Mastered tracks come back longer?

[Parametex]

Gents,

I would love your opinion on this:

I received an album back from perhaps the most highly regarded mastering house in UK just to notice that the tracks have gained extra length.

When i say gained length I do not mean there would be extra silence of some sort in the beginning or the end, no. What I mean is that if I sync the original track and the mastered track so they are in phase in the beginning of the track and then compare the end of the original track and the mastered track I can clearly see how the mastered version has gained length by some milliseconds.

All this would not be such and huge deal but I am now referring to electronic music where keeping time is pinnacle of importance. For example one track is originally tight and solid 129 bpm and after mastering it is 128.985 and clearly and audibly of out sync with the original.

We sent the original file as 44.1 KHz 24bit and received the files back 44.1KHz 16bit so I do not believe this has happened in downsampling either.

I would love your professional input on this.

Thanks you,
Px

[PBM]

Originally Posted by Parametex

Gents,

I would love your opinion on this:

I received an album back from perhaps the most highly regarded mastering house in UK just to notice that the tracks have gained extra length.

When i say gained length I do not mean there would be extra silence of some sort in the beginning or the end, no. What I mean is that if I sync the original track and the mastered track so they are in phase in the beginning of the track and then compare the end of the original track and the mastered track I can clearly see how the mastered version has gained length by some milliseconds.

All this would not be such and huge deal but I am now referring to electronic music where keeping time is pinnacle of importance. For example one track is originally tight and solid 129 bpm and after mastering it is 128.985 and clearly and audibly of out sync with the original.

We sent the original file as 44.1 KHz 24bit and received the files back 44.1KHz 16bit so I do not believe this has happened in downsampling either.

I would love your professional input on this.

Thanks you,
Px

What did the mastering house say when you raised it with them?

[Rob Murray]

Bring it up to the ME, I'm sure they'll help you make it exactly as you want it. (which in this case is clearly truer to the original) It would appear there was enough of a clock discrepancy between their pitch D/A and catch A/D in their analog loop to create this difference.

[Parametex]

Yes bringing it up with the ME is of course and option but before I do I wanted to get more professional opinions to know where to stand on this.

Is this normal in your books? Should I be able to expect the timing to be the same?

Thank you,
Px

[wado1942]

This has always been normal since the dawn of mastering. He obviously used some analogue component and the DAC/ADC clocks weren't a perfect match. It really doesn't matter unless you have somebody higher up the ladder than you insisting the songs stay the exact same length.

[Parametex]

Originally Posted by wado1942

This has always been normal since the dawn of mastering. He obviously used some analogue component and the DAC/ADC clocks weren't a perfect match. It really doesn't matter unless you have somebody higher up the ladder than you insisting the songs stay the exact same length.

Sorry, can you elaborate a little bit...

'Has been normal' as in time drifting is normal? Or expecting same timing back is normal?

Higher up in the ladder than me? Do you mean a label or higher up in the mastering house...

Just to clarify...

Thanks,
Px

[Joelistics]

Sounds like some tape print issue. That's all I can think of off the top of my head, but yes take it up with them.

Are you able to identify if it's a gradual drift? Might also be a glitch somewhere...

[Parametex]

Originally Posted by Joelistics

Sounds like some tape print issue. That's all I can think of off the top of my head, but yes take it up with them.

Are you able to identify if it's a gradual drift? Might also be a glitch somewhere...

The drift is gradual and not a glitch as it appears in all of the album tracks.

I will take the matter to the ME but I simply wished to know if this is something that is reasonable to assume.

Thx,

Edit: typos

[wado1942]

Originally Posted by Parametex

Sorry, can you elaborate a little bit...

'Has been normal' as in time drifting is normal? Or expecting same timing back is normal?

Higher up in the ladder than me? Do you mean a label or higher up in the mastering house...

Just to clarify...

Thanks,
Px

Sorry, I meant the length of the songs being slightly different is normal, whether they used tape or not. The only way you could expect them to be identical is if it was done totally "in the box" or with slaved digital hardware.

I remember working on a movie and the director was totally freaking about me working on tape. He insisted that everything be digital from that point forward so the time would be accurate as they planned on playing my rough mixes on set for timing cues. I told him that my machine was accurate within 1-frame over 33 minutes, the length of a full reel. I tested this myself several times, but he insisted that it would drift. So, I digitized it and worked ITB the rest of the way so he could feel good about himself.... except that the CD player used on set and the camera didn't stay in perfect sync with the ITB mixes anyway. It's just not realistic to expect everything to run at the exact same speed. Even the best crystal clocks in the world can drift up to a frame per day.

By "higher up the ladder than you" I meant like a psycho A & R guy put on your project. The mastering house is just another place where a job gets done.

