Sound quality difference with higher sample rates

[jwd65]

Hi All,
I imagine this has been covered ad nauseum (especially when the 002 debuted) but wondering what people's thoughts are regarding an increase in audio quality when using higher sample rates.
To my ear, I hear very little difference in tracks I've recorded at 44.1 and 96. I imagine acoustic instruments and vocals would particularly benefit from higher sample rates but I'm still not hearing much difference.
Your thoughts?

J.D

[er1c]

Higer sample rates are usful for internal processing (plug-ins are better accruacy in higer sample rate)

[bb_aus]

the human ear con not hear the full benifit of 96KHz (and above, i cant not see any point of 192KHz at all) as mentioned above its usful for internall processing and the higher frequencies are clearer. and yes things like acoustic tracks you will hear more of a difference than say with rock or other music. however if you ar going to be putting ur finished product onto cd you will have to down sample to 44.1KHz any way
i never use 96Khz i do not see the point as there are only relitive small advantages that you will actually hear (when you will have to down sample anyway) and it required twice the system resorces
i work generally work at 44.1 in 24-bit or sometimes 48/24
the advantage of recording in 24bit far off sets the extra storage space and does not require faster processing
however with saying that i would not buy a main piece of equipment that did not support 96Khz bacause im sure at some stage i will be dealing with 96KHz and the 96K converters are a generally better quality (i stress the generally while comparing similar rage of converts, lots of people still happy using their 001 though)

[protein]

If you've got Reason you can do a little experiment and change the resolution while it is playing back a song. If you play something and switch between 44 and 48 you can hear something happening to the to the sound. The top seems more defined and I think everything seems to have a little more depth. Almost like it's come a little more into focus.

Interestingly if you put the resolution up to 96 there isn't much of a change. On my set up anyway.

[xeetstreet]

Higer sample rates are usful for internal processing (plug-ins are better accruacy in higer sample rate)

I think you may be confusing oversampling with increasing bit depth. Increasing bit depth within a plugin allows for an increase in internal dynamic range. Oversampling is typically used for interpolation (as in the case of a soft synth).

Theoretically, you could capture freqencies up to 48kHz (Nyquist frequency) when recording at 96k. If you high pass your recordings at 18-20kHz, you could make a song that only your dog could hear.

[Riad]

I would have to be careful of that, because my dog is a nasty critic. And I have a NyQuil theory, but I think that's a different thread.

[Captain_Pants]

The only person that can hear and mix songs at 192khz and in 24 or 32 bit is Bob Katz.

[Matt Whritenour]

the human ear con not hear the full benifit of 96KHz (and above, i cant not see any point of 192KHz at all)

i think u are confusing frequency response with how many times a piece of audio is sliced up per second.

it sounds to me like you are stating that 192KHz is pointless because the human ear can only
hear 20Hz-20KHz (it can actually go a little higher with harmonics), but 192KHz sample rate just means the audio is sliced up 192 thousand times per/second which allows for a clearer more acurate signal to be stored onto your harddrive.

[Phil O'Keefe]

i think u are confusing frequency response with how many times a piece of audio is sliced up per second.

it sounds to me like you are stating that 192KHz is pointless because the human ear can only
hear 20Hz-20KHz (it can actually go a little higher with harmonics), but 192KHz sample rate just means the audio is sliced up 192 thousand times per/second which allows for a clearer more acurate signal to be stored onto your harddrive.

Actually, bit resolution has more to do with S/N ratios and amplitude accuracy... sample rate is more determinative of the maximum limit of high frequency reproduction. The Nyquest theory specifys that the maximum reproduceable frequency (short of aliasing) is 1/2 of the sample rate. In THEORY, a 96 KHz sample rate can give you a 48 KHz HF response. Does that make an audible difference? Or more accurately, is the increased sample rate, in and of itself, the source of any audible differences? THAT issue is still a matter of considerable debate. I tend to think it is more a function of increased converter accuracy and quality than it is the actual sample rate reproducing frequencies that are beyond the limitations of human hearing, but then again, I am no expert in psychoacoustics. But IMO, if you could build a "perfect" 48 KHz converter, I don't hink you'd be able to hear a difference between that and a perfect 96 KHz converter - but since no one has made a perfect converter yet, I can't test that theory.

[Captain_Pants]

Hmmm...

Actually Phil, You are wrong.
Roy Howell makes the perfect converters in his basement. They should be at NAMM this year.