HydrogenAudio

CD-R and Audio Hardware => Vinyl => Topic started by: tinpanalley on 2011-06-29 23:08:19

Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-06-29 23:08:19
I'm having trouble figuring out which of the two is better for me to create a vinyl recording at the highest volume without clipping.

I capture a record via my preamp into Audacity, and I get what seem to be relatively low levels even though I make sure the levels are right when I record (Peak: -6.4db, RMS -24.6db). Am I doing something wrong? Or are those levels right? These recordings don't seem as loud as mastered CDs.

Anyway, to get the most accurate sound possible, should I use normalize or just increase the volume? If normalizing makes everything sound the same at a peak level, doesn't that completely screw with the dynamic range of the music?

Thanks!
Title: Normalize Vs Volume increase
Post by: DVDdoug on 2011-06-30 00:43:31
Audacity's Amplify effect has an option of setting the peaks to 0dB, which is the same as normalizing.*   

IIRC, Audacity's Normalize, adjusts both channels separately which can upset the left-right balance.  On the Audacity forum, they seem to recommend Amplify .

Normalizing or adjusting the volume doesn't affect the dynamics,  unless you push it into clipping.  With clipping, you are getting a nasty kind of dynamic compression where the average volume is increased without increasing the peaks (plus you're getting distortion).

If you've got some loud "clicks" & "pops", these defects may represent the loudest peaks.  So, you'll want to normalize after de-clicking.

Quote
(Peak: -6.4db, RMS -24.6db). Am I doing something wrong? Or are those levels right?
That's about right (before normalizing).  You can go a little higher, but if the peaks "try" to go over 0dB, the analog-to-digital converter will clip (distort). 


The ratio** of RMS to peak is determined by the dynamics of the music & recording.    I usually check the peak level after recording, and if it's at 0dB, I assume it's clipped, and I re-record at a lower level.    (I use a different program, but I believe Audacity shows clipping as red in the waveform.)

These recordings don't seem as loud as mastered CDs.  Most modern popular CDs are highly (dynamically) compressed to make them constantly loud (see Loudness War (http://en.wikipedia.org/wiki/Loudness_war)[/color]).  Compression is a way of increasing the average/RMS level without increasing or clipping the peaks.  This is why some people prefer older vinyl over the remastered & over-compressed CD version.

Audacity has a compressor, so you can get a "louder" or "more modern" sound if you wish.  Or if you want it listen louder, you can crank-up the playback-volume a bit and enjoy the dynamics! 




* You can actually "normalize" to other peak values, such as -1dB, but setting the peak(s) to the "digital maximum" of 0dB is the most common.)


** Decibels are logarithms.  So the difference in dB is a ratio,  and it stays constant when you amplify (multiply)  or attenuate (divide).  i.e. If you boost the peaks from -6dB to 0dB, the RMS level will also be boosted by 6dB from -24 to -18dB.
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-06-30 01:06:16
Thanks so much for the thorough reply. So, 2 questions:
1. What do you recommend I do with my settings when I capture vinyl, for someone like me who is all about the purity of the sound as it is reproduced by the stylus and cartridge?
2. I have Sound Forge and have decent knowledge of sound tech. Should I be using that instead of Audacity or is there some other program which would be even better? I'd rather learn a more complex program and capture my vinyl correctly than use a novice program that isn't going to reproduce sound as faithfully.

THANK YOU!!
Title: Normalize Vs Volume increase
Post by: DVDdoug on 2011-06-30 01:52:21
Quote
1. What do you recommend I do with my settings when I capture vinyl, for someone like me who is all about the purity of the sound as it is reproduced by the stylus and cartridge?
CDs  are 44.1kHz, 16-bit.  These settings are more than adequate for vinyl.  There's no harm in using 48kHz (or higher), and if you've got a 24-bit soundcard there's no harm in recording at 24-bits. 

But, don't buy-into the "audiophile nonsense" about analog vinyl having "infinite resolution".  That's like saying a tape measure has infinite resolution, but a digital caliper is limited.  The noise on a record (and in the preamp) limits the resolution...      You can (usually) hear noise between tracks on a record, and you can tell if the record is spinning or stopped, or if the stylus is in the groove or not (especially with headphones).  You can't (usually) tell if a CD is paused/stopped, because it's very quiet, and has greater dynamic range & resolution than a record.   


Quote
2. I have Sound Forge and have decent knowledge of sound tech. Should I be using that instead of Audacity or is there some other program which would be even better? I'd rather learn a more complex program and capture my vinyl correctly than use a novice program that isn't going to reproduce sound as faithfully.
The recording software doesn't affect sound quality.  It just sets-up the soundcard & driver for the proper format and routes the digital data to your hard drive.

Sound Forge may have better tools for noise reduction or other enhancements in "post production" (after you have a digital audio file on your hard drive).

The hardware is a different story.  Some cheap soundcards/soundchips can generate audible noise.  (Distortion and frequency response are usually not issues.)  If you suspect that your soundcard is contributing noise,* you might look for a better "audio interface".    You can get interfaces with line-level inputs, and a few have built-in phono preamps, which you don't need.  (But, watch out for USB soundcards...  Most "regular" USB soundcards only have a microphone input and line outputs, and are not suitable for high quality recording.)

* I'm talking about recording noise.  If you are using your computer for playback, that could be an issue too, but you really want to avoid adding noise to the digitized recording.
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-06-30 02:05:18
CDs  are 44.1kHz, 16-bit.  These settings are more than adequate for vinyl.  There's no harm in using 48kHz (or higher), and if you've got a 24-bit soundcard there's no harm in recording at 24-bits.


So there's no need (getting the dbs I showed you earlier) to do any normalizing and volume increasing if what I want is the most faithful capture of sound from the record?...within reason
Title: Normalize Vs Volume increase
Post by: AndyH-ha on 2011-06-30 02:55:10
There is no "need" to amplify or normalize, but there is no downside either. Neither the S/N ratio nor the distortion will change in any noticeable way. The quantization distortion of one, or a few, digital transforms will not be even close to audible, but if you work in floating point format and convert to 16 bit only as the very last step, you will minimize even that.

If any part of your playback system has significant noise, it is more likely to be before the analogue volume control. In that case, amplifying your recording digitally will supply a higher S/N ratio at the speakers.
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-06-30 03:09:28
But doesn't normalizing in making all loudness equal, in essence, affect the intended range of volume from track to track? Because it makes the entire album "loud" by the same amount even if one song is intentionally meant to be quieter?
Title: Normalize Vs Volume increase
Post by: tropicalfish on 2011-06-30 05:15:07
But doesn't normalizing in making all loudness equal, in essence, affect the intended range of volume from track to track? Because it makes the entire album "loud" by the same amount even if one song is intentionally meant to be quieter?

If you amplify your entire recording (either all in one track, or separate tracks) by the same amount, then you will still get the intentional volume differences between each of the different songs.
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-06-30 08:31:07
If you amplify your entire recording (either all in one track, or separate tracks) by the same amount, then you will still get the intentional volume differences between each of the different songs.


Perhaps this is a naive question, but why then after normalizing, does the waveform of the entire album side have consistently similar peaks and valleys through the whole side such that all the songs look like they are the same loudness and why, on the un-normalized version, can you clearly see which song has more loudness and which is softer?
Title: Normalize Vs Volume increase
Post by: db1989 on 2011-06-30 12:31:04
tinpanalley, your observation matches tracks that have been independently normalised by separate factors, rather than tropicalfish’s scenario of the album being normalised as a single entity or its tracks all being amplified by the same degree.

Audacity's Amplify effect has an option of setting the peaks to 0dB, which is the same as normalizing.*   

IIRC, Audacity's Normalize, adjusts both channels separately which can upset the left-right balance.  On the Audacity forum, they seem to recommend Amplify .

It still does that? Wow, that is dumb.

So if you wanted to remove DC offset, you’d have to use the Normalize dialog but with normalisation off and only DC offset correction on, and then go into Amplify and adjust it to a peak of 0 dB? That’s some great workflow right thar, even ignoring the potentially damaging effects and simply illogical nature of per-channel normalisation.
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-06-30 17:51:09
Well here's what I have... I think this will help you guys advise me.

I start with my turntable and go into a phono preamp. This one, in fact, the USBPhonoPlus: http://www.artproaudio.com/pro.....amp;id=128 (http://www.artproaudio.com/pro.....amp;id=128)

Then I go, via USB, into my laptop where I have Sound Forge 10 and Audacity, but Audacity seemed to be easiest to get my head around (even though I do sound editing with Sound Forge..go figure). Then I tend to use Brian Davies' Click Repair to try to clean things up.

That's the extent of my setup. Am I missing important gear? All I want to do is faithfully capture my records.
Title: Normalize Vs Volume increase
Post by: tropicalfish on 2011-06-30 19:25:37
Record with Audacity.
Go to Normalize, and check ONLY remove DC offset (or DC offset correction). Uncheck normalize track to ____dB

Then after that, Amplify. It'll automatically find the right amount to amplify to get your highest peak to 0dB.

