[AndrewOld]

Thing is, others on this forum have said the same about the 2Go - that uncompressed (WAV) files sound better than FLAC (lossless but compressed)…

Then Chord also have a problem with their FLAC implementation in that product. FLAC and WAV files hold identical information in different formats. They should sound the same. End of story. You shouldn’t make sweeping statements about a format when a particular implementation is defective. Take it up with Chord. They are not exactly renowned for delivering competent software and they have had many problems with interference and RF with other products.

 

[Progisus]

Wav is uncompressed. Flac requires decompression. Depending on where that is done there may be sound implications based on computing power.

 

[jarnopp]

It’s obvious: double the storage requirement!

I meant with regards to sound quality? WAV on the SD card vs your streaming setup.

 

[edwardsean]

Then Chord also have a problem with their FLAC implementation in that product. FLAC and WAV files hold identical information in different formats. They should sound the same. End of story. You shouldn’t make sweeping statements about a format when a particular implementation is defective. Take it up with Chord. They are not exactly renowned for delivering competent software and they have had many problems with interference and RF with other products.

This has nothing to do with Chord.

 

[earnmyturns]

The Naim Audio people I've spoken to are adamant that "WAV sounds better than FLAC"

Translation: "Our streaming processor is so underpowered (or buggy) that it cannot handle the trivial FLAC decompression algorithm." This is really total BS. There's no reason for a properly designed streaming processor to mess up the translation from an asynchronous FLAC stream to well-timed PCM (typically over I2S) to the DAC. if the slight additional computation needed for FLAC decompression creates more electrical noise in the system, that's just bad hardware design. Sorry, I'm tired of overpriced hardware vendors blaming perfectly lossless protocols for their corner-cutting.

 

[earnmyturns]

Thing is, others on this forum have said the same about the 2Go - that uncompressed (WAV) files sound better than FLAC (lossless but compressed)

Thing is, way too many hardware vendors under-design and under-test their streaming products. There are zero technical reasons for FLAC to underperform WAV, except for bad design or bad implementation.

 

[dcp10]

I meant with regards to sound quality? WAV on the SD card vs your streaming setup.

Ah, ok - sorry. No, I haven't tried this, because as I noted in my earlier post, I'm not confident I can personally tell the difference here. What I *can* detect, however, is an improved presentation when I used a "remote" streamer (in my case a Naim device) connected directly to the Hugo 2 (rather than the "local" streamer, the 2Go).

 

[dcp10]

Thing is, way too many hardware vendors under-design and under-test their streaming products. There are zero technical reasons for FLAC to underperform WAV, except for bad design or bad implementation.

I think you're being a bit unfair here.

There's no reason to believe that Naim or Chord - or anyone else - is not processing FLAC files correctly. I'm sure they're doing this perfectly! The problems are much more subtle, and possibly relate to RFI or other "ephemeral" causes - related to bolting on yet more electronics right next to the DAC. Bear me out here...

Ever since the 2Go was released, there have been users pondering sound quality related to the SD card playback. For example:-

1. Darko, in his video review, thought that playing music from an SD card in the 2Go "sounded better" than the exact same music played from his iPhone connected directly into the Hugo 2. Why? Perhaps because mobile phones are intrinsically "noisy" devices? Perhaps because of RFI? Perhaps because they provide a less-stable streaming platform?

2. Various users on this forum consider that playback from playlists - controlled via Bluetooth and the GoFigure app - sounds better than the same music controlled via WiFi and mobile phone apps. It has been suggested that disabling the WiFi in the 2Go may improve the sound. Again, is this an RFI issue?

3. Other users on this form (@MarkParity, @SteveHulk, @manueljenkin) have suggested that playback of WAV files sounds (subtly) different to - and possibly better than - the exact same music played from FLAC format. Why?

4. I believe that feeding the Hugo 2 via an "external" streamer provides better sound quality than the "local" (2Go) streamer bolted on. Again, why? Over on the Naim Audio forums, many speculate that Naim provides a more stable streaming platform than other devices, which could impact the DAC stage - timing? Or is it again, separating "noisy" components - the "firestorm of electrical activity" that the Roon creators sought to avoid (as quoted by @SteveHulk).
.
Please don't misunderstand me: I'm not saying that the 2Go isn't a great device; I think it's a fantastic way of "feeding" the Hugo 2 on the go. As a "transportable DAC", this is a great add on. But it isn't perfect; it's a very small device, bolted onto the front of the Hugo 2, using USB input. And it isn't even clear that this is a Chord design: the circuit board seems to be designed by a third party (Disign Consultants Ltd), so we may not be getting the ultimate in audio engineering...

