|
|||||||||
|
Thread Tools | Search this Thread |
February 19th, 2005, 04:02 PM | #1 |
Regular Crew
Join Date: Apr 2004
Location: S.F., California
Posts: 61
|
16-bit vs 24-bit mixer
let's say my field audio setup was simply
DVX100a -> AT897 shotgun (records 16-bit, 48KHz) and i'm thinking of doing audio instead through a laptop... laptop -(firewire)-> Presonus Firepod (Mobile I/O) -> AT897 (records 24-bit, 96KHz) what would i be gaining, if anything, in sound quality? i'm asking if going from 16->24bit and 48->96Khz makes any difference since the same cheap $250 microphone is being used. does it itself have the sensitivity to benefit from a new mixer's higher digitizing settings. also anyone have any opinions on the Firepod? thanks |
February 19th, 2005, 04:18 PM | #2 |
Join Date: Jan 2004
Location: Stockton, UT
Posts: 5,648
|
Like anything else, the higher bitrate and higher sampling rate always equate to "better" because you are slicing the audio into thinner slices and giving it more bit depth.
The question isn't so much what the input format is, as much as how much computer horsepower you have and are wanting to use. We do *most* things here at 24/48k but on selected instruments and projects, we use 24/96k, and on rare occasion, 24/192k. The higher bit depth will eventually be dithered down to something else on delivery regardless, but the higher the quality you start with, generally the higher quality your output will be. But if you aren't using a good DAW, or if it has weak dithering tools, then you'll want to use a plug in, or you'll want to capture at the lower rate.
__________________
Douglas Spotted Eagle/Spot Author, producer, composer Certified Sony Vegas Trainer http://www.vasst.com |
February 19th, 2005, 04:30 PM | #3 |
Regular Crew
Join Date: Apr 2004
Location: S.F., California
Posts: 61
|
i guess specifically i'm wondering about the dyanmic ranges, because 16 to 24 bit is going from 96dB to 144dB theoretically. what's that mean for a cheap mic like mine and for a pro mic? will that mean a lower noise floor in the final product, and thus cleaner sound in the end, or is the effect barely noticeable?
|
February 19th, 2005, 05:52 PM | #4 |
Join Date: Jan 2004
Location: Stockton, UT
Posts: 5,648
|
The self noise of the mic is going to be substantially louder than the bottom of the dynamic range, it simply means you can have cleaner/greater dynamics. You can record greater signal, and process more cleanly Will it be a big difference in a live recording? Not likely.
__________________
Douglas Spotted Eagle/Spot Author, producer, composer Certified Sony Vegas Trainer http://www.vasst.com |
February 19th, 2005, 07:09 PM | #5 |
Regular Crew
Join Date: Mar 2004
Location: Palo Alto, CA
Posts: 55
|
The biggest difference won't be in the equipment if you don't have a good sound guy handling your micing and mixing, or at the very least, knowledge about how to get the best sound via a good foundation in micing techniques. I'm guessing your mic will be camera mounted?
To put it simply, your AT897 at the end of a stick handled by a good sound guy (or gal) will sound better than a $2k Schoeps mic kit mounted on your cam, several feet from the sound source (unless you're purposefully wanting to capture lots of room acoustics and ambient noise in your recording). I don't mean to be knocking your questions, and these are important questions, but I just want to underscore the point that skill and technique far outweight bits and Hz when it comes down to sound quality. That being said, if you've got good hands handling a mic boom, a better mic into a good preamp would yield higher quality recordings than chasing bits and Hz. Also, realize that when you go from 16/44(48) to 24/96, you've more than doubled the amount of storage you'll need to store that data, while, as DSE pointed out, increased the required CPU power needed to process that data. It may be a good idea to refer to a solid book on the matter, like those written by Jay Rose (via www.amazon.com) or Ty Ford (www.tyford.com). Both great sources of knowledge that may save you $$$ down the road of wrong equipment for the job. One last thing, realize that your proposed laptop setup will not be easily made mobile, since the Firepod does not work off of bus power. It requires an external AC power source.
