Jump to content

What's the industry standard sampling rate for pro studios - 48 or 96 kHz?


Recommended Posts

  • Members

If you're going to cds, work at 44.1 or 88.2

 

48k is only used for 48k-native hardware (ADATS) and video where the final file needs to be 16bit/48k.

 

Unless you have a kajillion-dollar acoustically-designed studio with equipment that can stand the extra grunt of 96k (1/4 the tracks, 1/3rd the plugs before BSOD), forget about it.

 

Higher bit-rate for more headroom is much more important than sample rate.

 

Just lock everything down on 24-bit (or 32-bit float)/44.1khz and forgetaboutit.....

 

 

;-)

Link to comment
Share on other sites

  • Members

I'd say you need to deside if the drive space, CPU resources, and converter quality are worth recording at higher sample rates. If you're going to be using alot of tracks and plugins, Or your converters, front end mics preamps, source, etc are going to really benifit from a higher sample rate or not. If you had state of the art everything why not record at the highest sample rate.

 

 

I usually record at 24/48 too, It saves drive space and has littel impact on the final dither to 16/44.1 vs going from 88 to 44. I couldnt hear enough improvement at highter rates to justify working with huge files. I do have an older 8 track card that could record at 48/16 but the playback was slowed down or speeded up or something. The newer cards I use dont have that problem.

 

 

I may even try 24/44 as others mentioned. I do record tracks to a separate drive so space isnt a huge issue, but it does bog down loading and exporting projects enough that its not worth the hassle. If I though I had something special to record or was earning bucks, then i'd go for 88 or 96 too eak out any added benifits there may be, contriversial as that may be.

Link to comment
Share on other sites

  • Members

Record quick experiments with every sample rate available to you - acoustic guitar, a stick on a cymbal, human voice, whatever else you feel will show off the high end well. Use your clearest mic and pre. Compare your results and use whatever sample rate sounds best (with my old TC Electronic SK48 for example, I could clearly hear that 48 kHz sounded better than 44.1 kHz).

 

If you don't hear a difference between the sample rates, or you can't decide which one is better (as is the case with my current interface, the Steinberg MR816), use 44.1 kHz. The files are smaller and you won't have to convert the sample rate to make a CD quality mixdown.

 

Typically, the differences between sample rates are nil when using higher quality converters. With cheaper converters, you will start to hear a difference and want to use a higher sample rate to get all the filtering distortion out of the audible frequency range. Use a high quality sample rate converter afterwards (Voxengo r8brain sounds great and is free, so that's my suggestion).

Link to comment
Share on other sites

Going from 16 bit to 24 bit makes a MUCH more audible improvement than going from 44.1kHz to 88.2kHz does. Of course, any time you increase the bit depth, or the sample rate (or both), it places greater demands on your computer system in terms of CPU requirements and HDD storage, but the 33% increase in HDD storage required for 24 bit (vs. 16 bit) is more than worth it IMO. The doubled HDD space for 88.2kHz vs 44.1kHz isn't in many cases - especially when you take into consideration the increased demands on the CPU when running at higher sample rates.

 

I have a reasonably fast dual core computer with lots of RAM and HDD space, and a lot of DSP assistance (Pro Tools HD2 Accel, SSL Duende), and although I could easily do sessions at 88.2 or 96kHz, I fail to see the point in doing so. Yeah, I'm definitely in the "make it sound as good as you can" camp, but I also try to be realistic... and the realities are that there are costs in terms of storage space and CPU resources, no one wants to pay us (making gear upgrades hard to justify), not many projects are going out to DVD-A, and as it is, the vast majority of people don't ever hear it sounding half as good as what I get to hear in the control room; where I'm listening back in a decent room on high quality speakers to material recorded at 24 / 44.1... the average listener is going to hear it on crappy earbuds via an MP3 player... by the time it gets to that point, any benefits from going to 96kHz will largely be lost.

 

99% of the time, I'm running at 24 bit / 44.1kHz. And so are most of the other engineers I know.

Link to comment
Share on other sites

  • Members

another question: by resorting to 44.1, do you try to avoid resampling? if not what's the best plug in for re-sampling? i ran tests involving several plug ins - wavelab's and sony's own, and third party - apogee - i couldn't hear any difference?

Link to comment
Share on other sites

  • Members

 

another question: by resorting to 44.1, do you try to avoid resampling? if not what's the best plug in for re-sampling? i ran tests involving several plug ins - wavelab's and sony's own, and third party - apogee - i couldn't hear any difference?

 

 

Not sure what you mean by resampleing. Are you talking about dithering down or are you talking about something like reamping. The term "resampling" means your sampling a sample or sampleing something a second time.

Link to comment
Share on other sites

  • Members

The only time you wont need to dither is when its at 16/44.1 any other combination yes if you're going to burn a wave file. Going from 88~44 is supposed to have better results then converting from 48,96, or higher. Its something to do with the math in conversion. I cant hear any difference, but maybe theres something there.

