Bricasti's Best DAC isn't a DAC?

[Miska]

I am extremely impressed with the performance and feature set of the M12. Especially the way it handles DSD in the analog domain. It's up there with some of the best one-box Preamp/DACs that I've heard. Bricasti has taken tremendous care to keep the analog and digital circuitry separate with each getting their own power supply and grounding. I also like how it uses a R2R resistor ladder for analog gain control. The USB interface is electrically isolated from the host computer, eliminating any grounding or power induced noise issues that could be transmitted to the M12 from a USB source. After spending a bit of time with it and going over the M12 in detail with Brian Zolner, I think they knocked it out of the park with this one.

 

Do you have any more detailed information about the DSD conversion stage etc? I would like to understand the differences to T+A DAC8 DSD which seems to be similar (but DAC-only, so no analog inputs). Since the T+A has a separate discrete DSD DAC and resistor ladder analog volume control (with a switch to bypass it and use fixed level output).

[Ciamarasound]

Do you have any more detailed information about the DSD conversion stage etc? I would like to understand the differences to T+A DAC8 DSD which seems to be similar (but DAC-only, so no analog inputs). Since the T+A has a separate discrete DSD DAC and resistor ladder analog volume control (with a switch to bypass it and use fixed level output).

 

From the manual

 

"The M12 features 2 digital audio conversion paths, one for PCM which utilizes the ADI 1955 converter, and the other is a true one bit modulator of our own design and unique to the industry.The DSD conversion is performed on analog level boards and is a true 1 bit analog converter followed by an analog post noise filter... When selecting NDSD the DSD setting, the DSD post noise filtering is done in the analog domain so there are no DSD filter setting. An artifact of DSD processing is the buildup of ultrasonic noise and with DSD 64 this noise starts at 24 kHz and rises to peak level at -50 dB at 50k and beyond. In the M12 set to NDSD DSD this is done with a simple low pass analog filter for the very best sonic performance."

[captfidel]

I disagree: if you don't want the linestage (or want to bypass it), just get the M1. It's cheaper and better. At least, better for most digital playback file types -- I'll come back to that.

 

One reason why the "linestage version" of their DAC (that is, the M12) is in the lineup is that not everyone is only-digital. Take me, for example. I love my digital, but I also have a turntable and a reel-to-reel machine. With the M12, I can use those analog sources without fussing with an A-to-D on top of a D-to-A -- the analog sources just "pass through" and hit the attenuator.

 

Another reason -- the output of that alternative one-bit modulator path (for DSD only) is at unity. Which means that there must be an attenuator somewhere downstream. Given that there must be one, I suppose that Bricasti might have concluded that they'd better not leave that to chance.

 

Sure, they could have made another converter product that has no attenuator at all, but remember that the M1 routinely gets used as a DAC + attenuator, or "DAC-direct" into their amplifiers. That's how they demo. Given that their M28 amps have the clever feature of a high-quality analog "attenuator" built into their input, that means that you can run the M1 (or any DAC or pre or source) directly into the amps and not have to use the digital volume control for anything other than "fine tuning" (<12dB, say) or muting. This arrangement is extremely transparent, and "up there" with the very best and most revealing I've heard in any demo anywhere.

 

But that alternative one-bit modulator path makes that arrangement untenable. The digital volume control is not available here. So, they'd need an off-board analog attenuator (which tends to sound better than digital volume controls, most especially when we're past the "fine tuning" stage). Given that one box is almost always better than two, they combined the M1 with the best attenuator they felt like making into a single box, the M12, which let them do both "traditional" DAC and that alternative one-bit modulator path at the same time. Ta da!

 

If that sounds like a lot of effort, it's because it is. Which is why the M12 is more expensive. But for those that have heard it, that alternative one-bit modulator path is extraordinary. Even with "only" 1x DSD. A back-to-back demo of the M1SE and the M12 was eye opening. For all DSD material, the M12 was clearly superior, with more 3-D depth and tonal richness. Now, for everything else, I'm pretty sure that I prefer the M1SE direct into the M28 amplifiers, though I'll have to get that gear in-house before I can do more than wave my hands at it.