[Franco]

To add what others have said, it's going to depend on your Mastering Engineer's chain (I know if you were my client, I would prefer you call me and ask me why, as opposed to asking people who have no idea what the chain is).

For my chain, all my masters have a bit of extra time at the tails, that's because of a small analog loop as well as compensation of DSP cards I use for processing (I add a tad bit of tail, usually about 30ms just to make sure I don't cut anything off while printing masters). I assemble sequences on another platform and it's at that stage where I tackle sequences or remove the extra time at the end if needed.

[squirreltrench]

I'm no expert, but I would assume that a Mastering House should be delivering to you exactly what you want. If you tell them exactly what it is about your tracks you want improved, then they should do that. If you give no guidance, then they have nothing to work from but their own tastes, which may be different than yours.

Since you didn't tell them that retaining the exact time down to the sample was important to you, then they weren't worrying about that aspect of your tracks; they were focusing on something else. Whether or not this is normal in mastering, I have no idea. I don't think there IS a "normal" in mastering. A mastering engineer is going to use whatever tools/process/chain he/she thinks will benefit the track the most. So clearly, in this case, the chain included something that threw the timing off slightly- as previous posters said, likely a slightly different clock rate in the DAC than ADC.

Basically, if this is really that important to you (of course this would only be perceptibly different when you ARE lining up your original track with the mastered track... which presumably no one else on the planet would be able to do), then simply tell that to the mastering engineer.

To me, the more important question is: how do you like what the mastering house delivered? Was it pure awesomeness is every other way?

Communication is key. Just tell the M.E. that sample-accurate timing is important to you, and he/she should be able to adjust their process accordingly, or at least open up a dialogue about what you are shooting for in mastering and what they are able to provide.

[Parametex]

Thank you all for chipping in on the discussion!

I'll be contacting the house to discuss it further.

Cheers,
Px

[Colin Leonard]

It's completely normal and not something to worry about if you like the way the master sounds. I mean, it shouldn't be by much at all, and definitely not audible unless you play the two tracks at the same time. How many milliseconds? If it's more than that I would say tape was involved.

[Joelistics]

I understand your concern and tbh I have never encountered this in my chain. For those who say it is normal, can you explain why you'd say this was?

Edit:
Hang on. A few MILLIeconds? I didn't catch that. No cause for concern. 1 millisecond=1 cycle at 1k. Difficult to even catch that visually in a DAW...

[dcollins]

It is normal for a D/A/D signal path. It can also be caused by asynchronous SRC. Nothing to worry about unless you are trying to sync the master up to something where very high accuracy is required.

Doubtful you would have noticed it without analysis.


DC

[grandmasters]

I notice this all the time using an analog signal path. Different ad/da's affect it too. It's absolutely nothing to worry about. I don't quite get why you would care if it sounds good.

[SASMastering]

There are many instances when the file length is not exactly the same as that delivered and for a variety of reasons as well.

If it is critical tell the ME when the job commences.

[mpdonahue]

Just a quick reality check here - This gives an effective sample rate of 44.094 kHz. Not really something I'd lose sleep over. Check the accuracy of your AD and DA converter, I suspect each be off by 3 or 4 Hz from the nominal frequency.
All the best,
-mark

[nms]

Where dance music destined for djs is concerned it's not a very good idea to use a DA and AD that aren't clock synced to each other for exactly this reason. Sounds like a simple matter of uncorrected clock drift that takes it off of the initial even integer bpm. The issue is that this causes the track to drift when mixed in a dj set with another track of the same bpm.

In contrast, I can route my Motu 828mk2 out through an analog chain and back in or back into my Mytek and it won't be off by even 1 sample when they're synced via wclk cable, so long as you truncate the beginning and end of the file in the same spot of course.

[Fat larry]

I remember when DJ's had to learn to beat mix by ear,Somtimes it was necessary to ride the pitch fader during a mix in order to keep things tight.

[IIIrd]

Originally Posted by Parametex

Gents,

I would love your opinion on this:

I received an album back from perhaps the most highly regarded mastering house in UK just to notice that the tracks have gained extra length.

When i say gained length I do not mean there would be extra silence of some sort in the beginning or the end, no. What I mean is that if I sync the original track and the mastered track so they are in phase in the beginning of the track and then compare the end of the original track and the mastered track I can clearly see how the mastered version has gained length by some milliseconds.

All this would not be such and huge deal but I am now referring to electronic music where keeping time is pinnacle of importance. For example one track is originally tight and solid 129 bpm and after mastering it is 128.985 and clearly and audibly of out sync with the original.

We sent the original file as 44.1 KHz 24bit and received the files back 44.1KHz 16bit so I do not believe this has happened in downsampling either.

I would love your professional input on this.

Thanks you,
Px

I take it that the pitch hasn't altered?

