ARE SUPER EXPENSIVE STREAMERS REALLY WORTH IT


Folks I am confused why some streamers need to be so eye wateringly expensive. I appreciate the internal basics need to be covered such as a high quality, low noise power supply and a decent processor speed etc..  but that is not rocket science.

So my question is could a decent streamer outputting its data stream via I2S to a good quality DAC receiving the I2S stream be a more cost effective way of rivalling let’s say a streamer costing 5k upwards.

I have heard and digested the argument for expensive streamers quality being centred around the management of the data timing via a quality clock circuit but there are very reasonable in relative terms, DAC’s out there that have dual super high quality temp controlled clocks within, at least the equal or arguably even better than the say a 5k streamer with some sporting dual high end DAC chips etc.

So could utilizing a good quality streamer and a separate high-quality DAC connected via I2S indeed offer significant benefits and potentially reduce the need for a very expensive streamer.

I say this with the knowledge that I2S is designed to preserve and separate the Signals so avoiding the timing issues connected with multiplexing. I2S (Inter-IC Sound) separates the music signal from the timing signal, potentially eliminating jitter or at the very least greatly reducing the possibility for the pesky music killing jitter which we all could agree would lead to improving overall sound quality.

Wouldn’t this separation ensure that the timing information is more accurately preserved, even when compared to a high price streamer, leading as clean or cleaner and more precise audio data output. With I2S, the DAC can use its own high-quality clock/s to synchronize the data, which will reduce jitter and improve sound quality.

Could this possibly mean that even if the streamer has a less advanced clock, the DAC’s superior clock can take over, ensuring best  performance.

So bang for buck would it not be advantageous to investing in a high-quality DAC and using a good but not necessarily top-tier streamer to achieve excellent sound quality without the need for an extremely expensive streamer. Surely the DAC’s performance will play a crucial role in the final sound quality.

Play gentle with the pile on please....................

nubiann

No. Start with an Ocean Digital  WR 10 and go from there-if need be. Paired to a great DAC , it's an astoundingly good piece of gear for an embarrassingly low cost. I know, we all think good= expensive...........

 

had mine for years, from Amazon, that's right. Don't be a snob......

 

the folks at O.D. are very nice and quick to respond too

@nubiann 

In high-tech computing, only the data integrity and timing matters.  However, in high-end audio, it’s more than just delivering the data, the data rides on an analog stream that often pick up noise.  Even one’s audio components can introduce noise.  

I’ve been following this thread with interest since I’d ordered a new streamer around the time it was first posted.  I stream music via Tidal Connect and my existing chain consists of a third gen Bluesound Node (which, for the $350 price, cannot be faulted) connected via USB to a Holo Cyan 2 which then connects via XLR’s to my Hegel H390 and then onwards to a set of Boenicke W8SE speakers through Veritas’ Apurta speaker cables (great performers for the money). 

I gave myself a $2K budget for a new streamer and settled on Holland based Magna Hi Fi’s Mano MKIII.5 streamer with its separate matching Farad 3 LPS.  Magna Hi Fi offers several optional upgrades and, after bathing in snake oil, I opted for their offered upgrades on the connecting cable for the two boxes, a fancy pants fuse and LAN isolator from Synergistic Research.  I’m disclosing these choices fully realizing it may elicit flack, but for me, it was in my $2K  budget (just, at $1900USD) and rather than trying to take it with me, I thought I’d have some fun spending the money in the pursuit of audio happiness. I think their base case unit retails somewhere around $1,200 and can be upgraded thereafter if down the road, if curiosity gets the better of you.

At $1,900, I’d consider this at the low end of “expensive” in the world of pure streamers (units without dacs).  The two boxes are matched and diminutive in size but once plugged into my system, and now connected via I2s (because, well I’m curious) I can say without question, it makes a difference. 

I don’t know why or how, but am really quite stunned at the improvement in sound quality.  Everything I stream sounds much cleaner, quieter and more detailed.  Am now a big believer in the benefits of a separate linear power supply. Also, I'd note their Level 2 silver cable works really well in a system like mine, which already has plenty of warmth.  Honestly, I would not have thought a streamer could make such a contribution to my overall sound quality but it's clearly a step forward. 

I’m using the Volumio software (they offer several options) which is proving a solid platform for me. One downside, if you like to stream Tidal Connect natively (as I do) the optimal way to do so requires Volumio’s subscription service.  Note you can stream Tidal via the Volumio software without the subscription, but to do so natively, via the Tidal Connect app, you need to pay the piper.

So the answer to your question, in my opinion based on my experience with the new gear I acquired is a resounding yes.  It not only made a difference, but a pretty significant one.  As always, YMMV.

Upgrading the power cord to your streamer might make a bigger difference than which I/O you use. I’m currently using an Eversolo A8 into a LaIV Harmony via I2S. I should have the Harmony uDDC Clock in a few days. I’m curious to see what difference that makes. Education is expensive.

