Skip to main content ↑ ↓

Notice

Please note that most of the software linked on this forum is likely to be safe to use. If you are unsure, feel free to ask in the relevant topics, or send a private message to an administrator or moderator. To help curb the problems of false positives, or in the event that you do find actual malware, you can contribute through the article linked here.
Topic: Ripping Mono Beatles CDs (Read 16451 times) previous topic - next topic
0 Members and 1 Guest are viewing this topic.

Ripping Mono Beatles CDs

Hello all,

I have been lurking here for a while and finally am posting.  I recently purchased the Beatles mono boxed set and am going through the process of ripping the CDs.  I have always used EAC but just discovered dBpoweramp and love it!  I am ripping to ALAC which, as you know, does not allow you to force a mono encoding.  As you also probably know, the two channels on the Beatles CDs are not bit-identical.  I would like to rip these in mono because of the size advantage (I've done tests with ALAC and there is a significant difference, though not 50%) and also because I want the recordings to show as mono in iTunes instead of stereo.  Who wants to rip two channels for mono recordings anyway? 

The question is how best to do this.  I discovered that dBpoweramp has the "channel mapper" DSP that allows me to pick only the first channel.  This seems to work well and give me the same sound as the stereo file at the same level (I think when I tried the "channel count" DSP, it gave me a hotter signal that clipped).  Then I came across this post [a href='index.php?act=findpost&pid=656222']here[/a].  Is there some other way I should be doing this so that I preserve all of the information and get an "extra bit" of data?  Can this be done by using a resampling DSP before converting to mono?  Or am I already doing it the best way?

Thanks for your help!

P.S. - All of the topics I have seen about this say that lossless encoders are smart enough to figure out when the source is mono and there won't be a difference, but there is a difference in this case.  Besides the size advantage, I really do want to end up with a mono file. 

Nick

Ripping Mono Beatles CDs

Reply #1 –
You say that the mono ALAC file is not 50% smaller than the stereo file. How much smaller is it?

Ripping Mono Beatles CDs

Reply #2 –
Quote from: pdq on
You say that the mono ALAC file is not 50% smaller than the stereo file. How much smaller is it?
As an example, "A Day in the Life" is 21,406 KB when doing a standard (stereo) rip, and it is 16,554 KB when just one channel is encoded by the above process.  So, closer to 30% but still significant. 

Ripping Mono Beatles CDs

Reply #3 –
If the difference had been small then I would have said that the two channels are very similar so keep either one.

This amount of difference suggests that there may be some kind of pseudo-stereo enhancement. Combining both channels into mono may or may not give you a desirable effect. You probably should try it and see.

Ripping Mono Beatles CDs

Reply #4 –
>I discovered that dBpoweramp has the "channel mapper" DSP that allows me to pick only the first channel.

In Channel Mapper you can manually combine the two channels, select 1 channel then click Change and add the 2nd channel, you would have to apply a reduction in volume to stop clipping or you could have DSP Effect:

Bit Depth: Floating Point
Channel Mapper: combine the two channels to one (can go over +-1.0)
Volume Normalize: Peak to Peak (to lower the volume to stop clipping)
Bit Depth: 16 bit (with dither - there is a beta test of dsp effects which offer dither)

Ripping Mono Beatles CDs

Reply #5 –
Okay - I tried also converting to mono using the same method except for mapping 50% L and 50% R channels to the mono file.  Interestingly, the result gave me a wav file with the same KB size on each.  I inverted one and added the two together, which appears to have given me nothing audible.  There is something there, and I did a frequency analysis in Audacity which gave me this result (not sure if it's relevant):



It seems it may not matter either way.  What interested me about that other post was the reference to an "extra bit".  Is there any reason to upsample, combine both channels, then downsample?

Ripping Mono Beatles CDs

Reply #6 –
I would guess that the high frequency difference is due to dithering the two channels independently. I don't know what would explain the low frequencies.

Ripping Mono Beatles CDs

Reply #7 –
Is it safe to say, if this is the difference between the two channels, that it wouldn't be audible?

I'm surprised the lossless encoder doesn't realize this and produce a file size similar to the mono.

EDIT: Also, based on the graph of what's there...would this be something I would want to add to my mono mix or keep out?  It's simple to do either way using the channel mixer.  I'm just trying to get the best chance at having good mono sound from the CD.

Thanks everyone!

