Handshape fingerprints outline vary. Anyone know why?

[Berneer]

Hi guys,

I am not thinking the following is an EoF issue so I figured I would post here.

During final testing of my newest CDLC (Blaze Bayley - Stare at the Sun) I am noticing something I haven't noticed before. Those finger prints in the game that show what handshape is needed to play a chord behave a bit differently based on a logic I don't yet understand. If you look at the GIF snippet below of my custom you'll notice a series of chords intermitted with palm mutes. Notice how the finger prints tend to either show up in game in 1 of 3 ways.

1. Number with a double outline around it.
2. Number with a single outline around it.
3. Number with no outline around it.

Does anyone know why the game does this and upon which logic or criteria it decides how many outlines a fingerprint must have?

http://i920.photobucket.com/albums/ad41/Berneer/CustomsForgeStuff/Questions%20for%20CF/Handshape%20varying%20outline/Handshape%20fingerprint%20outlines%20vary_zpso9hakqzl.gif



Here is what it looks like in EoF with representation below that shows what happens in game to better allow recognition of a pattern, which I don't see yet.

http://i920.photobucket.com/albums/ad41/Berneer/CustomsForgeStuff/Questions%20for%20CF/Handshape%20varying%20outline/Stare%20at%20the%20Sun%20verse%20hanshape%20outline%20variance%20questionJPG_zpsel9nc4cm.jpg~original

And at the end of this post is the pertinent XML code which doesn't seem to dictate number of fingerprint outlines since regardless number of outlines, the code seems to have the same nomenclature for each chord. Anyone see a pattern? Or can anyone provide an explanation?

I just checked an official song, REM - Losing My Religion, and within the first 30 seconds of that song I see all three types of outlines discussed above. It must mean something. Can any tell me what each outline type signifies..... looks like it is more a game rendering phenomena than and XML related one.

Link to pertinent XML

[iminashi]

I know that a double outline indicates that the following chord is a different one, but uses the same finger on the same fret.

 

A single outline just means to hold the fingers in place, I think.

Edited October 22, 2016 by iminashi

[firekorn]

I don't think there's much of anything that can control this apart from doing heavy fingering editing with ghost chord around to perfectly control how Rocksmith will handle that.

 

Outlines seems related to how long your finger will stay in that position :

- No outlines => will fastly change to a different position

- 1 outlines => will stay here for the next chord(?)

- 2 outlines => will stay here for more than 2 chords(?)

 

Note that the repeating chord doesn't seems to count toward the total.

[Berneer]

Thanks @@firekorn and @@iminashi.

 

Yours are the best explanations so far, but if you look at the linked image in the original post, a static annotated EoF view, you'll see that there are exceptions to even both your explanations... seems like the correct answer is still a bit of a mystery.

 

I suppose for now I've got to live with it though I wish I could control it since I find that passage not aestheticlly pleasing in my new custom which I am making every effort to make pristine. I'm being a even more perfectionist than usual since I will be sending my custom and video playthrough to the artist (Blaze Bayley, former Iron Maiden singer) since he was nice enough to personally send me tabs for his song.

 

If I understood the logic better then I would be happy to so some heavy-handed finger editing.

 

Thanks for you help guys!!! I really appreciate it.

[firekorn]

let's look at the finger on the yellow string first :

On the 128 measure double outlines cause there's more than 4 consecutive chord where the fingers stay in that position

Then it switch to 0 outlines because the finger will soon change position

On the measure 130, there's no outline cause it doesn't stay for more than 2 consecutive chord

On the 3rd and half beat of the measure 130, there's one outline cause there's 3 consecutive chord where the finger stay in that position

It switch back to 0 outlines before it changes fingering

On the 4th beat of the measure 131, i have no real clue here why it would be double outline instead of a single one as the other element would suggest...

 

TBH outlines is something that aren't really the most visible thing especially on really fast section and if you get everything else to look as neat as possible, i doubt this would make a difference.

 

Note that i would suggest to add some handshape marking when chords are repeating as it might help make things clearer (at least worth a look imo)

[Berneer]

I think both firekorn (proximity dependency)and iminashi (double outlines for different chords with same fingering) are onto something.

 

 

I am thinking is that the finger outline is based on how much time until the next change in position. I am thinking that magic number is between 0.5sec and 0.75 sec but using Bandicam and frame by frame verification of multiple examples in EoF I can probably zero in to the exact number. Using this and your conclusions above I think I am close to finding 5 or 6 rules that dictate the behaviour. I will report my conclusions later.

[Berneer]

OK, so I could have made this a rigourous analysis but I think I will content myself with the conclusions from the tests below to at least get a good approximate understanding of the RS fingerprint logic. Still a deep enough look to be typical of Berneer... :)  If you want to skip each test (observations and conclusions) you can skip all that and look that the Overall Conclusions section at the end of this post.

 

For the masochist. Click here to see all tests, obsersations and test conclusions in a Word document.

 

The document above is complete.