I am a big fan of the French high-end brand Metronome Technologies. They pay a lot of attention to the power supply in their streamers and DACs. Not cheap, but worth it, as I know from my own experience.

@vthokie83 Nice thorough explanation. For those contemplating using I2S interface very valuable to have dac that provides the ability to use async (use dac internal clock) with that connection, with this one can determine the value of slave or internal clocking.

 

Also, there are streamers out there with I2S outputs, as to how they're  optimized is the question. Going this route one would be bypassing need for DDC and the input to that DDC, in most cases this will be usb.

nubiann

This is a very long read, ignore if you wish. I have made this a longer more complete post on purpose, since it comes up several times each month.....and it gets old re-posting the same information in different ways. Most of this is from my personal experience and education, and not some form of conjecture. This also is a long post, so some discrepancies may exist….I’m not an electrical engineer (I’m mechanical and industrial)

I'd also suggest (highly) that anyone watch the video from "A British Audiophile" as I think Tarun does a nice job of explaining a bunch of these principles https://www.youtube.com/watch?v=3tKWWs-OMxw

As you can see from all the varying posts, there is disagreement as to the benefits of I2S, the use of a DDC, better DAC versus better streamer, which clock is most important, and whether I2S/SPDIF/AES-EBU/USB is the best connection to be used. In the end we all just want the cleanest, jitter free, most accurate, best clocked data to be used.....regardless of where it comes from, that is what provides the best sound quality. Many of these posts are not incorrect (though many are), because in that poster’s experience it is true.

There is NO sweeping generalization that can be made, the answer in almost all cases will be "it depends" or YMMV.....and to say that I2S is most important or unimportant connection is also incorrect, it depends. I however (correctly or incorrectly) feel that it is important to understand the underlying architecture, to learn why these variations exist.

And then we have external clocking devices that set a universal clock to be used across multiple devices, so that it remains a constant. I have no experience with external global clocks, so I'll make no comment.

Below is a post I made a few weeks ago about a DDC and DACs that another poster was asking about, that explains some of what I'm talking about. You can also see (from posts here and elsewhere), that the better the streamer and the better the DAC, the less need (or no need) for a DDC. But I will say that in each chance I've had to use I2S (all else being equal), it has been the best connection; sometimes by a lot. I would venture to say that  most highly regarded inexpensive streamers (IE Bluesound, Wiim, iFi, Eversolo), could benefit from a DDC….many of us have found this to be the case.

However the more expensive (better) DACs and streamers are optimized for different connections. Some sound best with USB, some AES/EBU, some SPDIF......it varies by product and how they were designed, where the effort and expense was put. In other words the highest quality devices basically do internal DDC functions without the need of an external unit. This is why you see posters saying USB or I2S or AES/EBU or SPDIF is the best, in their experience; that is true. But it is not universally true.

On the lower end (read that as cost if you wish), none of the connections are properly optimized as that would add to cost. To say that I2s output from an average streamer negates the need for a better DAC would not be true. In fact that could be the worst case scenario, as the clock on I2S USUALLY comes from the sending device....garbage in, garbage out from a lower end streamer.

In my personal situation, I still own a Denafrips Pontus II DAC because 3 years later I still LOVE it's sonic signature; it still is one of the most fun, enjoyable, easy to listen to DAC I've ever heard, with excellent presence. It's connections (and clock) are not optimal, it's an inexpensive DAC at the $1,450 I paid for it, and there are better options (more expensive) available if level of detail and nuance is your criteria.

What does help my scenario (and many others), is that I/we use a DDC to take the signal from a cost effective streamer, convert it/clean it/de-jitter it/re-clock it, and send that much improved signal to the DAC via I2S where it remains clean. I2S has separate pins for the master clock, word clock, and data stream; this is known as a "synchronous" connection.

Other connections such as USB, SPDIF, and AES/EBU are usually "asynchronous" which means the master clock, word clock, and data are all combined in one interface. NOTE: Sometimes AES/EBU can be either “synchronous” or “asynchronous” depending on it’s implementation. This information must be separated at some point either by a DDC or by the DAC, into separate word clock, master clock, and data. Higher quality gear does this very well.

When we hear many (myself included) talking about the need or value of higher end gear, it is because they generally use much better power supplies and components to do this job very well. It is not to say that you HAVE to go that high, but if you have a “mid fi” or “hi fi’ system, the better gear will provide noticeable sound quality improvements. That is not a knock on cost effective gear, it’s a fact most of us recognize or have learned in listening.

Later this year I will be upgrading both my DAC and my streamer to rather expensive units, about $15,000 to $20,000ish combined for both units. In that instance, both the streamer and DAC will have optimized connections; though some will still sound better than others and what I'll have to discover. I suspect in this case that I may not have the need for a DDC any longer, in fact I expect that I will not need one. Also high end gear may have other benefits such as the ability to shut off certain aspects (ability to remain or turn off clocking in the DAC).

