Sample rate

[tech21man]

Mainly in the way the treble sounds in higher gain sounds. I definately hear difference and it is usually for the best, through my headphones mostly.

[newholland]

 

It's placebo effect for sure.   Sample rate means how many times your audio is measured per second. With the 44.1k rate the audio is being captured in 44100 slices per second. With 48k, the audio is being sliced 48000 times per second. THE HUMAN EAR IS NOT EQUIPPED TO NOTICE THE 3900/sec DIFFERENCE. PERIOD.   Bit-rate is simply how many volume registers you have between the noise floor and the headroom. It means you can record softer and louder parts with 24bit-rate than you would with 16bit-rate. The problem is, no one can even hear the noise floor at 16. And anything that would clip the headroom at 16 would be outside of most musical dynamics anyway.   People love to talk about {censored} they don

[guitarbilly74]

Better deal with it

Hearing differences in sample rate doesn't mean you have better ears. It means you have a weaker brain that is more subject to placebo.

Deal with it.

[newholland]

 

Hearing differences in sample rate doesn't mean you have better ears. It means you have a weaker brain that is more subject to placebo. Deal with it.

 

 

well-- what about on mixdown to mp3? don't you ever hear differences in treble as you go from a .wav to and mp3? i sure do-- i usually feel like the end result's more hashy on top-- and i mix to 320 for bands to review. it doesn't sound TERRIBLE.. but it sounds way LESS terrible going from a wav at 88.2 to mp3 than it does running 44.1 to mp3.

 

also-- depending on what you're using-- {censored}.. it's entirely possible with a 57 and a regular interface input that you'd hear ZERO difference. not necessarily the case though that it HAS to be placebo.

[guitarbilly74]

no-- it's not only about noise floor-- it's also about high frequency resolution. surprisingly-- you CAN hear artifacts-- even if digital theoretically goes to 20k-- you'll get grit and such way lower due to the fact that the computer has to truncate some information. as you make it sample MORE-- it truncates less because there's more 'data space'-- hence the larger file size. MOSTLY-- i'd agree with you-- but the way it seems to work is that if you give a longer decay time to something-- it sounds more natural- particularly in the high end. some biz about 'aliasing'. can't speak legit science about it to the details-- but i can say, on my interface-- the top end's better at 88.2 than 44.1.

 

The highest frequency on the 44.1K rate is 22kHz. On 48k rate it increases to about 30kHz.

 

The highest frequency the human ear can hear is.. well, 20kHz. 2kHz less than what the 44.1k provides. And that's a person with excellent hearing, not folks that have been standing next to guitar amps and cymbals for years

 

Enjoy your anecdotal facts and please continue to disregard science. Might as well add some Christian lyrics to your songs. Why don't record a track about the Noah's arc in 96kHz? That would be awesome in many ways.

[guitarbilly74]

 

well-- what about on mixdown to mp3?

 

 

mp3s utilize compression and dithering to squash the file to a manageable size. It's an entirely different subject than sample and bit rates.

[Kardula]

what's the sampling rate of analog recording?

[newholland]

The highest frequency on the 44.1K rate is 22kHz. On 48k rate it increases to about 30kHz. The highest frequency the human ear can hear is.. well, 20kHz. 2kHz less than what the 44.1k provides. And that's a person with excellent hearing, not folks that have been standing next to guitar amps and cymbals for years Enjoy your anecdotal facts and please continue to disregard science. Might as well add some Christian lyrics to your songs. Why don't record a track about the Noah's arc in 96kHz? That would be awesome in many ways.

 

uh.. okay billy. that you can only hear TO 20k, doesn't mean you can't discern BAD SOUNDING sounds above 12k. even if your hearing stops at 20k-- you CAN hear sound, and there's a definite difference in my experience. what anecdotal or any other evidence are you offering besides 'digital theory'? THEORY is one thing, but nobody's going to show me on an oscilloscope that treble is better or worse. human ears are a hell of a lot more sensitive than a lot of 'theorists' propose.

 

and what exactly that has to do with christian lyrics etc... hrm.. no idea there, but i think you may've just started drinking early today.

