POLL: What Sample Rate/Bandwidth do you use for recording?

[Tim Brown]

Okay guys - What Sample Rate and Bandwidth do you use for Recording?


Somebody just told me he records at 24-bit/44.1khz

And that kind of surprised me - I was positive he was using at least 96 khz since he's recording in Neuendo (I just didn't see the point in spending that kind of cash for a software like that and then just recording at 44.1khz.)

Tim

[Newbie-Doo]

16-bit/44.1khz mainly because my pc setup runs the best at that setting. Since MP3s are recorded at 16-bit/44.1khz anyway it would seem to be overkill to user higher rates (this is where Joe Techno-shmoe explains why higher bit-rates and khz make a 16-bit/44.1khz mp3 sound better).

[Massive Master]

Every poll I've seen for years (not just on the forums) put around 75-80% of full-time industry pro's at the target rate.

A surprising few record higher than 44.1 (for CD) or 48kHz (for video). At 24-bit of course -

If you can't get 44.1kHz to sound absolutely fantastic, there's no hope for getting it to suddenly sound "right" at 96kHz.

[Tim Brown]

Originally Posted by Massive Master

Every poll I've seen for years (not just on the forums) put around 75-80% of full-time industry pro's at the target rate.

A surprising few record higher than 44.1 (for CD) or 48kHz (for video). At 24-bit of course -

If you can't get 44.1kHz to sound absolutely fantastic, there's no hope for getting it to suddenly sound "right" at 96kHz.

Massive Master,

I figured that since a lot of the newer gear would do 96k and 192K that most people would be using those higher rates. That really surprises me - and here I've been worrying about how I'll be able to afford the jump to 192k.
Whew! That takes a lot of pressure off of me in buying A/D converters.

Thanks for that bit of information!



Tim

[BigRay]

24/44.1 unless I am doing a DVD-Audio or my project will see life as an SACD.

[SouthSIDE Glen]

Originally Posted by Tim Brown

And that kind of surprised me - I was positive he was using at least 96 khz since he's recording in Neuendo (I just didn't see the point in spending that kind of cash for a software like that and then just recording at 44.1khz.

Would you say the same thing about someone recording in CubaseSX?

The audio section in Nuendo is the exact same program code as CubaseSX. The only reason Nuendo costs more is because of the extra features it includes that target it for the broadcast and pro video post-production industries. It has noting to do with any increased audio quality over a $500 program.

96K is useful if your incoming signal is worthy. If you're using quality miking technique in a good room to record dynamic acoustical music though great pres and in thru a top-shelf converter, *then* you have something worthy of 96K. If you're using amateur miking technique to record squashed and distorted death metal through economixer-quality preamps and coming in through an average-sounding converter, then all 96K is going to do is wastefully eat up disc space and CPU cycles to give you nothing more than a high-resolution copy of an amateur recording of squashed and distorted music.

And that's all not even considering the quality of sample rate converter (SRC) when all is said and done.

48k is more common amongst those doing audio for video for surrendipitous mathematical reasons as much as anything else, but for those doing audio only, unless one has an A-list SRC, whether the cost of conversion from 48k down to 44.1k is worth paying for the marginal increase that 48k gave to begin with is debatable, at best.

Which leaves us with 44.1/16 or 44.1/24. The general consensus these days is that one should record and mix in 24 bit, which is the natural capacity of the software and hardware these days. To constrain that unnecessarily would indeed have an adverse effect on the quality of the sound regardless of the quality of signal chain or recording style.

G.

[Track Rat]

I track at 24/44.1. While the stoarage and processing power isn't the issue it once was, for pop music, I just don't hear the diiference in the higher sampling rate. Now I hear a big difference between 16 and 24 bit.

[Tim Brown]

Originally Posted by SouthSIDE Glen

Would you say the same thing about someone recording in CubaseSX?

The audio section in Nuendo is the exact same program code as CubaseSX. The only reason Nuendo costs more is because of the extra features it includes that target it for the broadcast and pro video post-production industries. It has noting to do with any increased audio quality over a $500 program.

96K is useful if your incoming signal is worthy. If you're using quality miking technique in a good room to record dynamic acoustical music though great pres and in thru a top-shelf converter, *then* you have something worthy of 96K. If you're using amateur miking technique to record squashed and distorted death metal through economixer-quality preamps and coming in through an average-sounding converter, then all 96K is going to do is wastefully eat up disc space and CPU cycles to give you nothing more than a high-resolution copy of an amateur recording of squashed and distorted music.

And that's all not even considering the quality of sample rate converter (SRC) when all is said and done.

48k is more common amongst those doing audio for video for surrendipitous mathematical reasons as much as anything else, but for those doing audio only, unless one has an A-list SRC, whether the cost of conversion from 48k down to 44.1k is worth paying for the marginal increase that 48k gave to begin with is debatable, at best.

Which leaves us with 44.1/16 or 44.1/24. The general consensus these days is that one should record and mix in 24 bit, which is the natural capacity of the software and hardware these days. To constrain that unnecessarily would indeed have an adverse effect on the quality of the sound regardless of the quality of signal chain or recording style.