QUOTE: "You are in slightly different territory right now than my and Soix experience, as your Venus 15th DAC already uses the high end OCXO (oven controlled crystal oscillator), and Fred's Terminator 15th uses dual OCXOs. So if you were to add a DDC, it would have to be the Denafrips Hermes or Gaia so you don't downgrade the clock.

Typically (though not always) the DDC sets the clock, and the DAC is the slave to that clock when using I2S. With USB and the other connections, the DAC clock is most frequently used.....in my experience (and a bunch of others), I2S is the superior connection. I've found the same in other systems that I've helped friends set up.....using various Denafrips DDCs. The more expensive the streamer, there is possibly less of a need for a DDC; possibly.

Part of the reason that the Pontus is improved with the Iris, as the Iris has a TCXO (temperature compensated crystal oscillator) which is an upgrade from the Pontus' standard FEMTO (voltage controlled) clock (short for Femtosecond, or one quadrillionth of a second). Standard FEMTO clocks are not temperature or oven controlled, so their accuracy varies with temperature changes. The clock upgrade however is only part of the advantages of a DDC.....the Denafrips Iris ONLY has USB inputs; no AES, RCA or optical SPDIF, the Gaia and Hermes allow for those.

The other advantages of a DDC are: isolates the input signal (USB, AES/EBU, RCA or optical SPDIF), filters the signal both galvanically (magnetic) and optically, and re-clocks it with the superior crystal.

Digital devices use I2S internally to transmit data, using I2S to transmit from the DDC to the DAC allows for one less conversion. There is some debate on this, as I2S correctly was never designed to be transmitted over a cable.....but here we are. Also it is correct that there is no standard for I2S pinouts, but many use one of two more common formats.....and if not, most devices allow for I2S pin configuration. Most Chinese products use a similar I2S pin configuration. In my experience I2S cables should be of very high quality and as short as possible…..this is somewhere I’d absolutely look to spend money on better cables.

If it were me (and I will be upgrading soon), I'd buy an Innuous/Lumin/Aurender/Grimm/etc high end dedicated streamer and see where you land. With one of these high end streamers, you may find you do not need a DDC......and you can always add one later if you wish.

I have now shortlisted some streaming products from Aurender and Lumin, maybe I might add HI FI Rose also. I am still looking at options for DAC’s that maybe a possible synergistic match for these products as well as the rest of my system.

Cool! Plenty of choices out there. The two you have are two of the most popular in this forum. In addition to names I gave above, add Innuos, EMM Labs, PS Audio, Melco, Moon, Audiolab, Auralic, Wattson, Cambridge Audio, Cyrus, nobody here ever talks about Naim, rarely Linn- who invented the high end streamer. I guess it all starts with budget. 

Here is one cool "network player" that is totally upgradeable and customizable- buy the modules you want and don't pay for ones you don't- check out the review: 

Aqua Linq

 

To the OP:  Your emphasis on the clock is a mistake, imho.  The impact on sound from a quality clock/reclocking is easily measured (jitter).  The only source I know of that consistently measures jitter in streamers is HiFi News.  So I put together a small database of 13 streamers measured by HiFi News over the last few years; the two streamers costing £1k or less have jitter levels at about the same levels as the 11 higher cost streamers (up to £33k), and these are vanishingly low levels of jitter.

So that leaves only one other area of potential difference among streamers:  power supply-related noise.  And as some others have mentioned, DACs vary widely with respect to their sensitivity to noise; some are largely immune.  Unfortunately, HiFi News measurements of this - thru signal-to-noise ratios - is less comprehensive.  However, what measurements they have done on this issue suggest, again, that you don’t need to spend a lot of coin on a streamer.  There are well-engineered noise-free streamers available for £1k or less.

Finally, my own experience of streamers over the last 20 years, including a fair number of direct A/B comparisons, has confirmed what the data show.  Isn’t it great when the science and our experience are consistent?

there is a reason why people buy a $7000 Sean Jacobs power supply for their Chord Dave.

Yep

From Antipodes: 

"a clock stage is only as good as the circuit it is used in and the power supply that powers the circuit. And the re-clocking stage needs to receive a low-noise signal with high-bandwidth, before it can re-clock the signal accurately. If the signals before and after the re-clocking stage are polluted by noise below the bit-rate (such as from employing linear power supplies) then the re-clocking will be largely ineffective. Anyone that claims that digital is 'just ones and zeroes' has not applied their knowledge to demanding real-world problems."

All the best super-expensive, no compromise streamers have gone to huge lengths to develop their power supplies, which is a large portion of their expense. 

@russbutton  I’d encourage you to at least demo a decent streamer from a place like Crutchfield, Amazon, etc. that allows for returns rather than blindly holding on to the belief that it’s all just ones and zeros.  Hundreds of us here have found that’s absolutely not the case for audio — not even close, so unless we’re all somehow wrong and fooling ourselves it’d be worth your while to at least try. 

