File-based audio playback - compressed vs. uncompressed

[bobbasrah]

Funny you say that Stan i find listerning to headphones in the snow makes them sound eerily different

I know that P in snow is silent

[bobbasrah]

Ok, so after that brief diversion of talking wet, we have eatablished that on two different computers on two different systems there were no detectable or notable differences here. So apart from clutching at various straws, the principal variables are resolved, but what is causing this elsewhere?
Stan and Andre have both noted differences, but has this likewise been tested on different computers and audio systems, and if so what are we left with as a cause?
The player ?

[AlexM]

Ok Alex, so with async DAC the timing control is from the DAC, that I understood, but does that mean that the data transfer method is changed from that which would apply say for a drive?
Sorry if that sounds a bit thick

No, not at all.

There is no timing information encoded within a FLAC/WAV file, other than symbolic information about sample rate, bit depth etc. in the header. Music stored on a computer’s hard-drive is transmitted using packet-based bus protocols that don't have to be syncronised to the data rate of the encoded file. The timing of the internal busses of are to ensure that the data lines flip at the same time, and their logic levels are read at the same time - nothing to do with music.

Only at playback when audio is streamed through the D/A converter does timing information, and hence jitter, matter. For a DAC this is usually under the control of the local oscillator, along with and reclocking by PLLs or sample rate converters.

USB interfacing has it's own problems, but ayncronous mode USB is employed to remove the dependency of the DAC on the host computer, i.e. allows the local clock to be decoupled from the input clock.

Stan - what USB reciever chip do you use in your DACS?. I'm sure you can give us chapter and verse on this.

Alex

[AlexM]

You lot make I larf.
The very notion that your going to "explode" any audio myth on an internet forum is as delusional as many of the claims made by the hard core subjectivists.

I'll let you lot exercise your keyboard warrior skills and go do something more useful and more entertaining.

You're probably right, but I'm interested to understand what lies behind the subjective experience that people have. There isn't a worthwhile argument that FLAC compresses without throwing data away, but if the subjective comments are real then there is something wrong with the players or other implementation problems.

Anyway, please don't let me keep you from your more useful and entertaining pursuits!

[bobbasrah]

The frenetic pace seems to have cooled somewhat just as it appeared that the variables were being narrowed down.....

So Alex, while there is a short pause, see if my understanding is clear on this....
The data is transferred to the usb port (in the case examined so far) and onward as normal, but it is at the delivery end (DAC) where timing issues may arise as the data is converted into a digital audio stream (with timing incorporated), for conversion by the DAC into analogue?
This is complicated if the DAC is subject to control by the computer ie non-async?
So this is a data buffering and controlled feed issue ?

[AlexM]

Bob,

I think that the trouble is that there are too many variables to correlate with the subjective observation, i.e. My DAC is jitter insensitive, yours is, mine uses SPDIF from a SB Touch, yours is via async USB, firewire or something else.

I wanted to go through something of a process of elimination to work out which elements of the chain might be influenced by flac encoding/decoding. There is no inherent reason why playing an AIFF, WAV or FLAC copy of the same LPCM data should result in a different data stream being output from the source.

My DAC (or CD player actually) resamples everything to 24/384khz, and the data is deeply buffered to remove jitter. Maybe that is why I can't tell the difference whether FLACs or WAV files are sent to the streamer, even if you accept that local processing overhead has some impact on the timing of the spdif stream from the SB touch.

It would be interesting to see if there is a difference WAV to flac when using the SB Touch via a USB memory stick plugged in the back - at least that takes the server out of the equation. I'll give this a go and see what happens.

Cheers,
Alex

[Ali Tait]

Even the Touch buffers though.

[Welder]

How about approaching this from another angle and researching who says they can hear a difference between Flac and WAV and then consider why it might be in their interest to make such a claim.

Course, if I was manufacturing/selling/promoting a Dac or even another piece of digital Hi Fi equipment and suggested I could hear the difference between a Flac and a WAV file using my equipment might not that generate even a small bit of curiosity in prospective buyers.
Not that I'm a cynic or anything.
Of course, the way to demonstrate its bullshit is to sit the person who claims to be able to tell the difference down with the kit they claim to able to hear the difference on and play them Flac and WAV files.

[bobbasrah]

John, though you had better things to do that listen to this mystery tale ???? Nose itching???? Or it's your inner cynic?

Fair enough on your trial Alex, that is why I have been waiting for over 2 hours now to hear from Stan and Andre what they used to further refine the culprits for the difference they noted.
I think Andre mentioned he was using SB, and noted a difference, although his following statement regarding FLAC compression puzzled me as it implied FLAC was being used in spite of the difference I assumed in favour of WAV. Better go back and re-read that I guess...

Given that there is no theoretical difference between FLAC and WAV, and this is borne out by my own experiments, the only logical conclusion I can come to is that it is some other process, and not the container itself.

PS - I play async usb out to the DAC, but don't have SB or other setup as have no need for them.

PPS - Andre's post - As i stand now i can hear a difference between the original & FLAC burn, with the streamer i use now i'm happy at that & have no desire to try others (less potential headaches) .I did however try different encoding levels when ripping, i couldnt detect any difference so chose to keep the Default level...
My recollection was incorrect.

[Alex_UK]

My boss used to complain that our monthly sales figures were terrible. I stopped using Zip to send him the Excel spreadsheet, and there was an immediate night and day difference in the quality of our figures! Quite remarkable.