advertisement


WAV better than FLAC due to increased processing load on the CPU of the latter?

item said:
I've claimed nothing controversial to a reasonable person: everything I've experienced points to a subtle audible difference between different file formats in a low-noise system. Hundreds have found something similar. Placebo? Collusion? Imagination? Maybe. But there's a plausible pattern, and something different is actually happening during playback, so - like Naim and Linn - I feel it would be dumb to dismiss it out of hand.
Click to expand...
Even dumber to claim you can hear a difference then, when confronted with having to prove it, admit you can't.

I'd use the words 'feet of clay' but don't think anybody believed you to begin with.
 
Here is a statement from a Linn engineer:

"We have done extensive measurements on power supply disturbance recently, and have compared results for both FLAC and WAV streaming. Our findings are as follows :

1. If we measure the power rail that feeds the main processor in the DS we can clearly see identifiable disturbance patterns due to audio decoding and network activity. These patterns do look different for WAV and FLAC - WAV shows more clearly defined peaks due to regular network activity and processing, while FLAC shows more broadband disturbance due to increased (but more random) processor activity.

2. If we measure the power rails that feed the audio clock and the DAC we see no evidence of any processor related disturbances. There is no measurable difference (down to a noise floor measured in micro-volts) between FLAC and WAV in any of the audio power rails.

3. Highly accurate measurements of clock jitter and audio distortion/noise also show no difference between WAV and FLAC.

The extensive filtering, multi-layered regulation, and careful circuit layout in the DS ensure that there is in excess of 60dB of attenuation across the audio band between the main digital supply, and the supplies that feed the DAC and the audio clock. Further, the audio components themselves add an additional degree of attenuation between their power supply and their output. Direct and indirect measurements confirm that there is no detectable interaction between processor load and audio performance."

See here.

So, if you do it right, there need be no difference.
 
I still uphold that the only possible reason for a difference could be designers and designs that are simply not up to the task, no doubt the designers of those substandard products with woefully inadequate PSRR will try to tell you some patently bullshit line about their equipment bring more revealing, when the opposite is in fact the only logical answer.

No one is surprised with the outcome here and the dailies of boastful snake oil sellers to back up their claims with deeds.
 
sq225917 said:
I still uphold that the only possible reason for a difference could be designers and designs that are simply not up to the task, no doubt the designers of those substandard products with woefully inadequate PSRR will try to tell you some patently bullshit line about their equipment bring more revealing, when the opposite is in fact the only logical answer.

No one is surprised with the outcome here and the dailies of boastful snake oil sellers to back up their claims with deeds.
Click to expand...

+1

"WAV sounds better than FLAC" = "I don't know how to design equipment that plays FLAC properly"
 
In my system wav sounds better than flac when using Squeezebox Touch. It is just a cheap good streamer. I have a nice Super Teddy ps and applied TT 3.0, so it is much better than standard SBT, but generally construction is the same.

Maybe Linn, Naim? and others did some research in this area (i hope so) and their devices are constructed in such a way, that there is no difference.

It means for me - everybody should make their own experience with own system not believing in some general opinions as flac=wav.

:)
 
Linn has some interesting papers.

Budget system - router, computer, streamer:

DSBudget4.png



http://docs.linn.co.uk/wiki/index.php/DS_Budget_System

For one entry system isp router, second router, computer, nas and streamer is adviced:

Entry5.1.2.png



http://docs.linn.co.uk/wiki/index.php/DS_Entry_Level_System

For one high-end system a DEDICATED SWITCH for streaming connection is adviced.

Are they so stupid :D to take a dedicated switch ?

High6.png



http://docs.linn.co.uk/wiki/index.php/DS_High_End_System

:D:D:D
 
Do they say why they recommend four separate network controllers in a high end system? The reason I ask is that bandwidth (while the kids are downloading HD porn) and flexibility/budget may be the reasons for these suggested configurations. We are talking about networks after all, not analogue audio componentry.
 
mudlark said:
Fascinating, but completely wrong. I think it would be difficult or impossible to detect any difference between flac and wav in any system.

