[MUL] Shovel Knight - "The Schemer (Tinker Knight Battle)" by Dekkadeci

Started by Zeta, June 11, 2016, 04:49:12 PM

Previous topic - Next topic

Zeta

Submission Information:

Series: Other
Game: Shovel Knight
Console: Multiplatform
Title: The Schemer (Tinker Knight Battle)
Instrumentation Solo Piano
Arranger: Dekkadeci

[attachment deleted by admin]

Dekkadeci


Watch NSFPlay play this song:
(OK, you'll probably have to skip to 0:57:36 for "The Schemer")

Jake Kaufman's complete Shovel Knight soundtrack source code is here: https://dl.dropboxusercontent.com/u/56073318/Famitracker/Other%20Songs/Shovel_Knight_Music.nsf

Yup, I took the .nsf at Song 23, transcribed several channels' notes into Musescore, and omitted some other channels' notes. (IMO, those other channels' notes are rather quiet and make the song require two pianists to play.)

...Ugh, I hate how features are lost when I export the Musescore MusicXML and try to read it with Finale Notepad, even after editing the MusicXML directly with Notepad++ to fake it as a Finale Notepad original (instead of a Musescore original). It took out the parentheses in the highest left hand B flat in Bar 18, and it also took out the trill on the right-hand C in Bar 8. (The trill has been notated with 32nd notes--they're not exactly what the composer put down in the NSF, but there's no way I'm notating quintuplet-based notes (3:2 ratio between notes) for the trill.)

Speaking of parentheses, transcribing Famitracker channels' notes raw results in tons of ugly hand collisions. I'm not sure how to solve them. Granted, the right-hand octaves are very demanding to play, so perhaps discarding all their bottom notes may be the best solution. What do you all think?

The .mid is exported from Musescore - can anyone help me quiet down the upper clef octaves (or whatever they end up becoming) in Bars 21-36 in the .mus version, just like the .mus and .pdf direct in the text? Right now, all I can hear in the .mus version are the right hand notes (except when they're not the highest notes played).

And yup, those are left hand crossovers in Bars 24 and 32.

Latios212

Whoa there. This sounds nice but is quite literally impossible to play as written. "Ugly hand collisions" and "very demanding to play" are understatements. Try sitting down at a keyboard and working out how exactly you intend for this to be played by an actual person. Page 1 should be workable after some adjustments are made but page 2 is just ridiculous.
My arrangements and YouTube channel!

Quote from: Dudeman on February 22, 2016, 10:16:37 AM
who needs education when you can have WAIFUS!!!!!

Spoiler
[close]
turtle

WaluigiTime64

Yeah, as Latios said, page 2 is absolutely ridiculous. I think reducing the octaves in the right hand to single notes, and lowering the left hand an octave should do the trick.

Actually, lower the entire left hand by an octave. Then page 1 should be fairly playable.
My Arrangements (All Outdated)
My Compositions (All Outdated)
Quote from: WaluigiTime64I strive for second place and I will fight for the position.

Dekkadeci

Updated all 3 files after trying a few trial runs at the piano.

The only things I did was discard notes and re-add parentheses to collided notes (the parentheses were in the Musescore version but not previously the Finale Notepad version).

All right-hand octave passages in Bars 21-36 have been turned into single-note passages. Accompaniment notes have also been stripped from much of the piece.

The crossovers in Bars 24 and 32 are still there (I'm not sure whether pianists would prefer crossing over or prefer passing the accompaniment to the left hand at those points).

I'm not a fan of transposing the left hand because it's the melody in Bars 21-36, and I think it ruins the effect in the rest of the piece.

EDIT: On several other listens, I see your point, WaluigiTime64--where the heck are the solid enough B flat below Middle C (in Bar 13) and the A flat below Middle C (in Bar 14) (approx. 0:08-0:09 in the topmost video) coming from when I listen to the soundtrack? The closest I see in the .nsf are the 16th notes in the accompaniment.

Dekkadeci

Bump--I adjusted all 3 files about 2 weeks ago to improve playability. As I suspected, written-out tremolos instead of RH octave passages in Bars 21-36 sound good enough.

Bespinben

Collisions, my friend.

You got a sonicallly sound piece of music, but the forced cramping of 16ths runs within 4 measures of a system, along with little breathing space in between the treble and bass staves, results in many unyieldy collisions of dynamics and accidentals against many of the notes and beams of the score.
Quote from: Nebbles on July 04, 2015, 12:05:12 PM
Someone beat Bespinben to making PMD music?! GASP!

MLF for Chatroom Mod next Tuesday

Dekkadeci

Quote from: Bespinben on July 04, 2016, 02:10:04 AMCollisions, my friend.

You got a sonicallly sound piece of music, but the forced cramping of 16ths runs within 4 measures of a system, along with little breathing space in between the treble and bass staves, results in many unyieldy collisions of dynamics and accidentals against many of the notes and beams of the score.
Not sure how to solve the collision problem anymore--I did find one more note collision and changed out all 3 files just now to put the collided note in Bar 31 in parentheses and eliminate the hand collision in Bar 11 (and delete the collided note in Bar 12 by eliminating the accompaniment).

