Sample Rate Conversion 96kHz --> 44.1kHz Artifacts

[NystagmusE]

http://src.infinitewave.ca/

This is highly technical kind of comparison test and I don't pretend to understand it completely, but it is interesting. The webpage offers a technical comparison of the effects of different types of Sample Rate Conversion algorithms on control test signals.

Of interest are the Weiss Saracon 1.0 and the iZotope RX Advanced 1.05 (High Steepness) sample rate conversion algorithms.

I wish I could hear what those sound like.

Also, it looks like GoldWave or Wavosaur are not good choices for sample rate conversion.

Giel, what is your opinion on all of this?

[sinbad]

It looks like the R8brain freebie from Voxengo is among the best

[NystagmusE]

Yeah that is ironic about the r8brain. :)

[evon]

What about Sound Forge's Converter?

[NystagmusE]

From what I could tell from one of the charts, SoundForge has a lot of aliasing noise during the frequency sweep. It's not a good thing. It's essentially tonal noises added to the original sound content...frequencies that aren't supposed to be there.

[evon]

But what about the antialiasing features?

[NystagmusE]

Even the anti-aliasing converter for SoundForge 9.0 Anti-Aliasing seems to have quite a bit of aliasing according to the chart on that test website.

It does have less aliasing than the SoundForge 9.0 Non-Anti-Aliasing converter, but not less than a converter such as the Weiss Saracon 1.0

http://src.infinitewave.ca/help.html <-- this explains a little bit

[evon]

Thanks for the link, this explains a little more indepth than the original link at the top of the post. I am not able to read all of it now as I am currently at work.