Sample Rates 44.1 vs 96 (+) ?

  • Thread starter Thread starter Edisonsarm
  • Start date Start date
E

Edisonsarm

New member
SO - tracking at several different studios now and getting mixed input with people I work with. How much difference audibly am I going to get when listening back to my masters when things are done in 44.1 vs 96 or more? Should I deal with using up more memory and getting to use less plug ins etc w my protools rigs in exchange for a possible fidelity boost?
 
From what I understand (and my knowledge is gained from reading resource material and whatever I pick up on these forums - have pity on me :D ) setting a sampling rate higher than 44.1 will prevent aliasing of frequencies above 20khz (or, in reality 22khz). Frequencies that go above the nyquist frequency (1/2 the sample rate) will 'fold over' and become audible unwanted sound.

Setting a sampling rate higher than our 'human discernable' audio spectrum (nominally 20hz - 20khz) allows for upper harmonics from synthesisers, possibly some subliminal brightness from crash cymbals, tube-driven amps etc to be captured without causing interference through alisasing. [edit2 - if you're recording an orchestra, there are arguably heaps of ultrasonic (harmonic) frequencies that happen naturally with strings and brass, so using a higher sampling rate to capture the live performance will preserve some of the original 'character' of the instruments]

[edit - I believe most current software should have a built-in anti-aliasing filter to kill frequencies outside the sampling range to prevent this from occuring. But please correct me if I am wrong]

But the recording (either in mixdown or track by track) needs to be dithered down to 44.1 for playback on CD, so I haven't really found the need to limit myself to 8 track recording by setting a higher sample rate that will kill my computer's CPU when everyone will be listening to the track in 44.1 anyway :)

Dags
 
Last edited:
You shouldn't have to worry about any alias frequencies when recording at 44.1. I'm pretty sure any decent piece of equipment has a low pass that will drop off frequencies above 20K, and recording at 44.1 will put you just over that. What you will gain when recording at 88.2 or 96 will be better clarity, but you have to decide if it's worth using the extra hard drive space and processor power. If you do want to record at a higher sample rate, my advice would be to record at 88.2 if you plan on putting it on CD, it's cleaner math from 88.2 to 44.1 as everything is just cut in half.
 
I can recall reading somewhere that recording using higher sample rates can also help with spatial representation of the individual instruments - probably only really applies to orchestral recording though - as the extra psychoacoustics in the ultra-high frequency range allow for microsecond delays between sound sources to be accurately captured therefore....<snore> :rolleyes: :D

Dags
 
Most of the high-end converter designers have said "If you can hear a difference between different sample rates, your converters are broken."

In real life, some converters sound better at some sample rates than they do at others. It has more to do with that converter than it does with the sample rate. That's why the experiences are all over the board - there aren't that many people with more than one set of converters to test, so they assume that if their converters sound better at a certain sample rate, all converters sound better at that rate.

whether or not tha math is cleaner converting between 88.2k and 44.1k than 96k to 44.1k depends on the type of SRC being used. Most of the better ones upsample into the Mhz range and resample at the target frequency, so it doesn't matter.
 
WELL - up until now I've just been converting through my digi 002. BUT I just bought an API a2d mic pre (which also converts - bypassing my digi 002... and from what I understand... is a better converter). So, in my case I do have 2 to compare. It's a tough call... but 44.1 leaves me so much more editing and mixing options... I'd love to keep recording that way, again I just don't want to sacrifice fidelity.
 
You won't sacrifice fidelity.
I guess if what you're hearing being played/sung sounds fine once it hits the converters and is recorded onto the HD, then there's no reason to suspect there's any problems.

Dags
 
I've always done it at 48kHz. Dunno why really. It sounds no better at 96k, but then I can't hear a difference between it at 44.1 either.

Bottom line - 44.1kHz as a sampling rate is unlikely to ever be the limiting factor in your recordings.

That 88.2 into 44.1 maths thing is bogus, btw. If you tried cutting out every other sample you might not like it, because it just isn't that simple. It's just one of those things that does the rounds on the forums every so often.
 
You must log in or register to reply here.
Back
Top