Does anybody in here use 24/96 recording?

Thanks Dave. I have been a 96k (and above) skeptic, and still am for most normal audio recording situations. However, what you say makes sense from a sound FX POV.

Cheers.
 
I still say this: if you record at 96kHz and slow it down to 1/2 or something, there mic has to pick up (flat) to above 40kHz if here really is a difference compared to a 48kHz or 44.1kHz recording pitched down 50%. I am curious to hear what microphones you guys use?
 
The advantage to high sample rate recording is not exclusively related to the additional bandwidth that the process offers, but rather to the improved in band performance of both the anti-aliasing and anti-imaging filters. This is the exact reasoning that most high performance D -> A converts use over sampling DFs (digital filters) in front of their converters. I do a lot of projects (music) at 192 kS/s data rates and the improvement is noticable, but with diminishing returns as the sampling rate increases.

The mention of "slowing it down" is likely a misunderstanding of a SRC (sample rate conversion) process. Here, much of the improved in band performance can be retained by the use of a very good down sampling algorithm. Recording at high data rates (above 1 FS, 44k1 or 48k) and then applying the proper down sampling prior to producing the delivery media can be a great compromise between ultimate performance and media capacity.
 
It's not just about recording ultra high frequencies. It's about the number of times a waveform is sampled in one second. The more samples, the more natural something is going to sound when I time stretch the hell out of it. This is important for things like guns when you do the whole "bullet time matrix thing", or anything else like that.

I'd say, most of the time, my sessions are at 48Khz, cause that's what the production sound was at. If I want to design something specific that I've recorded at 96Khz, then I'll have a specific session for that that I'll design in, and then import that into my main session that's at 48Khz.

-Dave
 
Exactly what I was going to say, well almost. We always record 24 bit but only go to 96 when we think we will be doing a lot of pitching down. Foley is one place where we have been doing a lot of 96 for that reason. We could hear the difference between 48 and 96, but it was very slight. Not enough to justify for portable recording unless we knew we were going to pitch.
SK

I thought I had quoted... I was referring to using 96K for FX
 
What you get is smoothens. When you start pitching stuff down it can get "gritty". Part of the reason that a number of the biggest designers still use synclaviers is that they can sample at 100K and so pitch down very smoothly.
SK
 
No your missing the point. More samples per-second give you a more accurate picture of the waveform. The Nylquist number is for the Minimum sampling frequency that will capture a given sound frequency. At that maximum you are only using two samples to define an entire wave. If that wave is a sine wave it is being recorded as a square wave or at best as a triangle wave. In the D/A converters there are filters that round off the high harmonics of that wave and approximate it back to a sine wave. If you are recording at 96 you have that same wave defined by four samples. A lot less filtering is required to get it to look like the original. So the fact that the max frequency goes up is not necs. the point. THe high end is more accurate with 96k. So when you slow that recording down and that top end ends up in the middle it is much smoother than if you recorded at 48. If your not slowing down the sounds then it's not that noticeable, especially since most peoples hearing rolls off between 12 and 15 K. But when that 20K you didn't hear becomes 10K, you sure can hear it.
SK

Again thought I quoted - I must be asleep. This was for Petrus.
 
khmuse said:
The mention of "slowing it down" is likely a misunderstanding of a SRC (sample rate conversion) process.
Click to expand...



Actually no I mean slowing it down. Soundminer and the E4, and actually many samplers lower the pitch by slowing down the playback. This was a tried and true technique with 1/4" tape. That and playing stuff backward. Now were talking real fun!
SK
 
I'm going to do some car recordings for a video game coming up, and we are going to be 96Khz 24 bit all the way.

-Dave
 
Dave, I've never done that kind of sound fx creation and was going to answer 24bit 48khz (for DVD), but thanks for this explanation. I'll play with that soon, very interesting to learn...

When I record into the computer, I record at 32bit floating point.
 
Here in LA, 96k recording is becoming the standard for the top engineers. Many of these guys will tell you that depending on the material (vocals, acoustic instruments, drums) there is a noticable difference. Not to mention that for archivin purposes it's great. It won't be long before there's a standard of some sort for higher definintion recording onto DVD (blue ray or something). it will be nice to have the origional sessions recorded at 96k.
 
You must log in or register to reply here.
Back
Top