Go Back   Winamp & Shoutcast Forums > Community Center > General Discussions

Reply
Thread Tools Search this Thread Display Modes
Old 12th May 2006, 01:54   #1
nukesgoboom
Member
 
Join Date: Oct 2005
Posts: 50
Would re-encoding a 44.1khz mp3 to a 48khz mp3 do anything useful or is it a waste?

i know that 44.1 khz means that the audio is being sampled 44100 times a second, so would increasing that to 48000 times a second make any difference at all, seeming as it is already encoded at 44.1 khz? ive noticed that it does not change the file size at all.

i know that re-encoding a mp3 file to a higher BITRATE is useless, but im wondering if re-encoding to a higher FREQUENCY would be useful?
nukesgoboom is offline   Reply With Quote
Old 12th May 2006, 02:00   #2
gaekwad2
Foorum King
 
gaekwad2's Avatar
 
Join Date: Jul 2003
Location: bar2000
Posts: 11,457
no

Both reencoding and resampling will hurt quality.
gaekwad2 is offline   Reply With Quote
Old 12th May 2006, 04:33   #3
Nico1980
Senior Member
 
Join Date: Apr 2006
Posts: 142
Yes, can hurt quality BUT could give you more bright in some sounds. Test it and you will notice what im talking about. Hi-Hat of the drums, cymbals and this type of sounds will sound brighter on some recordings (in new ones, maybe oldies not).
I recommend it if you like music bright...
Nico1980 is offline   Reply With Quote
Old 12th May 2006, 04:35   #4
dlichterman
Forum King
 
dlichterman's Avatar
 
Join Date: Mar 2001
Location: Where Hell Froze Over
Posts: 2,466
No, if its already at 44.1 then going higher will not do anything. It is only beneficial if you re-encode from the origional uncompressed source(the CD).

Software is like sex: It's better when it's free.-*-If at first you don't succeed; call it version 1.0-*-Guess the band from pics game
dlichterman is offline   Reply With Quote
Old 12th May 2006, 04:35   #5
mikm
Major Dude
 
mikm's Avatar
 
Join Date: May 2001
Location: somewhere else
Posts: 1,289
Wouldn't it be better to brighten music using a DSP designed to brighten music rather than doing something stupid like resampling/reencoding?

powered by C₈H₁₀N₄O₂
mikm is offline   Reply With Quote
Old 12th May 2006, 04:58   #6
Psythik
Got his CT back
and didn't pay $10
(Forum King)
 
Psythik's Avatar
 
Join Date: Aug 2002
Location: Phoenix, AZ
Posts: 3,583
I thought that CDs were natively encoded in 44.1KHz, so ripping it to 48 would do nothing more than waste disk space?

This is a sig of some nature.
Psythik is offline   Reply With Quote
Old 12th May 2006, 05:01   #7
mikm
Major Dude
 
mikm's Avatar
 
Join Date: May 2001
Location: somewhere else
Posts: 1,289
In any event, I doubt that most people notice a significant difference between music at 44.1khz and 48khz.

powered by C₈H₁₀N₄O₂
mikm is offline   Reply With Quote
Old 12th May 2006, 05:55   #8
dlichterman
Forum King
 
dlichterman's Avatar
 
Join Date: Mar 2001
Location: Where Hell Froze Over
Posts: 2,466
I guess I stand corrected

Software is like sex: It's better when it's free.-*-If at first you don't succeed; call it version 1.0-*-Guess the band from pics game
dlichterman is offline   Reply With Quote
Old 12th May 2006, 06:08   #9
swingdjted
DRINK BEER NOW
(Forum King)
 
swingdjted's Avatar
 
Join Date: Feb 2006
Location: Northern West Virginia
Posts: 10,074
Send a message via AIM to swingdjted Send a message via Yahoo to swingdjted
Quote:
Originally posted by dlichterman
No, if its already at 44.1 then going higher will not do anything. It is only beneficial if you re-encode from the origional uncompressed source(the CD).
This is the most true statement of the above posts.

Your ears, if normal, can hear the frequency range from 20Hz (low bass) to 20KHz (20,000 Hz - extremely high treble)

People with very good (and rare) ears can maybe hear highs up to 30 or 40 KHz, but again, that is rare.