I'm curious about all this - it's a learning experience, an audio adventure. But for my purposes - trying to get the best out of my music, for a not-too-unreasonable price - I'm beginning to feel that this "all in one" system may be too much of a compromise. As they say, "your mileage may vary".

 

[dcp10]

Translation: "Our streaming processor is so underpowered (or buggy) that it cannot handle the trivial FLAC decompression algorithm." This is really total BS

Please see my comments above. This has nothing to do with failure to decompress FLAC files properly. I think we're talking about much subtler aspects, which are difficult to quantify ("just because you can't measure it doesn't mean to say it isn't there" - it's knowing what to measure that's the problem). It may relate to RFI caused by the higher amount of electronic activity - who knows?

But I wouldn't knock Naim: they have some of the best audio engineers in the industry. They're not driven by bean counters - they are passionate audio enthusiasts who actually listen to their products. And listen hard. I've visited their factory, and I can see the dedication in everything that they do. Even crazy things like loosening audio cables; using flexible power mounts; using vibration isolation inside the case; using massively-oversized transformers; the list goes on… It's just a shame that they've chosen a different approach to DA conversion from companies like Chord and dCS.

 

[gto88]

l know 2g can play music sent from other players through WiFi, but if it can
play music file on local NAS, that would be great.

 

[earnmyturns]

I think you're being a bit unfair here.

There's no reason to believe that Naim or Chord - or anyone else - is not processing FLAC files correctly. I'm sure they're doing this perfectly! The problems are much more subtle, and possibly relate to RFI or other "ephemeral" causes - related to bolting on yet more electronics right next to the DAC.

Completely fair. If their designers can't do the work to keep RFI out of the DAC, or whatever it is that makes FLAC sound less good, they aren't doing their jobs to the fullest extent. It may we be that achieving the required level of isolation would cost more, and their beancounters don't allow them to do their best, but it's not as if the gear we are talking about is cheap.

 

[dcp10]

Completely fair. If their designers can't do the work to keep RFI out of the DAC, or whatever it is that makes FLAC sound less good, they aren't doing their jobs to the fullest extent. It may we be that achieving the required level of isolation would cost more, and their beancounters don't allow them to do their best, but it's not as if the gear we are talking about is cheap.

But you were suggesting that Naim and Chord weren't decompressing FLAC properly. They are.

As far as "required level of isolation" is required, I don't think anyone in the industry has really got a handle on this, so it's unfair to demand this at this stage. You can't expect this equipment to be perfect - we're just showing you directions in which improvements can be achieved.

 

[dcp10]

I suspect a lot of the problem is that hard-headed engineers are focused more on what they can measure, at the expense of what they can't measure. Generally, Naim comes out quite well here, since their engineers really do listen to their devices - but again, it's difficult to fix what you can't physically (as opposed to emotionally) measure.

As far as Chord is concerned, the Qutest - which shares much with the Hugo 2 - does have galvanically-isolated USB, so it would be interesting to know if these "sound better" than the Hugo 2 (anecdotal evidence from reviewers and forum comments, seem to bear this out). I don't know whether this "fixes" other RFI issues or not. Anyone here compared a Qutest and a Hugo 2?

 

[earnmyturns]

But you were suggesting that Naim and Chord weren't decompressing FLAC properly. They are.

As far as "required level of isolation" is required, I don't think anyone in the industry has really got a handle on this, so it's unfair to demand this at this stage. You can't expect this equipment to be perfect - we're just showing you directions in which improvements can be achieved.

If one thinks carefully about the data and electronics path from a streaming processor to a DAC, there are multiple places where things could go wrong. Even at the software level, you are dealing with realtime processing when you are driving I2S or another synchronous protocol, and bugs in realtime processing -- even in critical safety applications -- are not unknown. Then there's clocking, EMI, ... But all of this is very well known from a technical point of view, it's not as if we are dealing with quantum computing or even GPS-level clock accuracy. So, if there's a reproducible loss of quality in FLAC vs WAV, it is caused by one or more flaws that are well within current science and engineering capabilities. I don't know why you are trying to excuse the industry for their mediocrity.

 

[muski]

With wifi, the 2GO2YU used with an optical fiber and Hugo2 are perfect if both are used with battery.

This is a really interesting way to use the 2Go...

Makes me wonder if an optical connection might cure the pops and clicks that impacts some users?

I'd always suspected that it was an issue related to the USB interface.

If anyone can test/confirm/deny, I'd be very curious.

cheers,
muski