Holo Audio Red Streamer
Dec 5, 2022 at 7:24 AM Post #166 of 1,916
@Zaurux is your Allo USBridge Sig using stock CM3+ or modified and use CM4?

The Allo USBridge Sig running CM4 sounds better than the stock CM3+ in my system.

I think the Holo Red will be better than the USBridge Sig because it separates the Ethernet interface from the USB interface. The USBridge Sig also uses separate Ethernet chip (Ethernet to USB2) but it shares the same USB interface to either the orginal CM3+ or the modified CM4.
 
Dec 5, 2022 at 7:27 AM Post #167 of 1,916
About the clock.
Am too stupid to follow the technical details but didn't GoldenOne mention that for USB connection only the DAC clock does matter anyway?
 
Dec 5, 2022 at 7:50 AM Post #168 of 1,916
About the clock.
Am too stupid to follow the technical details but didn't GoldenOne mention that for USB connection only the DAC clock does matter anyway?
Technically, yes, it does. However, many companies and users still believe the better upstream USB clock will help to improve further the best downstream. I don't 100% believe this either. But like that, then the stock Pi 4 will technically work just fine with a good USB DAC like the Spring or May that have very good clocks and be galvanic isolated from the upstream USB device.
 
Dec 5, 2022 at 8:15 AM Post #169 of 1,916
...the Spring or May that have very good clocks and be galvanic isolated from the upstream USB device.
And I don't know about Holo whose new USB interface is said to be of high quality but for the Denafrips and my Musician, from memory, there is a circuit that detects the 5v to establish the USB link before cutting and switching to its own 5v supply.
 
Dec 5, 2022 at 9:35 AM Post #170 of 1,916
What's the stock OS btw?
It's Holo's own one. I'm not sure what it's based on (I assume it's not made from scratch)
Have you also tried the Farad powered Mano MK3?
I've not unfortuantely
How does it compare to the Sonore Optical Rendu?
Sonore Optical Rendu is a USB only device, no synchronous outputs so jitter isn't a factor. As to noise, I'm not sure as I've not had one in to test.
A better device to compare the Optical Rendu to would probably be the Intona 7055-C as they are much more similar in use case
And are you buying one for yourself? Or are you going to use something else?
I will most likely ask to purchase this unit yeah
The Red can't be used as both, a streamer & DDC...it's either or only?
Either. You can connect it via USB to use as a DDC, or connect via ethernet to use as a streamer
Better than the Pi2Design Mercury V2?
It's better than the Pi2Design Mercury v1, I don't have a V2 so can't compare there (though afaik the main digital circuits didn't change between V1 and V2).
But also a fair bit more expensive so not necessarily a direct competitor. Mercury is still fantastic for close to half the price of a RED
As I mentioned before, what is the comparison of Red Hollo vs Zen Stream (pure USB) ?
Both of them have ridiculously low noise USB outputs. If this level of noise is limiting a DAC in any way then the DAC itself is beyond incompetently designed.
The main differentiator for me is features. Zen stream has a 384khz sample rate limit which for HQPlayer users is a problem.
If you don't use HQP/upsampling though, then either will likely provide a near identical result if just using USB. They're both excellent. (And Zen stream is cheaper so it'd make sense to go for that)
For the moment, iFi audio has not upgraded its kernel, not even to the level of the Neo and so the limitation is still to DSD256 (while most of the new dacs and R2Rs, exceed it).
Yeah this was the main reason I wouldn't get a Zen stream personally (just for my use case, for many others this won't be an issue). I've no idea why they limited the output to 384khz/DSD256, because even with stock linux kernel you can do PCM768khz/DSD512.
The cynical side of me assumes its a limit put in place to provide a reason to buy the more expensive Neo Stream which CAN do 768khz, but that's just my hunch. There could be a perfectly valid explanation.

Regardless of reason, it is frustrating, especially when the product has a dedicated HQPlayer mode.
Am too stupid to follow the technical details but didn't GoldenOne mention that for USB connection only the DAC clock does matter anyway?
IMO yes
With I2S/SPDIF/AES, audio is streamed at a constant rate, and an audio rate clock signal is provided to the DAC from the source device and this is used to run the DAC itself. (Some may have PLLs/Buffers but still). As a result, the clocks and implementation in the streamer/DDC have a direct, measurable and audible impact on what you get out of the DAC.
I can even show small measurable differences by changing the cable!