 

 

Thank you Part-Time Audiophile for your reply, it was helpful and informative. Perhaps I can trouble you a little more for your opinion/thoughts on the following:

 

Your explanation on the product positioning makes sense, it also would imply that the pure path DSD decoding will never make it to the M1 as that would require essentially all of the extra HW available in the M12 (with the possible exception of additional analog inputs). Which brings me back to my original posting that the best DAC from Bricasti isn’t their DAC, but their source controller – as it does everything the M1 does and adds an additional (and spectacular) option for DSD decoding.

 

In closing your previous posting, you said you thought you’d prefer the M1 direct to the M28, I am not sure I understand the difference between that versus the M12 direct to the M28? It would seem that with the M12 you gain some things without losing anything. Is it perhaps the addition of analog gain/attenuation in the M12 that can make for a less “pure” signal than the M1 straight into the M28 – hence your preference?

 

In my case, I too have analog sources (turntable, reel-to-reel), and a nice BAT tube pre-amp that I love as my controller. Additionally, I have 8000+ albums of PCM about half of which are well above Redbook. I also have a smaller but growing collection of DSD recordings. I will be connecting a new DAC (either the M1 or M12) to my preamp and using that as my “controller.” I am therefore trying to decide between the M1 which is the simple to understand choice or the M12 with its better DSD decoding option but additional circuits in the signal path before things get to my preamp. Do you have a recommendation on this (I am quality and sound driven – not price driven)?

 

On another and unrelated question, I was wondering why the Ethernet on the M12 (and I presume some future upgrade on the M1 as well) does not support the highest level of sampling offered by the device. To use the max sampling rate, USB input is still required. This doesn’t really make sense to me as even slower wired Ethernet speeds are more than adequate for the larger file sizes. This is really “old hardware” and so there are plenty of off-the-shelf components for the DAC-side of the connection. For me, who uses a NAS+Roon, this means that I must insert a conversion in the chain to USB with something like a microRendu. It seems like a completely unnecessary step. Any idea why the Ethernet on the Bricasti is capped?

 

Thank you again for your previous, thoughtful reply.

[Part-Time Audiophile]

Thank you Part-Time Audiophile for your reply, it was helpful and informative. Perhaps I can trouble you a little more for your opinion/thoughts on the following:

 

Your explanation on the product positioning makes sense, it also would imply that the pure path DSD decoding will never make it to the M1 as that would require essentially all of the extra HW available in the M12 (with the possible exception of additional analog inputs). Which brings me back to my original posting that the best DAC from Bricasti isn’t their DAC, but their source controller – as it does everything the M1 does and adds an additional (and spectacular) option for DSD decoding.

 

It is, unfortunately, not straightforward. To get levels matched, mutes on fades in and out, mutes on stop -- all that gets handled by a PCM DAC no problem (well, the problems have all been ironed out), but here it has to fall to the ladder attenuator, as this cannot be done digitally on a DSD stream. In fact, if you don't do these things and just wing it with the DSD signal, you risk sending some seriously nasty crap directly through to the rest of the electronics -- high-frequency snaps and pops, all the way to fuse-blowing surges. It's a problem. I remember the old Lampi DSD DACs had some issues like this -- and I blew several fuses before realizing it was the gorgeous-sounding DAC. Whoops.

 

Anyway, the M12 combines these 2 streams (PCM and DSD) seamlessly with no noise, ticks etc when playing any source sample rate from 44.1 to DSD 128 on the USB, but does so by altering the circuit and path.

 

The argument goes: if you don't need it, you're probably better off without it. Which is why there are two products, and not just one.

 

In closing your previous posting, you said you thought you’d prefer the M1 direct to the M28, I am not sure I understand the difference between that versus the M12 direct to the M28? It would seem that with the M12 you gain some things without losing anything. Is it perhaps the addition of analog gain/attenuation in the M12 that can make for a less “pure” signal than the M1 straight into the M28 – hence your preference?

