HOLO Audio Spring DAC - R2R DSD512

[pkane2001]

23 hours ago, chsas75 said:

 Then USB module is with xu208 xmos. 

It bring question; how to implement xu208 to existing Holo Spring DAC?

 

Use an external DDC that implements xu208 with I2S connection.

 

[pkane2001]

1 minute ago, chsas75 said:

 

Ive used this crutch with Sotm sms-200, now im changing my source and non usb solutions greatly decrease range and facilities fora reasonable price($1-2k). No USB no Roon

 

Something like Singxer SU-1 DDC takes USB input, uses XU208 internally with excellent clocks and a good galvanic isolation and outputs via I2S connection to Holo Spring. 

[pkane2001]

12 hours ago, chsas75 said:

 

I used their young brother F-1, SU-1 better over i2s but their power scheme is not fine and +1 cable. Realy source + regen + converter + power for all too long usb train.

Soundaware, Shinrico and Cayin can escape it but they dont know about Roon@Tidal

 

I don't use any Regen products, USB cable goes directly into SU-1, from there a short, inexpensive HDMI cable (0.3m) into I2S input on the DAC. Yes, it's one more box and one more HDMI cord. This works well and sounds good to me. Power supply on the SU-1 can be replaced easily  with an external 5VDC one, but the internal LPS is good and can be upgraded with better, low noise voltage regulators. 

 

[pkane2001]

1 minute ago, GUTB said:

Am I doing something wrong, or is this a case of people who've never heard high end audio?

 

You’re doing something wrong.

 

[pkane2001]

Try HQPlayer for volume control. Set X20U volume to 0dB.

 

My x20pro sounds amazing with no preamp. So does Holo Spring. I also use both without a headphone amp with HD650s in balanced mode — also great sound.

 

Just be careful about the HQP volume control, it’s easy to set it to maximum without intending to.

 

[pkane2001]

18 hours ago, Miska said:

Yeah, if someone uses HQPlayer (or A+ or something else) to upsample to 352.8/384k and/or DSD, then NOS mode is what should be used.

 

Jussi, so what's the reason to use HQPlayer upsampling/filtering and then Holo Spring in NOS mode? Is it that HQPlayer has a better implementation? Higher precision? More filter options? What's wrong with Holo Spring filters or upscaling algorithm? 

 

I have mine configured as you recommend: HQPlayer converting all content to DSD512, using xtr-mp-2s filters,  HoloSpring in NOS mode.

[pkane2001]

57 minutes ago, ted_b said:

Yes, HQPlayer's filters and modulators are 10x better than most dacs, let alone Holo's.  One of Holo's biggest strengths is being able to bypass their filters...i.e NOS mode.  Many dacs do not have that setting.

 

10x better in what sense? I understand the theoretical benefit, but what's the practical one? I certainly don't hear a 10x better sound quality in NOS mode 

[pkane2001]

3 minutes ago, ted_b said:

Pkane2001, I think software upsampling vs hardware upsampling has been discussed ad nauseum.   I find it to be the same for a lot of "integrated vs separates" logic; the "separates" (in this case software upsampling using powerful cpus) are built for the task, the "integrateds" have to be jack of all trades, given a price point.  In some cases, the integrateds have no bypass or coexist capability..in the case of dacs like Holo they do.   Oversimplified, sorry.

 

Thanks, Ted, I get the general appeal of the software-based approach. I was curious about the details as to what @Miska found to be inferior with Holo Springs filters compared to HQPlayer.

 

[pkane2001]

37 minutes ago, Miska said:

Here's wideband output spectrum 0 - 22.05 kHz sweep with 44.1/32 source.

 

Upsampled to 352.8k PCM using built-in chip:

 

Upsampled to 352.8k PCM using HQPlayer:

 

Upsampled to DSD using built-in chip:

 

Upsampled to DSD512 using HQPlayer:

 

 

You can see that with PCM there are about three images still visible around multiples of the 352.8k sampling rate, not completely eliminated by the analog filter (completely typical amount, no surprises).

 

With DSD512 from HQPlayer you can see that there are no images or other extra visible from the digital source, just analog noises. There are no images, because HQPlayer runs digital filter up to 22.5792 MHz rate (real 512x filter), and multiples of that rate are completely eliminated by the analog filter.

 

 

Someone also asked earlier about the Jtest-24 results for the DAC's built in USB interface:

Cannot really get much better than that easily...

 

 

That is really excellent performance, Jussi. Thanks for sharing these!

