What speed audio recording?
- Thread starter Jet Nine
- Start date
When recording audio separately to a digital recorder, there is no "speed". There is bit depth and sampling frequency which for film and video is typically 24bit/48khz or 24bit/96khz.
John Willett
Veteran
In old money, 7.5 ips or 15ips on a Nagra VI-STC.
In modern terms 24-bit / 48kHz minimum.
Noiz2
Sound Ninja
Speed doesn't really come into your scenario. There are cases when you shoot at one speed and finish at another, typically shoot on film finish on video, and there you need to deal with changing the speed of the sound to match. You are all at "video" speed so whatever you set the recorder to would work BUT 48kHz is what digital video wants so that is what you should go for. More bit depth is good, though you will probably finish at 48/16 since that is the standard now. Higher sample rate is pretty much a waste on dialog and it doubles your file sizes.
Noiz2
Sound Ninja
Actually the sample rate is the "speed", and when you need to change speeds on the production end that is how you do it. But your right in that other than those special cases audio plays back at the same sample rate it was recorded at so it's "real world" speed is the same at all sample rates.
John Willett
Veteran
Sorry, I have to disagree here.
It is always best to record at 24-bits. It gives you a safety headroom and is always best if you are going to do any post-production at all.
24-bits gives you the safety you need and is better than the final result.
If you only record at 16-bits you don't have the safety headroom and any manipulation will make the result worse than it could be.
So I would always originate at 24-bits wherever possible.
However - it is not necessary to go higher than 48kHz as a sampling frequency - all 96kHz does is to extend the top-end frequency response from 20kHz to 40kHz.
You have several sample rates, or speeds, to choose from. Depending on how your product is released, you can choose the best rate. If your destination is Blu-Ray only, for example, you should probably stick to 48 kHz. If your sound will be released primarily as a CD, with some supporting video, you may want to select 44.1. If you are producing an audiophile product, or wish to future proof your product, you may choose "perfect fractions", for example, 88.2 or 96 kHz for more or less lossless downsampling to 44.1 or 48.
If your product is "internet and CD only", then you want 44.1, because internet doesn't care about sampling date, just DVD, Blu-Ray, etc.
There are other considerations that apply to surround sound and SACD which also boil down to the release medium.
44.1 to 48 conversion is a "bad fraction", but you can still convert from one to the other.
When in doubt, if video is involved, select 48 kHz, and if making a CD, select 44.1.
Owing to the cheap availability of flash recorders, dual-purpose material can easily be mastered at dual sample rates. You can y-split the mic cables, and feed two multitrackers set at 96 and 88.2.
However, the decline in the traditional CD format makes this a less attractive option unless you already have the gear. But if you were for example videotaping (videoflashing?) a recording session you could do that if you wanted to. Typically, in these situations, the audio is converted either in camera (less desirable) or in post, or going from an analog mix back through a converter set to 48 kHz. This last option gives you an analog mix as well as the desired sample rate; at the point of conversion, you can, if required, fine tune the sampling rate to compensate for measured clock drift of the camera(s), if no master clock was employed, or if the master clock itself has drifted.
If your product is "internet and CD only", then you want 44.1, because internet doesn't care about sampling date, just DVD, Blu-Ray, etc.
There are other considerations that apply to surround sound and SACD which also boil down to the release medium.
44.1 to 48 conversion is a "bad fraction", but you can still convert from one to the other.
When in doubt, if video is involved, select 48 kHz, and if making a CD, select 44.1.
Owing to the cheap availability of flash recorders, dual-purpose material can easily be mastered at dual sample rates. You can y-split the mic cables, and feed two multitrackers set at 96 and 88.2.
However, the decline in the traditional CD format makes this a less attractive option unless you already have the gear. But if you were for example videotaping (videoflashing?) a recording session you could do that if you wanted to. Typically, in these situations, the audio is converted either in camera (less desirable) or in post, or going from an analog mix back through a converter set to 48 kHz. This last option gives you an analog mix as well as the desired sample rate; at the point of conversion, you can, if required, fine tune the sampling rate to compensate for measured clock drift of the camera(s), if no master clock was employed, or if the master clock itself has drifted.
Noiz2
Sound Ninja
I think you miss read. THe quote your disagreeing with is "Higher SAMPLE RATE is pretty much a waste on dialog..."
I compleatly agree on bit rate.
I compleatly agree on bit rate.
Noiz2
Sound Ninja
Dr. D
We are on a film board here and one aimed primarily at very low budgets. So BluRay isn't on the table really and CD isn't either.
I'm not really disagreeing with anything you said but it's kind of mudding the waters to go off on audio recording for CD when the question was about setting up a recorder for production recording on a video. It's all fine info but has pretty much nothing to do with the OP's question.
We are on a film board here and one aimed primarily at very low budgets. So BluRay isn't on the table really and CD isn't either.
I'm not really disagreeing with anything you said but it's kind of mudding the waters to go off on audio recording for CD when the question was about setting up a recorder for production recording on a video. It's all fine info but has pretty much nothing to do with the OP's question.
Rippie
Veteran
I do know there can be issues with a particular time code stamp in certain editors.
The Sound Devices 552 when it doesn't see incoming timecode was originally stamped with a 30ND frame rate in its stamp.
I believe it was final cut, that when you went to drop your audio into a project with a different frame rate you could have sync issues.
Point being is, even if you're ignoring the timecode side of your recording what its stamped with can have consequences in post.
The Sound Devices 552 when it doesn't see incoming timecode was originally stamped with a 30ND frame rate in its stamp.
I believe it was final cut, that when you went to drop your audio into a project with a different frame rate you could have sync issues.
Point being is, even if you're ignoring the timecode side of your recording what its stamped with can have consequences in post.
Noiz2
Sound Ninja
It was/is FCP. But it's kind of convoluted and had more to do with the time stamp on the video. In the NTSC world you need to be really carful with the FCP settings if you work with any X.0 footage. I know there was a trick of wrapping the audio in a MOV file with a matching frame rate but that didn't always work.
But very good point since FCP has a number of quirks related to TC.
But very good point since FCP has a number of quirks related to TC.