[guitarbilly74]

uh.. okay billy. that you can only hear TO 20k, doesn't mean you can't discern BAD SOUNDING sounds above 12k. even if your hearing stops at 20k-- you CAN hear sound, and there's a definite difference in my experience. what anecdotal or any other evidence are you offering besides 'digital theory'? THEORY is one thing, but nobody's going to show me on an oscilloscope that treble is better or worse. human ears are a hell of a lot more sensitive than a lot of 'theorists' propose. and what exactly that has to do with christian lyrics etc... hrm.. no idea there, but i think you may've just started drinking early today.

 

yeah science and math is just "theory" but we all have to take it for a FACT that you say you can hear the difference, because well.. you said so. Gotcha. (that's where the christian joke comes in btw, they LOVE to do that {censored}).

 

Why don't YOU prove you can hear the difference for the extra 3900 slices per second on frequencies over 12k? You can't.

 

 

Not to mention that once you have a CD you have a 16/44.1k product in hand, no matter how high you recorded it. Does every cd in the world sound muffled to your dog-like ears? Please.

[zac503]

ITT: GuitarBilly relates information that is mostly technically correct, but fails to acknowledge that one man's own anecdotal experiences combined with a moderately simplified interpretation of the scientific aspects does not make him the worlds foremost expert in the subject matter of sample rates. In his attempt to defend his position, he ends up sounding like an ass as he tries to make his understanding of the subject apply to other people's experiences through the use of insistence.

[guitarbilly74]

 

ITT: GuitarBilly relates information that is correct and I don't have any way or knowledge to dispute it with real numbers so I am just going to call him an asshole to look good on the internet. .

 

 

Whatever.

[syscrusher]

 

there's sonic increases to a good degree-- but the factor that makes the difference is your converters. for a lot of interfaces-- you'd probably never hear a difference. speaking only from my own experience-- i bumped mine up from 44.1 to 88.2 and heard a WORLD of difference, especially in the granularity of cymbals and vocal top end. i tried 192-- but my computer didn't like the size of the files. If I were recording classical music with a stereo pair? no question-- i'd use 192khz. but for multitracking.. no way. I split the difference and do 88.2 on anything i'm multitracking, as it's gonna end up 44.1. as to the OP's question-- i know on SOME interfaces-- the limiting factor is other inputs being turned on. see if your control program lets you shut off other inputs-- and generally that'll give the processor enough bandwidth to let you bump up to a higher hz.

 

 

That's the same experience I had. I use 24/96 for everything. I find the high end nicer and clearer; less gritty. It's like more air to it for lack of better descriptive words. Another thing is, that I found I can have a better dynamic range between lower volume and higher volume with less to no compression or clipping.

[newholland]

yeah science and math is just "theory" but we all have to take it for a FACT that you say you can hear the difference, because well.. you said so. Gotcha. (that's where the christian joke comes in btw, they LOVE to do that {censored}). Why don't YOU prove you can hear the difference for the extra 3900 slices per second on frequencies over 12k? You can't. Not to mention that once you have a CD you have a 16/44.1k product in hand, no matter how high you recorded it. Does every cd in the world sound muffled to your dog-like ears? Please.

 

sigh.

 

you can have your experts and your numbers-- i actually listen with my ears. you also can't prove that you CAN'T hear the difference-- you're just unwilling to admit there's a difference because you BELIEVE more strongly that your ears are 'inaccurate' and some expert is probably right moreso than you. sorry about the inferiority complex you got there, holmes... must suck!

 

i didn't say anything but 'IN MY EXPERIENCE'.. that sir, is real. i'd say, do the work and find out for yourself to anybody. but the idea that 'yeah- don't bother-- it's placebo' as a reaction to the problem is intellectual sloth at its {censored}ing worst-- because a) i have done the experiment, and b) i'm not afraid to posit an opinion based on experience, and c) i stated it as such. no flies on me. where'd your opinion come from? the internetz?

[newholland]

in the interest of following this down the road though-- i will give billy this: if you're doing modern production with all close mics, loads of compression on each channel, a {censored}ton of buss compression-- sure.. what differences would you hear? probably very little. the MAIN advantage of a higher khz (in MY experience) is in the time domain in higher frequencies-- ie verb tails and room sounds. if you don't use room sound to your advantage... it's gonna be way less of an issue for you.

