TOPIC:

Hello,

when I have used a naim streamer (uniti or superuniti) for streaming it made a difference streaming wav instead of flac. wav did sound better. So I had activated on the fly transcoding from flac to wav in the UPnP server (minimserver).

I now have tried if it still makes a difference when streaming to a SHD Studio. => I cannot hear a difference. It sounds the same to my ears.

Did anybody else have tried this? What was the result of your test?

Greetings
Gert

There shouldn't be any difference since FLAC is lossless encoding. For it to sound different I guess there must a problem with the decoder.

Hello entripy,

certainly FLAC is lossless and so is the same as the wav data when comparing the resulting bits. But you can read of many streamers that sound better when feeded with there native bit resolution (frequency and bit depth). I did not believe this myself (because I know the meaning of bits and bytes) and so was very astonished about the better result of streaming wav to the naim streamers. (This was very clear to hear. I also asked some people what did sound better without telling them which stream is which format and without telling them what I did change generally.)

I do not think that the difference is because of faulty decoding algorithms. It probably has to do with the CPU or DSP having less to compute and transcode. Less computing may mean less electronic noise or less jitter in the resulting signal. I do not know the reason, but I know that there can be differences. The faster a CPU or DSP is the smaller the difference in sound will be I think.

I am asking if anybody did hear a difference, because if yes, I will do some further listening tests. Otherwise I say it does not make a difference wit the SHD.

Greetings
Gert

I use Audirvana, which converts my 16/24 Flac files to 32 / 192 signal before sending to My SHD which then processes that file and sends a 24/96 signal out to my amplifier.

Bottom line whatever we send to the SHD it will output a 24/96 signal to the amp. Some people believe ( I am one ) send the best signal you can to the SHD for it to process .

Theoretically 24/96sounds the same no matter how it gets there so this may be why you are not hearing a difference.

My system is a Win 10 PC using- Audirvana software,sending a signal to the SHD via upnp network connection, then out to a Crown XLI 1500 to Klipsch La Scala's

Hope this helps

Some info about FLAC and WAV.
Even FLAC without any compression doesn't sound the same as WAV, because the available sofywareplayers are also sounding different!!!
Compressed FLAC files have to be decompressed by software and/or hardware, which will "change" the sound for sure as well!
Foobar is sounding different as for instance Audirvana which is very beloved by in use by "high-end" users.

Remark: Yes and upsampling/oversampling could improve sound quality, even the best in DSD, but you can't use that format on the SHD!!!

Best sound in combination with SHD will be available with music files with 24Bit-96KHz, preferable in WAV or FLAC format !!!
WAV is in principle better as compressed FLAC, but flac gives easier and better metadata handling with cover/pictures.

Cheers!!!

Are you sure that decompressing a flac file changes its sound? Like unzipping a computer file changes it from the original file?

FLAC uses a lossless compression scheme that, when decoded, returns a bit-for-bit identical waveform that is indistinguishable from the original, be it WAV or whatever. Reports of sonic differences between FLAC files and the original WAVs are all based on sighted, easily biased opinions. Under properly-controlled double-blind testing conditions, these "audible" differences magically disappear.

As for sample rates: The miniDSP units resample every digital input to the "working" SR of the device. This resampling is asynchronous, which among other things means that resampling occurs even if the incoming sample rate "matches" the internal rate of the miniDSP device. Because of this, there is no "best" sample rate to send to the device. It's not a problem anyway since the ASRCs used are very precise, with artifacts appearing down in the -120dB neighborhood if I remember correctly - possibly even lower. We simply cannot hear that stuff, period.

A voice of reason! Thanks, Jim.

"Reports of sonic differences between FLAC files and the original WAVs are all based on sighted, easily biased opinions. Under properly-controlled double-blind testing conditions, these "audible" differences magically disappear"

No, sorry, this is not true for all streamers. Theoretically it is true, but there seems to be more than only bits and bytes in some solutions. In case of the old naim streamers the difference was clearly noticeable. To be sure I had done a blind test with my two sons and they only needed a few seconds of listening to tell the difference without knowing what I did change in the system.

A well built streamer should not make a difference. And with minidsp SHD Studio and my external DAC (integrated in speakers) I cannot hear a difference.

Here’s my theory on this. First, it’s given then a decompressed flac is 100% the same as the source wav. But adding a streamer into the mix might complicate things. What if the streamer had a terrible implementation (or insufficient hardware) such that either the flac decompression couldn’t be performed in a sufficiently timely manner or there was inadequate stream buffer to compensate for the flac decompression performance. In that case, you’d think that you would either have some weird timing effects or the streamer would be forced to drop data to maintain time sync.

Isn’t that why RAAT (Room’s streaming protocol) is so complex? I thought it took into account all kinds of in-depth streamer info in order to maintain timing and quality. Hence, why it’s taking forever for the SHD to pass certification.