Whatever interface one uses they all can be less or more depending on one's optimization. While I2S has some inherent advantages that doesn't mean it will always sound better than usb  or whatever. Clocking and power supplies to the clocks is of utmost importance.. Three categories of clocks, femto, txco, oxco, in theory the order of quality least to most is femto, txco, oxco. In reality a well implemented femto clock could outperform a less well implemented oxco clock. The quality of the clock itself and more importantly the quality of the power supply to that clock matters a lot.

 

The above holds true whether comparing I2S implementations or comparing usb vs I2S implementations. The most optimal implementation for either of these interfaces would be a well implemented oxco clock in both streamer and dac. Good luck trying to find that in either streamers or dacs. space limitations for the power supply alone make it very difficult to implement. Femto clocks are easiest to implement with these space limitations, cost considerations probably play a role as well.

 

So, wide variability in clocks and interfaces and we haven't even gotten to differences in motherboards, processors, RAM, operating systems, linear power supplies, music player apps. All have a bearing on sound quality.

 

And then we haven't even gotten to differences in optical vs ethernet, switches vs routers, digital cables, modems, linear power supplies to all these devices, and I may be forgetting some things.

 

Some believe you can just slap in any streamer and get the same quality of sound with a fully optimized streaming setup. I and many others have experienced otherwise so believe whatever you want to believe.

Some would say... buy a Lexus and be happy with your journey

Others would say... why buy the Lexus when you can save money and buy a Kia Soul,... after all you are just going from point A to B

Others would say... why buy the Kia when you could buy a bicycle and get there much cheaper and you might be healthier.

To each his own and be happy.

Good analogy @2psyop Exactly. Nobody goes to the Lexus forums and calls them all "luxury-fools" for spending there money this way, but plenty do this on audio forums. 

 The consensus strongly suggests just as most parts of the chain, best performance is often achieved when both the streamer and DAC have high-quality clocks, ensuring minimal noise and jitter throughout the entire signal path, will lead to better audio quality. 

Yes, but all of your Q and A above talks about clocking, and implies the clock is the only part that matters, and overlooks the most important part of a streamer: If the power supply(s) are insufficient not much else matters, and the concept that the clock can just "clean everything and make music perfect" is incorrect. 

From Antipodes: 

"a clock stage is only as good as the circuit it is used in and the power supply that powers the circuit. And the re-clocking stage needs to receive a low-noise signal with high-bandwidth, before it can re-clock the signal accurately. If the signals before and after the re-clocking stage are polluted by noise below the bit-rate (such as from employing linear power supplies) then the re-clocking will be largely ineffective. Anyone that claims that digital is 'just ones and zeroes' has not applied their knowledge to demanding real-world problems."

All the best super-expensive, no compromise streamers have gone to huge lengths to develop their power supplies, which is a large portion of their expense. 

Grimm, Antipodes, Taiko, Xact S1 have developed these power supplies all in-house, these are not off the shelf parts some Joe can buy off the street and think he can make the same streamer himself. Same goes for the computer boards. 

Not sure how Aurender goes about their power supply but their website states this: 

N30SA’s unique hybrid power supply delivers a pure DC current to flawlessly power the digital audio output board, while a separate linear power supply chain feeds clean power to the non-audio adjacent components of the machine. 

Of course we all understand power supplies in amplifiers matter, but there is a reason why people buy a $7000 Sean Jacobs power supply for their Chord Dave. Quality power makes a huge difference in all components. 

However, I will look at streamers very closely and try to invest in a very good streamer with a femto clock, strong Power supply electronics and thoughtful design, because I now recognize it can also make a significant difference in the overall performance.

Okay, your last paragraph addresses this. But just go and listen to a no compromise streamer and hear it for yourself instead of "believing it" due to responses or trying to rationalize it. 

Most times in life you get what you pay for. 

 

+1 ghdprentice and I might add.

Some would say... buy a Lexus and be happy with your journey

Others would say... why buy the Lexus when you can save money and buy a Kia Soul,... after all you are just going from point A to B

Others would say... why buy the Kia when you could buy a bicycle and get there much cheaper and you might be healthier.

To each his own and be happy.

@nubiann  A well thought out response. Good luck with your search.

Many thanks, nice if you to say so. Whatever the decision,  I will come back and say what I opted for but I wont review because there will be a danger of confirmation bias no doubt.

No offense intended but the length of your reply and it’s style smell of AI, am I correct?

No offence taken Bruce.

I have been overwhelmed by the extent of engagement on the subject of my enquiry. So I used AI as a cross referencing tool as I waded through the depth of knowledge deployed in response.

I am a Sony Trained engineer so my "indoctrination" into the Japanese way and tendency to scrutinize is almost second nature.

I am also a global quality Auditor, so my style of writing can come across in an overly logical fashion or somewhat formal report style but throughout my dialogue here, my deductions and questions generated are very much my own curiosity, as dumb as they might seem to some expert contributors.

Most people have been understanding here and for this kindness, I am very grateful. The advice has helped and tempered my over enthusiasm.