[chrispsullivan]

Try recording something on a midi keyboard over 20khz and see if you can hear it. Here's a hint. You can't.

[zac503]

Whatever.

 

 

New Holland has demonstrated more than once on here that he knows a thing or two about recording and recording techniques. He's considerate and often offers insight into the process which he backs up with details about how he completed the work and often posts high quality clips to prove that it's not all bull{censored}.

 

You're typically a cool guy, but your argument here makes you sound like a first year college student that just learned a neato fact that he feels like he has to bring up in any situation that is tangentially related. The truth is, that "not possible" and "works 100% of the time" are concepts that rarely hold up in the real world.

 

You're argument so far is that it's not possible to perceive a difference in the outcomes of an audio file by adjusting the sample rate, which, by and large, I agree with. However, I don't understand your insistence in mocking people who have made contrary claims by posting detailed examples of what conditions they perceived a difference. At the very least, New Holland has done a much better job of explaining the parameters of his opinion, whereas you've only managed to say "nuh uh".

[OverDriven]

But it's Joe... You know he knows everything about everything. His setup is probably just better than yours, and he will tell you where to draw the line.

 

Actually, there is a difference when running synths at higher sampling rates (as I mentioned). EVEN when downsampling afterward.

 

http://www.gearslutz.com/board/gear-shoot-outs-sound-file-comparisons-audio-tests/582822-running-your-software-synths-higher-sample-rates.html

 

If you can't hear the difference here then it's time to get your ears checked.

 

Dave, you're still a pathetic excuse for a man. Not only do you search for opportunities to call me out, you do so without knowing what you're talking about at all and simply going by someone's opinion because they're contesting what I said.

 

Same sorry {censored} you pulled in the JCM800 thread where you failed to realize that the very quote you posted to "back up" your claims about the diodes in the 2210 and 2205 actually verified that you were wrong. It wasn't even worth posting a reply at the time, but you apparently don't know that the very function of a limiter is to create clipping. That's what it does - clip high amplitude waveforms beyond a set threshold. The quote said that all gain came from the tubes, which is true. Gain is the amplification of a signal. Apparently you aren't aware of the difference between gain and clipping. If you shove more signal through a limiter, it eventually becomes a square wave, creating jagged distortion. The fact that you don't even have a grasp on simple concepts such as this shows that you're completely unqualified to comment on almost anything audio related. BTW, this was also confirmed to me in a conversation with Dave Bray, who did work on my 2210.

[OverDriven]

 

Maybe you can help me with another question I have as you have given useful info on the topic of recording. I am aware of the loss in quality you get when publishing a mix as an mp3 file, and obviously traditional CD's don't use the mp3 format, so how do I get that CD quality in final publishing?

 

 

Sure man. Look into mixing down to FLAC. Mixing down to WAV is a complete waste of space. FLAC is actually better due to it's ability to store metadata.

[Markdude]

Try recording something on a midi keyboard over 20khz and see if you can hear it. Here's a hint. You can't.

 

 

wat

 

MIDI isn't audio.

 

Anyway, that argument is silly since basically any sound that isn't a sine wave has undertones too.

[Markdude]

Isn't the point of recording at a higher sample rate supposed to improve the sound quality when the finished song is transferred to a lower sample rate, like when the music is transferred to a CD? I'm still very much a recording n00b, but I remember reading something like that in an article somewhere.

I've heard this a lot of times too - that even though the final mix will get downsampled to 44.1, the audio algorithms can work more efficiently while mixing/recording at a higher sample rate and the final mix ends up sounding better as long as downsampling was the very last step. Or something. I dunno, I've never been able to tell any difference. I feel like people who obsess over this stuff and chase it wouldn't be so concerned if their recording/mixing skills in general were better. It's not going to be the magic thing that makes your mix sound pro.

[Rad Machine]

I blame wikipedia for this argument...