G.

Glen,

I wasn't necessarily talking software or hardware, but I don't see the point in purchasing something designed to do 192khz (hardware), if you are only going to do 44.1k? why go for the "extreme" if you are not going to use it's full potential.

Is the RME Fireface really worth the $800 price difference over a Presonus Firepod if you don't even intendto take advantage of the the advantageous features (ability to do 192khz)?

That is why it just kind of shocked me - He I was thinking I was going to perhaps have to figure out a way to shell out another $15K to $20K to move up to recording on a PC, when I can actually get away with a lot less if I don't worry about the 192khz spec. That is essentially what I'm getting at.

And for what it's worth, I intend to go with Cubase as my software.



Tim

[Massive Master]

I'd buy a *really good converter* period. If it happens to sound "really good" at 192kHz isn't an issue (few I've heard actually do anyway, but that isn't the point).

But even the better converter designers (Dan Lavry, David Hill, etc.) pretty much tell you that if you can actually hear the difference between the sample rates on their converters, something is broken - Assuming you aren't recording dog whistles or something (which obviously would take a chain of gear that few people have in the first place).

Don't get me wrong here - I know there are plugins out there that throw calcs at 192kHz. I'd go to say that it makes a difference in the way it handles the signal.

But bringing the signal in... You'd need an entire chain that's able to handle the higher frequencies before it'd pay off. And then once it pays off, you can't hear it anyway. Most people I know can't even pick monitor whine out of a mix when it's right there in front of their face.

[mshilarious]

Originally Posted by Tim Brown

Glen,

I wasn't necessarily talking software or hardware, but I don't see the point in purchasing something designed to do 192khz (hardware), if you are only going to do 44.1k? why go for the "extreme" if you are not going to use it's full potential.

Is the RME Fireface really worth the $800 price difference over a Presonus Firepod if you don't even intendto take advantage of the the advantageous features (ability to do 192khz)?

It's entirely possible that 192kHz will sound worse than 96kHz. The increase in speed brings a decrease in precision; the additional resolution is unnecessary.

However, that doesn't mean that RME isn't worth the money. RME makes good gear, and the difference could easily be noticeable at 44.1kHz.

[SouthSIDE Glen]

Originally Posted by Tim Brown

why go for the "extreme" if you are not going to use it's full potential.

An analogy that I find often helps explain this situation is that the sample rate is very much like a TV monitor. 44.1k would be a Sony Trinitron CRT tube displaying in standard NTSC (or PAL, for our EuroFriends ) resolution. 96k would be like a nice wide-screen HD monitor.

Now, if we're watching, say, Star Wars III on DVD, it will look great on the Trinitron, but will look even more spectacular on the HD widescreen. In this case, the Star Wars DVD would be the analog of a top-notch pro tracking job through the best mics, pres and converters.

On the other end of the spectrum, let's say we're watching an original episode of The Honeymooners with Jackie Gleason et al, broadcast in standard analog from a tower on a building 50 miles away and we're receiving that signal through a pair of rabbit ears. The wide screen HD TV now looses any advantage it once had over the old school Trinitron. The picture will still be in poorly contrasted black and white, will still have a 4:3 aspect ratio, and will still be filled with static. It won't look any better on the bigger TV other than the fact that you'll be able to pick out even more fault with the content because the static and distortion will be in even finer resolution, even if the signal itself isn't. It'll still look like crap on both, but on the HD the crap will be more apparent.

So unless and until one is prepared to receive digital HD content, that big screen HD monitor - which probably cost about ten times as much as the "obsolete" Trinitron - is both money and technology wasted. In the case of high-resolution recording, it's also a waste of computer resources.

As far as the RME being better than the Presonus, yes it probably is, but not *because* it has a 192k option, but because it's better at 44.1. Is it worth $800 just for that part of it? It can be. I have my eye lustfully upon a UA2192 converter that's only 2 channels and that costs $1800 more than the 8-channel Firepod. Definitely far more expensive, even more so than the RME. But IMHO, if one has the budget, it's more than definitely worth the cost. I'm just saving up my pennies. And I'd still only be using it at 44.1k most of the time even though it too can go to 192k.

The quality of converter is far more important than teh speed at which it can convert. A great converter at 44.1k will beat out a dime a dozen converter at 192k every time.

G.

[slidey]

24bit @ 44.1 for me

[NYMorningstar]

I've noticed a difference when changing from 16/44.1 to 24/44.1 but I've noticed an even larger difference when I upgraded my converters.

[Tim Brown]

Originally Posted by mshilarious

It's entirely possible that 192kHz will sound worse than 96kHz. The increase in speed brings a decrease in precision; the additional resolution is unnecessary.

However, that doesn't mean that RME isn't worth the money. RME makes good gear, and the difference could easily be noticeable at 44.1kHz.

Now that is interesting - I would have thought that the increase in speed would have increased the definition and precision, because you are using more surface to store less information (ala 2" 24 track at 30ips, vs 15ips.)