Therefore, I would assume the difference from 44.1 to 48 KHz would not be recognizeable to the human ear.

Also, keep in mind that speakers that can produce these tones at the far low and high end are painfully expensive and also rare.

Don't forget to live before you die.
swingdjted is offline   Reply With Quote
Old 12th May 2006, 10:15   #10
gaekwad2
Foorum King
 
gaekwad2's Avatar
 
Join Date: Jul 2003
Location: bar2000
Posts: 11,457
No, the most true statement is
Quote:
Originally posted by Psythik
I thought that CDs were natively encoded in 44.1KHz, so ripping it to 48 would do nothing more than waste disk space?
Resampling from 44.1 to 48kHz will give you as much additional treble as resizing a picture from 441 to 480 pixels will give you additional detail.

Unless you use a crappy resampler and like the sound of IM distortion that is.
gaekwad2 is offline   Reply With Quote
Old 12th May 2006, 15:46   #11
xzxzzx
Forum King
 
xzxzzx's Avatar
 
Join Date: Aug 2002
Posts: 7,254
Quote:
Originally posted by Nico1980
Yes, can hurt quality BUT could give you more bright in some sounds. Test it and you will notice what im talking about. Hi-Hat of the drums, cymbals and this type of sounds will sound brighter on some recordings (in new ones, maybe oldies not).
I recommend it if you like music bright...
No, it won't. It really won't. Anything you hear will be either artifacts from a crappy resampler, or your imagination.

Resampling and reencoding it to 48 KHz will only make it sound worse, mostly due to reencoding it.

Freedom of speech is the basic freedom of humanity. When you've lost that, you've lost everything.
1\/\/4y 34|<$p4y 1gp4y 33714y, 0d4y 0uy4y? | Roses are #FF0000; Violets are #0000FF; chown -R ${YOU} ~/base
The DMCA. It really is that bad. : Count for your life.
xzxzzx is offline   Reply With Quote
Old 12th May 2006, 16:06   #12
nukesgoboom
Member
 
Join Date: Oct 2005
Posts: 50
thanks very much for all of the informative replies. you certainly showed me that resampling the frequency will not do anything if the original frequency is lower.

i was interested in this, because it seems changing the frequency to a higher sampling rate does NOT increase the filesize, so why doesnt everyone just use 48khz, or 99khz, or even higher? the filesize doesnt change. is a higher sampling rate only able to be used with specific hardware or something?

also, i read that most headphones, even fairly expensive ones, only reproduce sound of up to 20khz or 25khz, so even if my music is at 44.1khz i wont hear it, but i will say that they still sound great.

you know? whats the point, the cd audio is 44.1 khz but most headphones are capped at 25khz!!!!
nukesgoboom is offline   Reply With Quote
Old 12th May 2006, 16:16   #13
gaekwad2
Foorum King
 
gaekwad2's Avatar
 
Join Date: Jul 2003
Location: bar2000
Posts: 11,457
You need a sample rate of at least (a little more than) twice the difference between the lowest and highest frequencies you want to reproduce.

Therefore 44.1kHz can 'only' reproduce audio up to 22kHz.

If you use an uncompressed (or even a losslessly compressed) format higher sample rates lead to bigger files.

If you use lossy compression with a fixed bitrate then of course the file size will remain the same (remember: bitrate = number of bits (or thousand bits) per second).
Quality will at best be the same but since most if not all lossy codecs aren't tuned for high samplerates it will most likely sound worse.
gaekwad2 is offline   Reply With Quote
Old 12th May 2006, 16:29   #14
xzxzzx
Forum King
 
xzxzzx's Avatar
 
Join Date: Aug 2002
Posts: 7,254
Quote:
Originally posted by nukesgoboom
i was interested in this, because it seems changing the frequency to a higher sampling rate does NOT increase the filesize, so why doesnt everyone just use 48khz, or 99khz, or even higher? the filesize doesnt change. is a higher sampling rate only able to be used with specific hardware or something?
That's incorrect. The file size does change. Twice the sampling rate means twice the file size. Note that Mp3s typically filter out high-frequency sound, so while an Mp3 won't be any larger, it won't have any of the data that a higher-frequency file would have.