 

 

OVERALL CONCLUSIONS

 

1. Handshape fingerprints begin aligning with a subsequent chord/handshape 0.5sec prior to it, taking 0.4 sec to transition and being fully transitioned 0.1sec before the subsequent handshape. This transition will always begin even if within a preceding handshape! All transitions, be they appearing, disappearing or transitioning fingerprints take 0.4sec to complete.

 

Visual Substantiation: https://youtu.be/dAQM2XM_zy4 (All links: Timer bottom left, 30fps, 3 frames = 0.1 sec)

Chord 1/2/3 --> 0.8sec/0.9sec/1.0sec sustain --> No/Single/Single outline

 

 

2. Different notes and, by extension, fingerprints of a handshape behave independently from one another based their individual circumstances if the different strings are played on different frets. If the different strings are played on the same fret, see Conclusion 6.

 

Visual Substantiation: https://youtu.be/YIGQlWVBFwI (Blaze Bayley - Stare at the Sun, measure 128)

Different outlines on different notes of a chord.

 

 

3. Single outlines around fingerprints occur if two criteria are satisfied:

i) Another chord exists after the current one. (even if 5 secs after, perhaps even all the way to the end of the chart, but more than 5 sec was not verified)

AND

ii) The handshape lasts 0.9s or more without changing, even if it spans multiple chords.

 

Visual Substantiation:

i) https://youtu.be/fvLwazOnCEQ --> No chord exists after the last chord 5sec after the previous chord

Chord 1/2/3 --> 1.0sec/1.0sec/1.0sec sustain --> Single/Single/No outline

ii) See visual from Conclusion 1 and see following for example spanning multiple chords:

    https://youtu.be/OC-_OZGtMIE (Blaze Bayley - Stare at the Sun, measure 132)

Single outlines remain on multiple chords more starting at least 0.9s prior to next different chord - first two A7/D6 chords are start > 0.9 sec from later chord A4/D4 hence single outlines, however the last A7/D6 chord starts 0.4s before the A4/D4 chord hence no more outline.

 

4. Double outlines around fingerprints occur on notes of a chord that are common to the next different chord (as noticed by @@iminashi ).

 

Visual Substantiation: https://youtu.be/sv8gBs4IAWs (Blaze Bayley - Stare at the Sun, measure 128, chord 4 to measure 130, chord 1) -  A4/D6 x 3, A4/D4 x 1, A4/D6 x1, A7/D7 x 1

Double outline remains on the A4 for 4 chords, but not 5th (last A4/D6) since the next different chord has no common notes with it.

***VERY INTERESTINGLY, note that A4/D4 chord has the double outline on A4 despite starting only 0.8sec before the A7/D7 which seems to suggest that unlike the single outlines requiring at least 0.9sec between different chords, the double outline either is not time dependant or is time dependant with a smaller time requirement than 0.9sec. My tests performed so for do not allow to confirm which is the case.

 

 

5. No outlines around fingerprints occur on notes of chord:

if the current handshape is not at least 0.9sec long

OR

no follow-up chords exist after (5 sec or maybe even in the rest of the section or even chart - need more testing)

 

Visual Substantiation: See visual from Conclusions 1 and 3

 

 

6. If adjacent strings are barred for a given chord and each note, with the above logic would each possibly have a different number of outlines, the barre will be applied with both (I assume 'all' if >2 string barre) notes in the barre surrounded but the number of outlines of the note that has the most outlines.

Caveat 1: I have only verified the above conclusion with a case of  a barre chord with one note satisfying the criteria of double outline and the other note in the barre satisfying the criteria of no outlines with the result being both barred fingerprints being surrounded by one larger double outline. 

Caveat 2: I have not verified what happens on same fret notes on non-adjacent strings.

 

Visual Substantiation: See visual from Conclusion 4

A4/D4 chord have one double outline surrounding both fingerprints, even if D4 on the next different chord is not in common with it since the next chord is A4/D6.

 

 

7. Single outlines around fingerprints remain until there is 0.5sec till the end of a handshape, though the fingerprints (and numbers in fingerprints) remain in position until the end of the handshape unless a subsequent chord is within 0.5sec.

 

Visual Substantiation: See visual from Conclusion 3i at 25% speed in YouTube.

 

 

8. Single notes that are between two handshapes will interrupt (make disappear) outlines until, at the very latest, 0.1sec before the said subsequent handshape, however fingerprints and their number remain despite the single notes

Caveat: I assume if the time interval between the last single note and the chord is greater than 0.5s, no interruption will occur since the outline will not have appeared before the single note(s) in the first place.

 

Visual Substantiation: See visual from Conclusion 4

The palm muted A4 single notes preceding the A4/D4 chord interrupt the barre double outline leading up to the chord by quickly disappearing suddenly 0.16sec before the first palm-muted single note and reappearing suddenly instantly after the 2nd palm muted single note, doing so 0.1sec ahead of the A4/D4 chord.

 

 

Overall Caveats:

1. I have not run exhaustive tests to make 100% sure that the above statements are correct without exception.  It is possible that some of my conclusions may be premature.

2. I sorta use the words 'chord' and 'handshape' interchangeably throughout this post even though I only authored chords and imposed no handshape markings in EoF. I'm assuming the behaviour for both manually and automatically created handshapes to behave the same. 