If you use normalize, then all the tracks (if you record each track separately) will lose the intentional volume differences and they will appear/sound as if they're the same loudness.
Title: Normalize Vs Volume increase
Post by: db1989 on 2011-06-30 19:43:25
Then after that, Amplify. It'll automatically find the right amount to amplify to get your highest peak to 0dB.

If you use normalize, then all the tracks (if you record each track separately) will lose the intentional volume differences and they will appear/sound as if they're the same loudness.
As will also occur as a result of your suggestion to amplify to 0 dB, the only difference being that this option is free of the other’s idiosyncrasy whereby it treats both channels separately (as I noted in the past and as is still the case according to DVDDoug).
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-06-30 20:31:21
Perhaps I should have also stated that the way I record is, I play and record one whole side, and then I play and record the other. I end up with 2 WAVs (SideA.wav and SideB.wav) and then when I take them into Sound Forge I get this (I shrunk it so it wouldn't be so big)
(http://img200.imageshack.us/img200/8183/samplewav.jpg) (http://imageshack.us/photo/my-images/200/samplewav.jpg/)

Do those dbs look fine then?

How do I know if my sound card can handle 32-bit or not? Audacity clearly shows it as an option, does that mean it can do it?
Title: Normalize Vs Volume increase
Post by: AndyH-ha on 2011-06-30 20:51:59
A recording made with proper equipment should not have any DC offset to fix. The DC offset is your evidence of a hardware problem in the recording chain, or a bad soundcard. I have seen a DC offset in a few LP recordings that was on the LP itself, part of what I got off the disk, not introduced by my recording process.

I have no idea if Audacity or Sound Forge contain tools that can easily tell you if there is anything to concern yourself with, but that isn't a strange or uncommon software function.

Removing the very low frequencies (a rumble filter) will eliminate small amounts of DC offset. Any recording from a TT has a major amount of very low frequency noise you will be better off without anyway.

Normalizing to 0dB, or any other particular value, will not effect the dynamic range of the recording. Doing each channel independently could alter total dynamics but will not change the dynamics within the channel, no matter what the factor (of course that doesn't apply to something really stupid like working in integer format and amplifying enough to drive many peaks into clipping).

Some people like to break a recording into individual tracks and work on each separately. I find it more convenient to record both sides of the LP into one file and keep it that way until I am done with all processing. Normalizing, at the proper time, is then a single operation against the entire album. There is no possibility of altering the album's dynamics.

My LP recording chain is of fixed gain, which is to say I cannot adjust amplitude in the recording step. I normalize almost all recordings, usually to -1.0dB. I suppose there aren't many DACs these days that will clip the analogue output for peaks at or very near 0dBfs, but there also isn't anything to be gained by pushing to 0dB.

Normalizing should be done after any and all filters are applied, especially if you intend to use 0dB. Even when the filter is removing something (e.g. a rumble filter), it may push some peaks past 0dB, which means clipping. Of course if you do the more reasonable thing and work in floating point format there will be no clipping and you have the opportunity to apply a negative amplification to bring the peaks back down below 0dB before finally converting to integer (i.e. to the 16 bit final product), but it seems silly to apply a positive amplification early on just to have to reverse it later.



32 bit, floating point, is a data format, not a hardware format. Any soundcard can record into floating point format, the recorded data is the same as from recording into integer format, but is more convenient to work with. Regardless of all the discussion about 16 bit being enough, the convenience, the lower noise level, and the much smaller quantization errors from any post recording processing all are markedly in favor of using floating point unless you have really puny computer resources.



Nothing looks unusual in your recording. If the other side is much the same, if looks like you can apply about 6dB of amplification, but using actual program measurements to choose the factor is the proper way.
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-06-30 20:55:53
Ok, how about if I just control that from the phono preamp? What should my music be levels be peaking at? I always thought it was 3/4 of the way up the meter which ends up being about -12db. Is that wrong?
Title: Normalize Vs Volume increase
Post by: mixminus1 on 2011-06-30 21:01:16
I'll second AndyH's advice on recording both sides into one file - and keeping them as one file - until you're done with all processing.  It makes things such as normalization much easier.  Once you've got everything cleaned up to your satisfaction, and the levels where you want them, then you can break it up into individual tracks.

I use Cool Edit 2000's cue function to create regions, and then do a batch export to FLAC files - works great.  It's been awhile since I've used Audacity's label function, but I think you can get it to do something similar, and I'm pretty sure Sound Forge can do that, too.

I'll also point out that in the screenshot you posted, you have the fairly common situation of the needle drop producing a pop or tick louder than any of the actual music.  As such, you want to make sure you edit those out *before* you normalize, or the level of the music will still be several dB quieter than it should be.
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-06-30 21:06:17
I basically already do what you're suggesting mixminus. I record each side and end up with 2 WAVs, 1 for each side. Then I trim off the start and end. Then I do click removal and that's where I am now with trying to understand normalizing vs increasing volume. Then I turn the final edit into several tracks and into flacs. Then I do my metadata, just the way I work.

But at what point am I supposed to be concerned with making a 24-bit recording? Or am I just not able to do that?
Title: Normalize Vs Volume increase
Post by: AndyH-ha on 2011-06-30 21:24:42
The gain is normally set to leave plenty of headroom when doing live recordings. Setting to reach about -20dB on expected peaks is common. This leaves that 20db to capture the unexpectedly higher peaks without clipping. If you knew for certain what the highest peak would be, you would not need to leave any headroom.

The easiest, most convenient, and most reasonable thing is to record into a floating point format. You will thus capture the maximum resolution of which the soundcard is capable, and with the lowest quantization error,  and not have to fiddle with or worry about it thereafter. Stay in floating point until everything is done. As a last step, or next to the last if you intend to pull the recording apart into separate tracks, convert to integer. Here, 16 bit (preferably with dithering and good noise shaping) is perfectly adequate.
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-06-30 21:30:25
Stay in floating point until everything is done.


So you're saying I shouldn't export to Sound Forge but rather do all my normalizing, etc in Audacity? Seems that when I export both sides to Sound Forge I end up with a 16bit file. Or are these different kinds of bits? Sorry for all my confusion. 

Format                          : Wave
File size                        : 216 MiB
Duration                        : 21mn 23s
Overall bit rate                : 1 411.2 Kbps

Audio
ID                              : 0
Format                          : PCM
Codec ID                        : 1
Codec ID/Hint                    : Microsoft
Duration                        : 21mn 23s
Bit rate                        : 1 411.2 Kbps
Channel(s)                      : 2 channels
Sampling rate                    : 44.1 KHz
Bit depth                        : 16 bits
Stream size                      : 216 MiB (100%)
Title: Normalize Vs Volume increase
Post by: DVDdoug on 2011-06-30 21:36:24
Quote
How do I know if my sound card can handle 32-bit or not? Audacity clearly shows it as an option, does that mean it can do it?
There are no 32-bit analog-to-digital converters.  And the "rumor" is, that most 24-bit soundcards are only accurate to about 20 bits.  Your driver may be converting to 32-bit as you record, but of course that does not increase the true resolution...  It just makes a bigger file.   

You'll have to check your device specs to see if it's 16 or 24-bits.  Most soundcards (and interfaces) are 16-bits, and I didn't see anything on the ART website "bragging" about 24 bits.

Audacity (as far as I know all audio editors) work internally "behind the scenes" at 32-bit floating-point (or 64-bit), no matter what format you load-in.    As I understand it, DSP (digital signal processing) is "easier" in floating point.    From the user point of view, floating point allows you to do things that might boosts the peaks over 0dB (like boost the bass, or mixing) and you don't have to worry about clipping, as long as you normalize before saving in an integer format.        Floating-point also allows you to reduce the volume (temporarily) without loosing any resolution.    (If you reduce the volume in integer format, the least significant bits are truncated or rounded-off and you loose resolution.)
Title: Normalize Vs Volume increase
Post by: mixminus1 on 2011-06-30 21:40:05
Audacity has the option to export to bit depths greater than 16 - you have to select "Other uncompressed files" from the drop-down in the Save dialog and then click Options...

From there you can select a 32-bit float WAV, which Sound Forge *should* be able to read.

"Worst-case" scenario, you may have to export as "only"  a 24-bit PCM WAV, but that would still give you more resolution than you need for any editing/processing you might do.

Of course, just recording in Sound Forge in the first place would simplify things quite a bit - what is it about Sound Forge's recording process that you don't like?
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-06-30 21:43:49
Of course, just recording in Sound Forge in the first place would simplify things quite a bit - what is it about Sound Forge's recording process that you don't like?


Oh, I LOVE Sound Forge, I use it for audio editing in filmmaking. I just saw so many people using Audacity for vinyl captures that I thought, "geez, maybe I'm an idiot sitting here trying to use Sound Forge!".
Title: Normalize Vs Volume increase
Post by: AndyH-ha on 2011-06-30 22:15:09
Audacity is FREEware.

