How to make latency compensation work? OR - suggestions for a small interface that DOES work?

[Guest]

I'm using a Roland UA-3FX interface with Audition 5.5 on windows 7.

If I load in a drum loop on one track and then record some bass on another track (monitoring direct though my interface), the recorded bass is slightly but noticably delayed from the drums.

As I understand it, the ASIO driver and Audition, between them, should be working out any latency and compensating for it.Any settings I can fiddle with? Can't find anything.

I did try using the ASIO4All driver which does have adjustable latency compensation in its settings, but this either wasn't enough or didn't do anything at all (not sure which!)

If it's an issue with the driver for the UA-3FX, can anyone else recommend a good small interface which can record in time?it's a shame 'cos the UA-3FX actually sounds great.

Ivan

[Paul_Ferguson]

This means that overdubbing ought to be done in 3.01 rather than 5.5, yes?

[Guest]

therealdobro wrote:

This means that overdubbing ought to be done in 3.01 rather than 5.5, yes?

Sounds like it - if you can. Although the support guy I spoke to said he was getting the same in AA3, in contrast to Wild_Duck's results (although to be honest that conversation was slightly hard work, and I didn't pursue that point)

Unfortunately I don't have AA3 to try.  I did ask the support guy if there was any chance of getting it to try, but no dice. So I'm still stuck at the moment (as, if the issue is a win7/64, is anyone else on win7 who doesn't have 3 to drop back to, or wish to drop back to it)

[TheJackAttack-95Tmje]

I did actually run the tests on Win7 x64 with Au 3.01.  It worked just fine at all buffer settings with repetitions to verify results.  Notice, 3.01 so I have done the free update from Adobe's site.  Only CS5.5 exhibited the errant behavior.

[Guest]

Oh yes, thanks to JackAttack as well  (i ran out of 'helpful answer' clicks to give, sorry!)

[Guest]

If the "Adobe support guy" was using asio4all as the driver as he said he was, he wasn't actually testing a proper asio driver, just an overlay on the wdm driver (although my tests elsewhere indicate there is more to asio4all than that)..

I meant to remark on that earlier, especially as Adobe developers may be reading this. I would have thought that support should be checking on genuine asio before, possibly, testing with asio4all.

[SuiteSpot]

CEP 2000 runs fine here on Win7 x64 as have various other versions of CEP on other Win7 pcs

Been running a firewire interface for about 8 years without issue and we constantly are recording takes while monitoring previously recorded tracks without any hint of timing issues (other than the players themselves LOL)

[Bob_Howes]

Are the players in question commuting from Queensland?  I've been having the same problem for the past couple of days. 

Sorry...I'll return you to a serious discussion now.

[Paul_Ferguson]

topomorto wrote:

In that case, by lagging a step behind me, you're now a step ahead of me.  See, I can't run CEP 1.2 on Win 7.  Some plugins ditto.

          Do you have one of the versions of 7 (Professional / Ultimate / Enterprise) that has an XP mode? Any joy running it that way?

Way late - sorry - how did I miss this one?  No, I'm running Home Premium, so maybe that's why I hadn't heard of XP mode? 🙂   So now I'm going through the painful process of bringing old session in CEP into the present so that they'll archive better.

[TheJackAttack-95Tmje]

XP Mode is not all it's cracked up to be in my opinion.  One is better off using compatibility mode most of the time but for long outdated programs.

[Guest]

As far as I could tell, your original file was mono, and all the subsequent files stereo. So just to check, I generated a mono square wave and ran the test again, recording to stereo with exactly the same result - the sound returns approximately one sample before it plays.

If no-one else comes along, I could try the demo CS5.5 on another machine, but I'm reluctant to waste the trial instance before I really need it.

PS Steve's suggested utility says "any computer running XP". We are fighting Windows 7!

[Guest]

Wild_Duck, thanks. It didn't occur to me that there would be any latency difference stereo to mono - thanks for checking.

The thing is, in the simple use case where you are recording a second track next to a first, and monitoring direct (not through the software), it shouldn't matter what the actual latency is - what you record should be in time with what you were hearing when you played it... otherwise, you don't really have a 'multitrack', just a bunch of unrelated single tracks!

Hopefully a couple of other people can check this out, When I have a bit more time, I'll test on another machine, and maybe with some other software on this machine.

[Guest]

OK, so...  I am getting similar problems with Ableton Live, too. CS5.5 is off the hook for now!

Thanks for the help.  Wlll keep you updated!

[SteveG_AudioMasters_]

If you look at the ASIO4ALL FAQ, it gives you some information about how to feed info back to them, and also a link to a diagnostic tool. Maybe the way to go, and I'm sure that they'd appreciate the information anyway.

[Guest]

I've replied to this existing thread: http://mtippach.proboards.com/index.cgi?action=display&board=likelynotabugbutthereisaproblem&thread=...

In the meantime, time to look for the best workaround!

Best I came up with so far was to put a negative delay (31.92ms) on the recorded track, solo it, then bounce the master to a new track..

[Guest]

Pretty mild in Cambridge too... cold enough for the cat to moan about being thrown out, but it's quite sensitive!

It was recommended to me to try Cockos Reaper, and lo and behold...

    Absolutely spot on, to within a sample or two.

    That was with the native driver. The ASIO4All driver was still problematic, again with the sliders making no difference, So I'll write off my ASIO4all installation unless I can get it working in reaper.

[Guest]

Yes, Reaper is very good, and very good value. I use it a little and it's my son's main audio application. It has the advantage that, if you have bought Audition, you can import that as an editor when you need its features.

Here, as I said, both Reaper and Audition have no latency problem. If Reaper works well for you, that probably rules out a usb problem, but it does leave the question of what is causing the Audition latency issue.