Ripping Mono Beatles CDs

Reply #8 –
Quote from: ntm on
I inverted one and added the two together, which appears to have given me nothing audible.


I tried the same thing for "A Day in the Life". If you normalize the result, you will hear noise, but no musical content.

Since it is very low volume, I'm pretty sure there will be no audible differences between left, right and the mix of both.

Ripping Mono Beatles CDs

Reply #9 –
So is there an advantage to dither both channels independently?

Also ntm, if you're using lossless for archiving, wouldn't you want identical content to the source? Why not encode to AAC or MP3 otherwise?

Ripping Mono Beatles CDs

Reply #10 –
The dither might account for the HF noise (around -84dB), but what about the LF stuff around 1000KHz and -63 dB

Ripping Mono Beatles CDs

Reply #11 –
Quote from: andy o on
Also ntm, if you're using lossless for archiving, wouldn't you want identical content to the source? Why not encode to AAC or MP3 otherwise?

My main reason for using lossless is for sound quality, since HD space is cheap, I don't see a need to use lossy encoders.  I do want to preserve the source material, but I don't take that to the extreme.  I don't see a reason to keep stereo copies of mono material, and if I lose some dither noise while converting it back to mono, I'm not going to lose any sleep over it.  Lossless for me is about sound quality primarily.  Also being able to transcode from one codec to another is an important thing to me, depending on which codecs my portable players support.  This is hardly an issue of "well if you aren't going to keep it absolutely identical to the source, then you may as well use MP3" to me.  Some people may feel that way, but I don't. 

Is it possible that the noise is not unlike what would happen if you recorded a mono LP with stereo equipment?  Perhaps it's a combination of noise from the tape machine and dither noise that was not identical in each channel.  That's one reason why it's better to record old mono analog stuff in stereo and then subtract what isn't the same in both channels (presumably noise), right?

Ripping Mono Beatles CDs

Reply #12 –
Bump - Any help would be appreciated! Really trying to figure out where this noise is coming from and the best way to get mono audio from these CDs.

Thank you all!

Ripping Mono Beatles CDs

Reply #13 –
Personally I would just leave it as stereo and stop worrying about it. Surely the size difference isn't that big a factor?

Ripping Mono Beatles CDs

Reply #14 –
Here's another analysis of the "A Day in the Life" from the mono box set.

The remaining stuff after doing "invert mix paste" (= left minus right) in Audition:


The spectrum after normalizing the above file to 0 dBfs (= about +60 dB change):


When listening to it I can't hear any residue of the original track. It is clearly just high frequency noise (hiss).

A 30 s sample of the track in various formats is here:
http://www.hydrogenaudio.org/forums/index....showtopic=77836

For this sample the size differences are as follows (FLAC v.1.2.1 -8):

Stereo Mix (from the Stereo Box Set)  3.02 MB
Mono Mix (from the Mono Box Set), saved as a standard stereo file  1.94 MB
Mono Mix (from the Mono Box Set), the left channel saved as a mono file  1.58 MB

I'd say that creating "true" mono files would not be worth the effort. In addition, a stereo file is more compatible with various SW and HW setups. For instance, sometimes a mono file plays only through one of the speakers (one of a stereo pair or the center on a multichannel system). I have experienced this when an ASIO driver has been in use. Naturally this can be fixed by changing the playback settings, but then also the other files will play through the "forced" setup and may require changing the settings again.

Ripping Mono Beatles CDs

Reply #15 –
So, it's a 100% mono transfer, with stereo dither to 16-bits at the last stage.

If you add the channels together, you can average out the dither noise a little, dropping it by 3-6dB. You need 17-bit to save the result though - dithering or rounding(!) it back to 16-bits would defeat the point of the exercise! (though it would still reduce the high frequency noise).


Arguably, if you could hear the noise, it would probably be better to keep it in stereo anyway: 6dB louder, but spatially located well away from the signal of interest. Both these factors would make it easier to hear the noise, but it wouldn't be sitting on top of the music - it would be well away from it. In practice you can't hear the dither noise at all, so it's irrelevant what you do with it!

Cheers,
David.

Ripping Mono Beatles CDs

Reply #16 –
Quote from: Alex B on
Mono Mix (from the Mono Box Set), saved as a standard stereo file  1.94 MB
Mono Mix (from the Mono Box Set), the left channel saved as a mono file  1.58 MB

I'd say that creating "true" mono files would not be worth the effort.