If may also be adequate for some things but it is hardly best at much, perhaps nothing.

Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-06-30 22:16:45
Ok, I'm trying Sound Forge now. Aiming for -6db as a peak. The preamp's little light is telling me it's clipping, but Sound Forge says -7.0 has been the max.
Also, it won't let me do 32-bit float, only 24-bit.
Title: Normalize Vs Volume increase
Post by: AndyH-ha on 2011-06-30 22:25:17
It may be true that most editors work in floating point, regardless of what your source is (this is not true of the several I use). but if you don't do everything at one go, without saving anything until everything is finished, there will still be conversions back to integer every time you write to disk, with the resulting increased quantization error. That may be ok if you have no choice but, as I wrote earlier, unless your computer resources are puny, it is easier and more reasonable to just record in floating point and never leave it until you are done - with everything.



Sound Forge is really so primitive that it cannot record in floating point format? Can it convert to floating point after recording? The only downside of this (done one time, as the first step after recording) is the time it takes.

There are people who will point out that some LP processing software only works in 16 bit integer. This is true of a few. WaveRepair is one that is (otherwise) very good at certain things but if you are not using such a program, there is no reason to restrict yourself to its limitations.
Title: Normalize Vs Volume increase
Post by: mixminus1 on 2011-06-30 22:36:53
Ok, I'm trying Sound Forge now. Aiming for -6db as a peak. The preamp's little light is telling me it's clipping, but Sound Forge says -7.0 has been the max.

By "preamp", do you mean the USB phono pre?  If so, back off!

You definitely don't want to clip the preamp - if you can capture at 24-bit (which it sounds like you can in Sound Forge), the levels you've been using (-12 dB peaks) are just fine.

I would be rather surprised if Sound Forge didn't work at floating point internally, even if it can't save it to disk (although that is rather strange).

When you say it will only let you do 24-bit and not float, do you mean when you open up a new project, or when you try to save/export something to WAV?
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-06-30 22:39:57
No, just that when I choose 32-bit (IEEE float) it tells me
"An error occurred while opening an audio device. An unsupported media type was requested.
USB Preamp (USB Audio Codec) does not support 32-bit floating point input."
Title: Normalize Vs Volume increase
Post by: tropicalfish on 2011-06-30 22:51:23
Then after that, Amplify. It'll automatically find the right amount to amplify to get your highest peak to 0dB.

If you use normalize, then all the tracks (if you record each track separately) will lose the intentional volume differences and they will appear/sound as if they're the same loudness.
As will also occur as a result of your suggestion to amplify to 0 dB, the only difference being that this option is free of the other’s idiosyncrasy whereby it treats both channels separately (as I noted in the past and as is still the case according to DVDDoug).

??

If you have multiple tracks (and you select all of them) and amplify them all at the same time, it will raise all of the tracks by the same amount until the highest peak in any one of the tracks meets 0dB.

Not sure what you mean. Normalize raises all of the tracks by different amounts so that the highest peak in each individual track meets 0dB.

I'm using Audacity 1.3.13


(http://i144.photobucket.com/albums/r183/mrairplaneman777/audacitynormalizevsamplify.jpg)
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-06-30 23:05:44
You see, that's what I thought.
Title: Normalize Vs Volume increase
Post by: tropicalfish on 2011-06-30 23:22:35
There are no 32-bit devices, so why use it? You might as well stick with 16 bit.
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-06-30 23:52:47
There are no 32-bit devices, so why use it? You might as well stick with 16 bit.


I don't know, I didn't suggest that, it's being suggested to me.
Title: Normalize Vs Volume increase
Post by: tropicalfish on 2011-07-01 01:15:23
So you want to record in floating point, and then convert to 16-bit at the end?

I can do that in Audacity and Reaper (at 32-bit FP).
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-07-01 01:28:58
I don't really care how I record. I just wanna get a sense of what I asked before about getting the best sound possible.

1. What do you recommend I do with my settings when I capture vinyl, for someone like me who is all about the purity of the sound as it is reproduced by the stylus and cartridge?
Title: Normalize Vs Volume increase
Post by: DVDdoug on 2011-07-01 01:53:30
Quote
1. What do you recommend I do with my settings when I capture vinyl, for someone like me who is all about the purity of the sound as it is reproduced by the stylus and cartridge?
I think that's been answered, and I think you are doing fine.

Don't your digital recordings sound identical to the vinyl when played back on the same equipment at the same volume?
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-07-01 04:08:33
All the 24 and 32 bit stuff still has me really confused. It's fine, I just need to do the research and studying myself. It's frustrating for me because as with video I have this thing where I NEED to know what is happening technically throughout the process but I don't have enough background or experience with audio to get my head around it.

There's an analog signal being turned into something digital via the preamp which is also boosting the signal. That newly amplified signal is then being sent digitally (as 1s and 0s) via a USB cable into a computer which takes the input and copies those 1s and 0s onto a harddrive. The software represents that digitized analog signal as a waveform which I can save in various containers (.wav, .flac, etc) and with various filters, settings, and processes applied. I get all of that (IF that was all correct). It's the bits I don't get.

I don't get:
- if the signal looks like it's clipping on the light of the preamp, why is it only hitting -7 on the meter in Sound Forge? Is there another variable somewhere? The Windows input level or something? Why aren't they equal?
- How does the bit amount affect the recording?
- Is the bit depth something the RECORDING HAS or something the SIGNAL COMES WITH from the turntable?

Look, I've got this great turntable and the guys at needledoctor.com helped me pick the best stylus and cartridge for my setup. I don't wanna get so-so sound. I'm the guy that will only use FLAC on his media player (a Cowon). I use gapless. I like my classical having the appropriately low levels. I own Sinatra's Capitol albums on mono because I don't like the duophonic crap that was produced in the 50s when the labels wanted everyone to buy a stereo hifi. I can't be ok with "yeah, that sounds pretty good". I wanna have the best I can and understand why it's the best that I can do with my vinyl captures. Does that make sense? Am I being too picky?    ...geez, whiney dude, huh?
Title: Normalize Vs Volume increase
Post by: AndyH-ha on 2011-07-01 06:06:22
There exist a continuously varying analogue signal from out of the analogue part of the recording chain. It is sampled by the ADC (in your case, inside the USB device). The bit depth is the precision to which it is sampled.

2 to the power of the number of bits is the number of different voltage values that can be output from the DAC (i.e. 2^16=65536 steps between minimum negative voltage and maximum positive voltage -- half of these for greater that 0, half for less than zero, minus one or two for the sign).

The ADC must “choose,” from these 65536 possible values, the one that is closest to the “actual” value of the signal voltage. The more bits it can use (i.e. 24 bits), the smaller each step, so the closer the choice will be to the real analogue voltage (as measured with infinite precision, or at least more precision than the ADC can manage).

24 bits is the limit of the hardware. However, no soundcard can actually do 24 bits. Thermal noise fills the lowest order bits due to the nature of not living in a extra cold part of the universe, so voltages very (very) near zero are lost in noise.

16,777,216 steps for 24 bit covers exactly the same range as 65,536 steps for 16 bit, so the quantization error (difference between “actual” voltage and what the ADC can record) is that much smaller (2^24/2^16). If fact it isn’t possible to hear the difference but when you do post recording processing, the same kind of errors occur in each calculation, so they add up.

There are a variety of 32 bit floating point formats but they are basically a 24 bit mantissa and a 8 bit exponent. This lets the data scale as necessary. Recording is not different between 24 bit and floating point but calculations errors are smaller and it is almost impossible to clip (unless you forget about >0dB values when you convert to integer).

Any format, 16 bit integer, 24 bit integer, floating point (32 and 64 bit varieties) works fine for LP recordings, which have so much intrinsic noise that they obscure the differences. Still, why not do as well as you can?
Title: Normalize Vs Volume increase
Post by: db1989 on 2011-07-01 13:25:03
Then after that, Amplify. It'll automatically find the right amount to amplify to get your highest peak to 0dB.

If you use normalize, then all the tracks (if you record each track separately) will lose the intentional volume differences and they will appear/sound as if they're the same loudness.
As will also occur as a result of your suggestion to amplify to 0 dB, the only difference being that this option is free of the other’s idiosyncrasy whereby it treats both channels separately (as I noted in the past and as is still the case according to DVDDoug).

??

If you have multiple tracks (and you select all of them) and amplify them all at the same time, it will raise all of the tracks by the same amount until the highest peak in any one of the tracks meets 0dB.

Not sure what you mean. Normalize raises all of the tracks by different amounts so that the highest peak in each individual track meets 0dB.

[pic]
Sorry; I hadn’t considered you were referring to tracks that were vertically stacked (i.e. not horizontally).
Title: Normalize Vs Volume increase
Post by: 2Bdecided on 2011-07-01 18:19:32
I have seen a DC offset in a few LP recordings that was on the LP itself, part of what I got off the disk
Come on Andy - think that one through properly and have a good laugh at yourself!  Hint: how would a DC offset be stored in a record groove? How would a cartridge detect it?

