bapu

bitflipper
However, a null test wouldn't be the best way to check timing. The notes (or, more accurately the several thousand samples that make up the initial transient and therefore the onset of the note) are going to be in exactly the same place regardless of whether the two files null. 

And so it may be said that the transient timing of the converted wav will always be as accurate (in time) as it was recorded?

I think it's more correct to say that the results of the converted wave will always be more accurate than the performance.

Issues like monitor latency and latency compensation will introduce much larger variances in timing that the performer may interpret as a shift from their intention or recollection of how they played.


best regards,
mike

bapu

bitflipper
However, a null test wouldn't be the best way to check timing. The notes (or, more accurately the several thousand samples that make up the initial transient and therefore the onset of the note) are going to be in exactly the same place regardless of whether the two files null. 

And so it may be said that the transient timing of the converted wav will always be as accurate (in time) as it was recorded?

Technically with Sonar's SRC, it shouldn't be off by more than a ridiculously small amount which is many orders of magnitude below what humans can perceive. Moving your head less than a billionth of an inch closer/further from the speaker will produce a greater change in timing.

IOW, from your perspective it should be 100% perfect. 

bapu

mike_mccue

Issues like monitor latency and latency compensation will introduce much larger variances in timing that the performer may interpret as a shift from their intention or recollection of how they played. 

Please define "much larger variances".


Are you saying that a listen on different monitors will perceive a timing shift for me?

Are you also saying that certain plugs, or combination of plugs, will issue me a perceived timings shift over the raw wav?


Are both of these examples still in the minute sub-sample category as posited in above posts?

Bapu I think Mike means phase differences due to driver placement in a speakers box. With near fields its an issue. Not one anyone has to worry about though.

I like the fact that we keep telling him not to worry about it and then give him a little bit more information...

This idea should keep his posting rate down from 48,000 per second to 44,100, no?

drewfx1


I like feeding hypochondriacs new things to worry about!

That makes me sick...

drewfx1


Be careful! I've heard sickness can screw up your timing! 

Maybe that explains my timing....

Karyn

bitflipper


Just to be clear: neither sample rate nor sample rate conversion affects timing. Every note will still be exactly Am, just as you played it.

Coffee hosed
 
 
 
 
 
 
 
 
 
 
 
Sorry

You should be sorry, he can't claim the sample rate conversion was the cause of him being out of tune now.

See what you did?

Besides if you'd have just told him he'd be OK in the first reply I'm sure this thread wouldn't have ended up being a 2 pager....

Karyn


Hey,  I wasn't the one that mentioned aliasing and sub-sampling...

True dat.

John

bapu

mike_mccue

Issues like monitor latency and latency compensation will introduce much larger variances in timing that the performer may interpret as a shift from their intention or recollection of how they played. 

Please define "much larger variances".


Are you saying that a listen on different monitors will perceive a timing shift for me?

Are you also saying that certain plugs, or combination of plugs, will issue me a perceived timings shift over the raw wav?


Are both of these examples still in the minute sub-sample category as posited in above posts?

Bapu I think Mike means phase differences due to driver placement in a speakers box. With near fields its an issue. Not one anyone has to worry about though.

Thanks John.

mike_mccue


8-)
Dude, I think you need to talk to an audio scientist about this.
8-)

N/A now.

Kumbh Mela

oh the Humanity 

bapu


And, unless you're Danny Danzi or Jimmy Iovine (in his youth), none of us will ever hear a real timing issue between a converted 24/48 to 24/44.1 track. 

LOL! I think you may be giving me a bit too much credit, but thanks for that. :) I don't think I can tell a difference on this...but then again, it's never been something I've ever tried to listen for. I have noticed a few little anomalies that may or may not be because of this. To be honest, it's both a blessing and a curse to be able to hear some of this stuff. A blessing because it's great to be able to hear things and fix them or notice problems, and a curse because sometimes, as Drew points out, there ARE cases when you think you hear something and it's just a false alarm. Also, when you don't know this stuff exists and then someone brings it to your attention, then you ARE listening for it and sometimes...well, again, you can think you hear it when in reality you probably don't.
 
If it's off by what Karyn, Noel and Drew mentioned....I seriously doubt I could hear it. However, it's worth me testing for the simple fact that I have heard a few things in the past as I've mentioned. In a few instances of me being so smashed with work, I've recorded a few things for people at 24/48 when they weren't using that bit/sample rate in their projects. There would be a few slight timing issues once they brought my stuff into their projects to where I'd have to show them where they needed to align something.
 
I always thought this was due to how some soundcards talk to others...but when you have a tempo map and are operating out of the same project template as someone else and then you send them a broadcast wave, it should be in sync. I've noticed at times that this is not always the case. Could it be I'm literally hearing this imperfection? Personally, I'd say I'm not and it's a soundcard issue between my stuff and whoever I'm working with. Let me ask this...can any of you literally hear this or is it something that you know in your mind is off a bit that may make you think you can hear it?
 
I never even knew there would be a timing difference so it's not something I've ever considered or tried to listen for. But you better believe I'm going to now. I still don't think I'll be able to tell as it's never been something that has stuck out to me in my own realm using my gear. I've done this several times and have never had a problem here. But like I say, I did notice some weird timing type things after I've given clients wave files that were different than what they were using. I still chalk it up as a soundcard communication though. There's no way I'm hearing 44 thousandth of a second. If I am, I seriously need help. LOL!
 
-Danny

The width of a transient like a drum hit or the start of a guitar note is many samples wide.  The absolute worse case would be a very high frequency percussive hit (like the crackles you get from a vinyl record) but even these are several samples wide.

Why might any of this matter?  Well the time between samples at 48k is smaller than the time between samples at 44.1k and thus running the two side by side (if you could) would mean that individual samples would rarely match perfectly in absolute time.

As the op asked for specific evidence,  assuming sample 1 of each starts at time 0:0  (m:s)

S(1)_44.1 = 0:0    S(442))_44.1 = 0:0.01
 
S(1)₄₈   = 0:0     S(481)₄₈    = 0:0.01
 
Thus, each 1/100th of a second the samples will align perfectly in time,  whereas between these points the samples will never match the same absolute time.
 
So for a total worse case with a single sample "spike" in the 48k stream, that sample will most likely not align perfectly in time with the corresponding sample at 44.1k,  BUT the most  the error can be is half the sample width at the target sample rate,  ie: 1/(44100 *2) = 0.00001134 Seconds early or late.
 
 
 
 
 
 
 
Why does the track I recieved from a colab partner not sync with my master track?
 
Drift in master clocks between two (or more) seperate systems.  This is why you have to sync digital equipment with a common clock (word clock) as free running clocks in seperate pieces of equipment will never run at the same speed.
 
If you send a .wav to someone to play along to, you've effectivly sent them a timing map of your master clock at the time you first recorded it. As long as they play along to that wav, their recording will match your timing and their resultant .wav will sync on your system to your master.
 
But, if you tell them "Give me 3 minutes of keyboards in Am at 132bpm" and they send you a .wav there's a good chance that by the end of 3 minutes it will be out of sync.
 
This has nothing to do with SRC.

Here is a website that claims to have collected SRC data from more apps than I can count:

http://src.infinitewave.ca/

I selected the "sweep" analysis and compared R8brain Pro, Minimum Phase to SONAR 8.5.

The results seem to close, but there seems to be enough difference to encourage curiosity.


best regards,
mike

;-)

Hey, at least you knew I wasn't writing about voice coil alignment or crossover compromises.

;-)

It is just Tannoy bapu.