 

The M1, which does PCM (and like all delta-sigma DACs, converts everything to PCM, including DSD) and does it really, really well, doesn't need any extra stuff. The M12 does that too, but inserts an active (I think) attenuator after the DAC. It's an extra step that the M1 doesn't have or need. Is that difference audible? Dunno -- haven't had the two to compare here at home, but based on the demos I've heard, it's close.

 

What's not close is the DSD path on the M12. That's different. But that's why you need that late-stage attenuator, to keep the two paths user-friendly. Given that it also gets you the analog inputs, I tend to think that the M12 is a better trade-off, even if it might trail the M1 in terms of "pure performance".

 

In my case, I too have analog sources (turntable, reel-to-reel), and a nice BAT tube pre-amp that I love as my controller. Additionally, I have 8000+ albums of PCM about half of which are well above Redbook. I also have a smaller but growing collection of DSD recordings. I will be connecting a new DAC (either the M1 or M12) to my preamp and using that as my “controller.” I am therefore trying to decide between the M1 which is the simple to understand choice or the M12 with its better DSD decoding option but additional circuits in the signal path before things get to my preamp. Do you have a recommendation on this (I am quality and sound driven – not price driven)?

 

I don't know that I'd chain preamps together. If you have or want to keep your pre, I'd probably recommend a "regular" DAC like the Bricasti Goldfinger (the M1SE that's been electroplated in gold. Yes, the plating matters. God help me, it does. I have absolutely no idea why. Increased mass? I'm baffled, but there you go).

 

On another and unrelated question, I was wondering why the Ethernet on the M12 (and I presume some future upgrade on the M1 as well) does not support the highest level of sampling offered by the device. To use the max sampling rate, USB input is still required. This doesn’t really make sense to me as even slower wired Ethernet speeds are more than adequate for the larger file sizes. This is really “old hardware” and so there are plenty of off-the-shelf components for the DAC-side of the connection. For me, who uses a NAS+Roon, this means that I must insert a conversion in the chain to USB with something like a microRendu. It seems like a completely unnecessary step. Any idea why the Ethernet on the Bricasti is capped?

 

This may be counter-intuitive, but sample rates don't really matter all that much to the final sound. Well, below a certain level at least. Sure, it's entirely possible that you might be able to perceive them or even prefer the higher rates, but pretty quickly there are other things that weigh more heavily. Said another way, it may be important, but that list of important things is pretty long and sample-rate beyond a certain point may not matter as much as "noise", "reflections", "setup", or "amplifier power delivery". In this case, Bricasti assures me that the Ethernet interface produces the cleanest and most engaging sound -- that is, it is their preferred interface. Is it because it has a better clock (lower jitter)? Less electrical interference? Better signal reconstruction circuitry? No idea. But whatever the reason, they really like it. Me? I haven't had the pleasure -- my Goldfinger doesn't have that interface, though I'm told they can add it. I'm also told they're going to be offering an off-board unit soon, too. Don't think Bricasti is looking at Roon-Ready, however, which is a bummer.

 

As for the technical limits on the Ethernet interface, no idea. I suspect it's the limit of the circuit they're using. Whatever. I'll check.

 

Thank you again for your previous, thoughtful reply.

 

Sure.

[barrows]

Two points: I doubt the M-12 will have any drop off in sound quality as a cause of the fact that it incorporates an analog attenuator, as this does not require any extra active circuitry versus what is already in the M1, all DACs already have at least one, if not two active gain/buffer stages and there is no need for an additional one just because it has an analog attenuator. Additionally, if you set an analog attenuator to no attenuation, there is nothing additional in the signal path versus it not being there at all. So, if you really prefer the idea of how the M-12 processes DSD, I would not hesitate going for it.

I would contact tech support at Bricasti and run these questions by them for a bit more clarity regarding the differences of the M1 and M12.

As to the ethernet board limitations, it is likely that Bricasti is using a third party Ethernet board, probably the same one Ayre is using in their new DAC, as developing a proprietary Ethernet technology is quite demanding unless one has a really, really good computer Mobo engineer on staff, cost of such development would push the retail cost of the product sky high in the case of a low volume product like high end audio. Making Ethernet work reliably for audio above 24/192 requires more processing power and memory, and the third party companies developing these boards do not see high end audio as their main market, they are selling many times more boards to consumer audio companies and hence have to make sure the OEM price is low.

