48 vrs. 96K resource requirements
- Thread starter Arranger
- Start date
Bongo Fiend
New member
Yes, precisely.
You probably won't be able to tell the difference by listening to the sound normally. However, using settings higher than CD-quality (e.g. 96KHz, 24-bit) you have more headroom for messing around with a sound without reducing its quality.
However the settings described above would use over 3 times as much hard disc space as normal CD-quality!!
Take it easy
You probably won't be able to tell the difference by listening to the sound normally. However, using settings higher than CD-quality (e.g. 96KHz, 24-bit) you have more headroom for messing around with a sound without reducing its quality.
However the settings described above would use over 3 times as much hard disc space as normal CD-quality!!
Take it easy
Thanks, Bongo Fiend. I had a feeling it would. Now I'm thinking of picking up a leftover Lexicon Core2 (I'd like 4 analog inputs) instead of a Layla24, at less than a third the cost. Will I miss the Layla24? I can also get a Core2 "daughterboard" chip that puts some effects right on the Lexicon board for "resource conserving" effects. Maybe it's a good idea, in theory.
Who cares about sample rate?
Hi ...
It isn't really worth it to record at 96kHz for the time being. The perceptible sound difference is very small .. if it even exists at all. Remember, the human ear can only really perceive frequencies up to 20kHz (give or take).
Currently bit depth is a more important factor. Record at 24bit if you can, as you'll get more dynamic range (effectively more volume "ticks") than at 16bit. You CAN actually hear this difference, and digital mixing will be easier and harder to over-load because of this extra range. Some software (like Steinberg's Nuendo) has 32bit float internal mixing, and it's nearly impossible to clip out a mix with it in my experience.
If you have the hard disk space, record in 24bit. Save the 96kHz recording for when DVD Audio is mainstream.
Regards,
-=(stu.macQ)=-
Hi ...
It isn't really worth it to record at 96kHz for the time being. The perceptible sound difference is very small .. if it even exists at all. Remember, the human ear can only really perceive frequencies up to 20kHz (give or take).
Currently bit depth is a more important factor. Record at 24bit if you can, as you'll get more dynamic range (effectively more volume "ticks") than at 16bit. You CAN actually hear this difference, and digital mixing will be easier and harder to over-load because of this extra range. Some software (like Steinberg's Nuendo) has 32bit float internal mixing, and it's nearly impossible to clip out a mix with it in my experience.
If you have the hard disk space, record in 24bit. Save the 96kHz recording for when DVD Audio is mainstream.
Regards,
-=(stu.macQ)=-
Bongo Fiend
New member
Re: Who cares about sample rate?
I agree that using a higher bit depth is more important than increasing the sample rate (as it is very common, for example, to take a quiet sample and turn up the volume, thereby drawing attention to any 'jerkiness' in the waveform).
However remember that:
(1) As we can hear frequencies up to around 22KHz, you actually need to record at at least 44KHz (you need to double the frequency to reliably record sounds at that pitch). Hence the figure 48KHz being often used (gives a little headroom).
(2) If you want to manipulate the pitch of samples, it's well worth recording them at a very high sample rate (especially if you're time-stretching, or drastically slowing samples down).
stu.macQ said:
I agree that using a higher bit depth is more important than increasing the sample rate (as it is very common, for example, to take a quiet sample and turn up the volume, thereby drawing attention to any 'jerkiness' in the waveform).
However remember that:
(1) As we can hear frequencies up to around 22KHz, you actually need to record at at least 44KHz (you need to double the frequency to reliably record sounds at that pitch). Hence the figure 48KHz being often used (gives a little headroom).
(2) If you want to manipulate the pitch of samples, it's well worth recording them at a very high sample rate (especially if you're time-stretching, or drastically slowing samples down).
96-48-44 .... nice lady )
Hi ppl. You got it all twisted up. When you're talking about the physical frequencies... yes.. the human ear can take it up to 20KHz, but we are talking about SAMPLING FREQuencies.... it's more of a technical difference. If we sample a simple analogue sound at a samplerate of 96KHz it happen that the sound is chopped 96 times a second. Now you get it? The higher the samplerate is the accurate the sound is. 10x for visiting FP. Cheers.
)Hi ppl. You got it all twisted up. When you're talking about the physical frequencies... yes.. the human ear can take it up to 20KHz, but we are talking about SAMPLING FREQuencies.... it's more of a technical difference. If we sample a simple analogue sound at a samplerate of 96KHz it happen that the sound is chopped 96 times a second. Now you get it? The higher the samplerate is the accurate the sound is. 10x for visiting FP. Cheers.
Hi ...
First of all, I'll have to say that I'm surprised that anyone has posted to this thread considering that the last post was in April of 2001. Still, I got an email alert about it, so here I am.
I'm afraid that you're mistaken, my friend. 96kHz is actually 96,000(!) times per second, not 96 times per second. Seeing as the effective frequency range of digital audio correlates approximately to half of the sample-rate, we can see that 96kHz audio is actually capable of producing frequencies higher than 40,000Hz. This is FAR beyond the Nyquist frequency and the limits of human hearing, and though it seems to be better on paper, in reality the supposed fidelity benefit is for the most part imperceptible.
Here we are, nearly 2 and a half years after my last post on this matter, and I'm STILL using 44.1KHz and 24-bit sample resolution. DVD-Audio is also still far from mainstream. I think that surround formats (5.1, 7.1) are a more important discussion these days.
Regards,
(stu.macQ)
First of all, I'll have to say that I'm surprised that anyone has posted to this thread considering that the last post was in April of 2001. Still, I got an email alert about it, so here I am.
I'm afraid that you're mistaken, my friend. 96kHz is actually 96,000(!) times per second, not 96 times per second. Seeing as the effective frequency range of digital audio correlates approximately to half of the sample-rate, we can see that 96kHz audio is actually capable of producing frequencies higher than 40,000Hz. This is FAR beyond the Nyquist frequency and the limits of human hearing, and though it seems to be better on paper, in reality the supposed fidelity benefit is for the most part imperceptible.
Here we are, nearly 2 and a half years after my last post on this matter, and I'm STILL using 44.1KHz and 24-bit sample resolution. DVD-Audio is also still far from mainstream. I think that surround formats (5.1, 7.1) are a more important discussion these days.
Regards,
(stu.macQ)
The nyquist frequency isnt any particular frequency, it's half of the sample rate. Sampling audio higher than this frequency will produce an alias frequency.
Although its undeniable that 24 bit is far superior to 16, just for editing etc, (think about it- 16 bit has 65536 possible amplitudes it can represent, 24 has 16 million. That's worth the hard drive space), 48 and even 96 k recording.. well, as ghizmos said- it will sample the waveform more accurately. But then again the more accurately you sample the waveform, the more flaws you'll hear in the tone. Pah, somebody buy me an Otari 24-track analog recorder. And a spare pair of heads for it, I can't come at 2 grand to replace them if necessary. I'll be your friend forever
Although its undeniable that 24 bit is far superior to 16, just for editing etc, (think about it- 16 bit has 65536 possible amplitudes it can represent, 24 has 16 million. That's worth the hard drive space), 48 and even 96 k recording.. well, as ghizmos said- it will sample the waveform more accurately. But then again the more accurately you sample the waveform, the more flaws you'll hear in the tone. Pah, somebody buy me an Otari 24-track analog recorder. And a spare pair of heads for it, I can't come at 2 grand to replace them if necessary. I'll be your friend forever