With USB this is not the case. Data is NOT sent in a constant stream. It's usually sent in in chunks, irregularly timed, and put into a buffer. The DAC itself then takes and converts this data at the rate instructed by its own internal clock. So long as there is data in the buffer, the source timing doesn't matter.
It's a bit like checking your mailbox every day at exactly 11:00am. So long as there is mail when you go to check, the timing consistency of the postman makes 0 difference. If he delivers it at 9:00am or 10:58am you will still get your mail at exactly 11:00am.

There is no explanation as to why a higher precision clock would provide improved performance with USB, nor has anyone or any company ever been able to provide any actual evidence of a USB source providing any change at all in timing/jitter performance.
Noise does matter however, and can have a demonstrable impact, but USB does not at all work in the same way as other transmission methods for audio. Your DAC is in control of timing, not the source.

You won't get a degraded result by using better clocks, so if you wish to do so, or if you have a product that does this and are happy with it, that's great. Being happy with your setup is what matters. But my view is that whilst noise from the USB source definitely makes a difference and the result you can get from connecting your DAC directly to your PC vs to a low noise streamer (or a good isolator) can be well worth it, USB 'reclocking' is just marketing.

I'd be very happy to be proven wrong on this as it'd be a pretty interesting area to look into, but so far there is no evidence I'm aware of that USB clocks can have any impact unless they're so broken that the other device cannot communicate with the host
 
Last edited:
Dec 5, 2022 at 12:25 PM Post #171 of 1,916
Dec 5, 2022 at 12:44 PM Post #172 of 1,916
It's Holo's own one. I'm not sure what it's based on (I assume it's not made from scratch)

I've not unfortuantely

Sonore Optical Rendu is a USB only device, no synchronous outputs so jitter isn't a factor. As to noise, I'm not sure as I've not had one in to test.
A better device to compare the Optical Rendu to would probably be the Intona 7055-C as they are much more similar in use case

I will most likely ask to purchase this unit yeah

Either. You can connect it via USB to use as a DDC, or connect via ethernet to use as a streamer

It's better than the Pi2Design Mercury v1, I don't have a V2 so can't compare there (though afaik the main digital circuits didn't change between V1 and V2).
But also a fair bit more expensive so not necessarily a direct competitor. Mercury is still fantastic for close to half the price of a RED

Both of them have ridiculously low noise USB outputs. If this level of noise is limiting a DAC in any way then the DAC itself is beyond incompetently designed.
The main differentiator for me is features. Zen stream has a 384khz sample rate limit which for HQPlayer users is a problem.
If you don't use HQP/upsampling though, then either will likely provide a near identical result if just using USB. They're both excellent. (And Zen stream is cheaper so it'd make sense to go for that)

Yeah this was the main reason I wouldn't get a Zen stream personally (just for my use case, for many others this won't be an issue). I've no idea why they limited the output to 384khz/DSD256, because even with stock linux kernel you can do PCM768khz/DSD512.
The cynical side of me assumes its a limit put in place to provide a reason to buy the more expensive Neo Stream which CAN do 768khz, but that's just my hunch. There could be a perfectly valid explanation.

Regardless of reason, it is frustrating, especially when the product has a dedicated HQPlayer mode.

IMO yes
With I2S/SPDIF/AES, audio is streamed at a constant rate, and an audio rate clock signal is provided to the DAC from the source device and this is used to run the DAC itself. (Some may have PLLs/Buffers but still). As a result, the clocks and implementation in the streamer/DDC have a direct, measurable and audible impact on what you get out of the DAC.
I can even show small measurable differences by changing the cable!

With USB this is not the case. Data is NOT sent in a constant stream. It's usually sent in in chunks, irregularly timed, and put into a buffer. The DAC itself then takes and converts this data at the rate instructed by its own internal clock. So long as there is data in the buffer, the source timing doesn't matter.
It's a bit like checking your mailbox every day at exactly 11:00am. So long as there is mail when you go to check, the timing consistency of the postman makes 0 difference. If he delivers it at 9:00am or 10:58am you will still get your mail at exactly 11:00am.