[michaelD]

I have just upgraded to the M28 (new Version demo) waiting for my own pair. I can not believe how good they sound with my M1 SE. My version is totally black with polished Aluminum feet and front buttons. I think they will ship this week. In the meantime I have a pair of the upgraded version M28 and really really like them. I just gave up the Parasound JC-1's that sounded amazing with my speakers but the M28's are so much better.

[captfidel]

Two points: I doubt the M-12 will have any drop off in sound quality as a cause of the fact that it incorporates an analog attenuator, as this does not require any extra active circuitry versus what is already in the M1, all DACs already have at least one, if not two active gain/buffer stages and there is no need for an additional one just because it has an analog attenuator. Additionally, if you set an analog attenuator to no attenuation, there is nothing additional in the signal path versus it not being there at all. So, if you really prefer the idea of how the M-12 processes DSD, I would not hesitate going for it.

I would contact tech support at Bricasti and run these questions by them for a bit more clarity regarding the differences of the M1 and M12.

As to the ethernet board limitations, it is likely that Bricasti is using a third party Ethernet board, probably the same one Ayre is using in their new DAC, as developing a proprietary Ethernet technology is quite demanding unless one has a really, really good computer Mobo engineer on staff, cost of such development would push the retail cost of the product sky high in the case of a low volume product like high end audio. Making Ethernet work reliably for audio above 24/192 requires more processing power and memory, and the third party companies developing these boards do not see high end audio as their main market, they are selling many times more boards to consumer audio companies and hence have to make sure the OEM price is low.

 

 

Barrows,

 

You make several good points, especially regarding the fact that the active gain stage in the M12 is probably no different than the one in the M1 and that only attenuation has been added. In the posting prior, Part-Time Audiophile remarks that it would not be advisable to chain two preamps together, to which I agree. I will try asking Bricasti about the wisdom of this based on their knowledge of the gain stage design in the M12.

 

I am trying to figure out how to get the full DAC features of the M12 (superior DSD processing) into my system while still using my preamp. Another poster made the point that I could use the reference volume feature to set a pre-set volume level which would essentially be unity out. If I read the manual correctly, the big knob would still act as a volume control and make it too easy to add gain and overload my preamp. It would be nice if the output could just be set to Unity, eliminating the volume control’s ability to impact the power of the output signal.

 

Finally, regarding the Ethernet speed cap. I’m not sure this makes sense to me. There are two operations as I see it, one to reassemble packet-based data into a file, the second to process the file. Packet reassembly is handled by the Ethernet interface with no extra processing needed. I wonder if it has to do with how Bricasti is handling DxD files? I note that the spec sheet for the AD1955 states its max sampling rate is 192kHz, therefore Bricasti is doing something tricky to support DxD. Maybe whatever mechanism they’ve build for that is not visible from the Ethernet interface?

 

CaptFidel.

[captfidel]

What's not close is the DSD path on the M12. That's different. But that's why you need that late-stage attenuator, to keep the two paths user-friendly. Given that it also gets you the analog inputs, I tend to think that the M12 is a better trade-off, even if it might trail the M1 in terms of "pure performance".

 

I don't know that I'd chain preamps together. If you have or want to keep your pre, I'd probably recommend a "regular" DAC like the Bricasti Goldfinger (the M1SE that's been electroplated in gold. Yes, the plating matters. God help me, it does. I have absolutely no idea why. Increased mass? I'm baffled, but there you go).

 

 

Yes, that might be the inevitable conclusion – Goldfinder here I come? This does bring me back however to my initial assertion: to be considered Bricasti’s best DAC a unit would have to include their best DSD decoding as well as their best PCM decoding – which today is only available in the M12 – hence their best DAC is a source controller – and probably not the best signal chain for those of us that use a preamp. :-(

[Part-Time Audiophile]

As to the ethernet board limitations, it is likely that Bricasti is using a third party Ethernet board, probably the same one Ayre is using in their new DAC, as developing a proprietary Ethernet technology is quite demanding unless one has a really, really good computer Mobo engineer on staff, cost of such development would push the retail cost of the product sky high in the case of a low volume product like high end audio. Making Ethernet work reliably for audio above 24/192 requires more processing power and memory, and the third party companies developing these boards do not see high end audio as their main market, they are selling many times more boards to consumer audio companies and hence have to make sure the OEM price is low.

