When people say preset standard has some bloat and extreme has less. what does this mean?

Bloat - generally - refers to the inefficient use of a given resource. In the case of lossy encoding, it usually means that (some) bits are spent in an inefficient way.

It's the other way around: --preset extreme may have more bloat. =)
edit: move -> more

This could be comparable to encoding two files with different encoders (say, Ogg vorbis with -q 6 and LAME -V 2 --vbr-new) which are roughly the same "target" bitrate ~ 192kbps. The mp3 came out to 268kbps while vorbis just 197kbps both may be transparent, just one uses these bits more efficently.

Well, well, well... this is something that should be proved. It could also mean that LAME made a better job to achieve the requested target quality (not bit rate) for a particullary difficult to encode piece.

Sergio

That could be true, this wasn't a difficult to encode sample, and vorbis certainly had many segments where the bitrate spiked above 400+ kbps, but since both these encoders are using VBR, the encoders always aim for a given quality (perceptual transparency) not bitrate, which is normal. One may have just had less difficulty encoding the sample. To sum it up: Bloat = Innefficent use of some bits that are available to encode a given audio sample.

But with MP3, "bloat" most often refers to the sfb21 problem which means that encoding >16kHz content leads to a massive bitrate increase. These bits are kind of "inefficiently" used because MP3 as a format is inefficient for encoding anything above 16kHz and can only do so by bloating the bitrate. Adding the -Y switch to the VBR modes avoids this problem, but leads to less high frequencies being encoded.

http://www.hydrogenaudio.org/forums/index....=0&#entry341703