I have now shortlisted some streaming products from Aurender and Lumin, maybe I might add HI FI Rose also. I am still looking at options for DAC’s that maybe a possible synergistic match for these products as well as the rest of my system.

Are you tending toward AI integration more in your daily life?

Thanks for your comments anyway.

Gerald.

 

No offense intended but the length of your reply and it’s style smell of AI, am I correct?

Having chewed my way through and taken on board the many interesting, informed, passionate and sometimes provocatively forthright opinions from within our community, I have been through a weekend of introspection, weighing up what I have learned and asking some questions of myself. I reached a few at the very least tenuous conclusions and knowing that there are many here with far more detailed knowledge and experience I will still risk sharing my thoughts with you all.

My reasoning seems to make sense at least to me but does it align with common practices common to higher-end audio setups?

Instead of following the I2S yellow brick road, I may now focus more on a synergistic combination of attributes approach but since mine is not a limitless budget I still need to find a balance between budget constraints and the relevant talents of the two components. In doing so, I have to be prepared for this particular voyage of exploration to conclude at the same port.

Further Questions & Answers (generated via maybe a still emerging appreciation of this whole new world).

Q: Could the answer to how much I spend on a Streamer vs a DAC be about a combination of one having a femto clock and the other having one or more OCXO clocks?

A: Yes, there could very well be?

Q: Is there an inherently better reason for utilizing femto in a Streamer

A: It seems accepting femto clocks are better at reducing Phase Noise which is more common in streamers as they have more fundamentals in common with a PC than other components in the audio chain. If a streamer has a high-quality clock (like a “femto” clock), it can provide a stable and precise signal to the DAC. This can reduce jitter at the source, however, the DAC still needs to re-clock the incoming signal to ensure accurate digital-to-analogue conversion.

Q: Why OCXO Clocks could be a good choice for DAC’s?

A: Temperature Stability: OCXO clocks provide excellent long-term stability due to their temperature-controlled environment, minimizing frequency drift as well as further reducing jitter, an OCXO clock could help the DAC produce a more detailed and natural sound, with better imaging and soundstage.

Q: Will the DAC Re-clock any incoming data from the streamer?

A: The DAC’s clock/s are crucial for re-clocking the signal it receives. Even if the streamer has a good clock, the DAC needs to ensure the timing is precise during the conversion process.

This means that even if the streamer’s clock is not as good, the DAC can still ensure high-quality audio output.

Q: Could both Clocks Matter?

A: Yes, a good clock (Femto better) in the streamer helps minimize noise & Jitter introduced at the source, potentially affecting the signal before it reaches the DAC.

A high-quality DAC with one or more OCXO clocks will re-clock the incoming data from the streamer, further reducing jitter and ensuring precise timing particularly influencing the digital-to-analogue conversion

Temperature Stability: OCXO clocks provide excellent long-term stability due to their temperature-controlled environment, which can be beneficial for maintaining consistent audio performance.

Temperature Stability: OCXO clocks provide excellent long-term stability due to their temperature-controlled environment, which can be beneficial for maintaining consistent audio performance.

Q: Does System Synergy Count here?

A: The consensus strongly suggests just as most parts of the chain, best performance is often achieved when both the streamer and DAC have high-quality clocks, ensuring minimal noise and jitter throughout the entire signal path, will lead to better audio quality. How large and consequential such improvements are and how they are quantified is not easy; I guess its system and probably subjective listening dependent at this level. I cannot envisage an opportunity to conduct such an in-depth auditioning opportunity. So I will need to narrow down any list by a combination of factors not all listening experience with my system.

If my budget allows, my approach may well be a femto clock in the streamer to handle initial signal processing and an OCXO clock/s in the DAC for final re-clocking, leveraging the strengths of both types of clocks in within their respective environments.

Overriding practical considerations

Budget Allocation: Depending on my budget, I may still need to prioritize spending more on the DAC, as I still maintain no one has debunked its almost paramount importance as more influential for the final audio quality.

However, I will look at streamers very closely and try to invest in a very good streamer with a femto clock, strong Power supply electronics and thoughtful design, because I now recognize it can also make a significant difference in the overall performance.

I was a believer in i2s fairly early on and on paper it makes sense, however.......

After trying both PS Audio and Topping DAC's with i2s inputs, connected to either a RPi streamer or NuPrime CD transport with i2s outputs, the results were not show stoppers and in all combinations a decent (Only £50) coax lead sounded better using SPDIF!
As in everything, its all about implementation and for what ever reason, i2s just didn't work for me using these items, so be aware that improvements using i2s is not a given. The HDMI i2s pin outs were compatible and correct, as indicated by the manufacturers diagrams, but it only ever sounded ok, even with a short but expensive HDMI lead, supposedly designed for such a connection.
I now use 75ohm coax with BNC connectors all round and I'm happy with that arrangement.