 

Hey there, Barrows! Always good to run into you.

 

As for the 1x DSD over Ethernet -- it has to do with their (custom) renderer, apparently. 1x DSD is what they were able to get stable, but won't be their last word on it. As development progresses, more progress on sample rates will be made and then become available.

[captfidel]

...my Goldfinger doesn't have that interface, though I'm told they can add it. I'm also told they're going to be offering an off-board unit soon, too. Don't think Bricasti is looking at Roon-Ready, however, which is a bummer.

 

An off-board unit? Is that like an external unit like a MicroRendu or something else?

 

I agree in that I also don't think Bricasti is looking at Roon-Ready, it IS a bummer. I wonder if they've even looked into it? During the last conversation I had with someone from Bricasti I left with the impression they had written it off presuming it was for Tidal streamers and therefore it didn't really matter to their market. I tried to make the point that streaming from music services was the least important of its features...once you've used Roon, it's hard to imagine a top-end DAC not offering Roon or something similar. dCS and so many others are Roon Ready these days. Given the low cost and low effort to become Roon Ready, it really seems like a no-brainer for a quality HW manufacturer.

[barrows]

Hey there, Barrows! Always good to run into you.

 

As for the 1x DSD over Ethernet -- it has to do with their (custom) renderer, apparently. 1x DSD is what they were able to get stable, but won't be their last word on it. As development progresses, more progress on sample rates will be made and then become available.

 

Hi Scott! So Bricasti has actually developed their own proprietary Ethernet interface... that is very interesting. In that case they very well may be able to get it going at higher rates, nice.

[cjf]

An off-board unit? Is that like an external unit like a MicroRendu or something else?

 

I agree in that I also don't think Bricasti is looking at Roon-Ready, it IS a bummer. I wonder if they've even looked into it? During the last conversation I had with someone from Bricasti I left with the impression they had written it off presuming it was for Tidal streamers and therefore it didn't really matter to their market. I tried to make the point that streaming from music services was the least important of its features...once you've used Roon, it's hard to imagine a top-end DAC not offering Roon or something similar. dCS and so many others are Roon Ready these days. Given the low cost and low effort to become Roon Ready, it really seems like a no-brainer for a quality HW manufacturer.

 

I emailed Briscasti asking them about plans on certifying the ETH Input as a ROON Ready device. As far as I can tell and as far as I know, ROON doesn't do DLNA so if I'm not wrong that means you would still need to use something like the mRENDU or similar in order to continue using ROON with the M12. That of course means your back to using the old reliable USB Input of the M12 and not the "better" ETH connection.

 

ROON pretty much owns the market in terms of playback software so it is curious why it was not included. I guess if they are making there own stuff it could no doubt be added later though.

 

I'm also curious what speed the ETH Interface is, 100mb or 1gb? I would have much rather seen an empty slot for use with a fiber SFP but I'm not sure how "noisy" those little guys are.

[captfidel]

I emailed Briscasti asking them about plans on certifying the ETH Input as a ROON Ready device. As far as I can tell and as far as I know, ROON doesn't do DLNA so if I'm not wrong that means you would still need to use something like the mRENDU or similar in order to continue using ROON with the M12. That of course means your back to using the old reliable USB Input of the M12 and not the "better" ETH connection.

 

ROON pretty much owns the market in terms of playback software so it is curious why it was not included. I guess if they are making there own stuff it could no doubt be added later though.

 

I'm also curious what speed the ETH Interface is, 100mb or 1gb? I would have much rather seen an empty slot for use with a fiber SFP but I'm not sure how "noisy" those little guys are.