 

 

As always, conversion the very last step before burning to disk after final Limiting and mastering is done. There is less quality loss applying effects to a file at the higher sample rate because theres so many more bits to work with and the resolution is always going to be better applying effects before you convert.

 

I worked in digital color reproduction and video and the same holds true there. A Raw or Bitmap file is always going to have more resolution than a JPEG. It wont look that much different from a causial view, but enlarge them or try to work with tools and see which looks like garbage.

Link to comment
Share on other sites

  • Members

 

Actually neither. The vast majority of professional installations doing serious audio for broadcast and or soundtrack for film are recording and mastering @ 24 bit 192khz.

 

 

Bull{censored}. Sorry to be so blunt about it, but seriously, that statement is utterly false. Oh, and BTW, have you actually listened to 192? There's a good reason why so few people are working at that rate, and it has nothing to do with storage.

 

To the OP: Professional studios will use any sampling rate you like. Professional producers and engineers will have varying opinions on this, and much of those opinions have to do with their working methodology among other things.

 

If you work with me as a producer, we'll be recording at 44.1k for the duration of the project. If you hire me to mix, whatever you send me will likely be downsampled to 44.1k before I even start to mix, UNLESS I know I'm printing to, and storing to, and mastering from analog tape. Then, I'll probably leave the sample rate as delivered.

 

Now, many PT users claim that the plugins sound better at 96kHz. I haven't verified this claim, mostly because I avoid PT like the plague. If true (and I don't actually doubt it), then there's certainly an argument for working at 96kHz.

 

Enjoy,

 

Mixerman

Link to comment
Share on other sites

  • Members

Bullshit. Sorry to be so blunt about it, but seriously, that statement is
utterly
false. Oh, and BTW, have you actually listened to 192? There's a good reason why so few people are working at that rate, and it has nothing to do with storage.


To the OP: Professional studios will use any sampling rate you like. Professional producers and engineers will have varying opinions on this, and much of those opinions have to do with their working methodology among other things.


If you work with me as a producer, we'll be recording at 44.1k for the duration of the project. If you hire me to mix, whatever you send me will likely be downsampled to 44.1k before I even start to mix, UNLESS I know I'm printing to, and storing to, and mastering from analog tape. Then, I'll probably leave the sample rate as delivered.


Now, many PT users claim that the plugins sound better at 96kHz. I haven't verified this claim, mostly because I avoid PT like the plague. If true (and I don't actually doubt it), then there's certainly an argument for working at 96kHz.


Enjoy,


Mixerman

 

As an aside, and I'm sure you get this a lot, but I just finished your book and fuckin' loved it. Can't wait for what's next. :thu:

Link to comment
Share on other sites

Dont apologize, simply state the reasons why you think it's

so?

 

Good suggestion. :) And one that you didn't really follow completely yourself. ;)

 

PT and then Logic are the defacto standards for production at high end studios

 

They're certainly in the running... but there is no single "de facto standard" today like there was when I was starting out with this stuff; in the late 70s, 2" 24 track analog machines were THE standard. Sure, there was other stuff being used (1" 16 trk., 2" 16 trk.), and variations on the theme (Dolby A, SR, dBx or no NR; 15 IPS / 30 IPS, +3 / +6 / +9 with high output tapes, etc.), but the 2" 24 track was "the" de facto, professional recording format from the 70s and on into the 90s.

 

Today, PT is probably the most popular DAW in pro studios, and Logic is also quite popular. It used to be very common to see them being used together for what each did best - Logic for MIDI and content creation, PT for recording, editing and mixing audio...

 

But there are a LOT of choices out there today, and what gets used on a particular project is usually a matter that is decided upon by the producer and the engineer, and not dictated by an "industry standard". If you're in a big room in Nashville, you're going to be more likely to see Nuendo being used than here in LA. Some folks prefer RADAR, and others still record to analog...

 

and the interfacing A to D's that yeild the maximum dynamic range are 24 bit 192khz

 

The sample rate has nothing to do with "dynamic range".

 

and are what most field recordists prefer and are basically required for surround mixing.

 

Wait... are you saying that 24 / 192 is required for surround mixing? :facepalm:

 

Every godamn Major motion picture I've seen in the last ten years has AVID and Dolby Digital and DTS all over their credits at the end.

 

And how does that pertain to our discussion of sample rates used in pro studios? :confused: Are you suggesting that if someone records the soundtrack at 24/96, they won't have DTS and Dolby Digital in the credits at the end of the film?

 

I also know that 95% of the tracks that get to mainstream radio are mastered at the same rate.

 

Source? IOW, how do you "know" this?

 

I hear this audio standard every day, there are alot of radio stations in the Bay Area that will reject any CD if it hasn't passed the "listening test", regardless of the payola involved.

 

1. Sure, a station may expect something to sound up to their standards before agreeing to broadcast it, but that doesn't mean it has to be done at 192 kHz to achieve that.