Cheers,
David.
Title: Normalize Vs Volume increase
Post by: 2Bdecided on 2011-07-01 18:25:40
I own Sinatra's Capitol albums on mono because I don't like the duophonic crap that was produced in the 50s when the labels wanted everyone to buy a stereo hifi.
Aren't the official Capitol CD releases already the properly mono?

Quote
I can't be ok with "yeah, that sounds pretty good". I wanna have the best I can and understand why it's the best that I can do with my vinyl captures. Does that make sense? Am I being too picky?    ...geez, whiney dude, huh?
What matters is whether you can hear a difference. You already have everything you need in order to find out. Go on, do some listening


Peaking 12dB down while recording at 16-bit is just leaving the territory where people here will tell you "there isn't a cat in hell's chance of hearing a difference", and heading into the territory where "you probably can't hear a difference but just maybe sometimes theoretically you might".

Recording at 24-bit won't do any harm, and will almost inevitably cause a placebo induced sensation that the sound quality has improved, if you believe that it should.


Two different cartridges are going to sound 1000x more different than the difference caused by any of this stuff.

Cheers,
David.
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-07-01 19:51:13
Thanks for the wonderful feedback. Just wanted to get to this first...

Aren't the official Capitol CD releases already the properly mono?


No CD has ever been produced of the Capitol Mono masters... sadly. Only the odd track here and there. The only thing out is the horrible remaster from the late 80s which attempted to "fix" the Duophonic masters and ended up adding bass and excessive reverb to his voice and killing some of the background instruments in the process. A new remaster cleaned those up but they're still just the duophonic versions cleaned up. The UK Capitol CDs are better than any of the Us  ones though which is sad. A bit like Vera Lynn's US CDs sounding better than the UK ones. That effort should be made here.  Anyway... different thread, I suppose.
Title: Normalize Vs Volume increase
Post by: AndyH-ha on 2011-07-01 20:36:36
Quote
Come on Andy - think that one through properly and have a good laugh at yourself! wink.gif Hint: how would a DC offset be stored in a record groove? How would a cartridge detect it?

The answer is in: how is a DC offset stored in the recording on one’s hard drive?
Title: Normalize Vs Volume increase
Post by: DVDdoug on 2011-07-01 21:46:53
Quote
It's the bits I don't get.
OK,  I'll try to simplify it a 'bit'... 

Digital audio is a series of numbers (or "values".).    Each number represent the amplitude (height) of the wave at any a single point in time.   You essentially "connect the dots" to create the wave-shape.  If you zoom-in with Audacity, you will see the wave represented as a series of stair steps*.  But don't worry....  When you play the audio file, the steps are smoothed-out by a low-pass filter following the digital-to-analog converter.    Some audio editors will show a smooth-filtered waveform...  I'm not sure about Sound Forge.)


A "bit" is a binary "digit" (base-2).   

With 8 bits, the highest number we can represent (or can count to) is 1111 1111, which converts to 255 decimal. 

With 16 bits, we can count to 1111 1111 1111 1111 (65,535 in decimal).
(We usually write the bits in groups of 4 because it's easier to read and its easier for programmers to convert to hexadecimal = base 16.) 

The actual format and conversion to decimal varies because one bit can be used for a the +/- sign.  The values in a 16-bit WAV file range from -32,768 to +32,767.  And with floating-point, some bits are used for the mantissa and some for the exponent, as Andy explained.

A 16 bit file does not go "louder" than an 8-bit file, because all of the formats are scaled.**  The 16-bit file, has more, smaller, steps = More resolution.    Floating-point is scaled differently and you can have a "taller" wave.  But when you play-back a floating-point file, it still has to go through an integer digital-to-analog converter, so it won't go "louder" in the real world either.


*  Of course, digital audio is quantized (digitized) in both amplitude (vertical) and time (horizontal).    The bit depth represents the amplitude resolution, and the sample rate kHz, represents the time resolution.

**  In mathematics, this "scaling" to adjust of the formats to have the same volume/amplitude would be called "normalizing", but in audio processing normalize has a slightly different meaning.
Title: Normalize Vs Volume increase
Post by: DVDdoug on 2011-07-01 22:12:22
Quote
Come on Andy - think that one through properly and have a good laugh at yourself! wink.gif Hint: how would a DC offset be stored in a record groove? How would a cartridge detect it?

The answer is in: how is a DC offset stored in the recording on one's hard drive?
We're getting a bit off topic here, but DC = zero Hz which cannot be stored on a record, or reproduced by a phono cartridge which requires movement to generate a voltage.  However, you can easily store "zero Hz", or a constant offset value in a WAV file.

But, I believe you can have an average-offset, or asymmetry on a record where the positive half-cycles are larger than the negative half-cycles, or vice-versa. 


BTW - GoldWave's offset correction algorithm can actually introduce a true DC offset component if you feed-in an asymmetric waveform.    I assume it works by adding/subtracting a constant value to/from each sample, so that the new-average of all samples is zero.    This algorithm can create a "click" if there s silence at the beginning/end of a file.  Since I discovered this, I fix offset with a regular-old high-pass filter.  (I believe the fix-offset algorithm has been corrected in the new GoldWave beta release.)
Title: Normalize Vs Volume increase
Post by: 2Bdecided on 2011-07-04 12:34:00
We're getting a bit off topic here, but DC = zero Hz which cannot be stored on a record, or reproduced by a phono cartridge which requires movement to generate a voltage.
And even if it didn't, what would the DC offset be relative to, given that the record groove itself is moving inwards the centre of the record, and the arm+cartridge have got to follow it!

Cheers,
David.
Title: Normalize Vs Volume increase
Post by: AndyH-ha on 2011-07-04 20:32:26
My experience is based on CoolEdit Statistics reporting a few percent DC offset on a few LPs over the years. These were isolated cases. I normally record three or four albums at a time. There was no DC offset on most of 700 to 800 disks. It is unlikely that something changed electrically in my system for one of those three or four LP recorded on some odd day, then immediately went right again when playing the next album, preventing further DC offset in the next recording.

I know it is not the electrical condition of the LP. What you seem to be saying is that if I used recording equipment with an offset, so the recording I end up with on hard drive measures with a DC offset because of that recording chain problem, just cutting the recording to LP would eliminate that offset. If this is true, I don't understand the mechanism.
Title: Normalize Vs Volume increase
Post by: 2Bdecided on 2011-07-04 23:34:59
I think you may think I might be being clever than I am (?!) ...

If this is true, I don't understand the mechanism.
If you recorded something with a 40000% DC offset, it would move the entire record groove about 1 inch to the left.

How would you know this when you played it back?

Cheers,
David.
Title: Normalize Vs Volume increase
Post by: AndyH-ha on 2011-07-05 08:38:02
Perhaps I am just too ignorant of what is involved. The waveform is displaced by some amount from zero. This is normally caused by a DC voltage on the analogue signal line. It exist in the analogue signal. When converting to digital it is retained as a numerical component of each sample value. The file can be copied elsewhere, within the computer or to another, and the offset does not go away.

Since there is a way to detect and remove it in an editor, there has to be a way to add it where it did not originally exist. In a digital recording it is just another part of what makes up the sample, in a sense like 60Hz hum. It doesn't matter whether it was actually caused by DC where there should have been none, or by something else.

Also, I don't seen any reason there could not be a DC voltage on the signal line going into the cutting amplifier, adding an offset to the recording being fed to it. However, I take it by your extreme example that cutting the signal to disk would effectively remove the offset. I can see there is no reference available on the LP but if this means the signal becomes what it would have been had there been no offset  to begin with, then either
What I 've found on a few LP recordings is actually something else that just happens to look the same or
my system somehow, very infrequently, glitches just for the duration of one LP.

I never gave it any thought before. The few times it was there I just removed it. Since most of my recordings show no such problem, I did not consider the odd, very infrequent case a sign that my equipment was acting up, any more than I thought the clicks and pops were the fault of my system rather than the condition of the LP.
Title: Normalize Vs Volume increase
Post by: 2Bdecided on 2011-07-05 14:08:45
Vinyl is an intrinsically AC-coupled format. It has no concept of DC.

Any DC selectively reported by your software is because your software's idea of 0 DC did not match your equipment's - maybe due to different timescales / time constants for removing DC (e.g. most software just sums all values in the whole file), maybe because there was some assymetry in response after DC was removed (e.g. distortion) which means different signals will appear to have different amounts of DC.

Cheers,
David.
Title: Normalize Vs Volume increase
Post by: pdq on 2011-07-05 16:14:41
Think of it this way. If there is a DC offset in the signal when the master for the LP is created, all of the grooves will be displaced a few microns in the same direction. When you drop the needle into the first groove, you have just applied a DC correction by placing it in the center of the offset groove.
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-07-06 22:37:04
Hey all,

I've been absorbing everything you guys have helped me with recently. Just did a new recording at 32-bit (my set up won't allow IEEE float, Sound Forge just says my USB device can't do it). All my peaks are at about -6.0 now. I have to bring the file over from the laptop's Sound Forge to my tower to finish processing. How should I save this WAV knowing that I will be doing some light click removal?

- 32-bit IEEE Float?
- 32-bit PCM?
(by the way what's the difference between IEEE Float and PCM? A huge can of worms?)
- 24 bit?

Thanks guys! I won't save until I've heard some thoughts.
Title: Normalize Vs Volume increase
Post by: DVDdoug on 2011-07-07 01:34:41
Quote
- 32-bit IEEE Float?
- 32-bit PCM?
Floating point has an exponent, so it can hold very-small fractional numbers and very-large values.  PCM can actually be integer floating point. 

ANY of these formats are fine.   With floating point you can go above 0dB, but de-clicking should only reduce peaks, not increase them.  Use any format that your de-clipping software supports.

Don't worry so much!!!!   Unless you're doing something that might cause clipping, the differences are completely meaningless.  It's like the difference in a car that has 10,000 horsepower and 1 million horsepower...  It's all overkill and you're never going to notice the difference driving to work. 

Quote
(my set up won't allow IEEE float, Sound Forge just says my USB device can't do it).
I think your hardware is 16-bits...      There are NO 32-bit or floating-point ADCs or DACs.  When the software & drivers allow recording in 32-bit floating point or 32-bit integer, its just a convenience to convert the data on-the-fly during recording.  If you record to 24 bits with 16-bit hardware, that's just a format conversion too...  You can't get more than 16-bits of resolution from 16-bit hardware.

It's not as convenient, but you can convert to IEEE 32-bit floating point later as a separate step, if you want/need that format for some reason.
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-07-07 05:06:48
Don't worry so much!!!!


But...but... what if my computer blows up and my vinyl starts to melt and leak and causes death and mayhem and destruction and death again all because of the wrong settings!?!?!?!?!?!?!?   

No, I know. It's just that when you've been sitting in front of the computer for 30 minutes reading and rereading all about settings and bit-rates and so on, you start to worry more than necessary. It's just capturing audio, but ideally one doesn't want to have to do it more than once. It's awesome though that you guys have been able to give me pointers from the most practical all the way to the most minutely accurate and technical. So for that and for making me even more of an audiophile than before, I thank you all whole-heartedly.

I'll post an image of the waveform I got.

Thanks!
Title: Normalize Vs Volume increase
Post by: AndyH-ha on 2011-07-07 09:31:59
I'm curious about this inability to record in floating point.. That format is not a hardware format, it is just the data container. It should be irrelevant how many bits come in, that is just where you are dumping them. I don't understand any reason there should be a limitation with any soundcard.

Could you be more specific:
Exactly what is the message?
Is it identified as coming from the application or the OS?
What is the OS?
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-07-07 17:56:16
Exactly what is the message?


I had listed it on a previous page...
"An error occurred while opening an audio device. An unsupported media type was requested.
USB Preamp (USB Audio Codec) does not support 32-bit floating point input."

It appears when I choose New Recording options in Sound Forge which leads me to believe it is application driven. Regular 32-bit is fine.
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-07-07 18:37:33
This is probably like asking if there is one way to draw a line, but is there any series of processes that are typically run on all record captures as a base to "clean them up" or is it too dependent on too many factors to give a general list?
Title: Normalize Vs Volume increase
Post by: AndyH-ha on 2011-07-07 19:20:06
There are many descriptions of the process published. Opinions differ, of course.

1. Rumble filter
2. auto declicking
3. decrackling, if indicated
4. manual declicking follow-up
5. noise reduction
6. normalization
7. convert to 16 bit

Several of those steps can involve quite a bit of variation, mainly related to the condition of the LP  and to the facilities of the software employed.
Title: Normalize Vs Volume increase
Post by: DVDdoug on 2011-07-07 20:36:54
Quote
but is there any series of processes that are typically run on all record captures as a base to "clean them up"
This page[/u] (http://www.delback.co.uk/lp-cdr.htm#clean_pops) has lots of helpful information. 

Of course, the main thing is noise reduction. There are3 or 4 types/categories of noise reduction, and these require separate steps.

1.  I always try to remove "snap", "crackle", and "pop".    (I don't use Sound Forge and I don't know about it's tools.)

2. "Regular" noise reduction, where you feed-in a sample of noise-only (i.e. a "noise fingerprint") can reduce hiss, hum, and other more-or-less constant low-level background noise. 

3. A rumble filter will remove subsonic noise.  And, a 50/60Hz notch filter can remove line-frequency hum, and in some cases may work better than the regular noise reduction filter.

4. A noise gate will mute the sound completely when the sound level falls below a preset threshold (i.e. between songs).    But, it's usually best to do this manually by muting between tracks and/or by adding a short fade-in/fade-out before/after the music starts/stops. 

Now... with noise reduction you can get artifacts (side effects), so it takes some trial & error.    Sometimes when I've done a massive amount of processing, I'll keep an untouched archive copy with all of the original analog & vinyl defects. 

Some clicks might not be removed, some musical sounds might get misinterpreted as clicks and removed, and sometimes the "repair" is worse than the original click.

The same is true with regular noise reduction, especially with quiet-delicate music.  You can sometimes get some strange "digital" artifacts.

And, even a noise gate can be distracting if the background noise suddenly goes dead-silent.

You might not want to do this, since you want to preserve the original sound, but a lot of older recordings have rolled-off highs, so I sometimes boost the highs with some EQ.

And as the last step (after making a full album-length WAV file) I normalize.


----------------------
For de-clicking, I use Wave Repair[/u] (http://www.delback.co.uk/wavrep/) ($30 USD).  Wave Repair does an amazing on most "clicks", and in the manual mode it only "touches" the audio where you identify a defect.  On the downside, it usually takes me a full weekend to clean up a vinyl transfer.    The record I'm working on right now is in terrible shape, and it's taking me even longer!!!!  And, I believe it only works on 16-bit files.







Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-07-07 21:04:16
Here's a comparison of the 32-bit, higher gained capture I did of one side of the same album. Better, right?

(http://img220.imageshack.us/img220/8623/comparisoni.jpg) (http://imageshack.us/photo/my-images/220/comparisoni.jpg/)
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-07-07 22:16:10
This sounds great to me as I'm hearing it even without any further processing.

I've learned a lot in the past week or two. And that is entirely credited to all of you who have taken the time to help me "get it". A lot about what I can hear, what I want to hear, and as an audio and music purist what I think I should hear.

I think my aim here with all this, with my attempts to capture my records into a portable format or for computer listening to not do any further damage to some very valuable records I have is the following: I'm not trying to turn the audio from the vinyl into a magnetic tape or digital studio master. I'm trying to faithfully capture the actual sound from the record. The experience of "putting the record on" at home which is, at its core (if you'll indulge me the temporary philosophical observation) why I put on the vinyl rather than the CD version of the same album. I put this version of Coldplay's Parachutes on (that the waves are from) rather than the CD because I want that sound. Don't we all, even if we have our own opinions of that band's quality and wouldn't be caught dead listening to Coldplay, listen to records for 'that sound'? For that matter, it's also why I remain contented with a 'noisier' vinyl copy of an Oscar Peterson record rather than going and getting some badly or even some decently remastered CD. We own turntables because we, each of us in our own way, like the sound. Isn't that it?

Click removal for surface damage is one thing; I WILL do that because it's not "supposed" to be there. It's surface damage. And the same goes for click removal on my older vinyl. But to a certain degree, I want the vinyl sound. Hence the point of listening to vinyl in the first place. when I transfer my 78s, I'm not going to do a lot of click removal because to a higher degree, it will kill the mono, higher frequency sound coming from the shellac. Hum caused by electrical fields that my ear wouldn't hear from my record player through speakers is another thing I'm willing to remove. But rumble? I don't know... If you listen carefully after a side is finished, there tends to be a distinct lack of rumble in the room when the needle comes off the vinyl. So, perhaps some rumble is fine? And normalising... if I get good levels on the capture, I just don't see the point in messing with the sound. I don't think I understand the concept of normalisation yet.

Anyway, just some thoughts. I need a good Click removal (Click Repair seems to work best for me), a good program to remove some electrical noise and I think that might be it. Am I crazy?
Title: Normalize Vs Volume increase
Post by: greynol on 2011-07-07 22:34:32
No you're not crazy, just know that 32 bits for vinyl is complete overkill.  Even 24 bits is overkill as a delivery format.