 

You are correct, Roon does not support DLNA. Roon’s position is that as a protocol designed some time ago to work for every device, DLNA appeals to the lowest common denominator and hence is sub-optimal when it comes to delivering a Roon-enabled audiophile experience. One of several technical points they make relates to “who” owns the clock – the DAC or the protocol. In the case of Roon, it’s the DAC. Regardless of the DLNA vs RAAT (Roon Advanced Audio Transport - Roon’s protocol) debate, to be Roon ready you must support RAAT (in addition to whatever other protocols you choose to support).

 

As I understand it, for Bricasti to be Roon ready, they have to add the Roon-supplied end point code to their firmware. With this in place, the DAC can recognize and communicate with a RAAT data stream in addition to DLNA, etc. This is relatively straight forward and requires little effort on the part of Bricasti. Per previous postings – makes sense. Not to mention the shared heritage of Bricasti and Roon – AKA Meridian.

 

More info on RAAT to be found here: https://kb.roonlabs.com/RAAT

[Miska]

I wonder if it has to do with how Bricasti is handling DxD files? I note that the spec sheet for the AD1955 states its max sampling rate is 192kHz, therefore Bricasti is doing something tricky to support DxD.

 

It can take in 352.8/384k in the "external digital filter mode" just like many other chips too. That's the way most DACs support those rates. It bypasses the on-chip digital filter because that's the rate the on-chip digital filter itself produces.

[bmoura]

Developing a proprietary Ethernet technology is quite demanding unless one has a really, really good computer Mobo engineer on staff, cost of such development would push the retail cost of the product sky high in the case of a low volume product like high end audio. Making Ethernet work reliably for audio above 24/192 requires more processing power and memory, and the third party companies developing these boards do not see high end audio as their main market, they are selling many times more boards to consumer audio companies and hence have to make sure the OEM price is low.

That's probably why Bricasti is developing their own Ethernet board.

 

It worked for Merging Technologies - their converters for the pro market (Horus, Hapi) and the home market (NADAC, NADAC Player) support Stereo and Multichannel audio over Ethernet using the Ravenna protocol up to DSD 256 today.

[captfidel]

That's probably why Bricasti is developing their own Ethernet board.

 

It worked for Merging Technologies - their converters for the pro market (Horus, Hapi) and the home market (NADAC, NADAC Player) support Stereo and Multichannel audio over Ethernet using the Ravenna protocol up to DSD 256 today.

 

 

Good to hear, it sounds like they are very forward thinking on this, very nice! Now we just need that board to migrate to the M1!

[barrows]

That's probably why Bricasti is developing their own Ethernet board.

 

It worked for Merging Technologies - their converters for the pro market (Horus, Hapi) and the home market (NADAC, NADAC Player) support Stereo and Multichannel audio over Ethernet using the Ravenna protocol up to DSD 256 today.

 

Yeah, sort of. But Merging did it not by using standard Ethernet, the developed their own transmission protocols as you note. additionally, the pro market is bigger than high end audio, and the Merging is stuff is very expensive for what you get (standard OTS dAC chips, etc).

[bmoura]

Yeah, sort of. But Merging did it not by using standard Ethernet, the developed their own transmission protocols as you note. additionally, the pro market is bigger than high end audio, and the Merging is stuff is very expensive for what you get (standard OTS dAC chips, etc).

Actually the Pro Market is smaller than high end audio. That explains why companies like dCS which started in the pro market and then migrated to pro + high end, are no longer in the pro market.

[Miska]

Actually the Pro Market is smaller than high end audio. That explains why companies like dCS which started in the pro market and then migrated to pro + high end, are no longer in the pro market.

 

Not sure about size of the market, but certainly the margins seem to be higher in high end market... Except for software which is vice versa...

[bmoura]

Not sure about size of the market, but certainly the margins seem to be higher in high end market... Except for software which is vice versa...

The two markets operate differently. The Pro Market is very price sensitive and once an engineer settles on a converter, it's hard to get them to switch and purchase another. Especially at a higher price.

 

In the High End home audio market, buying converters is a much more frequent occurrence with the price often influencing the perception of quality. I've talked with DAC makers who find they are criticized by audiophiles for lower priced products (how can this DAC be any good at that price?, etc.)

