Cubase transposes guitar during recording (AxeFX II)


When I record with my guitar -> axeFX II -> Cubase 6, the recorded file sounds like it has been transposed down a semitone when compared to what I actually play and hear when recording in real time. I recorded the exact same way into garage band and it worked perfectly so I am assuming the problem is to due with cubase. I am using cubase 6.0.7. Can anyone help please?

Almost certainly a sample rate issue.

What sr is your project set to, and what sr is axeFX set to?

Thanks for your reply. You were right it was a sampling issue. I had to change cubase to 48.000 khz to match the axe Fx

You should be able to set the Axe FX to 44.1 as well.

Is there a reason why I would chose 44.1 over 48? Which one is preferable?

The only reason for using 48k is when working for video or film. Standard for audio (CD and most mp3, aac etc.) is 44k1, so better be at that sample rate already, so you don’t have to convert the sample rate, and risk artifacts.

Ok that makes sense. I looked into it and the axe fx sampling rate is fixed at 48 and can’t be changed. I guess I’ll just have to convert it afterwards

It runs at 48k, because 48k is professional standard.

So is 441k.

Mmh, very strange for an audio device, and has nothing to do with professional or non-professional. Depending on other audio interfaces, you may be better off going analog to that and use 44k1 coming into the computer. This could result in better sound, cause SRC from 48 to 44.1 might introduce artifacts.

It does appear that the Axe FX II is locked to 48KHz

Korg Kronos. Every Kurzweil keyboard I’ve ever owned. ADATs. Huge amounts of DATs, albeit for a different reason than the others…

You design an embedded system to run at it’s optimal sample rate, which sounds it’s best. You shouldn’t expect to change the sample rate to match some recording device. If you did, none of these devices would even be useful, since you really shouldn’t be recording at 44/48 anymore anyway.

I get a lot of 44.1 projects to mix. Thing that’s similar about each and every one? Not recorded by professional tracking engineer. 88.2 and 96k are industry standards…and have been for about a decade for anything outside of electronica/dance. ANything you’re using a mic for should be at one of those. Which, really doesn’t matter much to my ear…many like 88.2 because it allows full varispeed function in HD TDM systems that 96k doesn’t…Larvy seems to think it’s ideal because of low end loss that he says starts above 60khz…eh…tomato, tamahto to me…interchangeable. Couldn’t tell the difference to save my life.

…same does NOT apply for 44 to 48 or 48 to 88/96.

However…greater SR discussion aside…The two solutions here are simple. Run Cubase at 48k IF your interface analog connections are not up to snuff (and HUGE numbers of inexpensive interfaces are not). If they are, plug it in analog.

I have a Steinberg UR824 audio interface. Are you saying it would be better to run the axe fx into that and use a sampling rate of 88.2? Would that introduce more latency into recording?

At the same buffer setting it would actually reduce the latency, but it will stress the computer out much quicker as you add instruments and FX so you may have to increase the soundcards buffer setting thus increase latency of the system sooner than when using a lesser sampling rate.

I’m thinking that won’t be really be a problem when recording guitar as all the effects will be handled by the processor on the axe fx, but i’m using a mid 2009 macbook pro so it might struggle with other things. Which option would you recommend?

I’d recommend using whatever works for you. I personally almost always work with 44.1KHz except when I dont!!!

Set your soundcard to 48 kHz and record digital.

I don’t think it mattes much for the AxeFX itself. It’s a 48k digital box…88.2 won’t make IT sound any better than 48k…it depends on what the rest of your recording are.

Mic’d tracks? 88.2 and go in analog for the AF.

If the rest are virtual instruments…if sample based, they’re likely 44.1 samples, so they’ll use more CPU to set the project to anything BUT that…and not see any (significant) benefit above that. If the only sources are like a drum sampler and the rest is guitar/bass/vocal, 48k will be fine–drum sampler resample without too much pain, IME. Then, all you guitars and bass come in digitally…

Btw…if a 2009 MBP shouldn’t struggle–put an SSD (actually 2 if you use lots of VIs)…with one able to handle 80+tracks of 88.2 audio…and the other handling any disk streaming…you shouldn’t run into issues until it comes time to mix–and that’s what Freeze is for. :wink: