Wave or Flac

Rip to FLAC and transcode to WAV when playing.

Best of both worlds, files are smaller and tagging is less challenging but you get the benefits of WAV when playing which does sound better on Naim streamers.

Richard

+1 but why not have a listen and see what you prefer

Graeme

I have seen this before how do you transcode to WAV from NAS ?

Such a question can arouse religious type responses. However I rip to FLAC and transcode to WAV when streaming to Naim for sonic preference reasons.

Simon

Bert, the transcoding is done by the UPnP server app rather than the NAS, unless you have installed the UPnP server onto your NAS system.

i use Asset which transcode most formats other than ALAC.

Simon

The best way forward is to try for yourself. I prefer WAV. Some people can hear a difference, some can't, I can. You may as well ensure any effort on your part is worth it before you adopt a regimen.

okey i understand but why not ripping from the begining to wave isn`t it  better than to rip  to flac and transcode to wave

WAV takes up considerably more disc space (and don't forget the time required for back up runs).

And tagging is limited. Even if your library software can deal with it, not every streamer will. FLAC is a lot more sophisticated in this respeoct.

Thats not actually correct. If both formats are at zero compression then usually flac is slightly larger due to the additional tagging information. I have all my flac at zero compression.

Graeme

Well Foxman, this is your way! 

But generally FLAC is used at a certain compression level (standard is level 5).

Using FLAC at the (non generic) level 0 is kind of defying its purpose and not rendering any advantages. I am sure you will be of a different opinion. So be it.

PinkHamster

Sorry if my reply came across wrong, it wasn't intended so . I totally agree with your comment, and to be honest i store them at zero compression for no other reason than worrying that the NAS may be working too hard to decompress and transcode the files At the same time and so affecting SQ, although i have no proof that this is the case.

Graeme

Graeme, also sorry, if I came across too strongly.

Actually you don't have to worry about the computing capacity on decoding FLAC files, even at highest compression level. This capacity is only required on ENcoding the files. This is part of the concept of FLAC.

I think it is more to do with my lack of understanding of the technology, more so at the beginning when i started ripping. Wrongly confused compressed lossless with lossy, and hence assumed that compressed flac lost some information.

the main problem is hi res albums as they eat disk space. It maybe a job for the future and convert to level 5 compression to reclaim some disk space.

Graeme

Go for it!

Herbivore greets canivore

FLAC is very easy to decode and can be done on small portable players, it was designed to be so.  After all, you only compress it once into, say, FLAC 5, but you're going to decode it lots of times, each time you play it.

A NAS has no problems whatsoever.  My NAS decompressing FLAC and transcoding it to WAV uses about 1% CPU on the process.

Given that file compression involves removing bites, does transcoding involve interpolation?

When you decompress you haven't lost any information.  Then the transcoding takes place.

Right. I can get my little brain around that, so what gets taken out of the file in order to compress it? 

FLAC is a bit like steno.

So instead of saying: "Darling, I love you, would you please marry me?", I would just say: "You woman, I man." As long as it is set, that the two mean exacly the same, the latter can be translated back one on one.

I really cant get my head around how you can compress something without losing something, but ill just add it to the long list of stuff i cant get my head around.

Does compression level 5 mean its 50% compressed, i mean if a file was 1Gb and you used level 5 on it would it be 500mb

No, FLAC 5 reaches a file size which is about 30% below the uncompressed WAV file. There is no 100% rule as to the resulting file size, it really depends on the complexity of the file content.

Well, look at zip files. Same thing!

Foxman.. This is all linked to something call information theory entropy developed in the 1940s. Essentially this means that when the mechanism used to convey the information is inefficient and contains random redundancy It can be compressed without losing information. That is the encoding contains more data than is required to convey the information.

In information theory, data and information are not the same thing. This is the concept applied to many technologies in our modern digital world.

Now with FLAC there are two key lossless reduction techniques used:

1) LPC; linear predictive coding. This kind of looks at the audio signal, and by its nature of being an audio signal, as opposed to a computer executable for example, predicts what future values might be, and if the prediction is not correct, then register that fact in the encoding.

2) Rice Golomb entropy encoding; this kind of looks at the above encoded data as a series of data, breaks the data down into a string of quotient and modulus remainder pairs.. thereby reducing wasted data by creating a more efficient encoding of the information.. In this case the LPC info.

http://en.wikipedia.org/wiki/Golomb_coding

So put simply maths has allowed FLAC to find a more efficient way to convey the PCM sample information rather than the inefficient way of conveying that same sample information as an actual series of binary PCM samples as used in a WAV or AIFF format.

Simon

As always Simon your explanations are exemplary, even if i may not understand it fully. I can kind of understand how it can compress and drop stuff that may not be required but how does it know what to put back when it decompresses.