[Tommy Horrible]

It's placebo effect for sure. Sample rate means how many times your audio is measured per second. With the 44.1k rate the audio is being captured in 44100 slices per second. With 48k, the audio is being sliced 48000 times per second. THE HUMAN EAR IS NOT EQUIPPED TO NOTICE THE 3900/sec DIFFERENCE. PERIOD. Bit-rate is simply how many volume registers you have between the noise floor and the headroom. It means you can record softer and louder parts with 24bit-rate than you would with 16bit-rate. The problem is, no one can even hear the noise floor at 16. And anything that would clip the headroom at 16 would be outside of most musical dynamics anyway. People love to talk about {censored} they don

[Mesa4x12er2]

 

Actually, there is a difference when running synths at higher sampling rates (as I mentioned). EVEN when downsampling afterward. http://www.gearslutz.com/board/gear-shoot-outs-sound-file-comparisons-audio-tests/582822-running-your-software-synths-higher-sample-rates.html If you can't hear the difference here then it's time to get your ears checked.

 

 

Honestly dude if we're talking about Soft Synths then it sounds like that has more to do with its algorithm than the actual sound. If you can create a sound and then hear the same sound once downsampled then it was possible to have this sound created at the same sample rate as it now. This is 100% proof the sound does in fact exist and is possible in the lower sample rate as it does exist in the lower sample rate. So what this means is there is something going on within the processing itself not the sample rate as you've just definitively proven that the limitation does not exist in the sample rate itself.

 

2nd you're not using some exponentially higher sample rate. You're talking about the difference between 44.1 and 48. The biggest determination on which to use is generally whether your audio is to be just audio or if it's going to video and only because these are the standards, not because one sounds "way better."

 

Anyway, your synth argument regurgitation ala Overdriven Wiki style still doesn't address anything regarding actually recording sound. You absolutely do NOT notice a difference between audio recorded at 48 and 44.1. I can hear some subtle differences on thing like reverb tails but that's at way higher sample rates and VERY subtle. Definitely not something that will make or break your recording or that you will even notice unless you are scrutinizing in a well treated room, with great monitors, through great converters. I'm willing to bet you're using a budget interface, through budget monitors, in an untreated room.

 

There is a much bigger difference in sound quality going from a wav to an mp3 than higher sample rate to lower. Hell the brand of guitar strings will have more of a noticeable effect on your recording than sample rate.

 

Most {censored} you will hear commercially was recorded as 44.1/24-bit and then dithered down to 16.

 

OD you're the same douche that was giving out recording advice talking about EQing in solo, which is ass backwards and detrimental to your mix. I'd put my money on the fact that when you saw this topic you googled it and found that synth thread and then ran in here like a parrot. It's pretty clear you have some sort of an inferiority complex and have something to prove. The noobs that don't know any better might take your cocky false confidence as truth. The rest of us just think you're a try-hard joke. I really do feel bad for you.

[Markdude]

 

OD you're the same douche that was giving out recording advice talking about EQing in solo, which is ass backwards and detrimental to your mix. I'd put my money on the fact that when you saw this topic you googled it and found that synth thread and then ran in here like a parrot. It's pretty clear you have some sort of an inferiority complex and have something to prove. The noobs that don't know any better might take your cocky false confidence as truth. The rest of us just think you're a try-hard joke. I really do feel bad for you.

 

 

EQing in solo isn't always a bad idea. Sometimes you can tell (in a full mix) that a certain track/instrument is bloated at a certain frequency range but it's extremely hard to tell WHERE when you're listening to the whole mix, and it helps to solo the track to find where the frequency buildup is. But yeah, I don't think you should mix while solo'd more than 15-20% of the time.

[Mesa4x12er2]

 

EQing in solo isn't always a bad idea. Sometimes you can tell (in a full mix) that a certain track/instrument is bloated at a certain frequency range but it's extremely hard to tell WHERE when you're listening to the whole mix, and it helps to solo the track to find where the frequency buildup is. But yeah, I don't think you should mix while solo'd more than 15-20% of the time.

 

 

Yeah if you hear something in it causing a problem like an offending frequency you can solo it to find it and cut it, but in general to create a mix you do not EQ in solo as that defeats the purpose of a mix which is to make everything work together. He was suggesting EQing in solo to get everything to sound good on its own which is a huge pitfall in mixing.