 

[pkane2001]

3 minutes ago, Miska said:

 

I guess it is mostly like Spring DAC in DSD mode (where you also have option for built-in DSD OS), but without the PCM side.

 

A PCM version of Cyan will also be available, as I understand it.

[pkane2001]

5 hours ago, Miska said:

So far I've been using only DSD side of the Spring (apart from measuring the PCM side too). While I bet there are many people who have been doing pretty much the opposite.

 

I've played with switching Spring to PCM mode for a while, and found that I still prefer everything upsampled to DSD512, so that's how I'm using it. NOS mode with HQPlayer converting/upsampling everything to DSD512 and acting as the volume control. No preamp for headphones or speakers. 

 

I'm also curious to see your measurements of Spring Holo, PCM and DSD if you can share them.

 

[pkane2001]

1 hour ago, GUTB said:

 

The I2S interface was only useful for bypassing the SRC stage anyway. Since there’s no SRC in the DSD version than USB is fine.

 

That's not the reason for I2S. It's to allow for external DDC or other I2S sources to connect to Holo with minimum overhead, bypassing other built-in interfaces such as USB, SPDIF, etc. Holo USB is fine, but that doesn't mean everyone wants to use it.

 

When NOS mode is selected, all upsampling/resampling is bypassed, so no SRC, regardless of what input you use. 

[pkane2001]

17 minutes ago, Superdad said:

I own both an SU-1 and Spring L3.  The primary benefit to using the I2S input is that it supplants the Spring's dirt-cheap [audio sample] clocks with the Crystek CCHD-575 clocks of the SU-1.

Beyond that there is not a lot particularly better about the SU-1's USB input versus that of the Spring.  (Both USB inputs are bus-powered and benefit similarly from better than noisy, leakage-filled computer VBUS.)

 

There's also the additional galvanic isolation that SU-1 provides. And the fact that you can power it from an external 5v supply (I use a battery). Of course, a lower phase noise oscillator is also a benefit.

 

[pkane2001]

1 hour ago, Bimmer100 said:

FYI. I have posted our new drivers for the HoloAudio Spring Dac. Improved performance with dsd and pcm.

version v4.35 available for download on our website now.  Please let us know tour thoughts on this new driver!

 

 

 

Thanks, Tim!

 

Is this different than than the 4.36 driver some of us have been using on the SU-1 thread? That one appears to be an improvement over the Holo Spring driver we had to use previously for native DSD512 support.

 

 

[pkane2001]

11 hours ago, Bimmer100 said:

Go ahead and compare two new drivers and you likely will find the difference is the Version number only. Both are nicely improved for audible improvements.

Thanks, Tim! That is very helpful.

 

11 hours ago, Bimmer100 said:

Personally I don’t believe the singxer drivers were as “dramatic” or improvement  as some have said/commented. But believe there is a small improvement on my system. Who knows, maybe it is a dramatic improvement depending on source and hardware in the usb chain. 

 

Totally agree. I heard a difference, and spent some time switching between the drivers trying to ensure it wasn't all in my head. In the end, I did prefer the new Singxer driver to the old Holo Spring one, but only by a small margin. The new driver seems to manage timing and buffering better, as even with the tiniest buffer size I don't hear any dropouts or stutter. This happened consistently with the older driver when using some of the more aggressive settings.

 

If the new Holo driver has similar improvements, it is definitely worth switching to it, especially at the exorbitant price of $0  

[pkane2001]

4 minutes ago, LoryWiv said:

Question: I use Singxer Su-1 --> I2S to DAC. Will it work correctly to have the Singxer & HOLO drivers installed on my Win10 PC simultaneously and just use the driver selection dropdown in HQPlayer to switch back and forth for comparison? Or shall I uninstall the Singxer driver before trying out the new one from HOLO?

 

Thanks,

I couldn't get the old Holo and new Singxer driver to co-exist. I had to uninstall one, then re-install the other to compare them. Not the best way to do a comparison, but I was curious. Don't know if the two new drivers can coexist. Sounds like they might, according to Tim.

[pkane2001]

6 minutes ago, elan120 said:

I have just installed the new driver without un-install the Singxer driver, and was able to play the Holo driver, but not able to select the Singxer driver via HQPlayer drop down menu.

 

I think this is because SU-1 gets a driver started automatically, when it first connects over USB. HQPlayer can't switch which driver is talking to SU-1. It's probably the last installed driver, as that's the one Windows associates with SU-1 VID/PID. I also found that after reinstalling the driver, I had to sometimes switch SU-1 off and on again to make the new driver recognize it.

