| Author
|
Mythbusters: Audio Production Special
|
Spindrift
Spindrift
Started Topics :
33
Posts :
1560
Posted : Sep 3, 2008 01:38
|
|
UnderTow
Started Topics :
9
Posts :
1448
Posted : Sep 3, 2008 09:49
|
Quote:
|
On 2008-09-03 00:24, UnderTow wrote:
And it is a bad idea to record signals that peak at or close to 0 dB FS as usually the last few dBs of the converter are not entirely linear so good gain staging means your digital noise floor will be above -144 dB FS.
|
|
Oops! That is not what I meant to write: The digital noise floor will be less than 144 dB below the signal level!
More later...
UnderTow |
|
|
Upavas
Upavas
Started Topics :
150
Posts :
3315
Posted : Sep 3, 2008 11:28
|
|
Elad
Tsabeat/Sattel Battle
Started Topics :
158
Posts :
5306
Posted : Sep 3, 2008 11:36
|
someone gonna comment about that samplerate change and the pitch change and that spcial relation between them?
|
|
|
|
Kane
IsraTrance Junior Member
Started Topics :
23
Posts :
1772
Posted : Sep 3, 2008 17:09
|
Quote:
|
On 2008-09-03 11:36, Elad wrote:
someone gonna comment about that samplerate change and the pitch change and that spcial relation between them?
|
|
I've always been sorta confused about this..I've experienced pitch change with sample rate change before, but what I don't understand why the pitch isn't affected when I have reaktor at 22050hz and all of my other audio at 44.1khz.
Anyone?
 You believe in the users?
Yeah, sure. If I don't have a user, then who wrote me? |
|
|
|
MadScientist
IsraTrance Full Member
Started Topics :
97
Posts :
1220
Posted : Sep 3, 2008 20:45
|
elad, its as simple as you got a 44.1khz wav, and now you insert it in a 88.2 project without src. result: the file will play back twice as fast as before, because it got just 44.100 samples in a second, but the project plays back 88.200 samples per second, so you re playing 2 secs of your wav in 1 sec. 
kane: I cant help you about that because I dont use reaktor that much, I could just imagine that reaktor syncs to the project sample rate without showing it.
https://soundcloud.com/hazak
"Have you ever had that feeling where you're not sure if you're awake or still dreaming?"
"Hmm, yeah... All the time, man - it's called mescaline. The only way to fly!" |
|
|
|
Spindrift
Spindrift
Started Topics :
33
Posts :
1560
Posted : Sep 3, 2008 21:39
|
Quote:
|
On 2008-09-03 17:09, Kane wrote:
I've always been sorta confused about this..I've experienced pitch change with sample rate change before, but what I don't understand why the pitch isn't affected when I have reaktor at 22050hz and all of my other audio at 44.1khz.
Anyone?
|
|
Samplerate conversion (``·.¸(``·.¸(``·.¸¸.·`´)¸.·`´)¸.·`´)
« .....www.ResonantEarth.com..... »
(¸.·`´(¸.·`´(¸.·`´``·.¸)``·.¸)``·.¸)
http://www.myspace.com/spindriftsounds
http://www.myspace.com/resonantearth |
|
|
|
UnderTow
Started Topics :
9
Posts :
1448
Posted : Sep 4, 2008 00:50
|
Quote:
|
On 2008-09-02 00:01, Kaz wrote:
I added the "neighboring frequencies" comment because we can, in fact, hear the difference between 44.1 and 88.2 KHz due to differences in character
|
|
That depends on the converters.
Quote:
|
You need at least double the maximum rate to get a frequency properly (one for the positive peak, one for the negative).
|
|
Not the peaks. Anywhere in the waveform.
Quote:
|
By doubling the level of interpolation, you make sure that even the finer characters of the sound are 100% accurate within your hearing range -
|
|
Level of interpolation? You interpolate to get a new value from exiting values. The term is not correct when you are talking about sampling. Also changing the sampling rate, increases the bandwidth, not accuracy.
Quote:
|
Technically, 80KHz sampling rate is enough,
|
|
60 Khz SR is enough.  It needs a filter with a curve equivalent to the natural roll-off of air. Good enough by my book.
Quote:
|
In films it was 24KHz, doubled, just because they wanted to be sure that they sound better than regular CDs.
|
|
It had nothing to do with sounding better than CDs. It just seemed like a logical multiple of having 24 frames per second in film.
Quote:
|
88.2/96 are just that doubled again, to insure that up to 22050/24000Hz, playback will be accurate down to the most minute detail
|
|
Again, it has nothing to do with detail.
Quote:
|
An anecdote for SHARC processors and most common DSPs: they work in 64bit (double-precision floating point).
|
|
The fastest SHARC chips do 32 bit float.
Quote:
|
That is the #1 excuse of DSP card manufacturers not to endorse CUDA (nVidia's programing standard), since graphics cards use only 32bit math (80-90% performance hit in double precision). The actual reason is of course that TC, UAD and the likes just enjoy it when you have to buy their hardware in order to run their unpiratable plugins (considering that a high end graphics card, even with an 80-90% performance hit, is still stronger than another quad-core CPU in these types of processes).
|
|
As you clearly argue yourself, competition is the only reason. The fastest GPUs are close to a 100 times more powerful than the fastest SHARC chips.