Quote:
Originally posted by nukesgoboom
also, i read that most headphones, even fairly expensive ones, only reproduce sound of up to 20khz or 25khz, so even if my music is at 44.1khz i wont hear it, but i will say that they still sound great.
It takes a sampling rate of at least twice the frequency of a sound to capture it.

Quote:
Originally posted by nukesgoboom
you know? whats the point, the cd audio is 44.1 khz but most headphones are capped at 25khz!!!!
44.1 / 2 = 22.05, and realistically, even less than that.

Freedom of speech is the basic freedom of humanity. When you've lost that, you've lost everything.
1\/\/4y 34|<$p4y 1gp4y 33714y, 0d4y 0uy4y? | Roses are #FF0000; Violets are #0000FF; chown -R ${YOU} ~/base
The DMCA. It really is that bad. : Count for your life.
xzxzzx is offline   Reply With Quote
Old 12th May 2006, 17:28   #15
fwgx
Rudolf the Red.
(Forum King)
 
fwgx's Avatar
 
Join Date: Nov 2000
Posts: 9,314
Quote:
Originally posted by xzxzzx
It takes a sampling rate of at least twice the frequency of a sound to capture it.
Not quite true. All digitisation of analogue signals is going to reproduce an aproximation of the original, you're never going to fully capture the sound. The reason that we sample at double the highest frequency is to avoid ALIASING, which was a theory devised by Nyquist.

Sampling always comes down to aliasing.

"We think science is interesting and if you disagree, you can fuck off."
fwgx is offline   Reply With Quote
Old 12th May 2006, 18:05   #16
xzxzzx
Forum King
 
xzxzzx's Avatar
 
Join Date: Aug 2002
Posts: 7,254
Quote:
Originally posted by Phily Baby
Not quite true. All digitisation of analogue signals is going to reproduce an aproximation of the original, you're never going to fully capture the sound. The reason that we sample at double the highest frequency is to avoid ALIASING, which was a theory devised by Nyquist.

Sampling always comes down to aliasing.
Aliasing is going to produce a lower frequency... Thus failing to capture the sound. No?

Freedom of speech is the basic freedom of humanity. When you've lost that, you've lost everything.
1\/\/4y 34|<$p4y 1gp4y 33714y, 0d4y 0uy4y? | Roses are #FF0000; Violets are #0000FF; chown -R ${YOU} ~/base
The DMCA. It really is that bad. : Count for your life.
xzxzzx is offline   Reply With Quote
Old 13th May 2006, 00:45   #17
nukesgoboom
Member
 
Join Date: Oct 2005
Posts: 50
why would dividing 44.1 by 2 mean that its ok for headphones to cap at 25khz? im sorry i dont understand. does that mean per earpiece? or do you mean that its alright for the headphones to cap at 25khz because the human ear cant detect sound higher than that? and it would be much more expensive to manufacture headphones that reproduce higher sampling rates?
nukesgoboom is offline   Reply With Quote
Old 13th May 2006, 02:26   #18
fwgx
Rudolf the Red.
(Forum King)
 
fwgx's Avatar
 
Join Date: Nov 2000
Posts: 9,314
You're never going to "capture" the sound anyway, that's the point. the Nyquist frequency and ailasing just means that you produce a digital form that's good enough to be almost unoticable to the original. It's not so much that it will produce a lower frequency but that information will be lost and be noticable.

I always like the analogy of a spinning disc with a hole in it. as the disc spins you get to see what's behind it. now imagine someone opening a door and appearing then disapearing soon after. If the person appears and disappears faster than the frequency of the spinning disc then you will not know that the person appeared during the time when you could not look through the hole.

Now suppose you can look through the hole and see the person for 1/5th of a second but you can only take one reading withing that 1/5th of a second, where do you do it? at the start or end of the period? if the person is entering or leaving the room at 1/10th a second then you will lose information because you will only note that he entered once, when he actually entered twice..

So it's not so much that we fail to capture the signal but that you fail to capture enough for us to notice. Even sampling at 44kHz isn't really enough to capture the sound because the sound waves are analogue and therefore effectivly have infinate resolution that will never be captured by a digitisation process - because that process will always involve sampling.