 

I still have to figure out how to make my custom look cleaner in the verse section. I am even considering removing the handshape definitions since they are somewhat obvious.

 

Thanks for your help @@firekorn and @@iminashi ! You are very kind and helpful.

[firekorn]

Okay so after playing some official and reading through the document you shared, i arrive at the same conclusion though you can probably put it in a simpler way.

 

The single outline is controlled by timing while the second outline indicate that the fingering should be kept no matter what is next (aka there's a chord that will use that finger in that emplacement soon) and of course when there's no outline it means that the fingering will change imminently.

 

Note that there's probably an "expiration time" so that the double outline don't appear when the next chord is too far away.

 

My guess to make it look cleaner would be to use Handshape or to put no fingering on the chord to avoid it appearing at all.

[Berneer]

 

The single outline is controlled by timing while the second outline indicate that the fingering should be kept no matter what is next (aka there's a chord that will use that finger in that emplacement soon) and of course when there's no outline it means that the fingering will change imminently.

 

 

Yes, I agree that is a great 1 sentence conclusion!

 

Since you wrote this post prior to me writing (completing) the above post, I figured your use of the word "simplify" meant I had to write a long detail overall conclusion... ;)

 

Note that there's probably an "expiration time" so that the double outline don't appear when the next chord is too far away.

 

This is probable. I'm too anxious to finish my custom to do any more testing to confirm this... :P

 

My guess to make it look cleaner would be to use Handshape or to put no fingering on the chord to avoid it appearing at all.

I agree. I am still not decided which of the two options I will choose. I was even thinking of a 3rd option of only applying fingerings to a each different chord once where it is not too messy.

 

Thank-you @@firekorn, it is very motivating to be bouncing off ideas off one another. Makes the extensive testing feel more worth it.

[firekorn]

@@Berneer I'll admit i would have never bothered to look at this detail but it can be very interesting in chord heavy part so it's nice to know what those indicators are for :)

 

For your custom you can try one option, see how it goes and try another one if it doesn't work for you. I find that seeing how it actually show up into the game is the best way to know if it's an efficient way of showing what you want the user to see. It's sometimes hard to get how easy it will be to read directly into EOF.

[Berneer]

I have noticed another little handshape fingerprint behaviour with marked handshapes whereby the fingerprint outline blinks.

 

In my latest custom (Blaze - Stare at the Sun) I have decided to take a seemingly complicated almost arpeggio passage and use concatenated handshape markings in EoF to simulate a partial handshape whereby fingers that stay fixed throughout all the handshapes have fingerprints that stay put and then those that change usher in a new handshape marking in EoF which suddenly displays which finger to use to play the varying added note. Each handshape is painstakingly placed 2ms apart, though EoF seems to prefer to output them as being 1ms apart, but I digress. This construct is quite effective in my opinion apart from this very insignificant but irritating blinking problem.

 

See the pre-verse section which has a two different sets of 24 so-called  "concatenated partial handshapes".

https://youtu.be/RPmWYdYTRtI (btw, How do we post a YouTube window in the pst rather than a link?)

 

You could see the reason for the blinking in the annotated EoF screenshot below.

EoF screenshot (sorry issues with Customsforge with posting images in the last week(s))

 

Thought I'd share this as an addendum to the Overall Conclusions in the above post, since it does add fingerprint behaviour information for the case of a "EoF manually marked handshape" using (CTRL+SHIFT+H). Interestingly it also seems to show that a double outline can disappear based on small time away from the end of a handshape which the above Overall conclusions post was inconclusive whether a double outline  could be governed also by a time based logic.

 

Oh well, I guess I have to live with it in my custom. If the price of 'concatendated partial handshaping' is tiny fingerprint blinks, then so be it.

[firekorn]

Yeah at this point maybe @@raynebc could help a bit but we would need ODLC that use the same kind of handshape marking which is pretty rare.

 

Walk This Way space out the handshape so they don't overlap. There's Six by all that remains that use an overlapping handhsape but it's very short. Don't have anything else that comes up on the top of my head but there's probably a bit more than those two.

 

@@Berneer for youtube video, if you want them to display, use the full link, not the shortened one that YT provide.

[Berneer]

Thanks firekorn.

 

I didn't think any ODLC that use concatenated handshapes like I did. I have merely used an idea in my custom that raynebc gave me about a year ago and I've called it here "concatenated partial handshapes" for the lack of a better term.

 

I wasn't writing this last post as a problem issue and this is not an EoF issue, but I wrote my last post more as an addendum to the "Overall Conclusions" in the earlier post, to inform anyone who wants to know, more about how the handshape fingering behaves but this time in a marked handshape, since a marked handshape can be a bit more elaborate than an automatic handshape defined for all chords.

 

You mention overlapping. I'm only concatenating, not overlapping though. But interesting that you found ODLC with an overlapping handshape. Chlipouni was looking for an example of this over a year ago but then raynebc and chlipouni decided to dismiss that construct.

 

Thanks for the YouTube tip.