Hi, I have several dozen old real media (.rm) audio files that I'm planning to convert into the mp3 format with a bitrate of 128k. The real media files have a sample rate of 16 khz. I'm interested to know if it would make sense to resample to 44 khz when creating the mp3s, or would the 16 khz mp3 sound just as the 44khz one? Thanks for the help.
EDIT: I'm not asking if a 44 khz mp3 file would sound better than the source .rm file, I know that is impossible.
I can't think of any advantage.
Unless you can distinguish an audible difference, the best transcode will be the smallest one.
If this requires resampling to 44.1kHz, then so be it.
I'm interested to know if it would make sense to resample to 44 khz when creating the mp3s, or would the 16 khz mp3 sound just as the 44khz one?
Is there any theoretical reason why mp3 (the format or an encoder like Lame) would create
new issues e.g. at 8 kHz?
Anyway, compatibility could be an issue. Not every DAC is fond of 16 kHz sampling frequency. Mine refuses to play anything below 44.1 kHz - which I solve by resampling those files on-the-fly in my audio player of choice, foobar2000.
Mp3 will apply a low pass filter even at low sample frequencies. So if you encode an 16khz file the lowpass filter will be lower than 8khz which means you lose some "high" frequencies.
This is what i noticed while encoding 22.1khz files. So as a solution i used the -k switch to encode full frequency. I'm not sure if that's the best solution though. If someone knows a better way, i would be very interested.