Audio not in sync

[ski]

Out-of-phase, eh? Ouch...

 

First let's rule out the obvious... there shouldn't be any plugins anywhere in the song you're doing this test in except for the gain plug. Make sure of that first (especially on your outputs). Then, making sure you've bypassed the gain plug you inserted on Track 2 (the one you're recording the loopback "Y" on), make a new recording of "Y".

 

If "Y" comes back upside down again, then we need to suss out whether the output of your interface is putting signal out of phase, or if it's on the input side.

[rockydoo]

Thanks for the reply. Unfortunately, I didn't have any plugins anywhere in the recording. How do I figure out if the problem is on the input or output side of things?

[ski]

How do I figure out if the problem is on the input or output side of things?

 

I was afraid you were going to ask that question...

 

The best way would be to look at the output waveform on an oscilloscope. I know, silly question, but would you happen to have one?

 

If not, would you have a multimeter of some kind?

[rockydoo]

I have a friend who has an oscilloscope, actually. I've got plans to hang with him on Friday, June 27th and we could look at it then. When you get a chance, will you describe what I should do when I get access to the oscilloscope? Thanks!

[silentH]

im trying to follow this procedure, and i think im all set on the theory of it, but im having some trouble working it out. im using a fireface 800 for my interface, and i believe the recording is early.

 

i'm good up to the point where im going to move the anchor point around. i split region Y at 2.1.1.1, and i've moved the start point to the left to give me some space to move the anchor. when i click and hold the anchor, the display reads 88200, which i assume is my first number, the initial sample position. but when i move the anchor, this number moves with it, if you understand me. so when i've moved the anchor to a position that gives me a greater null, when i click the anchor, it reads the same sample, 88200.

 

what am i doing wrong?

[ski]

i'm good up to the point where im going to move the anchor point around. i split region Y at 2.1.1.1, and i've moved the start point to the left to give me some space to move the anchor.

 

All good so far.

 

when i click and hold the anchor, the display reads 88200, which i assume is my first number, the initial sample position. but when i move the anchor, this number moves with it, if you understand me. so when i've moved the anchor to a position that gives me a greater null, when i click the anchor, it reads the same sample, 88200.

 

what am i doing wrong?

 

Hmmm... OK, which version of Logic are you working in?

[silentH]

sorry, i thought i put that in my sig.

 

anyway, it's logic studio 8, latest version.

[ski]

Thx for that info. I'm on L7 atm. I'll boot up L8 a little later today and get back to you. Need to see what's going on in L8, dont' want to give you the wrong advice since I'm in a different version atm.

[silentH]

yeah, at first i was having some trouble moving the region start point independent of the anchor point, too. i got around that by right clicking on the anchor and deselecting compensate region position.

[ski]

Silent H,

 

See? I didn't forget aboutcha...

 

The sample editor looks a little different in L8 than in L7. In L8's sample editor there are two numbers side by side just to the left of the toolbox. Now, you say that your recording delay needs to be positive (because your recordings are ahead). So in this case...

 

Forget about the number on the right altogether.

 

Click/hold on the anchor. Write down the left number. Move your anchor until you get null. Subtract the new left number from the first number. That's your recording delay setting.

 

Who luvs ya?

[rockydoo]

I managed to test my Digi 002R with an oscilloscope and found that, indeed, by the time signal gets to any output, the phase has been inverted. What isn't clear to me is what to do about it, if anything. Any thoughts or suggestions?

[ski]

I managed to test my Digi 002R with an oscilloscope and found that, indeed, by the time signal gets to any output, the phase has been inverted. What isn't clear to me is what to do about it, if anything. Any thoughts or suggestions?

 

I guess the first thing I'd suggest is that you call Digi and ask them W(TF) is going on. Tell them what you did to test it, etc. Second thing would be to, well, I guess, get into the habit of having a phase inverter plug (gain plug with phase set to invert on L/R) on your output(s). That way at least you'll hear the sound in phase (it could have an influence on how you perceive transients, and the argument for having positive-going waveforms push the speaker cones outward seems like a good one).

 

Question: maybe you've said this already, but have you tested to see if the waveform you're recording is in-phase or not? Or is this just an output phenomenon?

[silentH]

Silent H,

 

See? I didn't forget aboutcha...

 

The sample editor looks a little different in L8 than in L7. In L8's sample editor there are two numbers side by side just to the left of the toolbox. Now, you say that your recording delay needs to be positive (because your recordings are ahead). So in this case...

 

Forget about the number on the right altogether.

 

Click/hold on the anchor. Write down the left number. Move your anchor until you get null. Subtract the new left number from the first number. That's your recording delay setting.

 

Who luvs ya?

 

but still, when i move the anchor point, it brings the number 88200 with it. i cant get a new number... am i not supposed to turn off compensate region position? i have to do that in order to move the region start point independently of the anchor point, but are you saying that i dont need to do that?