I use an Eversolo DMP A-6 ME with LPS upgrade and BNC digital in/out conversion.
This Streaming DAC is very user friendly and sounds great. In my opinion, you would have to have a very good quality and costly set up to hear any benefit from a more expensive Streamer/DAC. My set up is a Pass XP-12 Pre into Neurochrome 686 Monoblocks with balanced XLR connections throughout.
I try to keep my box count to sane levels and feel certain combinations go well together. A Streamer/DAC combo is a natural choice in my opinion as is a Pre Amp/Phono stage, but having said that, I have the Pass XP-17 which is a separate phono stage. The one thing I don't want is a Streamer/DAC/Pre combo, as this is invariably a digital Pre Amp and the built in analogue to digital conversion, needed for my phono stage will be a joke.

My only complaint with the Eversolo A-6 is its only available in Black, other than that, it's a keeper until the next leap in technology, provides even better quality/features at a reasonable price.

God luck with your system building, I hope you find the synergy we all seek in this hobby, strike that, obsession! smiley

@soix  So what youʻre telling me is that you can transfer terabytes of data halfway around the planet, transversing dozens of routers, going from fibre to copper to fibre and wireless with 100% accuracy routinely, but you cannot reliably transfer 10 to 50 megabytes over a USB cable?   

I hear what youʻre saying about timing, but why should that be an issue?   Are you telling me that a DAC youʻre sending data to doesnʻt have a buffer of sufficient size to guarantee proper operation 100% of the time?  Timing shouldnʻt be an issue if your buffer size is large enough. 

Are people dropping thousands of dollars on DAC devices and the DAC engineers canʻt add enough memory for an adequate buffer?  That makes no sense.  Your typical CD contains about a half gig of data in wav format.  Seeing how cheap memory chips are, why is that a DAC cannot easily have a buffer big enough to hold all of the data of a whole CD?

Expensive has different connotations for different people.  Really good amps that cost more than you should probably spend are worth it.  Just my two cents.  

ARE SUPER EXPENSIVE AMPS REALLY WORTH IT?

@mclinnguy 

No.

Folks look for certain attributes in amps that supports their listening styles. Those attributes don't necessarily command high prices.

Some folks like 10-watt single-ended triodes. Those rarely cost a lot. You can even build them yourself.

10-watt amps are not my style, however. I like high-current big iron. That doesn't necessarily cost a small fortune either, on the used market at least.

Now, if you like that new-gear smell and you don't know a soldering iron from a clothes iron, then "super expensive amps" may well be your best bet.

Also, there is no need to yell, unless you were just emulating OP's caps style.

 

You rarely see people fighting about amps.

ARE SUPER EXPENSIVE AMPS REALLY WORTH IT? 

The less people understand something, the more they fight about it.

You rarely see people fighting about amps.

 

@grislybutter +another one. Maybe if some folks would use the search bar before reiterating the same old same old.....anyway....No one relishes a rebuke and many find it difficult to be open minded that someone else might have a clue. 

If I could I'd get The Merging Technologies stack, DCS Apex, T+A Elektroakustik SD 3100 HV or the Aurender MC20&N30SA.

Bits is bits.  Thatʻs the beauty of digital audio.  At the very lowest level of computing, all machines do is:

  • copy data from one register to another
  • compare data between two different registers
  • add data from one register to another and write out the result.

Everything a computer does is a combination of those three operations.  The ability to copy data from one place to another with 100% accuracy is the hallmark of computing.

@russbutton  Sure, if you’re just sending a Microsoft Word document that’s not sensitive to noise, timing, etc. it’s perfectly fine.  But audio is very sensitive to these things, so you can hang your hat on that it’s just ones and zeros but most here who’ve migrated from basic computers to dedicated streamers know that’s absolutely not the case, and until you’ve heard a dedicated streamer your “opinion” is just insular ill informed. 

I don't know if the Ps Audio has that capability, best to ask those familiar with unit. I'd also question whether you are indeed operating in sync mode, I was just reading this, https://forum.psaudio.com/t/ps-audio-i2s-clock/30748 According to my reading the directstream is using it's clock.

@sns - can you point me to where in the settings I would be able to set my Directstream to async i2s?

Edit:  I should note that I once tried operating the Directstream without the MCLK, but it failed to lock.  This was as I was building an IanCanada streamer, so wanted to see if I could dispense with the MCLK since the Raspberry Pi does not have an output for it.

 

@wtyamamoto Incorrect information on I2S,all  the dacs with I2S input I'm aware off give you the choice of sync or async, in other words you have choice between the streamer/DDC clock or dac internal clock.

 

For those who hear differences amongst streamers, many also hear differences between music player apps. Check out all the Aurender and Innuos users who prefer their native apps vs Roon. I also hear differences in the various apps available to me in my custom build streamer, ie. Roon, Euphony Stylus, HQPlayer. Roon and HQPlayer both require much processor resources, this will affect sound quality.

 

And I don't think anyone here is saying dacs not important, they are absolutely important!

I am always fascinated by these threads, same arguments and disagreements. You would think that with all the science and facts, it’s a black and white issue, either bits are just bits or bits can be "better" when the streamer is better.

