Is the DR database really accurate for vinyl?

You are correct - I was referring to Mr. Shepherd's comment on the SQ between his CD rip and his vinyl drop and his assessment of those.

I have seen my drops change by 1 or 2 notches in the DR rating, however. I think it was a combination of a better ADC and when I moved from an AT120e to my AT440Mla (I have redone a few of my 2011 drops for stuff I listen to often). I think it's the cart that made the primary difference as the 440 is definitely more detailed.

 

Here's an example of a flip-flop for me, btw. Snow Patrol's "One Hundred Million Suns" (2008). My first drop was just plain BAD, so much so that I would still listen to my MP3 version over the drop. Horrible mastering. Re-did my drop earlier this year using my latest equipment and it's still bad, but it finally got to the point where I preferred it over the lossy MP3 version. Barely. (CD version I had heard from a library copy and that one was really bad too).

I have a few other LP's where the lossy download copy / CD copy is still the go-to version. Old LP's and newer ones both - I go with what sounds best to me.

 

This video made me do an experiment with DR values and waveforms, the results are quite interesting.

First of all I picked a song with similar dynamic range to the one that's in the video. In this case Laura Cantrell's "All The Girls Are Complicated". I played back the song on my player and captured the player's analog output straight into my Edirol R09 recorder by an analog cable. Then I took a straight rip from the CD and normalized both the actual CD file and the analog capture from the player to maximum level. Here we have the wavforms of both of the files in an audio editor, the CD on top and the analog capture from the player on the bottom.

As you can see there's not a lot difference between the wavforms. The bottom one is a bit more quiet, but there's no obvious differences in dynamic range. Even when zoomed in on a drum beat, like below, the wavforms are again very similar.

But what does the DR meter say?

DR9 -0.10 dB -10.79 dB 3:03 ?-lauracantrell_fromdisc
DR10 -0.10 dB -11.69 dB 3:02 ?-lauracantrell_fromplayer

It adds a full dB of dynamic range and 0.9 dB of RMS. Not as extreme as in the video, but with the waveforms looking so very similar I'm still surprised at the significant difference.

Here's the same experiment repeated, this time with a much more compressed song, "Set Yourself On Fire" by Stars. Similar results, the only obvious differences occur at squashed peaks:


DR6 -0.10 dB -8.50 dB 5:39 ?-stars_fromdisc
DR7 -0.10 dB -9.31 dB 5:38 ?-stars_fromplayer

This would suggest that according to the Dynamic Range Meter, just playing back a disc already adds a dB of dynamic range and a nearly a dB of RMS? Very strange, where does this extra dynamic range come from?

 

I think you already answered it -- it's the analog out that's literally "colorizing" the sound, for the lack of a better word. What would also be an interesting test is, for a DAP or receiver for example, using the line out versus the headphone out as I'd guess you'd get different results just between those two outputs.

 

Great to see the discussion here. I think the video says everything I want to say on the topic, but a few extra bits of information to reply to some commenters here:

- Both files were lossless. The original source file was dithered down to 16-bit for the CD, whereas the file supplied for the vinyl cut was 24-bit, however it had the same limiting applied as the CD master, and so there will have been no benefits in terms of peak-to-loudness ratio (PLR), aka "dynamics"

- I *am* fierce opponent of the Loudness Wars, but my position is honest and therefore nuanced. So I would never say that higher PLR is always better - that's not true. This is one of the loudest songs on the album, which has an overall reading of DR10, iirc. I'm very comfortable with this choice, and my hand wasn't forced in the mastering in any way. DR8 for a loud song is easily enough to get a great sound, as I hope this album demonstrates - you can hear it on Spotify here:

http://open.spotify.com/album/4wxpLaoNDEhRgznUmozgvL

Or Bandcamp here

http://bandcamp.com/artemis

- In my opinion, mastering is about finding the "sweet spot" where the music is dynamic enough to sound fantastic, and controlled enough to translate well on the widest possible range of systems. I think I got it just right with this album.

- I did say that "some people like to call this 'brickwall' limiting", but the name tells you nothing about how such a limiter will sound. Used well, the effect can be inaudible. Here it wasn't inaudible, but it was beneficial, which is why I used it.

- I'm not a vinyl cutting engineer, but my work has been cut by Abbey Road amongst other places, and the comment often comes back "wish all masters were supplied like this - we didn't have to do anything to it".

- All the debate about how this was ripped, what equipment it was played on etc has nothing to do with the point I'm making. Any *apparent* increase in dynamics is an illusion, not a genuine benefit, IN THIS CASE.

Finally, to whoever said "I was surprised when I heard this, because it *looked* like a hard rock song" - thanks ! You just proved my point perfectly

Cheers,

Ian

 

Well I may have to revisit the video because I never heard him use the term brickwalled when referring to the mastering of this album.
When it comes to vinyl mastering, why do you assume that this is how he would master a recording for vinyl? My guess is if he was asked to master this recording for vinyl he would have mastered it differently. The label probably didn't want to pay for addition mastering specifically for vinyl. Mastering for a digital release is the priority now a days and vinyl is an afterthought.
Also why don't you have an idea how the CD sounds didn't you listen to it on the video? I understand it's YouTube but you still would have an idea of what the CD sounds like by watching and listening to the video. The sound quality isn't that bad.

 

Thanks for the info Ian, great video and I couldn't agree with you more!

 

Awesome experiment, and welcome to the forum.!

I have heard some say that simply pressing to vinyl and playing back adds either harmonics, literally from stuff "ringing", and or out of phase signals that are at different levels than the master or the CD.

I suppose there can be several characteristics at work here when playing vinyl.