"We think science is interesting and if you disagree, you can fuck off."
fwgx is offline   Reply With Quote
Old 13th May 2006, 10:48   #19
gaekwad2
Foorum King
 
gaekwad2's Avatar
 
Join Date: Jul 2003
Location: bar2000
Posts: 11,457
Quote:
Originally posted by Phily Baby
You're never going to "capture" the sound anyway, that's the point. the Nyquist frequency and ailasing just means that you produce a digital form that's good enough to be almost unoticable to the original. It's not so much that it will produce a lower frequency but that information will be lost and be noticable.
No, you are capturing the sound, but only in an interval as wide as the Nyquist frequency. You can reconstruct any wave inside that interval, what's outside is irrelevant.
Quote:
Originally posted by Phily Baby
I always like the analogy of a spinning disc with a hole in it. as the disc spins you get to see what's behind it. now imagine someone opening a door and appearing then disapearing soon after. If the person appears and disappears faster than the frequency of the spinning disc then you will not know that the person appeared during the time when you could not look through the hole.
That's why you have to lowpass (or bandpass in case of subsampling) the analog signal before digitizing it.
Quote:
Originally posted by Phily Baby
Now suppose you can look through the hole and see the person for 1/5th of a second but you can only take one reading withing that 1/5th of a second, where do you do it? at the start or end of the period? if the person is entering or leaving the room at 1/10th a second then you will lose information because you will only note that he entered once, when he actually entered twice..

So it's not so much that we fail to capture the signal but that you fail to capture enough for us to notice. Even sampling at 44kHz isn't really enough to capture the sound because the sound waves are analogue and therefore effectivly have infinate resolution that will never be captured by a digitisation process - because that process will always involve sampling.
Well, no recording method captures the signal without any degradation.
Analog recording simply has a 'built-in lowpass', it's not a sharp cut-off as with digital but in most cases this only means that frequencies below 20kHz will already get severely degraded.
gaekwad2 is offline   Reply With Quote
Old 13th May 2006, 16:01   #20
nukesgoboom
Member
 
Join Date: Oct 2005
Posts: 50
very interesting analogy about the spinning disc with a hole in it. the person behind the door would be the higher frequency that would not be detectable to the human ear, am i correct? since you cannot sample it, it doesnt matter anyway because you would not hear it? or have i misunderstood?
nukesgoboom is offline   Reply With Quote
Old 13th May 2006, 16:14   #21
gaekwad2
Foorum King
 
gaekwad2's Avatar
 
Join Date: Jul 2003
Location: bar2000
Posts: 11,457
You have to filter it away before sampling, otherwise you'd get an audible artifact. Apart from that, yes.
gaekwad2 is offline   Reply With Quote
Old 13th May 2006, 16:19   #22
deeder7001
Jesus Freak
(Forum King)
 
deeder7001's Avatar
 
Join Date: Aug 2002
Location: Tempe, AZ
Posts: 5,562
Send a message via AIM to deeder7001 Send a message via Yahoo to deeder7001
Quote:
Originally posted by nukesgoboom
why would dividing 44.1 by 2 mean that its ok for headphones to cap at 25khz? im sorry i dont understand. does that mean per earpiece? or do you mean that its alright for the headphones to cap at 25khz because the human ear cant detect sound higher than that? and it would be much more expensive to manufacture headphones that reproduce higher sampling rates?
when specs for headphones say 25khz, it's talking about the sound frequency and not the sample rate. you can use the same headphones with 96khz sample rate and they'll play it.

There is no sig.
deeder7001 is offline   Reply With Quote
Old 13th May 2006, 18:13   #23
xzxzzx
Forum King
 
xzxzzx's Avatar
 
Join Date: Aug 2002
Posts: 7,254
Quote:
Originally posted by nukesgoboom
why would dividing 44.1 by 2 mean that its ok for headphones to cap at 25khz? im sorry i dont understand. does that mean per earpiece? or do you mean that its alright for the headphones to cap at 25khz because the human ear cant detect sound higher than that? and it would be much more expensive to manufacture headphones that reproduce higher sampling rates?
Because a sampling rate of 44.1 KHz means that it can make a sound with the frequency of 22.05 KHz, at the most. Sampling rates are not the same as frequencies.