 

i really do appreciate the help, this is something that i've always wondered about on any daw, and i can see the latency- which is why its so frustrating to not be able to account for it!

[rockydoo]

I think it happens just on the outputs. A recorded waveform looks ok unless that recorded waveform is recorded from an output channel (as in the original test for setting RDC). I tested all the outputs (including the headphone out) and the oscilloscope showed them to be 180 degrees out of phase. Test results were the same in PTLE and Logic 8.0.2.

[m-m-m]

:

 

but still, when i move the anchor point, it brings the number 88200 with it. i cant get a new number... am i not supposed to turn off compensate region position? i have to do that in order to move the region start point independently of the anchor point, but are you saying that i dont need to do that?

In the sample editor you have to hold option to move the start of the region without affecting the anchor position.

 

I finally got mine done (after about 4 or 5 different techniques). I found it much easier to get everything done in the arrange window and to simply count the samples.

[ski]

@ rockdoo -- I'd call Digi and ask them what's up.

 

@ silentH -- I thought you knew how to move the start point to the left (without affecting the anchor point position). But in case I mis-read your post, do like m-m-m said, hold OPT and drag the start point to the left. Your anchor should remain where it is.

 

At that point, move the anchor manually to figure out the number of samples required to achieve null.

[silentH]

im still finding that it is not possible to move the region start point independent of the anchor point without first right clicking on the anchor and deselecting compensate region position, which adds some kind of little icon to the anchor graphic itself. it also makes it harder to line up the region in the arrange window, sometimes it wont let me get exactly at 2.1.1.1, it'll put the region at 2.1.1.1., with the little dot next to it to indicate that it's not exactly at 2.1.1.1.

 

whats really making me nuts now is that sometimes i can get it to null perfectly, but when i look at the waveforms, there are visually misaligned, and the number (88200) moves with the anchor point, so i cant do the math. sometimes the sample editor acts differently/shows a different display, even when i follow the exact same procedure. its very inconsistent.

[silentH]

now, when i try to split the region at 2.1.1.1, it makes the split in front of the playhead bar. this is madness- do i have a corrupt project file or something? this is the worst software behavior i've encountered with logic thus far.

[silentH]

okay, i just did the Z recording to test my work, and i think +89 samples did the trick. it's nulling immediatel after recording, without having to move the region around.

 

but im still having the problem where, i press split region by playhead, it's actually splitting the region about a sample behind the playhead's actual position. could this be some setting that i've inadvertently thrown off while doing this test?

 

finally, i've been doing this via the software loopback function of the fireface, but as far as i know, this should be exactly the same as connecting an actual cable between the inputs and outputs on the back of the device. if im wrong someone correct me, because i'd hate to have to redo all this when recording an actual external signal.

[ski]

Hi SilentH,

 

Wow, you're having lots of complications...

 

Region splitting past the playhead & inconsistent display of the waveforms: I'm stumped...

 

Compensate region position: I would like to think that the default setting (whatever it is) should be OK. In other words, you should set it so that the region does not change its position in the arrange page when you make edits to the audio file in the sample editor. You'll have to do a quick experiment to see which way that option should be set.

 

Hardware in/out of your fireface: I would think that this would be OK, but I can't say for sure. If you're getting total null then that's a good thing. But, if you were recording an analog signal (vocal, guitar, etc.) I wonder if the results would be the same. That's why, in my procedure, I specified that you actually cable output ----> input on your interface.

 

I'm going to suggest that you start with a fresh song and do the test one more time, looping output--->input with a cable. If your rec. delay setting of +89 samples is correct, you should get cancellation as soon as you playback your "Y" recording.

 

BTW, 89 samples at 48K = nearly 2 ms of delay (1.85 ms) which would definitely be enough to throw off the groove of what you record, for sure.

 

At 44.1K it's slightly over 2 ms of delay.

[silentH]

i just opened another project and the split function is working properly there, so im stumped too.

 

also, just to clarify what i was saying about the compensate region position setting: it's not that it's moving the region in the arrange window, (or it might be, but that's not the behavior im talking about) im saying that i cant move the anchor independent of the region start point in the sample editor itself without deselecting the compensate region position, which is not the default.

 

i will try a fresh project and test with an actual cable and get back to you. i really do appreciate you sticking with it and helping me sort this out.

[silentH]

alright, i tried it with an actual cable. (1/4 inch, unbalanced monster cable, maybe 6 inches in length) it looks like the latency is negligible: when i zoom in and look at both audio files in the arrange window, the waveforms line up almost perfectly; maybe a half sample off. the only noticeable difference between the two waveforms is some aliasing on the newly recorded one, which could either be the simple nature of a/d conversion, or it could be some loss in quality due to the unbalanced cables. either way, the latency is pretty negligible, as is the audio signal degradation, which is wonderful.

 

so now i know that if i need to run through effects or go outboard for any reason, i can do so without losing much audio quality and no need to worry about time alignment. and if i want a perfect audio copy of, say, the output from apple's dvd player or a youtube video, i simply adjust the latency +89 samples and record through the fireface's software loopback function.