Tim

[Mindset]

Yes, but you also bring out more flaws in clearer sounds. All the stuff you missed before, you can hear now. Plus it's more work for the cpu that really isn't needed @ cd quality.

[Massive Master]

Originally Posted by Tim Brown

Now that is interesting - I would have thought that the increase in speed would have increased the definition and precision, because you are using more surface to store less information (ala 2" 24 track at 30ips, vs 15ips.)

Yes, but even tape (and the gear attatched to it) don't have *near* the frequency response to take advantage of the response at 192kHz - Even 96kHz for that matter - You'd need a core sound that was tapping 48kHz, a mic that could pick it up, a preamp that could reproduce it, converters that could do a decent job digitizing it, etc. Otherwise, it's hype.

And I suppose the analogy would really be better off with word length - Running at 30ips really isn't changing the frequency response... It's changing the "resolution" on the tape - more information per area. With 16-bit's 65,000 possible values and 24-bit's 16.7 million possible values, I think that analogy would be more accurate.

[scorpio01169]

really....in the end it doesnt really matter, i mean does the end listener really care about what sample rate you used or what converters you used. if you get your music to sound understandable then thats what people care about, yeah all the technical talk about sample rates are impressive to us in the feild but the kid next door listening to the music could care less.

[NYMorningstar]

Originally Posted by scorpio01169

really....in the end it doesnt really matter, i mean does the end listener really care about what sample rate you used or what converters you used. if you get your music to sound understandable then thats what people care about, yeah all the technical talk about sample rates are impressive to us in the feild but the kid next door listening to the music could care less.

The end user 'does care about the quality' of the sound and you don't want to lose site of that. It's more than just getting it to sound understandable especially with the newer generations that are evolving that are so damn versitile and intelligent.

Did I mention spoiled? Generally they want the best of the best. You should try talking to the kid next door to get the real truth.

[scrubs]

Originally Posted by NYMorningstar

The end user 'does care about the quality' of the sound and you don't want to lose site of that.

Then why does most music coming out of major labels these days sound terrible...oversquashed, autotuned shite that will eventually be encoded at 128kbps and listened to with crappy earbuds?

Granted, a better base recording will stand up to this degradation with a little more finesse, but the reality is, most end listeners care less about audiophile sound quality than they do about whether the song is "kickin" or has lyrics tap into their teenage rage.

[NYMorningstar]

Originally Posted by scrubs

Then why does most music coming out of major labels these days sound terrible...oversquashed, autotuned shite that will eventually be encoded at 128kbps and listened to with crappy earbuds?

Granted, a better base recording will stand up to this degradation with a little more finesse, but the reality is, most end listeners care less about audiophile sound quality than they do about whether the song is "kickin" or has lyrics tap into their teenage rage.

I can't agree with you that most music coming out of major labels these days sounds terrible. I'd agree there is a volume war going on because some artists insist on it but I believe it's a fad and will go away when they figure out that people are less likely to continually listen to loud oversquashed sound.

You have a point when you restrict your audience to teenagers but eventually we all grow up and out of the rage scene. It's also not an either or question because most listeners care about both the sound quality and about whether the song is "kickin" or has lyrics.

[Fishmed]

I record mostly in 48/24 when I use my external HD recorder, but if I am doing a project on the PC, I will use 44.1/24. Space is not an issue because I back up projects to DVD and remove them from my hard drives once the project is over.

[amra]

I record at 24-Bit, 96Khz.
Why because I can...lol.
I have a powerful PC, and a terrabyte of hard drive space. I can't hear any difference, but it doesn't make my machine run noticeably different recording in this format either. Due to my (admittedly shaky) understanding of the Nyquist effect which states something to the effect of "unambiguous reconstruction is possible if the signal is bandlimited and the sampling frequency is greater than twice the signal bandwidth", there seemed to be at least a possibility that recording at higher than 48Khz might have some benefit.

Peace!
amra

[BigRay]

Originally Posted by amra

I record at 24-Bit, 96Khz.
Why because I can...lol.
I have a powerful PC, and a terrabyte of hard drive space. I can't hear any difference, but it doesn't make my machine run noticeably different recording in this format either. Due to my (admittedly shaky) understanding of the Nyquist effect which states something to the effect of "unambiguous reconstruction is possible if the signal is bandlimited and the sampling frequency is greater than twice the signal bandwidth", there seemed to be at least a possibility that recording at higher than 48Khz might have some benefit.

Peace!
amra

not really.
http://www.lavryengineering.com/docu...ing_Theory.pdf

[Dogman]

I think Audition defaults to 32bit 48 on my pc...and I have no idea if my junky equipement even really records at that.

[crosstudio]

i record in 24/48 with the belief that recording at what was the highest sample-rate possible would give me the best opportunity to create pollish-ready turds.

i tried 24/88.2 on a song few months ago but never completed the tracks. i figured it would be easier for the processor to cut every other slice in the mastering step.