With the exception of feedback, you are hard-pressed to provide objective evidence showing that a delivery format of 16-bits, properly dithered, cannot fully provide every audible nuance found on even the most pristine copy of vinyl.  If you like feedback, you should be able to capture whatever amount is present when listening while recording.
Title: Normalize Vs Volume increase
Post by: DVDdoug on 2011-07-07 23:38:30
Quote
And normalising... if I get good levels on the capture, I just don't see the point in messing with the sound. I don't think I understand the concept of normalisation yet.
It's simply a volume adjustment. It's no different than an analog volume adjustment.  It does not affect the "character" of the sound. 

It's generally not necessary, but there's no harm in it...    The main purpose is to get the volume more in-line with most digital recordings.  i.e. If you are playing your low-level digitized vinyl at a comfortable volume, and you switch to a more modern recording, you'll get "blasted", and you'll have to turn-down the volume.  There's more to perceived loudness than peak levels, so even after normalizing, your files will probably not be as loud as most (over compressed) modern recordings.  (And, there are other solutions such as ReplayGain, that adjust/match the volumes at playback time.)

Normalizing could become necessary if you're not able to get enough volume from your portable player.


...This might add to the confusion, and you're never going to hear the effects of this...  But, if you are rendering to a 16-bits from a higher-resolution format,  and you don't normalize first, you are not getting the "full advantage" (the full dynamic range) of all 16-bits. 
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-07-07 23:48:17
Couple of things:

It's simply a volume adjustment. It's no different than an analog volume adjustment.  It does not affect the "character" of the sound.

Then what is the difference to just doing a volume increase in Sound Forge?

Secondly...
if you are rendering to a 16-bits from a higher-resolution format, and you don't normalize first, you are not getting the "full advantage" (the full dynamic range) of all 16-bits.

...and...
you are hard-pressed to provide objective evidence showing that a delivery format of 16-bits, properly dithered, cannot fully provide every audible nuance found on even the most pristine copy of vinyl.

Aren't these two comments contradictory to each other?
Title: Normalize Vs Volume increase
Post by: greynol on 2011-07-08 00:03:22
No, they aren't contradictory at all.  DVDdoug's comment was (and should be) taken into account.

As has already been discussed to death, recording in 24-bit provides a greater margin between clipping and adequate SNR.  It also provides more than enough resolution to ensure that any post-processing (calculations are done at 24-bit or better), ultimately including normalization, will not degrade SNR.  Once this is all done, your finished product can be converted to 16-bits, and if done properly (which includes making use of at least 14-15 bits, but preferably all 16), will not produce any audible degradation.
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-07-08 00:08:26
Ok, my bad. I misunderstood. I have some people tell me I should never save less than 24-bit and other people who tell me it's a waste and that I should do EVERYthing at 16-bit because I'll never notice the difference anyway.
Title: Normalize Vs Volume increase
Post by: DVDdoug on 2011-07-08 00:39:18
Quote
Then what is the difference to just doing a volume increase in Sound Forge?
Right!!!    There is no difference.   In your 1st post, you said you had a peak of -6.4dB.  If you increase the volume by +6.4dB, your new peak level is 0dB and that's the same as normalizing to 0dB!  (Of course, every recording is going to be different before normalizing.) 

When you normalize the software automatically scans the file before calculating and making the appropriate adjustment.

Quote
Aren't these two comments contradictory to each other?
One of the  themes running through this discussion (and lots of discussion at HydrogenAudio) is that there are often true mathematical or measurable differences that are not audible differences. 

I know you want to make absolutely the best recording possible, even if you can't hear the difference.    But at some point,  these things aren't worth worrying about... Do you want to use a format that's 10 times better than your hearing, and 100 times better than vinyl?  Or 1000 times better than you can hear?
Title: Normalize Vs Volume increase
Post by: greynol on 2011-07-08 00:55:28
I have some people tell me I should never save less than 24-bit and other people who tell me it's a waste and that I should do EVERYthing at 16-bit because I'll never notice the difference anyway.

First, I try to shy away from the word should.  Second, I do not trust what people say unless they provide objective evidence.
http://www.hydrogenaudio.org/forums/index....showtopic=16295 (http://www.hydrogenaudio.org/forums/index.php?showtopic=16295)

I see no reason why one should limit himself to recording and processing at 16-bit unless there is some kind of constraint, at which case I would consider upgrading if it is feasible.  On the other hand, it's helpful to understand a little about resolution and the limitations of the physical media and hardware involved, without blindly believing that more always results in an audible improvement.
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-07-08 01:57:15
But at some point,  these things aren't worth worrying about... Do you want to use a format that's 10 times better than your hearing, and 100 times better than vinyl?


No, you're right. I absolutely don't care about that level of perfection which is why my 'final analysis' if you will was basically, look, I get that using more bits upfront minimizes processing decay and it doesn't cost anything. Fine, I'll do it. But processing more than what my naked ear can hear with click removal etc is silly to me especially since i want it to sound like a record anyway. So, I'm gonna keep it simple. There's a site I visit often where this guy makes vinyl transfers on a daily basis with all kinds of specs and I listen to the files and they sound like a CD without even the warmth or texture of vinyl and I think.... wait, what's the point? So he's processed the vinyl until it sounds like a CD. Ok. Why not just get the CD? But yes, you're right. There is a limit to how much mathematical perfection is worth it when essentially what you want is to enjoy the music.

...funny, my initial question was just about normalizing vs increasing volume. I think we can close this thread now as far as I'm concerned. Don't want anyone else feeling like it's been "discussed to death".
Title: Normalize Vs Volume increase
Post by: AndyH-ha on 2011-07-08 02:46:14
The Sonic Foundry Noise Reduction plugin 2.0 has been included as part of some versions of Sound Forge for a few years. I have no idea if it is totally integrated or still arrives as a DX plugin that can be used with like any other DX plugin.

Prior to that bundling, the plugin was only available as a separate item, more expensive than many editors. Without this plugin, the declicking and noise reduction facilities of Sound Forge are pretty primitive. If you have it, it can do a rather good job of cleaning up most LP transfers.

The older original vinyl releases are very frequently a different, and generally superior, mastering than later CD releases. That is one big reason for retaining the LPs, regardless of what you think about LP noise and distortion. Since few of the LPs I've worked on are much newer than 1980 (certainly fewer in number than those prior to 1960), I don't know much about albums released simultaneously on both formats.

Rumble filtering isn't very much about what you hear in most cases, it is about those very low frequencies, which are most definitely there in every recording from an LP, that eat up a significant amount of amplifier power, going nowhere. Sometimes there can be enough of this stuff to cause bass distortion. It is also claimed to sometimes be at a high enough level to case bass speaker damage, but it that isn't happening from playing the LP, it probably isn't part of the problem.
Title: Normalize Vs Volume increase
Post by: greynol on 2011-07-08 02:53:34
I simply wanted to let you know that I respect that you have a preference for the sound of vinyl and that 24-bits as a recording format (or interim storage format) is more than adequate and that 16-bits is adequate as a final delivery format.

Please don't take the "discussed to death" comment personally.  Hi-res is a point that is often argued with no difference in outcome.  I thought I'd just summarize the consensus.
Title: Normalize Vs Volume increase
Post by: AndyH-ha on 2011-07-08 03:04:07
I made an inquiry elsewhere. I got back that your experience of not being able to record in floating point, because of the hardware you are using, is indeed a peculiar aspect of Sound Forge. The program first queries the device and then sets format limitations based on it.

If you want floating point, you have to convert after recording. The results will be identical, there is just that extra step.
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-07-08 07:34:03
Ohhh!!! Well, why didn't someone just make it is as clear as what THAT guy said?!?!?
Title: Normalize Vs Volume increase
Post by: AndyH-ha on 2011-07-08 08:19:55
Because Sound Forge has such an unusual approach (and senseless, it seems to me) that nobody knew.
Title: Normalize Vs Volume increase
Post by: 2Bdecided on 2011-07-08 12:01:26
There are many descriptions of the process published. Opinions differ, of course.

1. Rumble filter
2. auto declicking
3. decrackling, if indicated
4. manual declicking follow-up
5. noise reduction
6. normalization
7. convert to 16 bit

Several of those steps can involve quite a bit of variation, mainly related to the condition of the LP  and to the facilities of the software employed.

The kind of person who believes there's some benefit to keeping vinyl transfers at 24-bit wouldn't run any automatic process on their recordings.

Many people who are closer to the sane "16-bits is more than enough for vinyl" end of the spectrum wouldn't run automatic processes on their recordings either.


None of the auto declicking or denoising algorithms are "transparent" (if you can even define that term for these processes) - they all have a "sound" to them when used aggressively - but some fall into the "do more good than harm" category - the one you suggested (NR-2) being the best one that I can afford. There's some newer stuff over $1k that may be better, and of course CEDAR ($5k+) - though even those products can be abused.

There is something to be said for listening to a raw transfer on decent quality equipment. It's a bit like listening to lossless audio. With lossy audio (and automatically restored audio), you never know if any faults you hear are down to the processing - and this thought can be distracting. Whereas with lossless audio, or unrestored audio, you know any faults are there on the original, so you might as well forget about them (or fix them manually, if possible).