 

There's no need to master differently for vinyl in most cases:

http://productionadvice.co.uk/vinyl-mastering/

I knew the album was going to be released on vinyl when I mastered it, and I knew I could do one master that could sound great on both formats. If it was ONLY for vinyl release, I would have passed on the mastering, since it would probably be more economic for the band to just get the cutting engineer to master it as well.

Ian

 

Very interesting to watch.

I assumed that something like this can happen, but I would have guessed that the difference in dynamic range would be about 1 dB or so. It is quite obvious to me that playing back a vinyl record can never give you a completely brickwalled waveform. So, the playback of Vinyl influences what ends up on a needledrop (waveform) and also what reaches our ears.

In the video, I like the sound of the vinyl a tad better, even if it is further away from the original file that went into the vinyl production.

But the numbers that we see from the DR meter have to be taken with some care for sure.

 

Thanks !

Yes, there's no question that cutting to vinyl (and playing it) changes the sound in many subtle (and not-so-subtle) ways - and that lots of people like the sound of the changes. They are just adding something that wasn't there before, though, not "revealing" something which was "hidden".

Ian

 

Thanks for replying and giving us more info on the source specifics.

I do disagree on one thing which is the needledrop specifics - from my limited experience I've found that the sound from the drop itself will vary between TT / cart / ADC etc. It's going to be a factor in any A/B test and it's going to "color" the sound in a different way. I call it aural trickery and you're calling it an illusion, but if our ears perceive something different and something that one thinks is "better" then that's really the end goal here.

I wish I could remember who posted this, but a forum member here called it vinyl's "secret sauce". There's just something about the distortions of vinyl that many of us find pleasing. I almost wish I could replicate that in an Audacity plugin as it could save me some money in the long-run.

 

Hi Ian,

thanks for clarifying. Could you by any chance post the spectrograms & spectrum curves of both CD & needledrop tracks (preferrably at equal perceived loudness, i.e. equal ReplayGain values) and equalization comparison graph too? Those would be interesting to look at...

 

Sorry, I didn't explain it clearly enough. YES different playback equipment and many other factors will affect the sound of vinyl each time it's played. All I'm saying is that the differences those factors introduce can't be measured accurately, so we can't draw any conclusions from them.

Maybe on another deck, on another day, with a different soundcard, that same piece of vinyl will play back in a way that doesn't measure DR12. It doesn't matter - all that's happening is "colouring" of the original source. If you like that "colour" - all well and good. Personally I'm rather partial to a little rumble

But if the measurement method is flawed, you can't draw useful conclusions form those measurements.

The TT meter *can* be useful for comparing lossless digital files, but that's about it.

Ian

 

My guess is there is a plugin somewhere for pro tools or something that can add that vinyl "secret sauce" into the audio.

 

Ian,

Welcome! Have you compared the CD and vinyl of the new Nine Inch Nails album? I'd be very curious to hear your thoughts.

 

I have a feature on my fairly budget priced ( but not bad sounding ) Yamaha Home theater receiver that I'm mostly using now that "Adds" some excitement and dimension to the sound.

It is supposedly meant to "Enhance" the sound of MP3 Files, but I have found after many days of experimenting that it adds a lot of dimension and soundstage to all music.

It is not so obvious that is sounds bad or calls attention to itself, but at times, I sense it and look around the side walls and the ceiling for speakers. I know it is an illusion and partially just an image enhancement, but it is by far the most listenable enhancement I have heard of this type. Certain sounds literally appear several feet beyond the speakers at times, and hang in the air.

 

I could, but they won't tell you more than you can already hear. The original recording was made at 44.1 kHz, so there will be no information above 20k-ish. The vinyl sounds a little softer in the high frequencies to me, other than that they're very close apart from some added noise and distortion in the vinyl.

Ian

 

I think some of the more extreme numbers can be helpful (and it's one of the reasons why I upload to the DRD website), so a LP drop at DR8 will tell you something's amiss for example. Or a DR14 on an LP rip, again versus a DR5 on the CD, will give you an indication that there's possibly something different between the two besides the format.

One thing I do is add a comment line if I feel the numbers are "lying" a bit, but I should also add my TT and cart too, just as a reference.

 

Thanks - I've done some quick comparisons and the vinyl sounds better (more dynamic) than the CD to me. I'm not ready to comment on the vinyl versus "audiophile" yet though, except to say that the vinyl has less bass on the rip I have.

Ian

 

Thanks Ian. The audiophile/vinyl comparison is the one I am curious about. If you get a chance, please PM me anytime!

I certainly agree that the vinyl has less bass.

 

My Cowon DAP has its Jet Effects as well - I use it, not heavily but I do tweaks until I find the spot that's the most pleasing to my ears. Essenitally its an EQ on steroids.

What's nice about some of my drops is that the dynamics are such that I can tweak these settings enough on the drop whereas with the CD rip with a DR5, for example, there's only so much I can do. I'm definitely no purist but I suppose that's pretty obvious - I want good sound, I don't need neutral sound and those two aren't always the same thing to me.

What I meant by an Audacity plugin was a way to bottle that vinyl "secret sauce" into a plugin where I could, for example, EQ a brickwalled CD to make it sound more like what a vinyl drop would sound like. In other words, "make my own vinyl secret sauce". I bet it's possible with enough trial and error.

 

I am at work, so I cannot watch this yet, but does this person provide downloadable wav files to compare?....or just the sound in the youtube video?

 

This 4dB DR value difference should manifest itself somewhere, IMO. Basically, the logarithmic difference between the 2nd max peak value & the RMS in the loudest 20% of your needledrop increased by 4dB compared to the same of the digital master. How?