IF .,. you have the rest of your WHOLE system components being suitably matched as high-end with inherent superior resolution performance capabilities ….. then … generally . “ Yes… it’s worth it”

BUT .,,,AND ITS A BIG “BUT’ .,if you don’t have the audio resolution components as your system , then it’s a very highly likely “ NO” …. It’s not worth the added  investment cuz gonna get you to audio OZ .

Listening to TV either YouTube or Netflix movies using the P1’s internal DAC over HDMI ARC sounds fantastic even though it’s just 2 channel stereo ...

Both YT and Netflix use lossy compressed audio - comparable to mp3 files. Many people are happy with that level of fidelity.

The simple answer is yes, expensive streamers are worth it.  There were significant differences in SQ every time I moved up the Aurender line.  More detail, blacker backgrounds, wider/deeper staging, denser imaging.  These improvements are due in part to more sophisticated power supplies, isolation, and clocking systems.  Associated with the purchase of an expensive streamer you need to invest in a high quality USB or SPDIF and “digital” power cord to optimize the benefit.  I recommend at least $1k on the digital interconnect and power cord respectively.  

@ddafoe What we hear is analog. We never hear a digital signal. Only analog. So when we are hearing differences, itʻs always in the conversion of digital to analog, or something farther down the audio chain.

Whether Iʻm streaming from a computer or a branded streamer, itʻs still the same bits going down the wire.

Now I will confess that a computer can go kablooey like any other device. I had a PC that was barfing (special It tech term) from time to time, so I replaced it. The new one has been flawless. With digital, either it works or it doesnʻt. Itʻs not like where you can get an impedance mismatch between a preamp and power amp and the worst thing is that it doesnʻt sound right. Either a computer is transferring bits correctly or itʻs broken.

@russbutton, have you ever tried different streamers to see if you hear a difference in your system?   

Like you, I also believe bits are bits, but I also have tried a couple streamers and other tweaks (LPS vs switching power supplies, fiber vs copper, etc.) and ultimately use my ears to decide if I can detect a difference.

Sometimes the more expensive piece wins and I keep it, sometimes the cheaper piece wins and I sell the expensive piece.   I'm certainly glad I didn't let that belief prevent me from experimenting with my digital front end and ultimately improving it for at least my ears & system without spending a fortune to do it.  e.g. I no longer use a custom Pi based SPDIF streamer for my primary system even though I was extremely happy with it.  I have found something that sounds much better to me, even though I believe the Pi was passing the bits along just fine...

A couple of years ago, I was fortunate to have the opportunity to compare Linn’s Next Generation DSM/3 streamer (2020/21 variant) and Klimax Organik Exaktbox DACs (2021) with previous models, the Klimax DSM/1 (2014) and Klimax Katalyst Exaktbox (2017).

What I found was that both the new streamer and DAC brought about significant improvements. However, they did different things.

The new DAC gave a smoother refined sound that was more pleasing to listen to. The new streamer gave the impression of being more rhythmic and musical.

In my opinion, the improvement from the new streamer was more fundamental and worthwhile. I should say that not everybody agreed with my view and some preferred the DAC.

@nubiann I'm with the crowd here that says DAC is far more important--clockwise, power supply-wise, galvanic isolation wise--everything-wise.  If you use Roon, Roon's own streamer the Nucleus or Nucleus + is excellent.

@russbutton +1

@soix Bits is bits.  Thatʻs the beauty of digital audio.  At the very lowest level of computing, all machines do is:

  • copy data from one register to another
  • compare data between two different registers
  • add data from one register to another and write out the result.

Everything a computer does is a combination of those three operations.  The ability to copy data from one place to another with 100% accuracy is the hallmark of computing.  If you canʻt trust a machine to do that, you should be very, very worried about your bank account.

 

When you stream music data from Qoboz, or whoever your source is, those bits are copied at least a dozen times between devices, routers, etc, over wired and fibre links (none of which are "Audiophile grade"), each and every time with 100% accuracy.  Once those bits get to your streamer, they get copied and moved around 3 or 4 times more before they go down the wire to your DAC or such.   And then you tell me that somehow, at the very last few inches of wire, the data transfer is NOT 100% accurate?

 

@soix ​​​​​​ @russbutton

You are both correct.

Streamers are nothing but computers, and often not very good ones (think Raspberry Pi 4 or 5) unless you’re in the mood for something like a $31,000 Taiko Extreme which, besides being an exquisite piece of kit, packs serious computing horsepower.

To @soix ’s point, quality power supplies, galvanic isolation, clocks, etc. do make a difference, and they should all be part of any serious audio PC build. The aforementioned Lucas Audio does just that for you, but by the time you’re done with him you’ll have spent near-Taiko money. You can also build it yourself; if so, expect to spend in the neighborhood of $4-5k on quality components.

