IPB

Welcome Guest ( Log In | Register )

 
Reply to this topicStart new topic
What is the effective bit rate of vinyl?, LP recording
klockworks
post Jul 14 2005, 07:27
Post #1





Group: Members
Posts: 3
Joined: 10-December 02
From: Golden, BC, Cana
Member No.: 4046



I would like to record my vinyl and would like to know what the effective bit rate is of a vinyl (LP) recording? Anyone? Or is this a silly question?


--------------------
Fix it . . . Just Fix IT!
Go to the top of the page
+Quote Post
kotrtim
post Jul 14 2005, 07:34
Post #2





Group: Members
Posts: 657
Joined: 4-December 02
Member No.: 3989



lossless .....900kbps?
actually what's your question, can't understand

This post has been edited by kotrtim: Jul 14 2005, 07:35
Go to the top of the page
+Quote Post
DreamTactix291
post Jul 14 2005, 07:40
Post #3





Group: Members (Donating)
Posts: 552
Joined: 9-June 04
From: A place long since forgotten...
Member No.: 14572



Vinyl being analogue would not have a bitrate like a digital source and cannot be compared in this manner.


--------------------
Nero AAC 1.5.1.0: -q0.45
Go to the top of the page
+Quote Post
dedalus
post Jul 14 2005, 08:00
Post #4





Group: Members
Posts: 22
Joined: 7-April 04
Member No.: 13304



I am sure he was just curious as to which bitrates are reasonable and which are overkill. Personally I would start with 16/44.1 waveform and compress with FLAC, if only to keep it consistent with your CD rips. You could then transcode into LAME, etc. if you like. 24/192 etc. are probably overdoing it. I am sure your tonearm, cartridge, phono stage, and ADC, are the real limiting factors here, so long as you select a reasonably wide bitrate.


--------------------
Medium Pimpin...
Go to the top of the page
+Quote Post
Axon
post Jul 14 2005, 08:01
Post #5





Group: Members (Donating)
Posts: 1984
Joined: 4-January 04
From: Austin, TX
Member No.: 10933



Actually, they are comparable via Shannon's Theorem, but not in a way that is particularly meaningful for determining what bitdepth and sampling rate to record at.

I answered the question on head-fi, and then the thread turned into mindless babbling by idiots who had no idea what they were talking about. Idiots.
Go to the top of the page
+Quote Post
cliveb
post Jul 14 2005, 09:13
Post #6


WaveRepair developer


Group: Developer
Posts: 838
Joined: 28-July 04
Member No.: 15845



QUOTE (klockworks @ Jul 14 2005, 07:27 AM)
I would like to record my vinyl and would like to know what the effective bit rate is of a vinyl (LP) recording? Anyone? Or is this a silly question?
*

Not a silly question at all. To determine the effective bit rate necessary to fully capture the information on a vinyl LP, you need to know the bandwidth and the dynamic range. A typical vinyl LP has a bandwidth of about 18kHz (when it's brand new, it might get up to about 22kHz). There is certainly some audio stuff above that, but it generally bears no relation to meaningful programme material (ie. it's noise and distortion). So let's be generous and assume a bandwidth of 22kHz: you'll need to sample this at 44kHz. The dynamic range of a beautifully pressed LP on virgin vinyl can get to about 65-70dB on a good day with a following wind, which equates to slightly less than 12 bits. So the bit rate required is 44,000 x 12 x 2 (for stereo), giving about 1030kbs. A more typical LP (18kHz bandwidth, dynamic range of 55dB) needs a bit rate of about 650kbs. For comparison purposes, the CD bit rate is 1378kbs.

As to what format to record vinyl in, there is little point in using anything higher than 44.1/16: exceeding this does nothing other than consume more disk space. If you are intending to do huge amounts of post-processing, you *might* consider recording at 24bit rather than 16bit to avoid any possibility of rounding errors during processing. But to be frank, even 16bit gives you vast amounts of space to do plenty of processing and still keep the rounding errors way below the vinyl noise floor.
Go to the top of the page
+Quote Post
klockworks
post Jul 14 2005, 14:20
Post #7





Group: Members
Posts: 3
Joined: 10-December 02
From: Golden, BC, Cana
Member No.: 4046



Thanks all ... Clive and Axon you hit it. Good to know.


