I'm encountering a problem. When I record something along to the rest of a project in Sonar X3 the audio is recorded a second or so behind the rest of the track. I checked to see if it was my playing, but it's steady and consistent throughout - the audio is recorded and locked into the project at a point in time behind where it should be.

It happens sometimes and I've played around with buffer settings to try and resolve it but I don't really know what I'm doing, or why this happens in the first place. It has happened when I'm recording 9 mics at once on a drum kit, but it's doing it now for just two mics. What is going on?
 
I know it is possible to grab all the tracks and manually slide them along to get them in time, but I don't like that and I don't want to do it because you're trying to figure out how you played rather than listening back and hearing exactly what you played. I don't like the end result of that.
 

Interface? ASIO drivers?
OS?
Settings?

See that bottom setting, "Use ASIO reported Latency"
That's the "jewel" right there.
SONAR takes care of it, if you're running WDM, you have to set that number; different for every buffer setting.
I don't have a laptop, if you're using the new/improved WASPI driver, someone else will have to help you.
Need more info!
T

In addition to enabling Use ASIO Reported Latency, you will usually need a small Manual Offset to compensate for unreported latency, but that's generally less than 50 samples. You can calculate it by measuring the actual roundtrip latency using the free CEntrance Latency Tester, and subtracting the Total Roundtrip reported on the Driver Settings page of Audio preferences.
 
     http://www.centrance.com/downloads/ltu/
 
But if the compensation error is truely on the order of a full second, something else is going on. There have been cases in the past where users reported that having metronome count-in enabled was causing audio to be laid down a measure late. I never saw that issue personally, but if you're still on X3, that could be it. Do you have the 'e' update installed? Try disabling the metronome count-in in any case as a test.

DeeringAmps
Interface? ASIO drivers?
OS?
Settings?

 
Sorry - I put all my system spec in my signature and forgot to set my profile so it always shows on my posts. Hopefully sorted now.
 
Intel i7-4710HQ - CPU @ 2.50Ghz
16 GB RAM
Windows 8
Sonar X3 Studio
RME Fireface UC
 
I switched from ASIO to WDM a while back on a recommendation. I keep getting conflicting opinions about which is better for running SONAR on my system.
 
 
 
 
 

ASIO; RME are the drivers by which all others are judged.
You can do the "loop back" test, it will be spot on; sample perfect.
Literally I never have a "sync" issue with either the UFX or the Babyface.
Oh, and did I mention; RME, ASIO, the BEST!
Make sure "Use reported Latency" is checked.
If you're using WDM, that's you issue right there.
T

Caught myself in a LIE! LIAR! pants on fire!

SONAR reports 214 sample offset

as you can see at 64 samples SONAR reports 246
CE nailed it at 247, still 1 sample off.
I've always done the loop back and then "null" tested the two audio files moving 1 to find the "best" null.
Still I am ashamed, 1 sample; I lied! Can you ever trust my judgment again?
Close enough for f'ing rock'n'roll.
Don't I wish I could play to 1 sample accuracy
T
Of course this a$$umes CE has it "right"

Gideon K
 
I switched from ASIO to WDM a while back on a recommendation. I keep getting conflicting opinions about which is better for running SONAR on my system.
 

Reason could be this quote from the manual:
 
The program Sonar from Cakewalk is unique in many ways. Sonar uses the so called WDM
Kernel Streaming, bypassing the WDM mixer, thus achieves a similar performance to ASIO.
[page 20 - UFX manual]
 
But this should be read in context (same page). RME recommends ASIO.
 
 

DeeringAmps
RME are the drivers by which all others are judged.

Oh yeah? Judge this, Babyface! 
 

 
 
 
 
 
 
 

WDM mode will always be off sync. ASIO from any decent brand should be close enough for rock and roll. 
Screenshot of loopback test. You want real bad results just use MME mode! 

16 samples? NOWAY!
What does SONAR report on the latency?
How much "load" til we have "rice crispies"?
You know snap, crackle, pop!
T