I often listen to my records - either directly, or looped through the NR-2 declicker (and sometimes, denoiser) in real time. Then I can pick how "clean" I want it to sound when listening. Actually picking optimum settings for preserving on CD is more time consuming, brings out the perfectionist, and requires "a decision", which isn't something I'm that good at!

Cheers,
David.
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-07-08 15:45:18
Because Sound Forge has such an unusual approach (and senseless, it seems to me) that nobody knew.


Actually, Andy, whenI made that posting it was because late last night, some dude had popped in and left this enormous posting of computer generated spam gibberish and was obviously cleaned out by one of the mods. It was pretty ridiculous.

Now, I guess that the fact that Audacity LETS me do it just means it isn't checking the hardware but then it can't really do 32-bit IEEE floating anyway, it's just not programmed to know what I have. Correct? Is Audacity totally a lame program? Vinyl capturing sites seem to love it?!
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-07-08 15:49:25
There is something to be said for listening to a raw transfer on decent quality equipment. It's a bit like listening to lossless audio. With lossy audio (and automatically restored audio), you never know if any faults you hear are down to the processing - and this thought can be distracting. Whereas with lossless audio, or unrestored audio, you know any faults are there on the original, so you might as well forget about them (or fix them manually, if possible).


Exactly, David. Thank you for the thoughts.
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-07-08 16:08:53
This is the info from a download from the site (which will remain nameless for legal reasons) where the guy posts vinyl transfers he does and the specs on them. They sound great to me... but too clean. I don't even know what half these processes or gear are. Although I undeerstand a lot more (especially the DR stuff) now because of you guys.



RCM Hannl ‘limited’ with “Rotating Brush”
Music Hall MMF 9.1 Turntable
Tonearm: Pro-Ject 9cc evo with Pure Silver Wires
Cartridge: Nagaoka MP-500
Brocksieper Phonomax (Tube Phono PreAmp)
E-MU 0404 external USB 2.0 Audiointerface
Interconnections : Silent Wire NF5
WaveLab 6 recording software
iZotope RX Advanced 2.00 for resampling and ditheringVacuum cleaning > TT > Brocksieper Phonomax > E-MU 0404 > WaveLab 6 (24/192) > manual click removal >
analyze (no clipping, no DC Bias offset) > resampling and dithering with iZotope RX Advanced 2.00
> split into individual Tracks > FLAC encoded (Vers. 1.21)

Vinyl rip in 24-bit/192kHz (presented in 24/96 & 16/44.1) | FLAC | m3u, cue & Tech Log
Artwork | DR Analysis | 875 / 274 mb incl. recovery | FSonic, FF & HF

Dynamic Range Analysis
———————————————————————————————-
DR Peak RMS
———————————————————————————————-

DR14 -0.11 dB -16.24 dB A1
DR13 -1.12 dB -15.71 dB A2
DR14 -1.38 dB -16.08 dB A3
DR13 -2.69 dB -16.62 dB A4
DR12 -1.50 dB -16.11 dB A5
DR12 -1.83 dB -15.25 dB B1
DR13 -1.63 dB -16.36 dB B2
DR12 -0.83 dB -16.37 dB B3
DR13 -3.09 dB -16.86 dB B4
DR12 -1.64 dB -16.24 dB B5
———————————————————————————————-

Number of files: 10
Official DR value: DR13
Title: Normalize Vs Volume increase
Post by: 2Bdecided on 2011-07-08 20:30:48
Nothing automatic there, just "manual click removal" - no idea why it would sound too clean (unless you have dirty tastes).

Cheers,
David.
Title: Normalize Vs Volume increase
Post by: DVDdoug on 2011-07-08 20:32:50
Quote
Is Audacity totally a lame program? Vinyl capturing sites seem to love it?!
As I think mentioned before, your choice of recording program shouldn't affect recording quality...  All of the "real work" is done by the hardware & drivers.  The recording program simply gives you a "control panel",  communicates between the driver and operating system, and routes the digital data from the soundcard (or interface) to your hard drive.

When it comes to editing & processing, there can be a difference.  For simple things like cutting, pasting, and volume adjustment, there shouldn't be a difference.  But, for more advanced things like noise reduction, there may be a big difference.

If you made a list of the top 5 or 10 open source programs of all kinds, Audacity would probably be on the list.  But, it's not on the same level as commercial programs like Sound Forge or Adobe Audition.    I use mainly use GoldWave, but there are some things that Audacity can do that GoldWave can't, and I use some other special-purpose audio tools such as Wave Repair.

The vinyl capture guys probably love Audacity because it comes FREE with various USB turntables.  I believe it also came free with your ART interface.

Quote
...where the guy posts vinyl transfers he does and the specs on them. They sound great to me...  but too clean. I don't even know what half these processes or gear are.
Different recordings?  Better or newer recordings in better condition?  The manual de-clicking probably makes a big difference too.  His equipment list shouldn't make that big of a difference, except some phono cartridges are "brighter" than others, and some preamps are quieter than others.  But, 1950s technology (vacuum tubes), and silver wires (or oxygen-free copper) don't really help.


I can't relate to "too clean"....  I know some people prefer the sound of vinyl, and this is the vinyl forum...  But, I prefer noiseless, distortion-free, crystal-clear audio...    I grew-up with vinyl and the noise always annoyed me.    Now, I only use my turntable to digitizing records that are not available on CD.
Title: Normalize Vs Volume increase
Post by: AndyH-ha on 2011-07-08 20:36:13
I 've paid no attention to Audacity for some years so I can't comment on what ways it might have improved. The declicking and noise reduction were not very good, probably as likely to be destructive as helpful. It was clumsy relative to what I'm used to, but then being used to something else counts a great deal when making comparisons. It does not come close to better programs in display quality where actually getting much information off the screen is the goal.

There is nothing wrong with its recording quality, but the same can be said of virtually every recording application in existence. The main thing about it is that it is freeware. If you can't, or don't want to, put money into something better, it is way more than nothing.

Recordings applications do get some information from the hardware, or at least from what the OS tells them about the hardware through the driver; I won't pretend to know what is involved at the code level. Audacity is normal, Sound Forge is abnormal in regard to not letting you recording into a floating point format. Using floating point is not about the hardware, it is about optimal DSP processing. The hardware characteristics are irrelevant from that viewpoint. If you want only to record and store, floating point would be an irrational choice. (From what I read, there is such a thing as a floating point ADC, and it can have certain advantages, but I've never seen anything about one being used in recording music.)

To each his own in regard to tastes and goals. The kind of person who believes there's some benefit to keeping vinyl transfers at 24-bit has something weird going for him but it is hardly the kind of weird as the person who blows up a bus station because god told him sinners were using the buses to escape cosmic vengeance, and I have nothing to say against him.

I occasionally do only manual declicking, perhaps totally irrationally even then, but this is quite infrequent. I know there is a real data difference from any automated declicking process, and I know it is easy to severely damage the recording with such tools, but mostly it is possible to get very good results without any obvious downside, if your focus is on the music. I'm only concerned about 'how much' for any given recording. If I can't tell a difference in A/B comparisons, or I can only tell a difference by paying very close attention during an A/B comparison, then the easy route to the music is the rational choice.

I've been buying LPs from thrift shops for some years and have obtained some marvelous music. For anywhere from twenty cents to two dollars I get an album that would cost considerably more new, and, in many genres, is often better than a new version redone to modern mastering practices. I find not the least charm in the "vinyl sound". Getting the recording nearer to what it would have been if it had never been cut to the media to begin with is my goal.
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-07-08 21:09:28
Getting the recording nearer to what it would have been if it had never been cut to the media to begin with is my goal.


I totally appreciate that perspective as well. Maybe you're swaying me a bit to see it that way. The ONLY counter I have for that is that, well, it WAS cut to that media. The Beatles WERE cut to vinyl. Robert Johnson WAS cut to 78s, and for the most part people like this were heard on AM radio on crappy speakers. I've played CDs of the Beatles that are mastered "correctly" to my parents who grew up with them and they say, "hmmm.... interesting. But it's not the way I grew up listening to them so it's like a different song to me. The rhythm and the beat and the singing is what got us back then. Not the perfect sound."

But yeah, how it sounded, let's say, in the studio has value as well. I think this is an endless debate which we don't need to continue here.
Title: Normalize Vs Volume increase
Post by: greynol on 2011-07-08 21:16:55
Um, no, The Beatles were recorded to tape and the audio from that tape was manipulated in various ways in order to prepare it to be cut to vinyl.  If you want to get closer to what was done in the studio, you go back to the tape.

As to the rhythm, beat and singing, they are still present on any version of the CDs no matter how they were mastered, no?
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-07-08 22:46:43
Um, no, The Beatles were recorded to tape and the audio from that tape was manipulated in various ways in order to prepare it to be cut to vinyl.  If you want to get closer to what was done in the studio, you go back to the tape.