:)
Click to expand...

Why wrong? If you don't take the steps Linn do, it's perfectly reasonable for the different kinds of noise they measure to differentially and discriminably affect playback in the ways they describe it might. There's no law of the universe that says FLAC and WAV do sound the same kn all systems, just an expectation on my part that they ought to in a well engineered system. In contrast to Linn, Naim Audios own engineers say on the Naim forum that WAV on their systems sounds and measures better than FLAC, from which I conclude that Naim stuff isn't well engineered.
 
tomek said:
Are they so stupid :D to take a dedicated switch ?
Click to expand...

A router will blindly route all network traffic to all connections. A switch will try to intelligently route traffic only where it is needed. Most devices called routers that you buy in the shops are actually switches, so there is no need to worry about it unless you are getting very obvious glitches or cut-outs. None of it will affect SQ.
 
striped67 said:
A router will blindly route all network traffic to all connections. A switch will try to intelligently route traffic only where it is needed. Most devices called routers that you buy in the shops are actually switches, so there is no need to worry about it unless you are getting very obvious glitches or cut-outs. None of it will affect SQ.
Click to expand...

A router is not blindly routing a networkk traffic. I was writng about it before - a router is a mini computer with embeded OS (mostly Linux) and EACH packet is examined for different criterias - where it has to go next, firewall, QoS.

One non managable switch is working on different network layer (MAC Adresses) and is just very fast forwarding packets in any direction. Fast in, fast out. In some more expensive switches one can configure some other options like QoS, virtual networks etc.
 
item said:
I'm probably the best authority on what I am claiming, so let's cut through the inflammatory rhetoric and put it plainly:

I've never claimed that CAT7-standard wire sounds better than CAT5. It is better built and better specified, and not a lot more expensive. What's not to like?

Like many others, I have claimed that I personally find a slight, on the threshold of audibility, difference between WAV and FLAC. In tests I've conducted on half a dozen listeners, in low-noise systems, those best able to differentiate such cues find the difference more palpable.

Under pressure, in front of an audience, acuity-suppressed, with my reputation at stake, blind AB-X'd into confusion, would the difference be demonstrable? Probably not: better ears than mine have failed to identify grosser differences in such tests. Later, relaxed, would it again become apparent? Probably: typically brain behaviour.

Testing me isn't the point: shooting the messenger doesn't change the message. Many, many keen listeners, in the the comfort of their homes and studios increasingly find there is a small difference: not making great claims for their brilliance, just communicating their experiences.

Seeking to connect an abstruse discussion of file formats with 'disseminating fear' and/or hyping equipment sales is patently, pathologically ridiculous. Elsewhere, this is being discussed rationally. Here, the conversation hasn't begun.
Click to expand...

How can the irrational be discussed rationally, Item?

And as for your dismissal of ABX, this is reason number 3 usually put forward by foo merchants as to why a test procedure which works in every other field is invalid in the world of hifi.

I'll bet even your wife heard the differnce as well!

Chris
 
A 'switch' or 'hub' broadcasts all packets to each and every attached node. Great for using tcpdump to snoop. Also fun with Etherpeg software which shows all jpegs passing through your local switch (pin the pr0n on the spotty bloke in accounts!)

A router has the ability to direct packets to separate interfaces, and pull off tricks like network address translation. It therefore has to be able to run a tcp/ip stack, and may or may not be broadcasting packets from one socket to another.

If your network DAC were badly designed it could allow signal on its network port to pollute the audio signal. But it would have to be very very very stupidly designed to do that, since from an audio p.o.v. a network port is wild screaming epic noise, or not working at all. This brings us back to the point which item never answers: if you can hear tiny voltage fluctuations on your DACs input which are not part of the signal, why can't you hear the screaming raging brutal sound of the signal itself?
 
