shortened the signal path...

[Miska]

But, going from 352.8/384 to (whatever, often MHz rage) will generally not be audible, as any artifacts produced by this OS step are so far beyond the audible range as to not matter

 

It depends on your analog filters. It can still have audible impact due to correlated intermodulation products as explained many times in other threads. Only if you have good and steep enough analog reconstruction filters that remove all the images at 352.8/384k and multiples, you don't have such worries. But so far most of the DAC's I've measured still leave quite a bit of images...

 

If you start from 44.1/24 with 8x oversampling, you need analog filter that rolls off to -144 dB by 330.75 kHz. IOW, you need about 36 dB/octave (6th order) analog filter. And then you need to forget hires, because it needs to begin to roll-off already at about 25 kHz. As a result, the phase response will look pretty bad too.

[barrows]

It depends on your analog filters. It can still have audible impact due to correlated intermodulation products as explained many times in other threads. Only if you have good and steep enough analog reconstruction filters that remove all the images at 352.8/384k and multiples, you don't have such worries. But so far most of the DAC's I've measured still leave quite a bit of images...

 

If you start from 44.1/24 with 8x oversampling, you need analog filter that rolls off to -144 dB by 330.75 kHz. IOW, you need about 36 dB/octave (6th order) analog filter. And then you need to forget hires, because it needs to begin to roll-off already at about 25 kHz. As a result, the phase response will look pretty bad too.

 

Yes, totally agreed. I was attempting to keep things a bit simple in speaking generally. Most commercial DACs (but certainly not all) are engineered with pretty strong analog side filters precisely to avoid out of band artifacts causing IM distortions which could be audible.

what is additionally interesting, and relevant to the NOS discussion, is the variation in DAC designs which are called NOS. some NOS DACs do not have much (if at all) for analog stage filters, and some have very complex analog filters in an attempt to reduce the inherent problems of NOS (like Zanden).

[Miska]

I was attempting to keep things a bit simple in speaking generally. Most commercial DACs (but certainly not all) are engineered with pretty strong analog side filters precisely to avoid out of band artifacts causing IM distortions which could be audible.

 

My experience is that there is typically 2nd order analog filter and in some cases 3rd order. But I have not seen anything higher. That is also apparent from the measurement results.

 

For hires that spec I said would need to be 8th order analog filter, that would allow pushing the corner up. IOW, four op-amps per channel, not counting I/V and possible output buffers.

 

what is additionally interesting, and relevant to the NOS discussion, is the variation in DAC designs which are called NOS. some NOS DACs do not have much (if at all) for analog stage filters, and some have very complex analog filters in an attempt to reduce the inherent problems of NOS (like Zanden).

 

Yes, quite many seem to have something like 1st order passive filter or no filter at all (filterless NOS).

[Miska]

What about switching between PCM to DSD or vice-versa?

 

It is the same always, 50 ms mute for first and last of DSD...

 

When switching occurs there's brief period when the OS (Linux or Windows) or and playback program may generates all '0' data

 

That shouldn't happen ever, but for example ESS Sabre will mute output when it detects same byte repeating.

[Altabay]

@Bimmer100, do you have any good news for us on the native DSD loud pop problem? I think you mentioned somewhere Jeff Zhu was looking at implementing a fix/workaround in the DAC. Any update? Can end users apply such an update or will it require sending the DAC to you or Jeff?

 

It's been said this problem doesn't happen with the Windows driver. Is it because the driver contains a workaround? Can it be added to the Linux driver to provide relief for the microRendu users?

[guymrob]

It depends on your analog filters. It can still have audible impact due to correlated intermodulation products as explained many times in other threads. Only if you have good and steep enough analog reconstruction filters that remove all the images at 352.8/384k and multiples, you don't have such worries. But so far most of the DAC's I've measured still leave quite a bit of images...

 

If you start from 44.1/24 with 8x oversampling, you need analog filter that rolls off to -144 dB by 330.75 kHz. IOW, you need about 36 dB/octave (6th order) analog filter. And then you need to forget hires, because it needs to begin to roll-off already at about 25 kHz. As a result, the phase response will look pretty bad too.

 

Correct me if I'm wrong, my understanding is hires recording at 352.8/384k (DXD) begins to roll of before half the Nyquist frequency, in this case 176.4/192k. I believe modern digital filter will able to take care the roll off beginning at 176.4/192K all the way down to 352.8/384k, but I'm not sure what is the attenuation factors these digital filters has, how many dB/octave? I'm sure nobody wants want to build a analogue filter to do that, as you mentioned above.

 

If over sampling digital filter is going to solve the 352.8/384k (DXD) while maintaining the hires content, then the final output stream is going to be 2.822/3072MHz, 8x OS of 352.8/384k. Not all modern DACs are designed to accept such a high data rate, especially the off selves DACs due to power consumption and heat issues.

[Miska]

Correct me if I'm wrong, my understanding is hires recording at 352.8/384k (DXD) begins to roll of before half the Nyquist frequency, in this case 176.4/192k. I believe modern digital filter will able to take care the roll off beginning at 176.4/192K all the way down to 352.8/384k, but I'm not sure what is the attenuation factors these digital filters has, how many dB/octave? I'm sure nobody wants want to build a analogue filter to do that, as you mentioned above.

 

For hires recordings you need about 100 kHz bandwidth, so the analog filter -3 dB point is usually placed at 100 kHz. Thus you have the band from 100 kHz to 252.8 kHz to roll-off to -144 dB. Thus you have octave and half, so you need analog filter of about 100 dB/oct to reach perfect reconstruction. That is 17th order filter.

[guymrob]

Can you tell me what is different between digital filter in the over sampling block vs the analogue filter right after being converted to analogue? I have an impression that over sampling digital filter has a much greater than -3dB roll off/octave, depending whether is a steep or gentle roll off. Thanks for valuable inputs.

[guymrob]

@Bimmer100, do you have any good news for us on the native DSD loud pop problem? I think you mentioned somewhere Jeff Zhu was looking at implementing a fix/workaround in the DAC. Any update? Can end users apply such an update or will it require sending the DAC to you or Jeff?

 

It's been said this problem doesn't happen with the Windows driver. Is it because the driver contains a workaround? Can it be added to the Linux driver to provide relief for the microRendu users?

 

I've contacted Jeff Zhu on my Auralic Aries Mini(Linux based OS) containing a loud pop when Aries Mini transport DSD256 natively (default), while DSD64/DSD128 is transported via DoP mode by Aries (default) so I don't experience the loud pop. He is looking at it. I also contacted Auralic, they too is investigating on the issue. It can be Holo Spring or the Auralic but I've to wait to get further reply from them.

[Miska]

Can you tell me what is different between digital filter in the over sampling block vs the analogue filter right after being converted to analogue? I have an impression that over sampling digital filter has a much greater than -3dB roll off/octave, depending whether is a steep or gentle roll off. Thanks for valuable inputs.

 

Yes, the digital filter can have as steep roll-off as you want. However, you always need an analog filter to remove image frequencies around multiples of the sampling rate. Oversampling digital filters are used to move those image frequencies further away from audio band, so that they can be removed with simpler (lower order) analog filter.

 

If we start with a 0 - 22.05 kHz sweep at 44.1 kHz sampling rate, then oversample it using steep digital filter to 352.8 kHz, this is how the R-2R DAC's output theoretically looks like at 1.4 MHz bandwidth:

 

You wouldn't need analog filter if you'd have infinite output sampling rate.

[guymrob]

The loud pop sounds that happens in Linux OS when playing back DSD in native mode as opposed to DoP is caused by the 'digital silence' at the starting of the track. I believed we have found culprit...

 

[PATCH] ALSA: usb-audio: Eliminate noise at the start of DSD playback. — ALSA Devel

[guymrob]

There's been some discussion at Roon on the nasty pop when playing back DSD...

https://community.roonlabs.com/t/loud-pop-noise-beginning-new-track-dsd-native-v1-3-200-alsa-fix-pending/20713

[lmitche]

The loud pop sounds that happens in Linux OS when playing back DSD in native mode as opposed to DoP is caused by the 'digital silence' at the starting of the track. I believed we have found culprit...

 

[PATCH] ALSA: usb-audio: Eliminate noise at the start of DSD playback. — ALSA Devel

Thanks for sharing this. This is great news! We can finally dump Windows and create a truly great DRM free OS environment for Roon/HQplayer!

[Jud]

The loud pop sounds that happens in Linux OS when playing back DSD in native mode as opposed to DoP is caused by the 'digital silence' at the starting of the track. I believed we have found culprit...

 

[PATCH] ALSA: usb-audio: Eliminate noise at the start of DSD playback. — ALSA Devel

 

Nice!

 

 

Sent from my iPhone using Computer Audiophile