Should I delete all collided notes in parentheses?

In the .nsf and my current files, there is a sharp dynamic change in Bar 4--should I change that to a diminuendo instead?

I'm also curious about which accidentals are collided. (The rests are in ugly positions in the score.)

Bars 29-35 do make the hands get very close to each other, I admit.

Pianist Da Sootopolis

Decent job, but there's quite a few things which need fixing.
Okay, here goes nothing. Glad I had coffee earlier.. (hehe I'm not sleeping)
Quite a few of these remarks are going to be personal preference for various reasons. Things you must/really should do will be bolded, and things you may want to consider will be italicized.

Formatting:
- Lower the first system, raise the musical direction marking (good on you for having one, by the way!).
- The interpretation note at the bottom of page 1 is redundant; most skilled pianists (which you would have to be to play this sheet in any case) would know to emphasize a melody line over octave tremolos. If you must make sure it is quieter, use a different dynamic in each staff, placing them above the top staff and below the bottom staff.
- Standardize your systems; you've got 6 measures in the second system and 3 in the fifth.

Arranging:
- First measure: Lower the accompaniment Cs to below their respective dyads. This gives a "fuller" texture otherwise not capable from one instrument, and also solves part of your clashing problem. In addition, let the bottom voices hold out with the use of a second layer.

- Measures 1-4: Replace the hairpin with a cresc. marking (for aesthetic reasons)

- Measure 4 (a): Replace the f p with a Fp (forte-piano). This makes it much clearer what you want. Better yet, since you have a crescendo before hand, you could do a sfp (sforzando-piano).

- Measure 4 (b): Get rid of the top c on the octave on beat 1.5. This allows for a cleaner texture in that little "passing" moment; the bit where the dominant note is repeated in different octaves. To allow this to be clearer to the performer, along with the fp/sfp (depending on what you pick or if a Mod deems it unnecessary), I would eliminate the octave; if anything, I would put it on the downbeat to give contrast.

- Measure 8, beat 2: make the C-Db tremolo a trill.

- Measure 9: Bring your LH down an octave to give both a better texture and a playable sheet

- Measure 12: The 3 note sequence leading into measure 13 is a 16th triplet.

- Measures 13 through 15: Like with the introduction, let the bottom voices hold out by use of layering. This is reflected in the original, as well.

- Measure 20: Make the last note of your RH match the value of the last note of your LH, since they're ending in sync and there isn't an obvious lingering of one voice while another drops out.

- Measures 21 through 23: This is where your sheet suffers most from cluttering. Drop it down an octave to not only detangle your hands, but also give a different timbre to the melody while you have the octave accompaniment.
Also, I know I said it already, but delete the performance mark (and replace it with two dynamic markings if you're so inclined).


- Measure 24: Drop the chromatic line down an octave for the same reason as above.

- Measure 28: Rewrite the RH part an octave down with an 8va sign, and hide the whole rest from the other treble staff for a cleaner transition. Also, increase the spacing a little on this system, as measures 29 through 31 are cramped. You have the same problem in the following system, albeit not as cramped.

- Measure 32: Ditto to the above.

That's about all I can think of right now. Hopefully this can help speed things up.

Btw, sorry if I come off as pissed; you did a good job! There's just some stuff that needs reworking.
what is shitpost

Dekkadeci

After looking at Pianist Da Sootopolis's comments, I re-analyzed my submission and found I've been transcribing the triangle wave channel wrong all along--in Famitracker/NSFImport, it's notated an octave higher than it plays. Due to this, I'll need to fix the files, and I believe it'll take several days to do so (including re-transferring the MusicXML from Musescore to Finale Notepad).

Pianist Da Sootopolis

Honestly you'll be better off re-editing things in notepad. It'll be annoying, but it's really a lot easier than re transferring and re formatting XML imports.
Take it from someone who did this for ages..
what is shitpost

Dekkadeci

Quote from: Pianist Da Sootopolis on July 09, 2016, 11:05:55 PMHonestly you'll be better off re-editing things in notepad. It'll be annoying, but it's really a lot easier than re transferring and re formatting XML imports.
Take it from someone who did this for ages..
I've read that Finale Notepad 2012 cannot let me change clefs, and since the first several measures will involve me putting the bass clef back on the lower staff, I'll need to re-transfer. Luckily, I have Notepad++ and the ability to directly edit (Music)XML...

...and I can also change the number of bars per system in the process.

Dekkadeci

Fixed the sheets' triangle wave channel transcription, then updated all 3 files. My replies to Pianist Da Sootopolis's comments are below in blue:

Quote from: Pianist Da Sootopolis on July 06, 2016, 12:46:32 AMDecent job, but there's quite a few things which need fixing.
Okay, here goes nothing. Glad I had coffee earlier.. (hehe I'm not sleeping)
Quite a few of these remarks are going to be personal preference for various reasons. Things you must/really should do will be bolded, and things you may want to consider will be italicized.