--------------------
Fix it . . . Just Fix IT!
Go to the top of the page
+Quote Post
CosmoKramer
post Jul 14 2005, 15:43
Post #8





Group: Members
Posts: 133
Joined: 3-January 03
Member No.: 4399



QUOTE (cliveb @ Jul 14 2005, 09:13 AM)
For comparison purposes, the CD bit rate is 1378kbs.
*


Hmm..

44100*16*2 = 1411200 = 1411.2 kbps.


We are talking about bits, not bytes. 1 kb = 1000 bits.
Go to the top of the page
+Quote Post
cliveb
post Jul 14 2005, 15:57
Post #9


WaveRepair developer


Group: Developer
Posts: 838
Joined: 28-July 04
Member No.: 15845



QUOTE (CosmoKramer @ Jul 14 2005, 03:43 PM)
QUOTE (cliveb @ Jul 14 2005, 09:13 AM)
For comparison purposes, the CD bit rate is 1378kbs.
*

44100*16*2 = 1411200 = 1411.2 kbps.

We are talking about bits, not bytes. 1 kb = 1000 bits.
*


Really? I stand corrected. I always thought the 'k', 'M' and 'G' factors represented 2^10, 2^20 and 2^30 in all circumstances, regardless of whether it's bits or bytes (except for the hard disk manufacturers who deviously use powers of 10, of course).
Go to the top of the page
+Quote Post
Frank Klemm
post Jul 14 2005, 16:43
Post #10


MPC Developer


Group: Developer
Posts: 543
Joined: 15-December 01
From: Germany
Member No.: 659



QUOTE (cliveb @ Jul 14 2005, 04:57 PM)
QUOTE (CosmoKramer @ Jul 14 2005, 03:43 PM)
QUOTE (cliveb @ Jul 14 2005, 09:13 AM)
For comparison purposes, the CD bit rate is 1378kbs.
*

44100*16*2 = 1411200 = 1411.2 kbps.

We are talking about bits, not bytes. 1 kb = 1000 bits.
*


Really? I stand corrected. I always thought the 'k', 'M' and 'G' factors represented 2^10, 2^20 and 2^30 in all circumstances, regardless of whether it's bits or bytes (except for the hard disk manufacturers who deviously use powers of 10, of course).
*



The 2^x idea is one of the most silly ideas.
It violates the idea of SI (all units use the same prefix, multiplication/division by
10^n is easy), generates a situation were noone knows exactly what k, M, G means.

It was a crazy idea and I actually use 1 KByte = 1000 Byte.
1 GBit/s = 1000 MBit/s = 1000 KBit/ms = 1000 Bit/Ás = 1 Bit/ns.

There is no real reason to call a hard disk with 121,4 * 10^9 Byte a 115,3 Gbyte
or 115776 MByte HD.

There is only a little bit benefits when you are talking about RAM and ROM chips.
But "128 MByte RAM-Module" is like "19 inch CRT" or "5 1/4 inch Slot" an idiom,
not a unit of measurement.


--------------------
-- Frank Klemm
Go to the top of the page
+Quote Post
CosmoKramer
post Jul 14 2005, 17:03
Post #11





Group: Members
Posts: 133
Joined: 3-January 03
Member No.: 4399



Ok, according to wiki it's still somewhat ambiguous
http://en.wikipedia.org/wiki/Kilobit

...but mostly when you talk about bits you use kilo = 1000. For instance, your 192 kbps mp3 is NOT 192*1024 bps. Your 512 kbps adsl is NOT 512*1024 bps.
Go to the top of the page
+Quote Post
Axon
post Jul 14 2005, 18:47
Post #12





Group: Members (Donating)
Posts: 1984
Joined: 4-January 04
From: Austin, TX
Member No.: 10933



The measured response of vinyl has in some cases measured at 38Khz. Some quadrophonic encodings used modulation with an ultrasonic carrier to store the surround information, and those records were certainly not unplayable after the 2nd spin. There also might be a use for ultrasonic recording to give pattern matching of pops/scratches but I'm not sure if those actually extend into ultrasonic.

That said, vinyl does degrade considerably in the high end over time, although the effect is lessened with proper cartridge/tonearm care. Also the SNR varies considerably over the frequency spectrum. And of course there's the high-end rolloff on the inner grooves due to lower linear velocity, which also reduces bandwidth. So all things considered, a proper application of Shannon's Theorem would in fact be quite complicated.
Go to the top of the page
+Quote Post
Matyas
post Jul 14 2005, 18:48
Post #13





Group: Members
Posts: 158
Joined: 8-August 03
From: Bratislava
Member No.: 8242



QUOTE (cliveb @ Jul 14 2005, 10:13 AM)
... The dynamic range of a beautifully pressed LP on virgin vinyl can get to about 65-70dB on a good day with a following wind, which equates to slightly less than 12 bits.
...