Quote:
Originally posted by Phily Baby
You're never going to "capture" the sound anyway, that's the point. the Nyquist frequency and ailasing just means that you produce a digital form that's good enough to be almost unoticable to the original. It's not so much that it will produce a lower frequency but that information will be lost and be noticable.
As geakwad2 said:
Quote:
Originally posted by gaekwad2
No, you are capturing the sound, but only in an interval as wide as the Nyquist frequency. You can reconstruct any wave inside that interval, what's outside is irrelevant.
You can capture and recreate the sound perfectly up to the frequencies of half of the sampling rate (less, of course, in practice) and the accuracy of the particular sample (16 bits or ~65536 discrete levels, less in practice), and of course, any inaccuracies introduced by the analog parts (the mic, the speakers, etc).

Obviously nothing is perfect, in order to get perfect reproduction, you need to break several laws of several types of physics...

Freedom of speech is the basic freedom of humanity. When you've lost that, you've lost everything.
1\/\/4y 34|<$p4y 1gp4y 33714y, 0d4y 0uy4y? | Roses are #FF0000; Violets are #0000FF; chown -R ${YOU} ~/base
The DMCA. It really is that bad. : Count for your life.
xzxzzx is offline   Reply With Quote
Old 13th May 2006, 23:36   #24
PulseDriver
w3 addict
(Major Dude)
 
PulseDriver's Avatar
 
Join Date: May 2004
Location: Norway
Posts: 1,806
You ear cannot hear frequenzies above 20KHz, and most samplerates above that, is just to keep soundquality stored as "better" than you can hear. In practice, media i stored up to 192KHz in some cases and most ASIO soundcards can sample 96KHz. It doesn't mean your ear can hear these frequenzies. So most headphones are above the soundquality you can dermain. The weakest side of headphones is rather in the other end of the frequenzy-scale, frequenzies below 20Hz.

When it comes to bits, if you imagine the the sound as a graphical wave, like you've probably seen in some audio-editing tools, the bits means how great resolution the dots make the straight lines go. Hard to explain, but between each bit, the sound is drawn in a straight line when extracted, or it can also be interpolated (made an average through each point to make the sound sound more smooth):

Sampling higher than original sound is, is just a waste of space, when it comes to both bits and frequenzy.

09 F9 11 01 9D 74 E8 5B D8 41 56 C3 63 56 81 C0
PulseDriver is offline   Reply With Quote
Old 14th May 2006, 00:01   #25
gaekwad2
Foorum King
 
gaekwad2's Avatar
 
Join Date: Jul 2003
Location: bar2000
Posts: 11,457
Imagining it as a stair-stepped wave is misleading though.
The waveform gets smoothed by the final lowpass, digital interpolation is only used to get this (analog) filter and its possible artifacts further away from the audible range.
(actually these steps are physically impossible anyway)

Below 20Hz isn't audible either btw, deep bass is between 40 and 80Hz (but indeed many headphones (and even more speakers) can't even properly reproduce that range).
gaekwad2 is offline   Reply With Quote
Old 14th May 2006, 00:07   #26
PulseDriver
w3 addict
(Major Dude)
 
PulseDriver's Avatar
 
Join Date: May 2004
Location: Norway
Posts: 1,806
It's not stairstepped. I never said horizontal or vertical lines, I said straight lines. Be it in any angle.

09 F9 11 01 9D 74 E8 5B D8 41 56 C3 63 56 81 C0
PulseDriver is offline   Reply With Quote
Old 14th May 2006, 00:16   #27
PulseDriver
w3 addict
(Major Dude)
 
PulseDriver's Avatar
 
Join Date: May 2004
Location: Norway
Posts: 1,806
forgot the damn attachment.. here it is.
Attached Images
File Type: jpg curve.jpg (14.2 KB, 313 views)

09 F9 11 01 9D 74 E8 5B D8 41 56 C3 63 56 81 C0
PulseDriver is offline   Reply With Quote
Old 14th May 2006, 00:20   #28
gaekwad2
Foorum King
 
gaekwad2's Avatar
 
Join Date: Jul 2003
Location: bar2000
Posts: 11,457
Straight lines like that aren't possible either (but I admit I misread it because it immediately reminded me of those misleading illustrations (often used to advertise x-times oversampling)).
gaekwad2 is offline   Reply With Quote
Old 14th May 2006, 17:17   #29
nukesgoboom
Member
 
