Head-Fi.org › Forums › Equipment Forums › Sound Science › A 96khz sample rate question
New Posts  All Forums:Forum Nav:

A 96khz sample rate question

post #1 of 12
Thread Starter 
Listening with IEMs I’ve acquired a habit of listening to flac files at 96khz 16bit. I haven’t come across anything that describes what I’m hearing. So I’m sorry if this is discussed in more detail elsewhere. Wading through various discussions I’ve not found an explanation. Also, I don’t want a sound quality debate.

What I’m curious about is why using a higher sample rate widens the soundstage and gives more instrument separation. When I listen to music at the 44.1 khz (CD) sample rate a band will seem to me to be in a tight arrangement and inside my head in comparison to 96khz sampling. At 96khz the instruments are more recessed and separated.

Does anyone know if what I’m experiencing has anything to do with the fact that the 96khz sample rate is a surround sound standard? None the less, there is definitely a greater separation of instruments at 96khz. I’m simply curious about why this might be happening, and what is different about the way in which 96khz sampling is handled?

post #2 of 12
Originally Posted by KMASCII View Post
What I’m curious about is why using a higher sample rate widens the soundstage and gives more instrument separation.
It does not as far as I am aware, nothing about 96K sampling should change the soundstage at all in and of itself, 96k sampling does a few things , it increases the bandwidth and makes filtering easier, due to the filtering differences it does push some noise out of band but it does not alter phase characteristics as far as I am aware.

Which track in 16/.44.1 and 16/96 are you using for comparison, are you sure they are the same mix , can you post snippets for us to analyze, cheers.
post #3 of 12
Thread Starter 
Yes, this is really curious for me... the widening occurs for ALL albums that I encode, and isn't characteristic of just one label, band, or whether HDCD or not. This is why I'd ask if it was my software, it's dBPoweramp. But I don't think that this should matter because codecs can't change?

My rips are CD-->Wav 16/96-->flac 16/96. I do the conversion to flac simply for space saving. Oh, and these characteristics are the same in both wav or flac formats. I'll try to grab some snippets and post them.

Thanks nick_charles.
post #4 of 12
So you're ripping 16/44 CDs to 16/96 WAV/FLAC?
post #5 of 12
Thread Starter 
Yes, they're resampled from 44.1khz to 96khz, so I'm curious why in a resampling would the soundstage be altered?

Here are the sound snippets:
From a regular CD, My_Brilliant_Beast.zip - FileFront.com
From an HDCD, The_Cars.zip - FileFront.com
post #6 of 12
You can also apply the resampling in real-time during playback, with a plug-in.
post #7 of 12
the difference might be caused by resampling to a samplerate that isn't a multiple of the source file's samplerate

and there's no reason to resample the files themselves you don't gain anything you can't bring back audio information that doesn't exist in the source file/stream
post #8 of 12
I'm with nick_charles. It's possible your DAC has better phase and/or amplitude linearity in 96Khz mode, particularly in the high frequencies. Since you're resampling from 44.1Khz files, that is the only possible reason I can think of.
post #9 of 12
My belief is that oversampling improves the signal to noise ratio, allowing low level ambiance information to become audible again. Information that was always there is now audible.(The least significant bits are raised above the residule noise floor.) Got no way to prove it, though.
I like nicks_charles idea also.
post #10 of 12
Googled it-

The Computer Music Tutorial - Google Book Search
post #11 of 12
96/16 and 44/16 have exactly the same bit depth, and signal to noise. The 96khz is the sample rate, and it means it has information up to 48khz, or approximately 3 times what most humans can hear. Since you're starting with 44k sample rates, which means the max that was actually captured on the disc is 22khz, that means that anything above that is noise generated by the sample rate conversion. Maybe this is giving you the illusion of greater soundstage, or maybe it's just a mind trick. It's definitely not from improved signal to noise ratio, tho.
post #12 of 12
Oversampling and upsampling are two different beasts although supposed to achieve the same benefit. For DACs with good oversampling upsampling can degrade quality and vice-versa.
Theta Digital - Upsampling
New Posts  All Forums:Forum Nav:
  Return Home
  Back to Forum: Sound Science
Head-Fi.org › Forums › Equipment Forums › Sound Science › A 96khz sample rate question