How do I determine how much OTB Latency?

[gongbass]

After being pointed to the 'Expert Sleepers-Latency Fixer' by this great forum I have a question that I'm simply lost on. I understand the process of inserting this plug on the "Send" track and how it works BUT I don't know the method for determining how much latency a track that I'm sending to outboard effects/compressors/EQ has.

Is there a simple way of finding this? I know PT and Logic have a plug that essentially does the math for you, like DP does with Plug Ins. I can't even imagine the time I'll save once I get this protocol down, so if anyone would be kind enough to explain the process, I'd be greatly appreciative! Thanks for reading everyone!

[Dan Worley]

There may be a plug-in that does all this automatically for you, I don't know.

Basically you just record the return and measure the difference between its starting wavefom and the original's.

Go to Setup > Time Formats and set it to show in samples. Go the the SE, turn off the grid and zoom in. Highlight the area between the start of the original's waveform and the start of the returns waveform and read the selection length (in samples). Divide that number of samples by the sample rate. The quotient is the number of milliseconds.

Oh, use a sharp transient soundbite to measure, like a stick click or something. It makes it much easier to see the waveform. Also, make sure you zoom all the way in or you'll get false results.

--
Edit: Changed "it's" (the contraction for "it is") to "its" (the possessive of it). I just hate that!

c-ya,

Dan Worley

[kassonica]

There may be a plug-in that does all this automatically for you, I don't know.

Basically you just record the return and measure the difference between it's starting wavefom and the original's.

Go to Setup > Time Formats and set it to show in samples. Go the the SE, turn off the grid and zoom in. Highlight the area between the start of the original's waveform and the start of the returns waveform and read the selection length (in samples). Divide that number of samples by the sample rate. The quotient is the number of milliseconds.

Oh, use a sharp transient soundbite to measure, like a stick click or something. It makes it much easier to see the waveform. Also, make sure you zoom all the way in or you'll get false results.

c-ya,

Dan Worley

That is the way to do it, but come on lets admit it, it's a PITA.

It's the one thing I've wished for in DP for a long time.....

[cuttime]

Didn't there used to be a calibration file included with the install disk around version 4.6 or so? Couldn't it still be used?

[gongbass]

Thanks guys. So this isn't really any different from lining up spikes manually. Its still a time consuming process. Although, if the OTB latency is the same every time than this plug is actually a very big deal. Problem is from what I can tell, latency differs depending on gear being used.

[Dan Worley]

I don't use any outboard boxes when mixing, but I would guess you'd only have to measure each piece once and that should do it. As long as you don't change the path it should be consistent each time (I would think).

Not sure if there's a tool for this on the install disk as cuttime suggested or not.

c-ya,

Dan Worley

[Kubi]

Works the same for every piece, and fwiw, my number hasn't changed since I measured it sometime in the DP5 era. Here's a bit from a thread on the topic from a few years ago:

To figure out my roundtrip delay, I measured everything every which way. My buffer setting had no influence on the roundtrip duration (And it's my understanding that it shouldn't. Buffer only influences things the app cannot anticipate, i.e. patch thru. But regarding playback from disk, and placing of recorded material to disk, the buffer delay will be compensated internally. AFAIK. I get confused.)

The Sample rate DOES influence the roundtrip delay. Haven't tested whether the bit depth does (I never use 16bit), but it logically should, too.

At 24bit, 44.1kHz my roundtrip is 120 Samples, again regardless of buffer setting, according to my test. So that's what my Latency Fixer is set to.(...)

To measure your roundtrip, make a two-pop (1kHz sine at -18 to -12dB for 1 frame (about 33ms) and insert it in a mono track. (I made like 30 in a row, every two seconds.) Make sure the wave starts at 0 and smoothly goes up from there. Send it out and back in and record to an adjacent track. Set levels to match as close as possible.

Magnify like crazy. (The new track's waveform will as a 'side benefit' show you a bit of the distortion your DA-AD chain introduces... you'll also see if your polarity is correct, not unimportant.) Then shift the new recorded track earlier until they are in perfect sync. Pan one left, the other one right and play back. When they play back good and loud and dead center, you got it. [sthg that's especially easy to check on headphones... note that since you will have some distortion due the roundtrip, and since you won't be able to match levels 100%, you won't be able to do a null test on this one.] DPs waveform display is amazingly accurate, so once you've lined up the waveform, the playback will just confirm what your eyes are seeing.

The number of samples shift necessary to get the recorded track earlier to line up with the original is your roundtrip delay. My number has worked for me in many critical applications.

[cuttime]

Welcome back Kubi!

[Dan Worley]

Welcome back Kubi!

+1000

c-ya,

Dan Worley

[James Steele]

Welcome back Kubi!

+1

[Kubi]

Thanks. Pleasure.

[kgdrum]

nice to see you back here Kubi

[MIDI Life Crisis]

I would have missed the return of the prodigal son! Welcome back!!!

[Shooshie]

Thanks. Pleasure.

I am so glad to see you back!

Shooshie

[Phil O]

Hi, Kub. And yet another welcome back.

Phil