Join Date: Oct 2005
Posts: 50
ah thank you it seems i was confusing the term sampling rate with the term frequency. thanks for point that out im sure i would have dug myself a hole with that little mistake, lol.

also thanks for explaining the headhpones oddness to me, since the human ear cannot detect (mostly) anything about 20khz then its fine. i was just curious why that was.

what is an ASIO soundcard? i have an audigy 2 zs, is that related at all??

nice explanation about the graphical wave for sound, PulseDriver. really appreciate that, now i kind of understand the 'wave' part. nice picture!

what is a lowpass? cutting off the lowest frequency? to save space?
nukesgoboom is offline   Reply With Quote
Old 14th May 2006, 18:05   #30
gaekwad2
Foorum King
 
gaekwad2's Avatar
 
Join Date: Jul 2003
Location: bar2000
Posts: 11,457
No, it's cutting off the high frequencies, in this case the ones above half the sample rate.

Lowpass = a filter that lets low frequencies pass through.
gaekwad2 is offline   Reply With Quote
Old 14th May 2006, 19:47   #31
Nico1980
Senior Member
 
Join Date: Apr 2006
Posts: 142
Quote:
Originally posted by xzxzzx
No, it won't. It really won't. Anything you hear will be either artifacts from a crappy resampler, or your imagination.

Resampling and reencoding it to 48 KHz will only make it sound worse, mostly due to reencoding it.
No, you are wrong.

Make a test and you will see.

Let me tell you that i have worked in many works related with audio.

The reencoding, if its well done will not be perceived and in SOME cases (the ones i explained before) you will 'feel' that some sounds are brighter.

The easiest way to know if im right or night is doing some test, easy!
Nico1980 is offline   Reply With Quote
Old 14th May 2006, 19:49   #32
Nico1980
Senior Member
 
Join Date: Apr 2006
Posts: 142
Quote:
Originally posted by mikm
In any event, I doubt that most people notice a significant difference between music at 44.1khz and 48khz.
Wrong.

It gives a very significante difference.

Just do a test...
Nico1980 is offline   Reply With Quote
Old 14th May 2006, 20:03   #33
gaekwad2
Foorum King
 
gaekwad2's Avatar
 
Join Date: Jul 2003
Location: bar2000
Posts: 11,457
Quote:
Originally posted by Nico1980
No, you are wrong.

Make a test and you will see.

Let me tell you that i have worked in many works related with audio.

The reencoding, if its well done will not be perceived and in SOME cases (the ones i explained before) you will 'feel' that some sounds are brighter.

The easiest way to know if im right or night is doing some test, easy!
You're wrong. I actually did test resamplers (for artifacts, not for audiophile mumbo-jumbo) and couldn't hear a difference.

But wait, of course that only proves my hearing is inferior to yours amirite?
gaekwad2 is offline   Reply With Quote
Old 14th May 2006, 21:43   #34
Guest232
Guest
 
Posts: n/a
I think "brighter" sounds, or a more "analog" sound, can be imitated with EQs and effect equipment, if they would do blind tests.
  Reply With Quote
Old 14th May 2006, 23:32   #35
ScorLibran
Resident Floydian
 
ScorLibran's Avatar
 
Join Date: Aug 2003
Posts: 6,217
Quote:
Originally posted by gaekwad2
You're wrong. I actually did test resamplers (for artifacts, not for audiophile mumbo-jumbo) and couldn't hear a difference.
He is indeed wrong. I've participated in dozens of independent double-blind ABX listening tests in the past few years, and in fact the guy who has administrated more of them than anyone else in recent times is a good friend of mine. (Meaning I've learned far more than my own direct experience has ever exposed me to.)

This type of testing is the only scientifically viable way to determine perceptual variance in audio signals. And it has proven that upsampling DOES NOT add audible high frequencies to an encoded audio signal. For essentially the same reason transcoding an MP3 @ 128kbps CBR to 320kbps CBR will not result in any better sound quality: Because an audio encoder cannot restore what was not there in the first place, and can't restore what was lost during any previous encoding.

As xzxzzx said, anything that you could differentiate would be only the effects of a poor transcode. Or placebo - which an ABX test eliminates.

