QUOTE(eofor @ May 29 2007, 16:50)
I have a Nokia phone (7710, Symbian S90) with Ogg playing software (OggPlay). I initially loved the fact that I could get twice as much music on the memory card by using Ogg @ 64kbps. The experience compared to mp3 however was not good at all, cpu loads are so high that the phone becomes unresponsive and the battery drains really, really quickly. Unless there's an efficient Ogg codec for these mobile platforms (like you can license from the big boys cheaply for WMA/MP3/AAC), phone manufacturers will not push Ogg, because nobody wants to advertise a music phone that lasts less than 4 hours playing the format.
This actually mirrors the situation of Rockbox consuming power far greater than iPod's standard firmware.
Someone said that it's not the codec's fault, it's the player's 'fault', i.e. the player's maker hasn't figured out how to turn of nonessential things in the hardware.
QUOTE(eofor @ May 29 2007, 16:50)
Also, encoding (using standard WinAmp, which uses an up-to-date Lancer build) is painfully slow compared to LAME. That's not meant as a attack on OGG as a format, but more as a reminder that building a great psymodel, defining robust specs and having impeccable "free software" credentials is not enough - for it to be a success as a commercial product, you need to provide OEMs with easy and fast implementations.
For me, I don't find Lancer to be far slower than LAME. aoTuV's standard build *is* a heck slower than LAME, but not Lancer.
Anyways, if the average user is given the option of:
- Fast encoding, but less music
- Slower encoding, but more music
I believe they will opt for the latter.
QUOTE(eofor @ May 29 2007, 16:50)
Microsoft understands this - even when WMA is even less popular than OGG and nobody likes MS to begin with, they provide such easy to implement and efficient decoders and simple, no-hassle licenses for OEMs that even every two-bit Chinese backstreet workshop implements it.
Ogg Vorbis has a simpler no-hassle license: It's public domain.