__________________
ELeven Feet Media, LLC - a nimble design & brand communications firm for growing companies http://www.elevenfeetmedia.com |
February 20th, 2005, 11:26 AM | #6 |
Inner Circle
Join Date: Nov 2004
Location: Baltimore, MD USA
Posts: 2,337
|
In a perfect world, where every stage is adjusted properly, there is a significant quality advantage to 24-bit over 16-bit audio.
EQ, effects or other manipulations are usually better done at 24 bit than at 16 because the audio is represented as more data. Perhaps even more important is the record level at the camera. With a good limiter on a mixer you can record at levels that are 5 to 10 dB hotter. The PD 150 we tried that on gave us audibly less hiss when recorded with a limiter at levels that peaked at -3dB. Before considering the use of an external audio recorder, make sure you're getting the most from what you already have. Many people record lower simply because they don't have, (or know how to use) a good limiter. It doesn't have to be that way. As for the quality of your $250 mic, the chain is only as strong as its weakest link. Mics, preamp stages, quality of A/D converters, sample rate and bit depth, quality of D/A converters and output buffer amps all come into play. Then there's the quality of the playback system (amps and speakers) and one's ability to tell any difference in the first place. Yes, ALL of these things make a difference. If YOU can't tell a difference, then it really doesn't matter. Someone mentioned on this board before, "Audio is easy. Good audio requires a lot more attention to detail." That sums it up pretty nicely. Regards, Ty Ford |
February 20th, 2005, 02:48 PM | #7 |
Inner Circle
Join Date: Jun 2003
Location: Toronto, Canada
Posts: 4,750
|
The practical difference might be the quality of electronics in the firepod versus the DVX100. The firepod might have a lower noise floor and wider frequency response. The wider frequency response won't make a difference for voice. Lower noise floor will give you more leeway in setting levels. The firepod might have slightly less distortion.
You can know things for sure by running tests on your equipment as to which is better (there is a possibility than the DVX100 is better). The DVX100 should definitely be good enough based on the dv.com's review of it. From a practical standpoint, I would record into the DVX100 so you don't have to mess with double system sound. Sound will come into your NLE in sync, and you do not have to deal with the laptop in the field. |
March 2nd, 2005, 02:48 PM | #8 |
Regular Crew
Join Date: Apr 2004
Location: S.F., California
Posts: 61
|
thanks for the replies everyone!
Ty, if your saying that realistically the 24bits will only help with EQ'ing and processing, then I think i'll record at 16bit because Adobe Audition upsamples to 32-bit, applies effects, and automatically downsamples back to the original bit depth. Harris, of course it'll be on a boom, that's just silly. And I have both of Jay's books. They're invaluable though it's always nice to get others opinions. The firepod has 8 XLR-ins, that's why I'm getting it and using a laptop. It's not a question of whether the DVX or firepod have better specs, that's not the main deciding factor. And i'll be running it and the laptop off a sealed car battery (unless I can find a better power option (which I hope exists because car battery's are tricky to gauge remaining charge)) Does anyone know if i'll be able to record at 24bits? My laptop is only 700mhz (Pentium 3) PC, though i'm going to max out the RAM. I hope i'll be able to get close to zero latency... |
March 2nd, 2005, 02:50 PM | #9 |
Inner Circle
Join Date: Nov 2004
Location: Baltimore, MD USA
Posts: 2,337
|
Allen,
the upsampling and downsampling procedures themselves, arguably, can negatively impact the audio due to rounding off errors, among other things. Good luck, Ty Ford |
March 2nd, 2005, 03:10 PM | #10 |
Join Date: Jan 2004
Location: Stockton, UT
Posts: 5,648
|
On a 700Mhz machine? Zero latency? Only if you're monitoring off the Presonus directly, and not routing it. And definitely not 24 bit. You can record it, but you won't be able to monitor it via the processing. And it goes without saying that you'll not be able to have real-time processor monitoring such as soft comps or eq's.
