Topic: latency issues while synching 2 iRiver sources....

[mikesalvo]

so, I have a recording that Im trying to synch. one SBD source, and one AUD. both were recorded on 2 seperate iRivers (120) from UA-5's. I used this method many times, and never had an issue. For some reason, Im getting major drift between sources. Whenever I match them, anywhere from 20-90 seconds later they drift from each other. Both are 44.1.

What the heck could possibly be causing this? 

[page]

so, I have a recording that Im trying to synch. one SBD source, and one AUD. both were recorded on 2 seperate iRivers (120) from UA-5's. I used this method many times, and never had an issue. For some reason, Im getting major drift between sources. Whenever I match them, anywhere from 20-90 seconds later they drift from each other. Both are 44.1.

What the heck could possibly be causing this? 

Just different clocks. When I mixed between a 722 and MR-1, the drift wasn't constant show to show so I always had to redo the math. I'd hazard a guess that you've been getting lucky in prior instances.

[stantheman1976]

so, I have a recording that Im trying to synch. one SBD source, and one AUD. both were recorded on 2 seperate iRivers (120) from UA-5's. I used this method many times, and never had an issue. For some reason, Im getting major drift between sources. Whenever I match them, anywhere from 20-90 seconds later they drift from each other. Both are 44.1.

What the heck could possibly be causing this? 

Just different clocks. When I mixed between a 722 and MR-1, the drift wasn't constant show to show so I always had to redo the math. I'd hazard a guess that you've been getting lucky in prior instances.

They start drifting 90 seconds into the recording?  That's not normal.  Even taking into account clock speeds between separate devices drift shouldn't start to happen until further into the recording.  Even when I sync two totally different recorders that used different sampling rates I rarely get more than about 1/2 to 1 second of total drift over an hour or more.  I wish I could help more but I can tell you for sure if the recording starts to drift after that short of a time then something is wrong with the recorder it's happening on.

[mikesalvo]

ya, this is REALLY bizarre and frustrating. Shouldnt the iRivers typically have no issues synching since they are on the same clock? Ive never had this problem using these2 recorders. Ive tried for HOURS and still cant seem to fix this matrix... 

[page]

so, I have a recording that Im trying to synch. one SBD source, and one AUD. both were recorded on 2 seperate iRivers (120) from UA-5's. I used this method many times, and never had an issue. For some reason, Im getting major drift between sources. Whenever I match them, anywhere from 20-90 seconds later they drift from each other. Both are 44.1.

What the heck could possibly be causing this? 

Just different clocks. When I mixed between a 722 and MR-1, the drift wasn't constant show to show so I always had to redo the math. I'd hazard a guess that you've been getting lucky in prior instances.

They start drifting 90 seconds into the recording?  That's not normal.  Even taking into account clock speeds between separate devices drift shouldn't start to happen until further into the recording.  Even when I sync two totally different recorders that used different sampling rates I rarely get more than about 1/2 to 1 second of total drift over an hour or more.  I wish I could help more but I can tell you for sure if the recording starts to drift after that short of a time then something is wrong with the recorder it's happening on.

my fr2-le and iriver drifted enough after 3-5 minutes that it made it unpleasant. My MR-1 and 722 didn't seem to drift for a good 20 minutes. Out of an hr, it's usually not off more then 22000 samples.

ya, this is REALLY bizarre and frustrating. Shouldnt the iRivers typically have no issues synching since they are on the same clock? Ive never had this problem using these2 recorders. Ive tried for HOURS and still cant seem to fix this matrix... 

Nope, all devices are best treated as individuals. You may get 2 722s, but they will have drift unless you use the WC or CLink functions. It's just a question of how much how fast.

As for fixing it, if the drift is reasonably constant, then the best guess is to treat it like any normal non-WC mix and stretch via resampling. Morst wrote an FAQ and directions (which are remarkable I might add, just make sure to read them a couple of times first) about it somewhere in the computer section.

[mikesalvo]

As for fixing it, if the drift is reasonably constant, then the best guess is to treat it like any normal non-WC mix and stretch via resampling. Morst wrote an FAQ and directions (which are remarkable I might add, just make sure to read them a couple of times first) about it somewhere in the computer section.

tried searching, couldnt find this. Any idea where it is? Thanks for your input btw! 

[runonce]

ya, this is REALLY bizarre and frustrating. Shouldnt the iRivers typically have no issues synching since they are on the same clock? Ive never had this problem using these2 recorders. Ive tried for HOURS and still cant seem to fix this matrix... 

Its not the same clock - just the same brand.

All sample rates are not created equally...

