GH6/GH5/BH .NOTE File Format

[raynebc]

Out of a set of 4 files I have, that and "No More Mr. Nice Guy" have 36 byte vox words entries. "Only Happy When it Rains" defines a fretbar size of 0 bytes (I don't know why it does this). "Dancin' with Myself" defines a vox note size of 256 bytes.

I'll be examine a document that describes the GHWT format next. Are there any other GH format variants I need to look into?

[raynebc]

Out of a set of 4 files I have, that and "No More Mr. Nice Guy" have 36 byte vox words entries. "Only Happy When it Rains" defines a fretbar size of 0 bytes (I don't know why it does this). "Dancin' with Myself" defines a vox note size of 256 bytes.

I'll be examine a document that describes the GHWT format next. Are there any other GH format variants I need to look into?

[codemann8]

[codemann8]

[raynebc]

I think I've noticed something: In "Only Happy When it Rains", where it is supposed to define the size and it defines 0 instead, the game probably assumes the default data size and considers 0 the first data value.

Ie. where the fretbar size is expected, and 0 is encountered, 0 is the first fretbar position. Likewise, when a timesig size of 0 is encountered, that's actually the first time signature position, and the following two bytes 0x0404 definitely would indicate the numerator and denominator of the time signature.

Edit: This chart does the same thing for notes, defining the first position where the entry size was expected. I'm not sure if there's some other parameter in the file that is supposed to indicate that the sizes are not defined, but perhaps there is.

[raynebc]

I think I've noticed something: In "Only Happy When it Rains", where it is supposed to define the size and it defines 0 instead, the game probably assumes the default data size and considers 0 the first data value.

Ie. where the fretbar size is expected, and 0 is encountered, 0 is the first fretbar position. Likewise, when a timesig size of 0 is encountered, that's actually the first time signature position, and the following two bytes 0x0404 definitely would indicate the numerator and denominator of the time signature.

Edit: This chart does the same thing for notes, defining the first position where the entry size was expected. I'm not sure if there's some other parameter in the file that is supposed to indicate that the sizes are not defined, but perhaps there is.

[codemann8]

Oh yeah, of course the first values are 0...I thought you meant that the size of the data was 0 bytes, which I found strange, but that's not the case I see

And yes, the timesigs and notes are exactly as I described in the OP
_________________

Frets and Fireworks GH:A Expert Tourney Winner
AAA-2 Season 5 Overall Points Champ & Playoff Runner-Up!!!
My FCs

[codemann8]

Oh yeah, of course the first values are 0...I thought you meant that the size of the data was 0 bytes, which I found strange, but that's not the case I see

And yes, the timesigs and notes are exactly as I described in the OP
_________________

Frets and Fireworks GH:A Expert Tourney Winner
AAA-2 Season 5 Overall Points Champ & Playoff Runner-Up!!!
My FCs

[raynebc]

No, I'm saying that the field that's supposed to indicate the size of the fretbar and the size of the timesig is given as 0. They're implementing a shortcut where the 4 bytes defining the size is omitted, and they go straight to defining the first 4 byte timestamp instead.

On a different issue, I've been asked if Guitar Hero charts have notation for double bass. Could it be involved with the "extra note" section? Perhaps the extra (unknown) byte defines whether the note is marked to require a hard velocity hit, or is double bass? Wikipedia's GHWT page indicates this game has "armored" notes that require a threshold hit force to be met to get the hit. Likewise, the wiki page for GH:Metallica indicates that the chart keeps track of double bass.

[raynebc]

No, I'm saying that the field that's supposed to indicate the size of the fretbar and the size of the timesig is given as 0. They're implementing a shortcut where the 4 bytes defining the size is omitted, and they go straight to defining the first 4 byte timestamp instead.