__________________
Douglas Spotted Eagle/Spot Author, producer, composer Certified Sony Vegas Trainer http://www.vasst.com |
March 2nd, 2005, 03:13 PM | #11 |
Contributor
Join Date: Sep 2003
Location: Santiago, Chile
Posts: 932
|
Allen, I see you seem to have been getting some very good advice. I would like to add a question? Do you need 8 XLR inputs and direct-to-disk? I ask because, well... that money can be much better spent on:
(1) getting a better microphone (2) a portable preamp with limiter and/or compression (3) if you insist on going double system, there is the possibility of using something like the Marantz CDR-300 recorder, a solid state or hard disk recorder like the Fostex FR-2 or even if you want to spend more money the Fostex PD-6, 6 channel portable recorder, so you don't have to deal with a preamp, a power module, a laptob, a battery...
__________________
Ignacio Rodríguez in the third world. @micronauta on Twitter. Main hardware: brain, eyes, hands. |
March 2nd, 2005, 04:34 PM | #12 |
Regular Crew
Join Date: Sep 2004
Location: Atlanta, Ga
Posts: 129
|
Once upon a time in a recording studio (project studio) our setup consisted of a Roland VM3100 Pro mixer running into a 750MHz PC using Steinberg NuEndo 1.52 on a WinXP box. 1Gig of Ram IIRC.
Running all 8 channels of audio in at the same time, we did not have any problems of latency or anything running various effects. We did not stack effects one on top of another, the most we ran were 2 effects on a track at once, and we did not run effects on all of the tracks. Anyway, the PC kept up w/o problems. Most of the time we were recording 8 drum tracks at once with efects such as reverb or flanging (just to experiment) on the snare & somtimes the bass drum. In realtime. We monitored this via headphones in realtime. I even monitored on headphones and ran a guitar in with only about 7 drum tracks (removed the bottom snare mic to make room for the guitar track). I had headphones on and played along with what I heard in the headphones in realtime. When we listened back to the 7 drum & 1 guitar track, there was no problem w/ the guitar being offtime because of latency in the headphones. IOW, it IS possible. We did "optimize" the PC before we installed this stuff, and had plenty of diskspace available. I think we used blackviper.com's guides to XP optimization, if I remember the site correctly. The tracks were all done at 16/48 if I remember correctly. Now as far as the question of how to run the audio... In the end, you are probably going to downconvert everything to 16/44.1 anyway, I imagine (that is standard audio definition for MiniDV & CDs. DVDs are a different story). What recording at 24bit gives you is headroom. If you record at 24 bit and do a lot of effects (flange, delay, reverb, compression, etc. everytime you apply an effect to the audio, the quality degrades ever so slightly. When recording at 24 bits, you have LOTs of headroom for degredation and can still have "CD quality" or better audio in the end. Record at 16 bits and apply lots of effects, etc. you will notice the degredation sooner. The biggest things here are: 1) Garbage in, garbage out. You can record great sound at 16 bits (or even 11 bits which is FM quality) and it will sound MUCH better than poorly recorded audio at 24 bits. It is better to concentrate on technique and equipment than bits. 2) When recording digitally, DO NOT hit to disk as hot as possible!!!! Traditionally in the analog world they used to record as hot (loud) as possible without clipping. WIth digital you do NOT want to do this! The reason is the math involved. You can turn UP the volume on digital audio, and it will sound the same as the lower volume audio (well, except louder!). However, take loudly recorded digital audio and turn it down, and you start losing your bits!!!!! In the digital worl it is better to record lower and bring things UP in the mix, which is the opposite of the analog world. A "best practice" is to record with your peaks at around -12db. That should give you plenty of headroom to turn things up during mixing. Hope some of this helps, Alex F |
| ||||||
|
|