But those of you who once used a mac mini or some such in your digital chain and then come here to denigrate PC network streamers on the basis of your one (admittedly dismal) experience, it’s like saying you drove a Nissan Leaf back in 2010 and it sucked, therefore all electric cars suck. Teslas don’t suck, neither do Lucids, and a $50 Raspberry Pi build will better a mac in terms of sound quality.

So I think it's time to give PC streamers their due. Done properly, they will rival the SQ of anything Aurender or Innuos make, plus they will give you unmatched software flexibility, upgradeability, and power.

All of your various streamers are just generic PC boards running a custom front end app on top of Linux.  Theyʻre just computers dedicated to running that one app.

@russbutton  No, that’s an oversimplification and there’s absolutely more to it than that.  Power supplies, galvanic isolation, clocks, etc. can all make significant improvements over a basic computer as most people here have found.  At least try a good streamer before you just dismiss them out of hand. 

All of your various streamers are just generic PC boards running a custom front end app on top of Linux.  Theyʻre just computers dedicated to running that one app.

 

But go ahead and drop $5k on something youʻll be trying to unload for $1800 in 3 years.  

My media manager is a miniPC ($130) I got from Amazon, with 128G drive (for Linux) and 8G memory.  I have an external USB drive ($100) containing the music data.  It runs Ubuntu Linux (free) and JRiver ($30).

When I furst upgraded my DAC from a Lumin X-1 to Lampi  Pacific with a Synoogy 8tb and Lumin U2 I was terribly disappointed in the Lampi.  I felt the sound of the X-1 was really close to the Pacific.  After researching many Servers/Streamers, I came across Lucas Audio, a small company out of Poland.  I talked to Lucas Domansky numerous times and he created a server for me.  

It was night and day better once I plugged the Lucas Audio server into my system.  Without a doubt, the money I paid for the Lucas Audio server was worth every penny/dollar that I paid.  I have since acquired the Lampi Horizon and the Aries Cerat Kassandra DAC's but I will never replace my Lucas Audio Server.  

They are both extremely important, but in my case, it is 60% Server and 40% DAC.

No one has really convincingly explained this.

And even after decades of research and millions of dollars spent nobody can explain how turtles can use Earth's magnetic fields to return to the beach they were born, after swimming thousands of miles away from it.

But they do. 

I was running Roon on a Mac mini using USB output to my Wavlet II DAC. I narrowed my dedicated streamer choices down to a HiFi Rose RS130 vs Aurender n200 and for me chose the HiFI Rose because it looks much better esthetically and supports Roon.  Noticeable sound improvement but not major. I’m still happy with my decision and can direct stream from Spotify now.

All SPDIF protocols and i2s utilize the clock of the streamer.  USB (asynchronous, which most DACs these days are) clocks/reclocks the signal internally.

SPDIF MIGHT be better if the streamer has great clock crystals/circuits, and asynchronous USB MIGHT be better if the streamer source has a poor clock.

Most modern DACs can handle a pretty high amount of jitter, and there's debate as to how much is too much.

Most modern DACs also have great USB inputs (most using XMOS chips; however, there is debate about how much "noise" USB circuits introduce to the signal and/or generate internally.

I have PERSONALLY not found a streamer that sounds better than my Raspberry Pi (now 5) running Moode Audio running USB to either my PS Directstream or my Okto DAC8, and that includes running my dedicated mini PC USB through my Matrix SPDIF 2 to the Directstream via i2s.  I THINK it sounds better than my Node 130 that I got because I wanted to try the "famous" Bluesound interface.

But I'm 62, and whereas I used to be able to hear a mosquito across the room, I can now only hear it if it's hovering right next to my head.

I guess I will just continue to ask dumb questions until I arrive at the right answers

@nubiann 

First, sorry about the above post. It's just that all-caps grate.

Second, your questions are thoughtful and on point imo. They are far less dumb than many of the answers you've been getting, if I may.

Third, since you ask, here is the truth about digital streamers.

- Buy a $15000 streamer if you can't afford a $30000 one. It doesn't really matter which one because at that price level they are guaranteed to enchant (they wouldn't cost that much if they weren't amazing - think about that for a second) and they are all created by geniuses, or should I say magicians - after all, anyone who can turn a botched Linux distro running on a lowly ARM Cortex into Gold has to be a little bit of one.

- Pick a team: USB or I2S (Teams SPDIF and AES are the audio equivalent of the Liechtenstein National Football Team). Acquire enough random knowledge about oven clocks, DDCs and HDMI pinouts to effectively defend your team in forum posts. In-depth knowledge of com and network protocols is entirely superfluous and will fall on deaf ears. Ditto references to the studio / mastering world, though it solved those issues eons ago with Dante, Ravenna, AES67 et al.

- Absolutely invest at least $1000+ in a USB / I2S cable. Digital bits are extremely small and wily. They easily escape lesser cables, causing annoying jitter. High-end digital cables effectively capture jitter and turn it into harmless vapor, resulting in the inky darkness of a 360 degree soundstage from which you can easily pick the exact location of Diana Krall's lips. You read it here first!