On a different issue, I've been asked if Guitar Hero charts have notation for double bass. Could it be involved with the "extra note" section? Perhaps the extra (unknown) byte defines whether the note is marked to require a hard velocity hit, or is double bass? Wikipedia's GHWT page indicates this game has "armored" notes that require a threshold hit force to be met to get the hit. Likewise, the wiki page for GH:Metallica indicates that the chart keeps track of double bass.

[codemann8]

[codemann8]

[raynebc]

[raynebc]

[Ring777]

what HEX editor are you guys using, i want to take a look at this.
_________________

Nothingness.

[Ring777]

what HEX editor are you guys using, i want to take a look at this.
_________________

Nothingness.

[raynebc]

There are probably ones with more functionality, but I use XVI32.

[raynebc]

There are probably ones with more functionality, but I use XVI32.

[codemann8]

010 Editor is the best! It's one I didn't mind paying for at all. Has very rich tools built-in and the ability to add on custom features. Can always try it out for 30 days first.
_________________

Frets and Fireworks GH:A Expert Tourney Winner
AAA-2 Season 5 Overall Points Champ & Playoff Runner-Up!!!
My FCs

[codemann8]

010 Editor is the best! It's one I didn't mind paying for at all. Has very rich tools built-in and the ability to add on custom features. Can always try it out for 30 days first.
_________________

Frets and Fireworks GH:A Expert Tourney Winner
AAA-2 Season 5 Overall Points Champ & Playoff Runner-Up!!!
My FCs

[LocalH]

Hex Workshop is also good, albeit commercial with a trial period.
_________________

MiloHax 2.0 Discord server - Everything GH/RB (RIP MiloHax 1.0 01/30/2017 - 04/04/2022)
O211: An Unofficial Scorehero Alumni Discord


Alakaiser sez: POST BECAUSE YOU HAVE SOMETHING TO SAY, NOT BECAUSE YOU HAVE TO SAY SOMETHING.

[LocalH]

Hex Workshop is also good, albeit commercial with a trial period.
_________________

MiloHax 2.0 Discord server - Everything GH/RB (RIP MiloHax 1.0 01/30/2017 - 04/04/2022)
O211: An Unofficial Scorehero Alumni Discord


Alakaiser sez: POST BECAUSE YOU HAVE SOMETHING TO SAY, NOT BECAUSE YOU HAVE TO SAY SOMETHING.

[raynebc]

The binary templates feature of 010 looks pretty useful, especially for working on dissecting a file format, but I don't think I'm willing to pay $50 for it.

[raynebc]

The binary templates feature of 010 looks pretty useful, especially for working on dissecting a file format, but I don't think I'm willing to pay $50 for it.

[raynebc]

Is there a comprehensive guide to the GHWT format as well? I found some information (gh_wt_midqb.html), but it doesn't seem to give complete header information, checksums, etc.

[raynebc]

Is there a comprehensive guide to the GHWT format as well? I found some information (gh_wt_midqb.html), but it doesn't seem to give complete header information, checksums, etc.

[codemann8]

[codemann8]

[raynebc]

I have to say that I hate that they put the song name into the un-hashed section names, making each checksum song specified. Looking at multiple .pak.ngc files though, the on-disc file path for the file appears to be included near the beginning of the file (ie. data/songs/mrcrowley.mid.qb.ngc"). Can I use this information to avoid having to look up the song name, common ID or other information from a .dbg file (ie. maybe this will allow me to expect having to look for the checksum for "mrcrowley_timesig")? If all information can be reliably parsed from just the QB file itself, I'd prefer it.

[raynebc]

I have to say that I hate that they put the song name into the un-hashed section names, making each checksum song specified. Looking at multiple .pak.ngc files though, the on-disc file path for the file appears to be included near the beginning of the file (ie. data/songs/mrcrowley.mid.qb.ngc"). Can I use this information to avoid having to look up the song name, common ID or other information from a .dbg file (ie. maybe this will allow me to expect having to look for the checksum for "mrcrowley_timesig")? If all information can be reliably parsed from just the QB file itself, I'd prefer it.