I've been wondering something. Is there a way to tell EOF where to position the .wav/ogg files?

 

It seems like this is random. But shouldn't importing position the waveform to start exactly on the very first beat marker?

 

This way, we'd be able to estimate BPM, then adjust the beat-map AND THEN add the leading silence (using beats, not milliseconds).

 

One difficulty I see is that EOF seems incapable of removing the brief 'nose' of silence that gets added onto mp3s (and oggs too) when they're created.

 

Wav files don't have this issue - for one song, I converted the mp3 to .wav, cleaned up the silence 'nose' so that it starts exactly at the first note/sound. Then I converted it to .ogg and imported that file directly into EOF. But EOF still showed that bit of silence.

 

Granted, the silence lasts for all of 50 ms, which probably isn't enough to throw things off.

 

Maybe I'm missing something here?

 

If I import an mp3 as is and add the 3000 ms as per the manual, then i have to drag the closest beat marker to match the waveform -- which throws off the lead-in beatmap. Maybe that's not an issue?

 

I've been wondering something. Is there a way to tell EOF where to position the .wav/ogg files?

The way EOF is designed, it has to store the audio files in the project folder. Allowing it to store files elsewhere would make it too easy for file paths to be invalidated, and it would require the path to be stored in the project file, something the original EOF author wanted to avoid in case people share project files to work on together. 

It seems like this is random. But shouldn't importing position the waveform to start exactly on the very first beat marker?

Importing what? Loading audio doesn't cause the beat positions to change unless it's the MIDI delay value tied to another OGG file you loaded previously and gave a different first beat position. 

This way, we'd be able to estimate BPM, then adjust the beat-map AND THEN add the leading silence (using beats, not milliseconds).

The author needs to set the position and length/tempo of the first beat before using the leading silence to add a specific number of beats worth of silence. Having EOF guess all of this automatically would probably lead to mediocre results. 

One difficulty I see is that EOF seems incapable of removing the brief 'nose' of silence that gets added onto mp3s (and oggs too) when they're created. Wav files don't have this issue - for one song, I converted the mp3 to .wav, cleaned up the silence 'nose' so that it starts exactly at the first note/sound. Then I converted it to .ogg and imported that file directly into EOF. But EOF still showed that bit of silence. Granted, the silence lasts for all of 50 ms, which probably isn't enough to throw things off.

In the past, it seemed like some kind of distortion being added to the chart audio and this was a bug with either LAME or oggenc. Is this distortion first being introduced when you select an MP3 file when creating a new chart, or when you add leading silence to a chart? 

If I import an mp3 as is and add the 3000 ms as per the manual, then i have to drag the closest beat marker to match the waveform -- which throws off the lead-in beatmap. Maybe that's not an issue?

If you add leading silence, the default option is to move the first beat marker position accordingly. The beat positions aren't automatically synced to the music, although you could try having Sonic Visualiser attempt to do so (See the manual description of the "File>Sonic Visualiser Import" function for more details). Otherwise I'll need more information about what behavior you're expecting.

When I say 'position' I mean within EOF -- not where the song is located on the hard drive.

 

Basically I'm wondering if it's possible to move the waveform within the chart.  So, instead of it starting at a position corresponding to beat 1:4:22, for example, I could reposition the wave to start precisely at 2:0:0.

 

I'm coming from the audio multitracking world, where this kind of thing is possible. I think that's where my confusion comes in.

 

In an audio program, the grid (beats/measures) is fixed and one moves the audio accordingly.

 

But in EOF, the AUDIO is fixed -- the grid floats on top of this and gets regenerated to match the audio.

 

 

From my experiments yesterday, it looks like all I have to do is reposition the first beat market to the begininng of the audio waveform and it redraws the grid (adds in new grid lines).

 

As for the 'nose' I'm speaking about: this is a problem with mp3 technology, I think. Converting from .wav to .mp3 adds that bit of silence. It looks like the ogg does too.

 

Since EOF converts these files back to .wavs anyway, is it possible to load a .wav directly into EOF without the mp3/ogg conversion? I think the audio will end up sounding better anyway.

EOF can't determine how much leading silence to add to cause the audio to line up a specific way, but if you determine how many milliseconds are needed, you can add that much leading silence.

 

As far as the chart audio, EOF requires the use of an OGG file internally. It's been this way since it was first created and changing it now would probably cause too many problems. If you want though, you can create the WAV file yourself outside of EOF and set the preference for EOF to not create a Rocksmith WAV file. Just beware that if you add leading silence to the chart in EOF, you'll have to manually replicate such changes to the WAV file because EOF won't be altering files external to the project.

Marty-  I'v been adding leading silence in Audacity before I even start the EOF project.  I export as an .ogg to use in EOF and a .wav to bring into Wwise and then cut it down to a 30 second clip with a fade in and fade out to use for the preview.  It seems to be the easiest way to do it to me.