 

2. The average PD at a radio station today is playing what corporate is mandating they play.

 

3. Even at independently owned stations, the average PD isn't going to be able to differentiate between something recorded well at 24/44.1 vs something done at 24/192, and their audience sure as heck isn't going to be able to do so after that material has passed through 40dB of compression and limiters before hitting the transmitters.

 

4. Maybe you missed the memo, but Radio Is Dead.

 

There is no way film studios can sum 90, 100, 120 or more tracks of audio, special effects, dialogue and ambient sounds without introducing 10 db of noise floor into the mix that aren't all recorded at least in the 120db neighborhood of dynamic range, and the only way to get that much headroom before noise is to use 24 bit 192khz converters.

 

Again with the dynamic range argument. That is a function of the bit resolution, NOT the sampling rate! In theory, the best you can get out of a 16 bit system is about 96dB, and each additional bit gives you approximately a 6dB increase... so for 24 bit, that goes up to ~144dB. However, no matter what you're running, the best converters out there right now can only give us about 120dB... and they'll do that at 44.1kHz as long as it is a 24 bit recording.

 

Another problem with your 192kHz argument is the decrease in track counts on the system. Yes, big budget films tend to use huge amounts of audio tracks in their production, and it is common for multiple "machines" to be synchronized for film work - one for dialog, one for Foley, one for the soundtrack, etc... But when you run a Pro Tools HD system at higher sample rates, that system becomes more limited in terms of maximum track counts. At 24 bit / 192kHz, a PTHD2 Accel setup maxes out at 36 tracks of audio. You CAN'T get more than that from ANY single Pro Tools setup at 192kHz, no matter how many I/O boxes or DSP cards you have installed. However, drop that sample rate down to 96kHz, and you can run 96 tracks, and at 44.1 / 48kHz, the maximum audio track count moves up to 192. Source. So if you're working on a film, where you need large track counts, and where you will also need to meet the requirements of DVD audio for the release, you're probably more likely to be running at 96kHz, if for no other reason than to have enough tracks available.

 

Unless you're not using any conversions at all and you're tracking EVERYTHING digitally: IE: Digital Mics, Digital interfacing,{like AES/EBU, Spdif, Firewire, USB, etc..} digital instruments and digital sampling. Almost all of the Video collections of major artists that are released on DVD or Blue Ray have the same standards employed.

Perhaps you know a better way? Or is it trade secret type stuff? Wouldn't want to ruin your scoop...

 

:whisper: Digital microphones use A/D converters.. they're just built into the mic itself. There's no way around using converters unless you are not going to record any "acoustic" sources (including vocals) - sound is compression and rarefaction of a porous medium (air), and that's analog... if you want to digitally record sounds your ear can actually hear, you're going to need converters. Even if you send the sound of a digital synth into your system via a S/PDIF interface, in order for your ears to eventually hear it, you're going to need an D/A converter...

Link to comment
Share on other sites

  • Members

Bullshit. Sorry to be so blunt about it, but seriously, that statement is
utterly
false. Oh, and BTW, have you actually listened to 192? There's a good reason why so few people are working at that rate, and it has nothing to do with storage.


To the OP: Professional studios will use any sampling rate you like. Professional producers and engineers will have varying opinions on this, and much of those opinions have to do with their working methodology among other things.


If you work with me as a producer, we'll be recording at 44.1k for the duration of the project. If you hire me to mix, whatever you send me will likely be downsampled to 44.1k before I even start to mix, UNLESS I know I'm printing to, and storing to, and mastering from analog tape. Then, I'll probably leave the sample rate as delivered.


Now, many PT users claim that the plugins sound better at 96kHz. I haven't verified this claim, mostly because I avoid PT like the plague. If true (and I don't actually doubt it), then there's certainly an argument for working at 96kHz.


Enjoy,


Mixerman

 

Dude, I LOVED your book :)

upd: sorry to hijack my own thread

 

after reading all of this, i think i should go back to 24/44 as opposed to 24/48.

Link to comment
Share on other sites

  • Members

Track @ 24/48 (rolling off everything above 16 kHz while tracking) and master OTB to 1/4 half-track analog reel-to-reel @ 15 ips... really! And then resample from reel-to-reel @ 16/44.1... or mix to 16/44.1 in the first place from the analog console.

 

You roll of the highs above 16 kHz because at 48 or below digital sucks with high frequencies... it always has and it always will. The analog tape in between the source and the 16/44.1 tempers the funky highs. Otherwise it would be better to roll off the highs at 14 kHz.

 

88.2 is great for tracking if you have the space, but still avoid conversion if at all possible and resample at the rate of the end product (44.1 for CD)

 

If you're doing any sample rate conversion at all you're part of the problem, but maybe not a bad person. Yes, I know that means 90% of the people in the music industry are part of the problem... and that's true... and it's a big problem... unfortunately.

 

Or forget every thing I said, do what you want but wear cotton in your ears. Either way. ;)

 

PS: Don't use plugins... just cut it out dammit!

Link to comment
Share on other sites

Archived

This topic is now archived and is closed to further replies.

×
×
  • Create New...