 

Quite interesting.

[Miska]

In the High End home audio market, buying converters is a much more frequent occurrence with the price often influencing the perception of quality. I've talked with DAC makers who find they are criticized by audiophiles for lower priced products (how can this DAC be any good at that price?, etc.)

 

That is very unfortunate, because there are number of devices that give very good performance for very reasonable price.

 

One difference is also that in high-end market, many times more money goes into fancy looking case than the components inside. In pro-audio market the box doesn't matter so much, just the contents, as long as connectors and the device itself have good durability and can take some beating without breaking down.

 

I can only wonder what would happen if I'd put a high-end 15k€ price tag to HQPlayer. I could also see a point in that, now if I spend 30 minutes answering to an email, the license price is already gone.

[Cebolla]

You are correct, Roon does not support DLNA. Roon’s position is that as a protocol designed some time ago to work for every device, DLNA appeals to the lowest common denominator and hence is sub-optimal when it comes to delivering a Roon-enabled audiophile experience. One of several technical points they make relates to “who” owns the clock – the DAC or the protocol. In the case of Roon, it’s the DAC.

Are Roon really still peddling this bogus 'technical point'? There is no concept of a 'clock' in UPnP/DLNA network audio file streaming.

 

Roon have to 'worry' about such things because their audio file player is at the sender end of the network, in the Roon Server. The audio file gets decoded & played before hitting the network & the RAAT protocol needs to account for that realtime digital audio signal all the way through the network to its receiving Roon Endpoint.

 

Contrast this with UPnP/DLNA, with the audio file player (aka renderer aka streamer) at the receiver end of the network. The UPnP/DLNA media server, at the sender end of the network, just provides audio files. Absolutely no realtime audio to deal with over the network, with it just transmitting audio file data, plus the posibility of having an actual direct internal I2S connection from player, where the realtime digital audio signal is produced, to its DAC - as presumably the M12 has. Can't get better than that as far as the DAC 'owning' the 'clock' is concerned!

[barrows]

Actually the Pro Market is smaller than high end audio. That explains why companies like dCS which started in the pro market and then migrated to pro + high end, are no longer in the pro market.

 

Check your data... Pro audio is huge these days, as many, many musicians are running their own recording (my GF is one of those musicians). Additionally, Merging is now leveraging both pro audio and high end home audio.

[barrows]

Are Roon really still peddling this bogus 'technical point'? There is no concept of a 'clock' in UPnP/DLNA network audio file streaming.

 

Roon have to 'worry' about such things because their audio file player is at the sender end of the network, in the Roon Server. The audio file gets decoded & played before hitting the network & the RAAT protocol needs to account for that realtime digital audio signal all the way through the network to its receiving Roon Endpoint.

 

Contrast this with UPnP/DLNA, with the audio file player (aka renderer aka streamer) at the receiver end of the network. The UPnP/DLNA media server, at the sender end of the network, just provides audio files. Absolutely no realtime audio to deal with over the network, with it just transmitting audio file data, plus the posibility of having an actual direct internal I2S connection from player, where the realtime digital audio signal is produced, to its DAC - as presumably the M12 has. Can't get better than that as far as the DAC 'owning' the 'clock' is concerned!

 

Interesting. I find better sound running DLNA vs. ROON (to be clear, I have not paired ROON with HQPlayer), using the µRendu as the endpoint. I would hope that RAAT still uses error correction at the endpoint (µRendu), if not, that is a really, really different way to do things which would result in much more sensitivity of things like Ethernet cables, routers, switches, etc.

 

Just another quick note on Merging and Ravenna: Chris C. has stated that Ravenna has some issues for audiophiles; I have never used it myself, so have no opinion. Anyone here using Merging/Ravenna for home audio and care to comment?

[bmoura]

Just another quick note on Merging and Ravenna: Chris C. has stated that Ravenna has some issues for audiophiles; I have never used it myself, so have no opinion. Anyone here using Merging/Ravenna for home audio and care to comment?

I have the Multichannel NADAC MC-8 here. What are the "issues for audiophiles" supposed to be ?