Formatting:
- Lower the first system, raise the musical direction marking (good on you for having one, by the way!).
Musical direction marking raised.
- The interpretation note at the bottom of page 1 is redundant; most skilled pianists (which you would have to be to play this sheet in any case) would know to emphasize a melody line over octave tremolos. If you must make sure it is quieter, use a different dynamic in each staff, placing them above the top staff and below the bottom staff.
I tried setting the upper staff to mp and the lower staff to f starting at Bar 21. I kept hearing the upper staff's dynamic in Finale Notepad, so I stripped the extra dynamics. I did remove the interpretation note, though.
- Standardize your systems; you've got 6 measures in the second system and 3 in the fifth.
Measures per system have been made more consistent.

Arranging:
- First measure: Lower the accompaniment Cs to below their respective dyads. This gives a "fuller" texture otherwise not capable from one instrument, and also solves part of your clashing problem. In addition, let the bottom voices hold out with the use of a second layer.
I wish I knew how to put in a second layer, especially with Finale Notepad...the accompaniment Cs now reflect the actual triangle wave part, though.

- Measures 1-4: Replace the hairpin with a cresc. marking (for aesthetic reasons)
It was easier to transfer the hairpin through MusicXML.

- Measure 4 (a): Replace the f p with a Fp (forte-piano). This makes it much clearer what you want. Better yet, since you have a crescendo before hand, you could do a sfp (sforzando-piano).

fp added. I don't like a sfp as much because I think that's too open to interpretation. (I normally interpret the sf part of sfp as increasing the dynamics about as much as an accent would.) I want to make sure that the initial dynamic is forte.
- Measure 4 (b): Get rid of the top c on the octave on beat 1.5. This allows for a cleaner texture in that little "passing" moment; the bit where the dominant note is repeated in different octaves. To allow this to be clearer to the performer, along with the fp/sfp (depending on what you pick or if a Mod deems it unnecessary), I would eliminate the octave; if anything, I would put it on the downbeat to give contrast.
The top C in Bar 4, beat 1.5 exists in the .nsf, so I left it in. However, it turns out that the lower note in the octave doesn't, so I removed it.

- Measure 8, beat 2: make the C-Db tremolo a trill.
Done--trills don't transfer properly in MusicXML between Musescore and Finale Notepad, so I had to manually add the trill marking in.

- Measure 9: Bring your LH down an octave to give both a better texture and a playable sheet
I used the triangle wave channel instead of some square wave channels now, so the LH is now about an octave lower

- Measure 12: The 3 note sequence leading into measure 13 is a 16th triplet.
The .nsf has an approximation of a 32nd-note sequence there, so I've left that section alone. (The .nsf's precise note ratio is 3:2, but I believe that's just part of the limitations of writing accurate Famicom music in Famitracker at 180 bpm.)

- Measures 13 through 15: Like with the introduction, let the bottom voices hold out by use of layering. This is reflected in the original, as well.
Again, I wish I knew how to layer with Finale Notepad...

- Measure 20: Make the last note of your RH match the value of the last note of your LH, since they're ending in sync and there isn't an obvious lingering of one voice while another drops out.
Done.

- Measures 21 through 23: This is where your sheet suffers most from cluttering. Drop it down an octave to not only detangle your hands, but also give a different timbre to the melody while you have the octave accompaniment.
Also, I know I said it already, but delete the performance mark (and replace it with two dynamic markings if you're so inclined).

As before, the performance mark has been deleted. I'm generally not a fan of changing the octave voicing of musical lines, though.

- Measure 24: Drop the chromatic line down an octave for the same reason as above.
I didn't previously realize this bar had such ugly hand collisions. I'm not a fan of changing the octave voicing of musical lines, though, especially prominent accompaniment lines like this one. I assigned all chromatic line notes to the RH, but as reflected in the MIDI, the chromatic line must be emphasized.

- Measure 28: Rewrite the RH part an octave down with an 8va sign, and hide the whole rest from the other treble staff for a cleaner transition. Also, increase the spacing a little on this system, as measures 29 through 31 are cramped. You have the same problem in the following system, albeit not as cramped.
RH put under a 8va sign. It's now more readable.

- Measure 32: Ditto to the above.
Again, I'm not a fan of changing octave voicings. Here, it's performer's interpretation whether s/he wants to put the chromatic line in the RH and the accompaniment in the LH at this point or to cross over instead.

That's about all I can think of right now. Hopefully this can help speed things up.

Btw, sorry if I come off as pissed; you did a good job! There's just some stuff that needs reworking.

Pianist Da Sootopolis

As far as the .nsf is concerned, honestly I'd ignore it and instead use what sounds best/can be played on a piano. An arrangement is almost never an exact transcription of a song note for note; often we have to make sacrifices because of the limitations of the piano (albeit there are few).
I'll have more later. Good work on those!
what is shitpost

Zeila

Layers are possible in Finale Notepad. There might be a keyboard shortcut for it, but you could just click on one of the numbers in the bottom left corner. There should be 4 possible layers available. If a 2nd layer is used, the 1st one automatically points upward (with the 2nd layer facing downwards). Also, you could change the stem direction by pressing L (lowercase is also fine), or break/combine beams by using a forward slash