I'm a psychosomatic sister running around without a leash.
ScorLibran is offline   Reply With Quote
Old 15th May 2006, 02:26   #36
PulseDriver
w3 addict
(Major Dude)
 
PulseDriver's Avatar
 
Join Date: May 2004
Location: Norway
Posts: 1,806
If you get more significat sound by converting from 44000Hz to 48000Hz it's because your recording equipment or whatever you use to convert it sucks so bad it adds noise at these frequenzies. Anyway, you cannot even hear frequenzies that high so what you say there is a load of bullshit.

What you hear could have been digitally applied upon the sound without changing the sample-rate.

09 F9 11 01 9D 74 E8 5B D8 41 56 C3 63 56 81 C0
PulseDriver is offline   Reply With Quote
Old 15th May 2006, 02:34   #37
mikm
Major Dude
 
mikm's Avatar
 
Join Date: May 2001
Location: somewhere else
Posts: 1,289
Quote:
Originally posted by Nico1980
Wrong.

It gives a very significante difference.

Just do a test...
The only place where sampling rates greater than 44.1khz are useful is for mastering /audio editing.

Read me

But I'll be quite happy to do a test. Provide me with two audio samples, one at 44.1khz and one at 48khz (or higher) where the ONLY difference is the sampling rate (no encoding w/ different tools, no mastering on different hardware).

powered by C₈H₁₀N₄O₂
mikm is offline   Reply With Quote
Old 15th May 2006, 05:21   #38
ScorLibran
Resident Floydian
 
ScorLibran's Avatar
 
Join Date: Aug 2003
Posts: 6,217
Quote:
Originally posted by mikm
But I'll be quite happy to do a test. Provide me with two audio samples, one at 44.1khz and one at 48khz (or higher) where the ONLY difference is the sampling rate (no encoding w/ different tools, no mastering on different hardware).
To be taken seriously, the onus is on him to prove he hears a difference with the results of an ABX test, not on you to prove that you don't.

He'll need lossless 44.1kHz and 48kHz encodings of the same audio sample, the means to convert them to WAV format (if they're not already WAVs) and an ABX Comparator.

I'm a psychosomatic sister running around without a leash.

Last edited by ScorLibran; 15th May 2006 at 07:42.
ScorLibran is offline   Reply With Quote
Old 15th May 2006, 06:44   #39
SSJ4 Gogitta
Followed by Gnomes
(Forum King)
 
SSJ4 Gogitta's Avatar
 
Join Date: Dec 2000
Location: West Virginia Googolplex: 10^10¹°° FB:/SSJ4.DominusDeus DeviantArt: DominusDeus XboX GT: A Wild Meeseeks Playstation 4: DominusDeus
Posts: 7,161
Here's a zip with 3 wav files I created in Sound Forge 7. They're all 15 seconds long, and all are the same 20 to 280 Hz sine wave sweep.

One is 44.1 kHz sample rate, one is 48 kHz sample rate, one is 192 kHz sample rate.

The rar file is 4.36 megs.

Download it here

SSJ4 Gogitta is offline   Reply With Quote
Old 15th May 2006, 15:56   #40
xzxzzx
Forum King
 
xzxzzx's Avatar
 
Join Date: Aug 2002
Posts: 7,254
Quote:
Originally posted by Nico1980
The reencoding, if its well done will not be perceived and in SOME cases (the ones i explained before) you will 'feel' that some sounds are brighter.
Yes, you might 'feel' that some sounds are brighter, but it has nothing to do with the audio file. See, it's actually impossible.

You're roughly saying that if you write the number 3 like this:

3

That makes it "more" of a 3 to a computer than this one:

3

But that's not true. They're both just "3". Everything else is in your mind.

Freedom of speech is the basic freedom of humanity. When you've lost that, you've lost everything.
1\/\/4y 34|<$p4y 1gp4y 33714y, 0d4y 0uy4y? | Roses are #FF0000; Violets are #0000FF; chown -R ${YOU} ~/base
The DMCA. It really is that bad. : Count for your life.
xzxzzx is offline   Reply With Quote
Reply
Go Back   Winamp & Shoutcast Forums > Community Center > General Discussions

Thread Tools Search this Thread
Search this Thread:

Advanced Search
Display Modes

Posting Rules
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off

Forum Jump