[807Recordings]

Originally Posted by nms

Where dance music destined for djs is concerned it's not a very good idea to use a DA and AD that aren't clock synced to each other for exactly this reason. Sounds like a simple matter of uncorrected clock drift that takes it off of the initial even integer bpm. The issue is that this causes the track to drift when mixed in a dj set with another track of the same bpm.

In contrast, I can route my Motu 828mk2 out through an analog chain and back in or back into my Mytek and it won't be off by even 1 sample when they're synced via wclk cable, so long as you truncate the beginning and end of the file in the same spot of course.

I would like to see the dj who can mix and notice a few milliseconds being off. In 25+ years of DJing I never seen it.

[squirreltrench]

Originally Posted by 807Recordings

I would like to see the dj who can mix and notice a few milliseconds being off. In 25+ years of DJing I never seen it.

You are absolutely right about a few milliseconds making no impact whatsoever on DJs in club use. However, the one place where I can see this making a difference is in remixing. Whenever I open a track in my DAW for remixing, I first determine the tempo. In older stuff, you can't just dial in an even integer BPM, it's going to come out at 123.5739 or whatever. Today's stuff: you dial in 124, and boom, you are locked in at 124.00000, thus making editing and remixing a breeze.

So if the OP is concerned about others grabbing his stuff for remixing, and being happy when they dial in the correct BPM and discover that it's 129.000000 and not 128.985, then it's a valid concern. However, even in this scenario, I would bet that most remixers would take a look at the track, dial in 129.00 and think they've got it right. Then when they are working near the end of the song, they'll wonder why their edits aren't landing precisely in the right place, if they even notice the drift.

[nms]

I was going by his comment that it was "clearly and audibly out of sync".

[807Recordings]

Originally Posted by squirreltrench

You are absolutely right about a few milliseconds making no impact whatsoever on DJs in club use. However, the one place where I can see this making a difference is in remixing. Whenever I open a track in my DAW for remixing, I first determine the tempo. In older stuff, you can't just dial in an even integer BPM, it's going to come out at 123.5739 or whatever. Today's stuff: you dial in 124, and boom, you are locked in at 124.00000, thus making editing and remixing a breeze.

So if the OP is concerned about others grabbing his stuff for remixing, and being happy when they dial in the correct BPM and discover that it's 129.000000 and not 128.985, then it's a valid concern. However, even in this scenario, I would bet that most remixers would take a look at the track, dial in 129.00 and think they've got it right. Then when they are working near the end of the song, they'll wonder why their edits aren't landing precisely in the right place, if they even notice the drift.

I am not disregarding this on the remix side but I am also pretty sure any of my drum machines would be off by a lot more also. Never had a problem remixing stems even off tape that I am sure where in the .0x% ranges. Still that said there have been times where stuff coming not exactly on time had lead to a lot of needless work hours in aligning (pre warp days).

[dcollins]

Originally Posted by Fat larry

I remember when DJ's had to learn to beat mix by ear,Somtimes it was necessary to ride the pitch fader during a mix in order to keep things tight.

I was thinking that same thing. The accuracy of two turntables must be much worse than this example of the two files......

DC

[jdg]

you got MORE music back then you sent... value added service IMO

[squirreltrench]

Originally Posted by nms

I was going by his comment that it was "clearly and audibly out of sync".

Right... I presume that what the OP means is that he took his original file and his mastered file, aligned the starts up exactly, then played them both at the same time. They were probably spot on with each other when he first hit play, and then by the end of the track, you'd be hearing the slight difference in timing. If it's a 4-minute song, they will be off by 28 ms by the end of the track, if I've done my math right (OP said original song was 129 bpm and mastered track came back at 128.985 bpm = 0.0116% slower).

So at 1 minute into the song, there is a 7 ms difference; at 2 minutes a 14 ms difference, at 3 minutes a 21 ms difference etc.

Now that I've calculated it out this way, there is no possible way a human being could ever tell the difference in timing between the two tracks, unless you lined them up at the same time and pressed play. That's why I stated earlier in this thread that the only person who will ever be able to do that is the OP, so it's probably not worth worrying about... unless you are concerned that remixers of the song will be off by 28 ms in their edits if they assume the song runs exactly at 129.00 bpm. That would still be hardly noticeable.

Maybe it's a 20 minute song, in which case it would be 140 ms off by the end... now THAT would be noticeable.

Yours mathemusically,
Morriss

[Jerry Tubb]

Originally Posted by jdg

you got MORE music back then you sent... value added service IMO

actual lol.

Cheers, JT

[ves]

Originally Posted by nms

I was going by his comment that it was "clearly and audibly out of sync".

QFT i have numerous tracks that drift audibly over a long period of time. after mastering. I am not sure why this is happens maybe because some parts of the tune are treated different maybe because of a small reverb? or maybe because the da / ad or oversampling etc. The important thing is that small drifting is happening when it shouldn't