UnderTow |
|
|
|
UnderTow
Started Topics :
9
Posts :
1448
Posted : Sep 4, 2008 00:53
|
Quote:
|
On 2008-09-02 11:59, subconsciousmind wrote:
But it is making sense to work in higher sampling frequencies in terms of the possibility of more accurate DSP Algorythms. Especially EQ-algorythms can be programmed more accurate for higher samplingfreqs, closer to analog.
|
|
Actually well designed EQs don't need oversampling. Dynamics (and other non-linear processes) do benefit from oversampling.
Quote:
|
As for the AD converters, some people praise higher samping freqs not due to the frequency range, but due to the fact that natural drawbacks from AD can be compensated by recording in higher sampling freqs.
|
|
The main drawbacks are the filters but even those can be done well at the base rates.
UnderTow |
|
|
|
UnderTow
Started Topics :
9
Posts :
1448
Posted : Sep 4, 2008 00:55
|
Quote:
|
On 2008-09-02 14:53, Kane wrote:
192khz is still higher quality..this is why Blu-Ray DVD audio is at 192khz.
|
|
It isn't higher quality. It is just better marketing.
UnderTow |
|
|
|
UnderTow
Started Topics :
9
Posts :
1448
Posted : Sep 4, 2008 00:59
|
Quote:
|
On 2008-09-02 17:09, Kane wrote:
I was just saying that audio at 192khz is technically higher quality than audio at 82.2khz,
|
|
No it isn't. It just has more bandwidth.
Quote:
|
just as video at 100 frames per second is technically higher quality than video at 80 frames per second, even though the human eye can only perceive about 60 frames per second.
|
|
It is always a bad idea to compare audio and video this way. The analogies don't hold. One reason they don't hold is that at 25 frames per second, there is visible motion aliasing. (Like the wheels of a car that turn backwards and forwards as it accelerates away).
The main difference is that with audio, we already have the full frequency and dynamic range of human hearing covered. With video we don't have the full human range covered.
UnderTow
|
|
|
|
UnderTow
Started Topics :
9
Posts :
1448
Posted : Sep 4, 2008 01:01
|
Quote:
|
On 2008-09-02 17:10, forumuser wrote:
Audiophiles could maybe not spot a SINGLE song either in 96khz or 44khz. But everybody can spot a complete production with eq, compression (digital) mixed in 96khz, or in 44khz.
|
|
What research are you basing this on?
Quote:
|
If its rendered into 96 or 44 in the end doesn't really matter anymore. 96khz or 192 really comes into play during processing and when many things come together.
|
|
It depends on the quality of the processing.
Yes some processing benefits from oversampling but you don't need to oversample the whole system. Localised oversampling makes more sense until we have much more powerful computers. (Well that isn't very far off really).
UnderTow |
|
|
|
UnderTow
Started Topics :
9
Posts :
1448
Posted : Sep 4, 2008 01:05
|
Quote:
|
On 2008-09-02 22:55, Kaz wrote:
I'm saying bits don't exist in an analog signal - the dynamic range of an analog signal is limited only by amplification,
|
|
Rubbish. You are still limited by the noise floor of the equipment (and the recording). If you boost the signal, you also increase the noise floor.
Quote:
|
and analog signals by definition already have an infinite sampling rate (continuous line instead of sampled dots).
|
|
By the time a digital signal is converted back to analogue, it is a continuous signal again.
Quote:
|
You lower the volume by X percent, you will lose no detail, the wave just has X percent less amplitude.
|
|
Please show me this analogue equipment that defies the laws of physics. I want to order it.
UnderTow |
|
|
|
UnderTow
Started Topics :
9
Posts :
1448
Posted : Sep 4, 2008 01:12
|
Quote:
|
On 2008-09-02 18:41, Kaz wrote:
192KHz, no matter how much you process, will not, in any way, whatsoever, give any advantages you can work with.
|
|
This is not true. Non-linear processes benefit from increased sample rates. The more oversampling, the less aliasing.
Quote:
|
an 80KHz sampling rate won't be making any difference in listening unless you apply digital processing to the waveform (as in digital EQs, karaoke filters, things like that) - if the output has been properly dithered.
|
|
Dithering has nothing to do with sampling rates. Dithering has to do with bit depth and changes thereof.
Quote:
|
A single repetition of a sinewave has two parts, one positive, and one negative, that's why a 20KHz signal requires at least two sample - one for the positive and one for the negative, otherwise, the system will not be able to differentiate it with a 10KHz sine wave. Hence, 44.1KHz sampling rates as a standard.
|
|
This is a bad explanation. You could have a situation where all sample points are on the positive (or negative) cycle of the wave. (Of course only for a certain time. At some point, it will slowly shift to the negative (or positive) cycles of the wave. This is when dealing with signals very close to 1/2 Nyquist.
A much better analogy is that you only need two points to fully define a circle and a sine wave is just a circle plotted over time.
UnderTow |
|
|
|
UnderTow
Started Topics :
9
Posts :
1448
Posted : Sep 4, 2008 01:19
|
Quote:
|
On 2008-09-03 01:38, Spindrift wrote:
They can run 32 bit fixed as well as 32 and 40 bit float:
|
|
Yes some do. ProTools also uses fixed point formats (Motorolla chips) except for the RTAS plugins they run natively on the PC.
UnderTow |
|
|
|