Switch and hub are not exactly the same thing.

http://en.wikipedia.org/wiki/Network_switch#Function
An Ethernet switch operates at the data link layer of the OSI model to create a separate collision domain for each switch port. With 4 computers (e.g., A, B, C, and D) on 4 switch ports, A and B can transfer data back and forth, while C and D also do so simultaneously, and the two conversations will not interfere with one another. In the case of a hub, they would all share the bandwidth and run in half duplex, resulting in collisions, which would then necessitate retransmissions. Using a switch is called microsegmentation. This allows computers to have dedicated bandwidth on a point-to-point connections to the network and to therefore run in full duplex without collisions.
Click to expand...
 
item said:
Testing me isn't the point: shooting the messenger doesn't change the message. Many, many keen listeners, in the the comfort of their homes and studios increasingly find there is a small difference: not making great claims for their brilliance, just communicating their experiences.

Seeking to connect an abstruse discussion of file formats with 'disseminating fear' and/or hyping equipment sales is patently, pathologically ridiculous
Click to expand...
The elephant in the room, right now, is revealed when you answer one simple question. Whose interests are most advanced by the possibility, unlikely that it may seem, that FLAC compression results in an inherently different sound?

Taking a glimpse around the current marketplace, there appears to be something remarkably laughable happening. Specialist audio manufacturers are producing toys styled "digital audio players" that do little different to a low-powered generic computer. Fit the latter with a decent sound card and you have a solution that has satisfied the world of professional audio for a generation. But of course, these days, you don't even need a sound card as USB has become the interface of choice for many users.

The cost differential between these two approaches borders on the immoral, IMHO. But whatever your view, there are clearly highly driven commercial interests at work here. The main requirement, at this moment, is for the domestic digital audio business model to differentiate itself from the economics of the general purpose computer. The consumer must be convinced that spending £nk on "dedicated" audio playback is buying him something more than can be achieved with a few hundred quid's worth of generic kit.

To achieve this end, certain interests are doing their utmost to endue computer-originated audio with the same kind of mystique that has, in the recent past, achieved the major own-goal of alienating the mainstream public from high quality audio, associating it with the 'phenomenon' of thousand dollar cables and three hundred dollar wooden feet.

Anyway, back to the point. The difference in computer load between FLAC & WAV is pretty minimal and difficult to determine. There's also an element of roundabouts and swings as the latter may well involve more disk effort to offset the necessary decoding of the former.

So it stands to reason to make the problem far more radical. Run a benchmarking program at the same time as playing music and determine the audible difference caused by the computational intensity of running the Fibonacci Series while listening. That way, any degradation caused by "computational noise" (whatever that is) will be instantly apparent. Saturate your CPU with other tasks while listening - does that make a difference to the sound?
 
tomek said:
A router is not blindly routing a networkk traffic. I was writng about it before - a router is a mini computer with embeded OS (mostly Linux) and EACH packet is examined for different criterias - where it has to go next, firewall, QoS.

One non managable switch is working on different network layer (MAC Adresses) and is just very fast forwarding packets in any direction. Fast in, fast out. In some more expensive switches one can configure some other options like QoS, virtual networks etc.
Click to expand...

Ok, so I'm talking about a hub, which I assume is what the Linn diagrams are referring to when they say router. The point that Linn seem to be making is that using a switch with all your devices attached is better than a hub since you are creating a network segment that only contains the audio data and won't get bombarded with all other network traffic. My point was that most consumer devices are switches anyway, so there is not much need to worry about it.
 
@Plutox, you have the nub of the issue right there. Anyone involved in selling computer audio who isn't primarily involved in selling computers is looking to gouge the customer by telling them there is a problem and then telling them they have an expensive solution to it.

Sites like Computer Audiophile had the chance to eradicate this shit from this new side of the industry before it started. Sadly that opportunity was not taken as media outlets decided to get into bed with their sponsors.
 
You must log in or register to reply here.


advertisement


Back
Top