 

so glad to have sorted this out. im definitely the type to get neurotic about esoteric minutiae when it comes to audio. you'll think im kidding, but i did not sleep as comfortably as usual last night...

[macphisto23]

Ok I just read through this whole thread- tons of information for a slow learner like myself.

 

I have been having problems with software instruments and 3rd party plug-ins. They record in sync with the project, but when I play back they are at least 1-2 seconds late in playback. Even though the midi matches with the audio regions perfectly.

 

So does this thread apply to me to get this problem fixed, or is it something different?

 

Help please

[ski]

macphisto,

 

It would only apply to you if you were sending audio out of your interface (to some outboard processing, say) and simultaneously recording it back into Logic on a new track. Is this what you're doing? And if so, what's your recording delay parameter set to?

[macphisto23]

macphisto,

 

It would only apply to you if you were sending audio out of your interface (to some outboard processing, say) and simultaneously recording it back into Logic on a new track. Is this what you're doing? And if so, what's your recording delay parameter set to?

 

I am not sure what I am doing

 

(I have to say that this is the first project I have done in Logic Pro 8, I just converted from Garageband.)

 

What I have done so far is-

 

Record 3 tracks of Guitars directly into my input 1 of the Fireface 800 inteface

 

Record 2 tracks of Vocals with mics

 

And now I created a new "software instrument" I first loaded a 3rd party Organ plugin in the I/O, I recorded it with my midi keyboard. All was in sync while I was recording but after I played back it was delayed 1-2 seconds. I then switched to a logic organ instrument but the same results.

 

I hope that helps, I am not sure if that is what you meant?

[amnestic]

ski--

 

So, I've been away from recording and such, as "real-life" (whatever that means!) took over, and I haven't had a chance to test out the recording delay test. But now that I'm back from exile, I'm pulling up my sleeves!

 

In case you don't remember me, I'm the guy with the Tascam (if you look on pg 3 of this thread, starting Wed May 28th).

 

You had mentioned to me:

 

Do your test between the RCA and 1/4" jacks. Use an RCA to 1/4" cable, two 1/4" cables with RCA adapters on one end, or dual RCA cables with 1/4" adapters on one end.

 

I looked over the specs of the unit (thanks for providing a link, BTW, very helpful Very Happy ) and I see that your inputs can be set to either line or guitar level. Set them to line level.

 

Next, I see there are gain pots on each input. You'll want to try and calibrate them so that the level of the signal you're playing out of Logic comes back in to Logic at the same level. (Post back if you want details).

 

I have realized that it seems that my Line Out has a gain pot, as well as my inputs, so this should be interesting!

 

However, I am having a hard time calibrating the levels as when I'm recording, there seems to be a weird display glitch occurring during the process:

 

 

http://i234.photobucket.com/albums/ee268/witkacy/Picture2.png

 

 

 

What's interesting is that the waves shown seem louder and more muddled than the original, but once I stop recording, the glitch fixes itself to show me how the soundwaves ACTUALLY look:

 

 

 

http://i234.photobucket.com/albums/ee268/witkacy/Picture3.png

 

 

 

 

I know this is off-topic, but is there any way to fix this?

 

On the plus side, at least I'm fairly close to having the levels match!

 

 

 

An addendum: So, it appears my lucky number was -177. Yikes! What does that say about my Mac or A/I? These numbers should be like a palm reading-- anything over 88 means "black clouds ahead"!

 

Should I expect some bad MIDI latency as well?

[amnestic]

Bump!

 

(An "jackass" move, I know. Forgive me.)

[ski]

Hey Amnestic and All,

 

I'm having a bout of "real life" myself and haven't been (and won't be) on the forum much for the next few weeks. [caution: shameless plug ahead!] Among other things I'm gearing up for a trip to LA where I'll be teaching classes on EXS-24 and Analog Synthesis as part of David's on-going series of workshops and master classes. I'll also be giving a talk at the Logic User Group meeting, so there's much to do.

 

As far as -177 samples goes, that simply means that your audio interface/driver combination is not reporting to Logic how much processing time it takes to do its thing, and so your audio is showing up late in Logic (I take it that the negative sign in front of 177 is what you plugged in to the recording delay parameter?)

 

As long as you plug in the correct value, all of your subsequent recordings should be aligned properly in Logic from now on.

 

As far as MIDI latency, it's hard to say. I don't know whether or not you should expect there to be a correlation between audio hardware/software latency and MIDI timing. That's a whole other subject that I've been meaning to address myself but just haven't had the time.

 

The glitchy waveform display -- I think that might be kinda normal.

[perversity]

Here's my report using an Apogee Duet, and the Artificial Audio Latency Bundle.

WIth PDC set to All and Off. I ran the 1/4 outs to the XLR ins.

[perversity]

and with the I/O plug, PDC off...