Hello.

I am looking at an Ipod 4g Nano.I will be transcoding preset standard files to q 2.2 vorbis for roughly 96kbps files.

This will about double the storage of the unit. I know the files will not be transparant but I find this quite listenable with vorbis anyway.

Are ther any commandline tweaks specific to mp3 transcoding that should be applied?

There shouldn't be. Just encode as usual.

Edit:
Sorry people, I didn't read that it's a Rockbox iPod

not specific to transcoding from mp3... but I suggest giving it a try...

--advanced-encode-option lowpass_frequency=19

default lowpass for -q 2 is 16.5khz, the above command would raise it to 19khz, maybe you will like how it sounds.. maybe you won't

As I said, use the defaults. If you raise the lowpass value too much, you'll create artifacts, since the bitrate you are aiming for can't cope with such a freq. range. It's for portable use, anyways...

At 96kbps? A 19k lowpass is somewhat questionable in general since you're just encoding noise, but doing it at 96kbps is just silly. You need that bitrate elsewhere.

Thanks for your replies.

Although the transcodes have lost a bit of sparkle, they are still quite good for a portable. Raising the lowpass to 19 has created artifacts.Not enough bits to spread around I guess.

I tried this with Nero AAC and Itunes as well.It would have been nice not to have to use Rockbox as it is still a work in progress.

Don't get me wrong, Rockbox is a fantastic bit of software but the team has not had adequate time to create database manipulation,tag readers etc.But mp3 to aac transcodes are only marginally better than mp3 to mp3, which is to say they sound pretty bad. Vorbis is the only thing I have tried that works,so Rockbox it is.

I was thinking more in terms of raising the impulse trigger thresholds a bit, perhaps to match the short block speed of the mp3 format. Would that save me any bits?

Just as a general rule of thumb, don't play around with parameters. The defaults sound good for a reason : they have thousands of hours of tuning in them; put those hours to a test, and it'll be fine.

You should also be careful about mentioning specific "quality" problems without proving it (or mentioning clearly that it's your opinion), otherwise a new member may come along and think it's objectively true. (TOS #8)

I've just a got Nano with Rockbox installed, and as a Vorbis user for years (at q6). I've finally started looking at lowering the bitrate, thanks to the recent Multiformat test. When transcoding from -V2 (or -aps) MP3 I find q4 good but q3 is about the lowest I could go. Q2 seems a bit lifeless to me.

This was when using aoTuV beta 4.51 anyway

Why not just get an mp3 player with a little more storage, they are getting cheaper all the time. I personally would never transcode due to quality loss. Mp3 at APS are very good quality. Transcoding all your files will take up a great deal of time and effort. Go ahead if you just like to sit in front of a computer for hours on end, but doing that is a complete waste of time and effort in my opinion.

There's a couple of reasons why I transcode to Vorbis for my nano:

1. It's natively gapless
2. I can't tell the difference between a 128kbits Vorbis transcode and original -V2 MP3, while listening on the bus, meaning more songs.
3. It's a Nano - it's tiny and fits in my pocket with ease (unlike my Karma)
4. I like the feeling of using Vorbis on an iPod

Good point.

Ok. I have transcoded perhaps 150 files to ogg 96,ogg 128,nero and Itunes at 96 as well as lame abr 128 and wma 128.

In my opinion,without formal abx testing, only ogg produced a listenable result.Although it is easy to tell the ogg from the original mp3, it is still very listenable on my system.Ogg 128 was very good, but still too big for the library I want to put on the Nano.

Mp3 to mp3 with lame abr 128 was hopelessly "smeared" sounding, as was wma at 128.Highly objectionable.

Either aac implementation was better but still artifacted and objectionable on all the files checked using VBR at approx 96kbps.I was just switching back and forth on winamp.

The artifacts did not appear to require abx testing.

I will post samples if desired but the problem was evident on each of the files I checked.They were classic rock genre for the most part.

Perhaps I should say the perceived artifacts of ogg were the least objectionable to me as a personal preference.

and the hardware itslef is flash based, which IMO is a plus...

Especially when you drop it every day like I do.