There is no explanation as to why a higher precision clock would provide improved performance with USB, nor has anyone or any company ever been able to provide any actual evidence of a USB source providing any change at all in timing/jitter performance.
Noise does matter however, and can have a demonstrable impact, but USB does not at all work in the same way as other transmission methods for audio. Your DAC is in control of timing, not the source.

You won't get a degraded result by using better clocks, so if you wish to do so, or if you have a product that does this and are happy with it, that's great. Being happy with your setup is what matters. But my view is that whilst noise from the USB source definitely makes a difference and the result you can get from connecting your DAC directly to your PC vs to a low noise streamer (or a good isolator) can be well worth it, USB 'reclocking' is just marketing.

I'd be very happy to be proven wrong on this as it'd be a pretty interesting area to look into, but so far there is no evidence I'm aware of that USB clocks can have any impact unless they're so broken that the other device cannot communicate with the host

That is the best explanation of timing and use of clocks I have ever read. I can't say I ever really understood it until I read that. Simply brilliant - thank you.
 
Dec 5, 2022 at 1:34 PM Post #175 of 1,916
Current price for the Mercury V2 and the Red are the same.

As for using these as DDC or streamer, it seems like the Mercury V2 won't have an issue with switching from DDC to/from streaming as the Red does, but the Red seems to be able to handle higher hi-res files.

Am I correct on this?
 
Dec 5, 2022 at 1:45 PM Post #176 of 1,916
Mercury V2 is not a DDC but a streamer. However one unique feature of V2 is it has buffered I2S supported by Metrum and Sonnet DAC and subjectively I don’t think you will be able to distinguish any difference in sound quality till 192Khz between Mercury and Red but I haven’t heard Red so it’s just my assumption. I’m not sure if mercury can do beyond that as I have never tried.
 
Dec 5, 2022 at 2:42 PM Post #177 of 1,916
Don't want to start a holy war here as it's clear this device measures well but how does it sound? I have a Spring Level 3 DAC fed by a SOtM 200 Ultra with a LPS. The computer that drives my HQP doesn't have enough grunt to do the exotic high DSD sample rates. I use USB - so if the DAC clock is the important clock here, and I don't want to use stratospheric upsampling, will the Red give me a better listening experience? If I'm not looking to convert a USB signal to (for example) I2S then what does this box give me? Genuine question - grateful for any insights.
 
Dec 5, 2022 at 3:07 PM Post #178 of 1,916
Don't want to start a holy war here as it's clear this device measures well but how does it sound? I have a Spring Level 3 DAC fed by a SOtM 200 Ultra with a LPS. The computer that drives my HQP doesn't have enough grunt to do the exotic high DSD sample rates. I use USB - so if the DAC clock is the important clock here, and I don't want to use stratospheric upsampling, will the Red give me a better listening experience? If I'm not looking to convert a USB signal to (for example) I2S then what does this box give me? Genuine question - grateful for any insights.

I think that's what everyone is trying to decide - how does it perform? I'm trying to decide whether to buy the Sonore Optical Rendu Lite or the Holo Red as an end point. I have no interest in the DDC functionality as that is provided for within my Denafrips ecosystem and I am happy with those components so they are staying. The Red has superfluous functionality for my use case, but I still want to understand where the SQ from it's USB out falls against the USB out of the Optical Rendu Lite.

We really need lots in the wild for feedback.
 
Dec 6, 2022 at 12:20 AM Post #179 of 1,916
Mercury V2 is not a DDC but a streamer. However one unique feature of V2 is it has buffered I2S supported by Metrum and Sonnet DAC and subjectively I don’t think you will be able to distinguish any difference in sound quality till 192Khz between Mercury and Red but I haven’t heard Red so it’s just my assumption. I’m not sure if mercury can do beyond that as I have never tried.

I was under the impression that the V2's USB ports could be configured for not only output but input as well.
 
Dec 6, 2022 at 12:22 AM Post #180 of 1,916
I was under the impression that the V2's USB ports could be configured for not only output but input as well.
That’s to connect SSD to play local files
 

Users who are viewing this thread

Back
Top