Best s/w to convert 44.1 to 48Khz

[I'd Rather Play]

To squeeze more performance out of my system I want to work in 44.1 and then convert to 48khz. What's the best software to do this?

[chris...]

To squeeze more performance out of my system I want to work in 44.1 and then convert to 48khz. What's the best software to do this?

What software do you have ? Most DAWs do this OK.

It may however end up being a load of faff for very little gain.

[ChrisCarter]

As Chris says... most current DAWs offer decent SR conversion as an option when rendering a mix down (to 44.1k, 48k, 96k etc.). But having said that... one of the best apps I've used to SR convert (mostly for batch conversion) is Isotope RX.

[G-Doubleyou]

Just work at 44.1 KHz not that big of difference in resources.

[I'd Rather Play]

I have to deliver in 48 khz. I read on some AV blog that 48khz uses 14% more resources than 44.1. Is this not true?

[dmills]

Rather depends on what you are doing...

I make it about 9% for things that are linear in samplerate, and roughly 18% for things that are quadratic in rate (Mainly some IIR type stuff), pretty much not worth worrying about, freezing a few tracks or increasing the buffer size will probably save more then that.

That said a single resample will not hurt anything, and is trivial to do, so do whatever works for your workflow.

Regards, Dan.

[Wonks]

It is always best to work in the final required format than increase the sample rate afterwards.

When increasing, the software has got to make stuff up that isn't there. It's a lot easier in software to accurately reduce sample rates and still keep as close to the original signal as possible than increase them.

[dmills]

When increasing, the software has got to make stuff up that isn't there. It's a lot easier in software to accurately reduce sample rates and still keep as close to the original signal as possible than increase them.

Ah, not really.

Both up and down by non integer rates are just a case of polyphase filtering, but the nice thing about going up in rate is that the potential for aliasing within the original passband is not there to anything like the same extent (The input is already band limited to within the passband defined by the lower sampling limit).

Going down you have the be VERY sure to have sufficiently attenuated anything in the original signal that falls outside the sampling limit for the new rate.

Now for video frame rates I might agree with you, but that is a totally different problem and visual analogies to audio sampling NEVER end well.

Regards, Dan.

[Mixedup]

As Chris says... most current DAWs offer decent SR conversion.

You can compare and contrast the SRC performance of different software at Infinite Wave.

You probably won't hear much difference — if it's a one-off most latest-version software is good enough — but some applications do a much better job than others of resampling, and some frighteningly recent versions are terrible. Eg check the difference in the Sweep, 1kHz Tone and Passband tests for Ableton Live 7 (awful) and Live 9.11 (rather better).