 

What I think Tim was talking about is using the Singxer driver talking to SU-1, while using new Holo driver talking to Holo Spring/Cyan. That would work, since these two devices have different VID/PID and can both be running at the same time.

[pkane2001]

1 minute ago, elan120 said:

This sounds like the case here...on the other hand, the hew Holo driver playing upsampled DSD512 seem to be quite good.  I will need to take some more time to see whether it is same or better than the Singxer driver.

 

Curious to hear if it's any better than the new Singxer driver... although I find this hard to imagine, since they are most likely based on the same XMOS driver template.

 

[pkane2001]

34 minutes ago, Bimmer100 said:

Depends on your gear. One of my setups I have a KTE spring and elac element amp to power elac uf5 slim towers. Another setup I use a iFi pro ican which actually can be used as a killer preamp  

 

we recommend you strongly consider XLR balanced with your amp to pre amp. Sound quality will be optimal and much better common mode noise rejection. But it’s not the end of the world if you have to use rca SE. however if you are in the market to buy new gear than choose balanced!  And I don’t want to start any argument with anyone about balance vs single ended.: spare me, spare us all.  My recommendations stand for xlr/balanced.

 

I'm not going to argue SE vs. balanced  as I use balanced from Holo Spring into my mono block, balanced amps. Works like a charm, and sounds better than with any of the preamps I've tried.

 

When driving amps directly from a DAC, you'll need to be careful about the software/PC volume control. I use HQPlayer volume control, but I had to disable all other sounds in the system, as these would bypass HQP and play at FULL VOLUME into the DAC and into my speakers. Not a pleasant experience when some notification pops-up on your PC at full blast! Since there is no hardware volume control in the system, it is very important to ensure that the software volume is always at a safe level.

 

 

 

[pkane2001]

4 minutes ago, Miska said:

 

Enable Safe Mode in the driver... Always make sure you don't have more than one player application running at a time. And if you use Roon as HQPlayer front-end, make sure you don't have the same DAC enabled as output zone in Roon...

 

Unrelated, but leave "Multicore DSP" in automatic mode (grayed), unless you specifically know that you need something else.

 

For Holo Audio, set "DAC Bits" to 28 or 24.

 

Jussi, does the 28 bit HQP setting make a difference compared to 24 with Holo Spring? I assume that the actual converter precision is around 20-21 bits (anyone measure this?)

[pkane2001]

1 hour ago, Miska said:

 

Yes, I think it's the same. Or even more so, because the Singxer SU-1 cannot even know how many effective bits the device behind it has... When the resolution cannot be determined automatically, output is dithered to the supported word length. Which for XMOS implementations is 32-bit; if the DAC then ends up truncating to 28- or 24-bit, the dither is lost. So for PCM outputs it is important to pay attention to this detail - either use the number of actual bits the DAC supports, or something less if the last LSBs are not linear anymore.

 

 

For what it's worth, I preferred the 23 bit setting in HQPlayer when feeding Holo Spring at 384KHz, via SU-1 and I2S connection. There was a very slight difference, but everything sounded just a tad smoother through my HD650s. I tried 28, 24, 23, 22, 21, 20, and 19 bit settings, switching frequently between them. 

 

Curious to hear what others prefer with this DAC (if anyone else can hear or measure a difference or if it's all in my head )

 

PS: this was tested on 24/192 content.

[pkane2001]

1 hour ago, Miska said:

I've been testing the Spring 2 L2 DAC with HQPlayer and with some interesting results.

 

If you wanna see some clean Jtest24 figures, here's one! Spring 2 running at DSD256, balanced output to the analyzer.

 

Source is my regular Xeon E5 workstation, running Ubuntu Studio 18.04 with it's normal graphical desktop etc.

 

IMD figure at DSD256 is also pretty amazing 0.00002%.

 

 

That looks great! Can you post one with a wider frequency range, say 20-20KHz?

 

[pkane2001]

2 hours ago, Miska said:

 

It is impossible to say why someone has some kind of subjective opinion.

 

I would first like to see how it changes things from technical perspective.

 

 

I see small differences in jitter and noise floor. Close-in jitter seems a bit lower with SU-1 in the chain, but the noise floor 10dB higher. Not sure why that is. These were captured using the same FFT settings, but in two separate sessions. I suspect they were both fed 32-bit data, and maybe SU-1 handles that differently than Spring. I'll have to repeat with 24-bit data at some point.

 

SU-1 feeding Spring 1 (level 1) via I2S cable:

 

 

USB directly into Spring 1:

 

[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

 

 

 

[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.