[page]

As for fixing it, if the drift is reasonably constant, then the best guess is to treat it like any normal non-WC mix and stretch via resampling. Morst wrote an FAQ and directions (which are remarkable I might add, just make sure to read them a couple of times first) about it somewhere in the computer section.

tried searching, couldnt find this. Any idea where it is? Thanks for your input btw! 

http://taperssection.com/index.php/topic,103031.msg1373822.html#msg1373822

Like I said, read it through a bunch of times first (and take some notes too). The first time I tried it, I did the math incorrectly 3 different ways before I figured out where I was misreading a section. Ever since then it's inconvenienced me maybe 5 minutes out of the entire workflow.

[SmokinJoe]

If you are running UA-5 > optical > H120 in 2 places, you are using the UA-5's clocks.  I've had good experience doing that, generally they are so close I don't need to shrink or stretch, just align.  Temperature could be a factor.  If one is in a hot bag, and the other on a cool bench, that could make the clocks differ.

To page's point, I find my MR-1 is the one with the drifty clock, and it's not linear.

There are a bunch of different ways to stretch/shrink one.  My suggestion is this:
- align the front, see how much they are off at the back (maybe like .300 seconds), and you want to shrink the bottom one.
- Then in Audacity, highlight that track and click "Effect -> Change Tempo".
- At the bottom of the screen it says "Length (seconds) from xxx.xx to xxx.xx".  Let's say it's 5000.00 seconds.  If you want it shorter by 0.3, change the right box to 4999.70
- Hit OK and let it churn.  Go line up the front again if necessary, and see how the back lines up.
- If you don't like it, edit -> Undo (twice) and repeat with a little different number.

[SmokinJoe]

One more thought... iRivers running optical in from the UA-5... if you have a flakey optical cable, you will get "skips" in time.    In other words there is a little bit of music missing.  On some other recorders you will get a drop out, or "dead air", but with an H120, it just won't write what it didn't understand, and it will just pick up where it can, and you have no freakin idea how much is missing.  I've had recordings with a fraction of a second missing, and I've had recordings with minutes missing.

The point being if you have one of these, you will drive yourself crazy trying to fix it.  The only fix would be to find those skips and insert dead air for the right amount of time.  Unless you found the missing Nixon tapes, it's not worth the effort.

[runonce]

One more thought... iRivers running optical in from the UA-5... if you have a flakey optical cable, you will get "skips" in time.    In other words there is a little bit of music missing.  On some other recorders you will get a drop out, or "dead air", but with an H120, it just won't write what it didn't understand, and it will just pick up where it can, and you have no freakin idea how much is missing.  I've had recordings with a fraction of a second missing, and I've had recordings with minutes missing.

The point being if you have one of these, you will drive yourself crazy trying to fix it.  The only fix would be to find those skips and insert dead air for the right amount of time.  Unless you found the missing Nixon tapes, it's not worth the effort.

^^^Very true...

I have had clock synced sources...that when you try to align - after some point, they will =suddenly= go out of sync...not drift, its very abrupt...

Took me a while to figure what was happening...and figure out where the samples were dropped.

Each source sounded fine on its own...the dropped samples werent noticeable to the ear...only when you tried to sync up.

I think I resorted to: sync, then export to the unsync point...resync, and then export the rest...seems like I had to do this a few times to get the whole show straightened out.

And it did turn out that I had a bad optical cable...

[F.O.Bean]

So, would yinz recommend dropping the iRiver and getting a Sony D50? You could move to 24bit AND get a more reputable recorder/bit dump, both at the same time

[runonce]

So, would yinz recommend dropping the iRiver and getting a Sony D50? You could move to 24bit AND get a more reputable recorder/bit dump, both at the same time

Hey thats y'uns out here in south central my PA brother!

And for sure - matrixing 16 bit sources is for the birds...thats why I roll OTF.

When your serious about multi-tracking you need those extra bits!

[F.O.Bean]

So, would yinz recommend dropping the iRiver and getting a Sony D50? You could move to 24bit AND get a more reputable recorder/bit dump, both at the same time

Hey thats y'uns out here in south central my PA brother!

And for sure - matrixing 16 bit sources is for the birds...thats why I roll OTF.

When your serious about multi-tracking you need those extra bits!

What does OTF mean

[runonce]

So, would yinz recommend dropping the iRiver and getting a Sony D50? You could move to 24bit AND get a more reputable recorder/bit dump, both at the same time

Hey thats y'uns out here in south central my PA brother!

And for sure - matrixing 16 bit sources is for the birds...thats why I roll OTF.

When your serious about multi-tracking you need those extra bits!

What does OTF mean

On
The
Fly