I am monitoring direct through the Tascam. I run my Fractal Audio Axe FxII into an XLR input. I am not hearing echo or latency of what i am performing while recording as much when I play back. Sometimes when I am doing the second or another track, though I am in time with the drum track, I can hear I am slightly out of synch with the other guitar.


My drummer has a 6i6 that he uses for triggering his drums and running Steven Slate. He let me borrow it tonight but after staying up late last night on this, sleeping about 2 hours, working today and just getting home from the evenings events I am beat. I am going to try it out in the morning. Any one have an idea of how I could post some examples?

I'm glad I read this thread because of what cowboydan posted I just went to the Tascam support page and well .well they have finally updated the drivers, last update was 2011 so I never even bother looking anymore.. So this might just breath so new life into the old girl. Not overly optimistic that it's going to be earthshaking improvement but nice to be able to still carry on with Windows 8 support. 
Thank you cowpiedan 

If you're hearing delay during playback, you can set record latency adjustment and forget about it. 
 
Go to Preferences>Audio>Sync and Caching.  Uncheck "Use reported latency" and set Manual Offset to zero.
Set up a click track or something ... patch an output back to an input and record.
 
Zoom in on the tracks ... you'll be able to see exactly how much RT latency you are experiencing by setting the timeline to show samples.
 
Go back to prefs and set Manual Offset to this value.
 
Now you can monitor direct and have your recorded tracks automagically adjusted to account for actual latency.  You should no longer hear the delay (other than timing of your actual playing) when you play it back.
 
 

+1 I was going to recommend testing your offset. Sonar and the drivers do this automatically and normally it's never an issue. 
So testing the function is one way to rule that out as being the issue. 
 
Just a small change to the above method. 
 
I would not change any setting first to see if the offset is being calculated correctly. This will change if you change audio buffer settings so it's best to leave it on auto if you can.
 
Use a short patch cable and connect an output back to an input and re record any transient audio track. The more visual the transient spikes the better. If you don't have one, make a audio track by bouncing a midi drum track. 
Record to a new audio track next to the original.
To check for clock drift you might want a whole 3 minutes.
Zoom in on the 2 tracks and see if there is any offset showing.
Right now your using your ears. That is flawed. You need to test the system first.
 
If the tracks line up all is good and you need to look elsewhere for what your hearing.  
 
 

SO I finally had the free time to test this today. I created a synth track w/AD2 set the tempo for 85bpm (dunno why I chose that tempo) and used the ride bell on every 8th note. I did about 5 minutes of audio and patched it into an input on my Tascam  recorded it. I then froze the synth track, zoomed in and viewed the timeline in samples. I checked the 2nd hit, the last hit and 2 in between. This is what I found:
1st check-sample of synth-16942
               sample of wave-19065
                        difference-2123
 
2nd check-sample of synth-6285162
               sample of wave-6287297
                        difference-2135
 
3rd check-sample of synth-11672469
               sample of wave-11674576
                        difference-2107
 
4th check-sample of synth-14620228
               sample of wave-14622339
                        difference-2111
 
I panned the frozen synth track & the recorded wave track L/R and I can hear a difference. I do not know what to do with this data, I see that the average difference is 2119, but how do I apply this? Thank you all for your help with this!

Sorry - I just noticed you posted back.  I'll do some math and post tomorrow
 
What sample rate are you running?

Thank you for your help. I have recently started doing my projects in 48k so they'll be compatible with my Fractal Audio Axe Fx II.

OK, you measured 2119 (avg) sample latency.  At 48K, that's about 44.14ms.  Hearable for sure.
At the bottom of this dialog, is the total roundtrip number in agreement with your measurement?

 
At the bottom of this dialog, is the previous number listed and "Use ASIO Reported Latency" checked?  If the numbers agree, but the box isn't checked, then checking the box should set you up.  If the numbers don't agree, you leave the box unchecked and enter 2119 in the manual offset box.  NOTE:  The Device dropdown selection never stays at what I select.  It doesn't seem to matter.

 
The downside to manual offset is that changing your driver settings will require you to change this number accordingly.  It's best if you're able to use Reported Latency.
 
I hope this makes sense  :)

Now I have my driver set as WDM, for some reason it seems ASIO doesnt work with the Tascam. I will check when I get home tonight. Thank you for all of your help.

ftf613
Now I have my driver set as WDM, for some reason it seems ASIO doesnt work with the Tascam. I will check when I get home tonight. Thank you for all of your help.

You really should concentrate on getting your interface to use its ASIO driver.
 
Have you installed the latest Driver V2.05 for Windows from Dec. 26 ?
 
What shows up in Sonar Preferences with respect to this driver ?
Can you supply some screenshots ?