vocal line differs on certain songs

[Mr. Tate]

Eh? I can't even begin to fathom how that would even begin to work since RB3 has all vocal parts merged into a single track as far as the audio goes, so how the game would even know when to be lenient in such a case is beyond me.

No it hasn't. RB3 songs have all the vocals parts in a single audio track when it comes to multitrack separation, but when it comes to authoring all harmonies have a separate dry vocal track, which is used for what I posted. Now, it's not that I think there's leniency, that's what the docs say and how it works. The implementation may be different with older songs, maybe not all songs have all the dry vocal tracks, but the scoring does get loose when there's difference between the dry vocal track and the MIDI notes. According not to me but to the documentation, I want to stress this because it looks like I'm submitting an idea I have.

[Mr. Tate]

HMX could be including metadata on such songs to notify the game engine to widen this margin, either for the entire song, or for only the sections where the singer is off key.

That's actually taken care by another switch for the whole song, not by the dry vocals comparing system.

[bclewis]

I wonder whether this "scoring loosening" is reflected visually by a change in the line width, or whether it's internal only. In any case, I think when people notice changes in the width, it's generally due to the vocal range scaling as discussed earlier.

[MofoMan2000]

I don't know whether you're talking about how strictly you must follow the pitch, or how thin the actual line seems to be.

Pitch strictness is defined by difficulty. On easy or medium, you can be several notes away from the target and still get "Awesome" on the phrase. On expert, however, if you are even a half-step away from the target pitch, you will miss it. Believe me, I've been singing on expert for years. I sang on medium to get the Bladder of Steel Award, because it was incredibly easy to do, and I barely put any effort into it.

How thin the guide line seems to be is defined by both difficulty and the vocal range of the song. Songs like Gimme Shelter have rather thin lines since it spans a few octaves. Songs like Song of the Century are largely in the same octave and the lines will be bigger. I think the thickness also varies with difficulty, in that easier difficulties give wider lines. I might be wrong about that though.

[bclewis]

How thin the guide line seems to be is defined by both difficulty and the vocal range of the song. Songs like Gimme Shelter have rather thin lines since it spans a few octaves. Songs like Song of the Century are largely in the same octave and the lines will be bigger. I think the thickness also varies with difficulty, in that easier difficulties give wider lines. I might be wrong about that though.

Yes, I think we've established these two points pretty well in earlier posts in this thread; Mr. Tate brought up a possible additional factor, which is that for some songs where the singer is often a little bit off key, the "pitch strictness" may be slightly loosened.

[MofoMan2000]

Mr. Tate brought up a possible additional factor, which is that for some songs where the singer is often a little bit off key, the "pitch strictness" may be slightly loosened.

I very highly doubt this. Try singing E-Pro by Beck, or Peace Sells (But Who's Buying?) by Megadeth. Either the charts are wrong (which I feel is likely) or they sing off-key. The workarounds for this have been to simply flub the chart so the singing is on-key, or to chart off-key parts as talkies. Both are acceptable, neither adjust the strictness of pitch beyond removing pitch altogether.

[bclewis]

I very highly doubt this.

Okay, but apparently (according to Mr. Tate), HMX documentation claims that they do. You're welcome to read the rest of this thread for details.

[Icemage]

No it hasn't. RB3 songs have all the vocals parts in a single audio track when it comes to multitrack separation, but when it comes to authoring all harmonies have a separate dry vocal track, which is used for what I posted. Now, it's not that I think there's leniency, that's what the docs say and how it works. The implementation may be different with older songs, maybe not all songs have all the dry vocal tracks, but the scoring does get loose when there's difference between the dry vocal track and the MIDI notes. According not to me but to the documentation, I want to stress this because it looks like I'm submitting an idea I have.

That's as may be, but if there's any sort of leniency (even in the dry vocals), it's so subtle that I don't think I've ever noticed it on any song, ever. At the very least, I don't ever remember playing a song with off-kilter notes where I didn't break combo when I expected to.

[Mr. Tate]

That's as may be, but if there's any sort of leniency (even in the dry vocals), it's so subtle that I don't think I've ever noticed it on any song, ever. At the very least, I don't ever remember playing a song with off-kilter notes where I didn't break combo when I expected to.

I have no idea about how loose the score becomes, I actually found out about this not long ago. Still, I wanted to share the information for the discussion.

[Mr. Tate]

for some songs where the singer is often a little bit off key, the "pitch strictness" may be slightly loosened.

I don't think that's necessarily only because of that. In example, you may have a live version with the crowd singing. If the singalong makes for a slightly different pitch, I suppose the comparison helps.