If the Edirol asio drivers are working well, there is no point in trying to use asio4all unless you want to try using multiple audio interfaces (which probably won't be satisfactory anyway because of sync issues between the clocks).

[Guest]

It has the advantage that, if you have bought Audition, you can import that as an editor when you need its features.

        I suppose there would be some sense in that, and as you say, it's cheap - although that's extra ££ on top of what I paid for Audition...

  If Reaper works well for you, that probably rules out a usb problem, but it does leave the question of what is causing the Audition latency issue.

       Seems fair to say that : it looks like if CS5.5 did what Reaper does, it would be working!

       Might be time for a call to support! There is a uk number,  0207 365 0735. I don't know what people's experience of them is....

If the Edirol asio drivers are working well, there is no point in trying to use asio4all unless you want to try using multiple audio interfaces (which probably won't be satisfactory anyway because of sync issues between the clocks).

          I was only putting ASIO4All into the mix to test the theory that the roland native driver wasn't reporting its latencies correctly.... as it turns out, that wasn't helpful, but It's quite difficult to eliminate all the possible variables working on just one system!

          I was considering grabbing a firewire interface from ebay but I have a dreaded Ricoh chipset so not sure if that's just going to be a different kind of pain.

          I'll probably carry on with the Reaper trial for a bit and see if it's stable on days other than the Lord's day.

[SteveG_AudioMasters_]

topomorto wrote:

          I was considering grabbing a firewire interface from ebay but I have a dreaded Ricoh chipset so not sure if that's just going to be a different kind of pain.

Did try one PCMCIA card with a Ricoh chipset just as an experiment, and that actually worked fine with the MOTU. It's not what they recommend though, and I have no idea at all as to how this would work out with W7. I only use the MOTU with the laptop - the desktop DAW uses an E-Mu 1820m which has it's own PCI card.

[Guest]

Hello again all, hope you and any seagoing craft are fine.

Got the hookup cables for the presonus Inspire firewire thing. Plugged in and it worked fine straight off the bat.

This is the driver I'm using....

Anyway, back to my latency compensation issues (are you bored yet? I am)

With Auditon:

Less lag than before, but still about 14ms, which is still unusable for anything that needs to sound in time.

With Reaper:

Much better than Audition - that's about 2.2 ms.

This is with the interface latency set at 12.5ms, so compensation is happening,

But not quite the accuracy that Reaper managed with the Roland USB interface.

But Reaper allows you to Fiddle About :

With a total of 100 samples worth of compensation added:

Close enough for jazz.

So we've ruled out driver specific issues, and it's not USB, cos the same happens on firewire.

So, I will probably try to contact Adobe in the UK...

[_durin_]

We've got a bug written to investigate this further.  With the basic output->input latency test, I'm seeing much higher-than-expected latency values at almost all sample rates on Mac OS X.  I'll incorporate the notes from this thread into the bug (I've already put a link to it) and look for topomorto's related bug so we have all the information together.

Thanks!  Y'all are awesome.

[Guest]

...and look for topomorto's related bug so we have all the information together.

Durin,

Your reply came up as I was talking to your support engineer Yatharth about my case (number 0182860328, if you haven't already found it ).

He was good enough to do the same test as us guys here have been doing and.... his win 7 x64 system also had similar problems with CS5.5 (he was getting 25ms uncompensated lag - he was using ASIO4All with his internal interface, and he also had the same result as me with ASIO4all insofar as juggling the 'latency compensation' sliders did nothing on his system - his lag remained constant).

Any more information I can give you? let me know ~

The only thing that I haven't mentioned in this thread so far is that I tried the same test on a Vista system, and was getting the same problems. 

TheJackAttack wrote:

Now however I'll be playing around in the preferences and menus looking for any clue though since this is built in compensation I doubt I'll find anything.

The support chap couldn't find any other relevant settings either. I was hoping he'd give me some secret "put this value in this xml file in this directory and say 3 hail marys" technique, but... no.

[_durin_]

There aren't any user-facing parameters for latency-compensation.  I'm seeing similar results on my Win7x64 system as well, and the bug has been assigned to a developer for additional research and fixing.

I'll let you know what we find, and if I can offer any advice for now other than manual adjustment.

[Guest]

_durin_ wrote:

There aren't any user-facing parameters for latency-compensation.  I'm seeing similar results on my Win7x64 system as well, and the bug has been assigned to a developer for additional research and fixing.

Thanks

... other than manual adjustment.

Is there a quick shortcut (or a way to make one) that would shift a clip by a number of samples (or milliseconds)?

At the moment all I have is :

-  zooming into the appropriate resolution, doing maths in my head, dragging a clip back in time to where it should be, and then zooming back out to see my arrangement

or

- putting a negative delay on the track, bouncing it, and then deleting the original track, and then moving the bounced track up to where the original track was in the first place...

Having to do either of these each time I record a clip cuts into the creative flow (and available time) so much that it's not really possible to use the program ATM.

[TheJackAttack-95Tmje]

At least on my system, all you would have to do is move the new clip equal to the sample buffer size of your interface. 

[Guest]

TheJackAttack wrote:

At least on my system, all you would have to do is move the new clip equal to the sample buffer size of your interface.

Sure, but unless there's a faster way than than the two methods I mentioned in my previous post, moving a clip by a number of samples - even to the nearest millisecond - takes a chunk of time to do, and isn't the sort of thing you can be doing again and again each time you record a clip when you're trying to put together a piece of music (not to mention being error prone - you have to remember to do it every time, and if you're moving it by hand, you have to get your floating-point subtraction right every time). I tend to build pieces up from short passages of a few bars in length, switching between instruments as I go, so this problem really prevents that way of working.

If there was a way to set things up so that a keyboard shortcut would do a 'move forward by x samples', that could be a reasonable workaround.