As to the rhythm, beat and singing, they are still present on any version of the CDs no matter how they were mastered, no?


A. It's an opinion, not fact. It's just how may parents, with uneducated ears feel, man.
B. I said they were "cut to vinyl", not that they weren't first mastered on tape. Is that wrong? They weren't cut to vinyl? What they were mastered on is also irrelevant to the average person who first heard them on record or radio. Or am I wrong in assuming that most people didn't have reel-to-reel tape machines in the average household in the early 60s?
C. I know you made an attempt to apologise for the comment "discussed to death" you made earlier that I took personally, and I let it go. But I have to say that the "um, no" at the start of your reply would be considered by most people to be a passive aggressive and kinda pretentious way of saying, "You're wrong and I'm going to make you look stupid for saying what you did." If you need to be right all the time then either shut my thread down or ignore me. I realise "Super Moderator" lends you a certain omnipotence, but I didn't ask for the attitude. And yes, it's attitude. Clearly not everyone can be as knowledgeable about audio as you and you need people to know it. I'm sorry, it's how I feel. Shut me down if you have to.

The rest of you, thank you ever so much for your kind attempts to help me understand. I apologise for my slow intake of all the info you've given me but I assure you it's been helpful to me and I appreciate it.
Title: Normalize Vs Volume increase
Post by: greynol on 2011-07-09 00:07:51
Denying that vinyl was the most common medium back then would be denying my own childhood.  While there is something to be said about songs as you remember them, I suggest you consider that it may have more to do with the editing than it does with the medium.

Regarding your temper tantrum, I made no attempt to apologize as there was nothing for which to apologize.  Assuming you're correct that I've given you attitude, what do you hope to achieve by giving back stronger attitude?  I have no intention on shutting you down, nor this discussion.  However, if you cannot manage to engage in a civil discussion then I suggest you refrain from posting.
Title: Normalize Vs Volume increase
Post by: Juha on 2011-07-09 09:31:27
Couple questions:

Would normalization be more secure because of the highest peak of source data is (automatically) checked before the final process (what I've learned, gain change in digital domain is kind of increasing/decreasing bits --> as (norm./ ampl.) process is done in 32/64-bit enviroment, after amplifying (without knowing the 'secure' amplification value) there might be clipping present you maybe can't hear until the data is transfered to its final 16-bit resolution (isn't 24-bits native driver resolution for many DACs ?)?.

Are there any advantages in amplifying or normalizing 24-bit data rather than 16-bit if the final resolution is 16-bit?

Juha
Title: Normalize Vs Volume increase
Post by: AndyH-ha on 2011-07-09 09:40:40
What one gets by normalization is independent of the bit depth (except that the quantization error is smaller at greater bit depths, just as it is with any other operation you might do on the data).

Now, if you mean should you do it while at 24 bit depth or wait until you have the 16 bit final format, do it while at 24 bits. One normally dithers when reducing the bit depth. Dithering is also recommended, for each operation, if working at 16 bits. Thus you would (should) have dithered twice if you wait until you've converted to 16 bit.

Of course, working in floating point gives you the ability to avoid clipping by normalizing to a reasonable level before converting to integer.
Title: Normalize Vs Volume increase
Post by: 2Bdecided on 2011-07-09 12:30:07
Um, no, The Beatles were recorded to tape and the audio from that tape was manipulated in various ways in order to prepare it to be cut to vinyl.  If you want to get closer to what was done in the studio, you go back to the tape.
On original vinyl, without damage, played on a decent turntable, sounds pretty close to the master tape anyway. Closer than the current stereo CDs (extra compression, 40 years of master tape degradation).

A beaten up vinyl, played on a cheap portable record player, sounds nothing like the master tape. That may be the sound many remember, and the sound most heard. It's certainly what was in George Martin's mind when he made those recordings. That's why lots of 60s pop music sounds so sparse on CD. That's why 2000s pop is so mushed on CD - producers know the replay equipment won't mush it up any more!

Cheers,
David.
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-11-12 22:24:51
If I'm going through a recording process with a mono record, where I'll be summing the two channels picked up by a stereo cartridge, where should the channel summing process happen? Basically, should I declick the raw 2-channel recording or declick-pop once the channels have been summed?

(Just in case, yes I know there are mono stylii, I know a stereo stylus on a mono record isn't true reproduction, I'm not buying a mono stylus, that level of fidelity doesn't interest me. Got heck elsewhere for even suggesting a stereo stylus on a mono record.)
Title: Normalize Vs Volume increase
Post by: mjb2006 on 2011-11-12 23:20:34
Why did you bump this thread? Your question has nothing to do with the topic.

Why are you choosing to sum the channels after [a href='index.php?showtopic=89597']you were essentially told that wasn't necessary[/a]?

This topic may also be of interest ([a href='index.php?showtopic=54633']click[/a]); your question is answered there.
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-11-13 04:09:53
Why did you bump this thread? Your question has nothing to do with the topic.
Actually it was part of the conversation in this thread that I started. P.S., my question isn't about whether to record stereo, my question is about when to do the declick before or after summing the channels.
Why are you choosing to sum the channels after [a href='index.php?showtopic=89597']you were essentially told that wasn't necessary[/a]?
I guess you've never made a mistake before and forgotten something after you already asked it months before, huh? Oh, and this guy disagrees with you:
"from the fairly large number of mono LPs I done, only a couple have not done better summing the two channels rather than choosing one channel. I believe both exceptions required choosing parts from each channel to make one best.
The summing does need to be done as the proper step, after declicking and (when needed) after decrackling."


This topic may also be of interest ([a href='index.php?showtopic=54633']click[/a]); your question is answered there.
Thank you.
Title: Normalize Vs Volume increase
Post by: botface on 2011-11-13 09:24:33
Basically, should I declick the raw 2-channel recording or declick-pop once the channels have been summed?

Most people will advise declicking then summing to mono. However, I have achieved better results with some 78's that were in very poor condition by recoding them mono in the first place - by summing the channels before the preamp and recording to a mono track. So, I'd suggest trying both of the approaches you mention and seeing which one works best with your source material
Title: Normalize Vs Volume increase
Post by: tinpanalley on 2011-11-13 21:24:07
by summing the channels before the preamp and recording to a mono track

Hey thanks for the reply. Appreciate it. But I don't think I have the capability of doing this. At least I wouldn't know what to use.
Title: Normalize Vs Volume increase
Post by: 2Bdecided on 2011-11-15 10:23:35
by summing the channels before the preamp and recording to a mono track

Hey thanks for the reply. Appreciate it. But I don't think I have the capability of doing this. At least I wouldn't know what to use.
It's identical to recording in stereo then converting to mono as the first stage in software. So there's no advantage to recording in mono IMO+IME.

I record, declick, decrackle and denoise 78s in stereo, then convert to mono as the last step. It works far better than converting to mono first (or capturing mono). The artefacts from the automatic restoration tools are pretty much the same either way per channel, but are reduced by about 3dB relative to the music when converting to mono as the last step, since they're not 100% correlated between the channels (and the music is, or should be).

YMMV - depends on the recordings and the restoration tools I guess.

Cheers,
David.
Title: Normalize Vs Volume increase
Post by: AndyH-ha on 2011-11-15 10:36:57
It also depends on the channel balance. Some cartridges, and set-ups thereof, are better than others. That is to say, the summing goes better or worse depending upon this. It seems to be pretty much the case that declicking first works better regardless.
Title: Normalize Vs Volume increase
Post by: botface on 2011-11-15 16:27:38
Yes, I wasn't suggesting recording in mono as the preferred method just pointing out that I have on some occasions got better results by doing so. Of course it should make no difference whether you mono in software or hardware but trying monoing first and comparing the result with monoing afterwards is what I was suggesting
Title: Normalize Vs Volume increase
Post by: 2Bdecided on 2011-11-16 12:43:54
It also depends on the channel balance. Some cartridges, and set-ups thereof, are better than others. That is to say, the summing goes better or worse depending upon this.
If I'm in a perfectionist mood, I check this before summing. I find the best approach is to adjust the levels while listening to the channels summed with one channel inverted. Set the balance for least wanted audio audible in this "difference" channel, then with that set, sum as normal (i.e. without inversion).

It's quicker to do that to say - especially with analogue controls, or analogue-like controls in the software. You can get close enough with CEPs channel mixer + real time preview, creating a new channel from L-R to check the proportions of each that are best.

Cheers,
David.
Title: Normalize Vs Volume increase
Post by: musicollector on 2011-11-28 03:30:14
Some people like to break a recording into individual tracks and work on each separately. I find it more convenient to record both sides of the LP into one file and keep it that way until I am done with all processing. Normalizing, at the proper time, is then a single operation against the entire album. There is no possibility of altering the album's dynamics.

Normalizing should be done after any and all filters are applied, especially if you intend to use 0dB.


I was one of those people! No more! I also like the idea of processing each file BEFORE breaking them up as it saves time and effort.

Thanks for the reminder.