Can anyone give a clue how did you calculate those 12 bits? And where did you get those numbers for the dynamic range?
Not that I wanted to argue, just understand.

Matyas

This post has been edited by Matyas: Jul 14 2005, 18:48
Go to the top of the page
+Quote Post
Woodinville
post Jul 14 2005, 18:59
Post #14





Group: Members
Posts: 1402
Joined: 9-January 05
From: JJ's office.
Member No.: 18957



QUOTE (DreamTactix291 @ Jul 13 2005, 10:40 PM)
Vinyl being analogue would not have a bitrate like a digital source and cannot be compared in this manner.
*


Um, didn't Mr. Shannon have something to say about that?

I did the math a while ago, and I don't remember the results, but if we assume 70dB SNR in the first of two channels (more on why that later), at least up to 1kHz, and then roll off to about 30dB at 20kHz (this is a very optimistic view of vinyl, but let's live with it), you can linearly interpolate this and then figure out what log2 of the SNR is.

For the second channel, we have to recall that the separation in vinyl rarely gets above 35dB, and that it's more like 15dB at high frequencies.

Now, vinyl can go to about 25kHz, but rarely has information on it above 16-20kHz due to the limitations of mastering equipment, etc. (Yes, I know about quad, but that's a somewhat redundant modulation scheme, it limits the baseband bandwidth, and it causes rather a lot of wear problems.)


--------------------
-----
J. D. (jj) Johnston
Go to the top of the page
+Quote Post
Woodinville
post Jul 14 2005, 19:03
Post #15





Group: Members
Posts: 1402
Joined: 9-January 05
From: JJ's office.
Member No.: 18957



QUOTE (Axon @ Jul 14 2005, 09:47 AM)
There also might be a use for ultrasonic recording to give pattern matching of pops/scratches but I'm not sure if those actually extend into ultrasonic.
*


This is fairly useful, if you can afford to sample at a high rate, detect the impulse noise errors (they tend to have a much wider bandwidth than real signal), and then downsample after you do your nonlinear processing.

But there is no point in long-term storage like that, unless you think you might have a better de-clicking algorithm some time later.


--------------------
-----
J. D. (jj) Johnston
Go to the top of the page
+Quote Post
Woodinville
post Jul 14 2005, 19:04
Post #16





Group: Members
Posts: 1402
Joined: 9-January 05
From: JJ's office.
Member No.: 18957



QUOTE (Matyas @ Jul 14 2005, 09:48 AM)
QUOTE (cliveb @ Jul 14 2005, 10:13 AM)
... The dynamic range of a beautifully pressed LP on virgin vinyl can get to about 65-70dB on a good day with a following wind, which equates to slightly less than 12 bits.
...


Can anyone give a clue how did you calculate those 12 bits? And where did you get those numbers for the dynamic range?
Not that I wanted to argue, just understand.

Matyas
*



1 bit = 6.02dB (plus a tiny bit) of SNR. SNR(in dB)/6.02=bits.

This post has been edited by Woodinville: Jul 14 2005, 19:05


--------------------
-----
J. D. (jj) Johnston
Go to the top of the page
+Quote Post
saratoga
post Jul 14 2005, 19:06
Post #17





Group: Members
Posts: 4901
Joined: 2-September 02
Member No.: 3264



QUOTE (Matyas @ Jul 14 2005, 09:48 AM)
QUOTE (cliveb @ Jul 14 2005, 10:13 AM)
... The dynamic range of a beautifully pressed LP on virgin vinyl can get to about 65-70dB on a good day with a following wind, which equates to slightly less than 12 bits.
...


Can anyone give a clue how did you calculate those 12 bits? And where did you get those numbers for the dynamic range?
Not that I wanted to argue, just understand.

Matyas
*



Its roughly 6 (6.2?) dB to a bit. The actual calculation is based on the average error from each level chosen during quantinization. On average you'll be off by half a level. You can use this to figure out the ratio of signal to noise. Google for the derivation. Its not intitive, but pretty simple once you see a diagram of how the process works.
Go to the top of the page
+Quote Post

Reply to this topicStart new topic
1 User(s) are reading this topic (1 Guests and 0 Anonymous Users)
0 Members:

 



RSS Lo-Fi Version Time is now: 21st August 2014 - 00:39