Over 18% of savings, that's not worthless, I'd say. That'd be 430 MiB saved over the whole boxset (2,317 MiB).

Ripping Mono Beatles CDs

Reply #17 –
Presumably the difference in file size between the stereo and mono versions is due to the dither being uncorrelated. Would someone try creating a stereo version in which both channels are exactly identical and see how that encodes?

Ripping Mono Beatles CDs

Reply #18 –
"Eleanor Rigby", from Revolver - (flac --best):

Code: [Select]
stereo:  12 MiB  111 KiB  872 B
mono:     7 MiB  751 KiB  597 B
right:    6 MiB  162 KiB  393 B
right x2: 6 MiB  166 KiB  668 B

I'm not seeing 18% of savings on this one, so it might not be worth it after all. Doh! I can't count for shit. It's more like 20%. To an obsessive personality such as mine though, it's neater to have a "true" mono file. I'm going to try very hard not to transcode the entire boxset

Ripping Mono Beatles CDs

Reply #19 –
I'm a little confused. I think I understand right and right x2, but what is mono and how does it differ from right?

I'm also surprised that stereo would be alsmost twice as big as right.

Ripping Mono Beatles CDs

Reply #20 –
Quote from: pdq on
I'm a little confused. I think I understand right and right x2, but what is mono and how does it differ from right?

I'm also surprised that stereo would be alsmost twice as big as right.
I think "mono" meant a straight rip off the mono CD, whereas "right" indicated a true mono file taken just from the right channel.  I suppose right x2 was the stereo file with both channels truly identical.

Ripping Mono Beatles CDs

Reply #21 –
Quote from: ntm on
Quote from: pdq on
I'm a little confused. I think I understand right and right x2, but what is mono and how does it differ from right?

I'm also surprised that stereo would be alsmost twice as big as right.
I think "mono" meant a straight rip off the mono CD, whereas "right" indicated a true mono file taken just from the right channel.  I suppose right x2 was the stereo file with both channels truly identical.

I see. So then stereo probably refers to a paeudo-stereo version in which the two channels are significantly different, explaining the larger file size.
 

Ripping Mono Beatles CDs

Reply #22 –
I hope this is clearer:

Code: [Select]
"Eleanor Rigby", from Revolver - (flac --best):

Straight rip from the stereo remaster (2ch):          12 MiB  111 KiB  872 B
Straight rip from the mono remaster (2ch):             7 MiB  751 KiB  597 B
Right channel of the mono remaster (1ch):              6 MiB  162 KiB  393 B
"Fake stereo" file with the right channel of the mono
remaster as both the left and right channels (2ch):    6 MiB  166 KiB  668 B

Ripping Mono Beatles CDs

Reply #23 –
Quote from: skamp on
"Eleanor Rigby", from Revolver - (flac --best):

Code: [Select]
stereo:  12 MiB  111 KiB  872 B
mono:     7 MiB  751 KiB  597 B
right:    6 MiB  162 KiB  393 B
right x2: 6 MiB  166 KiB  668 B

I'm not seeing 18% of savings on this one, so it might not be worth it after all. Doh! I can't count for shit. It's more like 20%. To an obsessive personality such as mine though, it's neater to have a "true" mono file. I'm going to try very hard not to transcode the entire boxset
LOL the main motivator for me was my obsessive personality as well.  After going through the trouble to procure the mono box, I didn't want stereo files and thought it was neater to have "true" mono files.  My iTunes/iPod plays mono files back just fine and so does my iRiver optical output to my DAC.  I don't use multichannel for sound usually but with the mono I tried it and, alas, one channel out of all speakers.  There may be some systems that have compatibility issues, but mine doesn't. 

So, ripping in true mono, we get a mono file, a (not completely insignificant) savings in file size, and the satisfaction of knowing it's true mono - to those of us who care.  With dBpoweramp it takes no more effort or time as far as I could tell to rip just one channel or a 50/50 combination of both.  I do wonder if there is any particular way that is best to do this as far as SQ goes and that "extra bit" theory.  Would it make more sense to end up with a 24 bit file to salvage the extra information we get from combining the channels?

Ripping Mono Beatles CDs

Reply #24 –
well its nice that your OCD got you 'true mono' and saved some disc space, but now you no longer have a 'true lossless' copy of the source.

hows that for an OCD paradox?