New thread 64 bit audio

Gordon Gibb · Posted: Wed Feb 21, 2007, 14:39 (GMT) Post subject: New thread 64 bit audio

I would love to know how often you are called upon, requested, required, or just feel the need - to deliver audio in 64 bit. My hardware and software only goes up to 32, and my stuff sounds fine. 16 bit is my default. I'm gotten some major gigs with what I have.

I would have to upgrade to higher-end software, and perhaps even hardware, to attain 64 Kpbs that no one, save for Voice 123, seems to ask for. And that's just so my demos will sound better on playback through the demo portal.

Any thoughts?

J.S. Gilbert · Posted: Wed Feb 21, 2007, 17:52 (GMT) Post subject: 64 bit is not 64kpbs

64kpbs stands for 64 kilobytes per second and refers to the streaming rate of the file. Generally this would be a designation for an MP3 file, such as 16 bit 64kpbs 128kpbs is generally considered to be CD quality for mono files and 256 for stereo. (I'm not positive but something like this)

Higher end recording systems will allow you to record at up to 192k and generally in 24 bit. There is still a lot of debate as to whether or not there is any advantage in recording voice-over any higher than 44.1k. Most of the tech heads I know seem to think not. However it is going to benefit youo to record in 24 bit. This refers to the length of digits that represent your sound. It allows you to record at a lower volume and still be able to normalize later with good results.

You would set your software or hardware or combination thereof to record at 44.1k and 24bits. Edit your file and then if it is for audition purposes, convert the file to a 64kpbs 44.1k MP3 file. This could be as simple as a function of "SAVE AS" in your software or using an external program to convert the file.

I hope this clears it up for you. I would also like to say that a little knowledge is a dangerous thing. Go get some books on audio engineering, talk to recording engineers, visit studios, offer a recording engineer $50 to come pick their brains. Or take some classes. It will be invaluable.

Gordon Gibb · Posted: Thu Feb 22, 2007, 03:01 (GMT) Post subject:

Appreciate the feedback, J.S. Thanks.....

Colin Campbell · Posted: Thu Feb 22, 2007, 22:50 (GMT) Post subject:

Hey JS... can you explain what "float" means in "32 bit float?"

Todd Ellis · Posted: Thu Feb 22, 2007, 22:59 (GMT) Post subject:

Hey Colin. my understanding of 32-bit float is just recording at a higher dynamic range. Personally, I never use it. It takes up more space, you can't save an mp3 @ 32-bit, and CDs are recorded at 16-bit anyway. If I'm way off base here - somebody call me down.

Colin Campbell · Posted: Thu Feb 22, 2007, 23:06 (GMT) Post subject:

Yeah, I just do 44.1 - 16 bit but I always wondered what the "float" meant on 32 bit. Adobe's manual doesn't say.

Tom Greenlee · **Voice Talent** Joined: 07 Mar 2006 Posts: 945

it's my understanding that the 32 bit float means that Adobe does all it's internal processing at the 32 bit level. It's also my understanding that the higher resolution you record at, like 24 bit 48K, allows you more leeway when processing such as compression and fx etc, before any artifacts are introduced.....once you have your final product, you convert the sample to 16 bit 44.1K....if you record at 44.1K 16 bit, it's great for straight recording and no processing....but you don't have as much room for any manipulation before degradation takes over.

J.S. Gilbert · Posted: Fri Feb 23, 2007, 18:42 (GMT) Post subject: basically yes

Tom's explanation pretty much sums it up. The process is somewhat along the lines of Pi, which is a number that never ends. Theoretically, when you take a sampling of sound and digitally convert it, you can represent that piece of sound by a number (potentially) as long as pi.

So at what point does it make sense to cut off the number of digits to get both a high quality representation of the sound and also something that doesn't take a Cray computer a month to calculate. for recording purposes most agree 24 bit, then truncated to 16 bit after processing. Most plugins operate at either 24 bit length or 32 bit length to provide high degrees of accuracy. Usually you see most of these numbers expressed as multiples of 8

Currently standards are starting to shift from 16 bit 44.1 to 24 bit 44.1 or 88.2. This is a function of increasing computer power and the adopition of Blue Ray and other higher capacity platform solutions. Major adoption is still a few years away (most believe).

If you guys are really introduced in more about mechanics of audio, etc. check into AES. www.aes.org AES is the audio engineers society. If their are chapters near you, you can probably find individuals who for small amounts of money would be happy to consult on soundproofing and other aspects of outfitting home studios. In my neck of the woods there have been really good meetings on mechanics of microphones and one is coming up on sound audio amplification.

This is a link to a page that talks about it floating bit numbers more in depth.in much greater detail. It's boring, so have a couple of cups of coffeee first.

http://steve.hollasch.net/cgindex/coding/ieeefloat.html