As topic suggests, I’m wondering if it’s still necessary to convert samples to 48kHz for use with the ER-301. The legacy advice seems to have been to convert all samples to 48kHz before using them with the ER-301, but it seems like alot of/most/all? units don’t care anymore, although I haven’t thoroughly tested.
What consequence (if any) will it have (in which scenarios and with which units) if using 44kHz samples? Will pre-hearing work?
I seem have misread everything, here are a few links that led me to confusion (and I know they are not “stating” I should convert to firmware samplerate, but to my ears hints at it being the best solution):
I still have to do performance tests on the subject, but my gut feeling tells me the conversion-on-the–fly is almost negligible with regards to CPU, usage, could you comment on the gut feeling @odevices?
So to reiterate: will the ER-301 handle samples in any sample rate gracefully (audition on correct pitch and usable in all units) with a (small) performance hit when playing back (converting on the fly)?
Both of those links state exactly the same thing that I stated in this thread. All sample player units (except the Raw Player) are always resampling on-the-fly even if the encoded rate matches the firmware sample rate.
The resampling process is not negligible, compare the Variable Speed Player (2.8%) with the Raw Player (1.1%) in your measurements.
Yes. The ER-301 will resample all samples on-the-fly for you by default. The only way to prevent the ER-301 from resampling is to use the Raw Player (*). Generally you would use the Raw Player when you have samples already encoded in the same rate as the firmware rate AND you do not need to re-pitch in real-time.
(*) The Raw Player used to be called the Native Player (as in it plays samples at the native sample rate).