Chlipouni

@@iminashi, 

  I understood your second request.

  Can you explain a bit more your first request ? Do you have an example ?

Thanks 

@Pc Plum,

  In the DDC Log file, you have the following warning :

04. Load useful informations from the XML input file
      => 266 ms
    - Song informations
      => 0 ms
    - Measures and ebeats
      => 30 ms
    - Sections
      ==> WARNING : Section noguitar (700.271) doesn't start on the main beat of a measure (new time : 699.455)
      => 3 ms

  So DDC modifies the last section position on the measure at time position "699.455".

  This measure contains the last note. This is why this note is not generated by DDC process.

  I have to change this logic in order to protect the last note (maybe by including the sustain in the computation).

Thank you for sharing the files ...

Can you provide a simple example EOF project file? I authored that example (an arrangement ending in an A6 single note linked to an A6 single note that unpitched slides to fret 2) and EOF exported both single notes:

        <note time="3.750" string="1" fret="6" sustain="0.250" linkNext="1" />
        <note time="4.000" string="1" fret="6" sustain="0.750" slideUnpitchTo="2" />

You can use the following EOF project : Project

Hi,

 

I have noticed a few times now that when I use DD it can chop off the last note of the chart

 

https://drive.google.com/file/d/0B2oZs0ta32KyOGNDejFrMk9ESkE/view?usp=sharing

 

becomes

 

https://drive.google.com/file/d/0B2oZs0ta32KyOXJjNFlmOUs4eU0/view?usp=sharing

 

i.e. it chops off the linknext unpited slide to finish the song.

 

 

I am using rstoolkit-2.8.2.0-273b0ba2-win (DDC 3.3).

 

Any ideas or fixes would be nice, I always like to finish a song with some style :D

@@raynebc,

  It seems that it is an EOF issue.

  I can reproduce it and EOF omits to generate the last note if the previous one has the linknext status

@@firekorn,

  You are right, we can add 5 more levels in the config file by defining only one change between two adjacent levels.

  However, the new levels will be used only if notes exist for these levels and if the merging process is not too high configurated.

  If I understand the main purpose of this request is to have more differences between songs in order to select them in Ignition ?

  Do we need to apply these new parameters in the default config of DDC ?

Thanks

@@firekorn,

  In order to generate more level of difficulties, modifying the config files is not sufficient.

  It is needed to add logic in DDC process, to split two adjacent levels if too many notes appear.

  For example, in a 4/4 time signature, 4 notes can appear at the same time if they belong to the same category and subcategory.

@@firekorn, my leading silence is only about 3 or 5 seconds, so I don't have to wait 10 sec before playing.

About the maximum number of levels of difficulty, I agree with you but I had to find a compromise when I started to design DDC.

Some users desliked to have too many levels (too fastidious to reach the highest level)

 

 

  About the first time position of a beat, I setted it like that because EOF generates the "0.000" beat by default. 

 

Not sure what you mean by that. EOF sets the startBeat and the first beat at the time where the beat map begins, not at 0.000.

 

It only matters if people want to use ODLC-like leading silence (10+ seconds of silence). I've tried it, but haven't noticed any merits in doing so myself. Anyway, Firekorn advises that people do that in his tutorial. And if you move the first beat from 10s to 0s, it will mean that about half of that silence will not be skipped like it is in ODLC.

 

It depends on how you use EOF.

In my case :

- I update myself the audio file to add the leading silence

- I update the GP Tab to add a leading measure (for that silence)

- I start my EOF project with these files and I move the measure number 2 to the beginning of the audio track

So, I set the first beat at "0.000" to manage these different ways to make CDLCs.

Did you notice an issue or a special behaviour in RS2014 due to this first beat position ?

Hi iminashi,

  Don"t forget that DDC is an automatic process. I'm positive that Ubisoft creates its DD levels by hand ...

  The whole DDC process is defined by measure, so the position of notes inside a measure is important.

  If I allow a phrase or a section to start in the middle of a measure, I have to separate the notes of this measure in two distinct sets.

  This feature has already been requested by Firekorn and if I want to implement it I will have to change a lot of things in the existing code.

  Nevertheless, I keep it in my todo list ...

  About the first time position of a beat, I setted it like that because EOF generates the "0.000" beat by default. 

I can't use any version of the DDC with the RS toolkit 2.8.4. With the RS toolkit 2.8.2 i don't have errors, even with DDC 3.3. I'm not following this forum, so maybe this problem was already solved.

Envoyé de mon SM-G920F en utilisant Tapatalk

Hi,

  Download the last release of the RS Toolkit and apply DD again on the psarc file (DDC tab).

  You can also use the "CustomsForge Song Manager" ...

@@raynebc,

There is no limit about the number of phrases that can be linked together.

In my opinion the easiest way to add this feature in EOF, could be to let the author adding a tag in the phrase name. Something like intro[g1] for phrase "intro" in group "g1". So all phrases in the same group could be linked together in the newLinkedDiff XML tag ...

Envoyé de mon SM-G920F en utilisant Tapatalk

@@firekorn,

  "LinkedDiffs" tag is not used anymore in RS2014.

  However, you can use the "newLinkedDiffs" tag to link two or more different phrases.

  When two phrases are linked, their level of difficulty will evolve similarly in the game.

  As an example :

  <newLinkedDiffs count="1">
    <newLinkedDiff levelBreak="-1" ratio="1.000" phraseCount="2">
      <nld_phrase id="7" />
      <nld_phrase id="14" />
    </newLinkedDiff>
  </newLinkedDiffs>

In this specific case, all iterations of the phrases id "7" and "14" will be synchronized in the game.

In DDC, I already use this when two phrases contain similar sequences of notes but not exactly the same (a diff percentage is computed to decide to link them or not).

I can't explain precisely the "levelBreak" and "ratio" attributes, as I always use the default values (these attributes need more investigations). In my mind, these attributes are useful only when the linked phrases don't have the same number of difficulty levels.

I agree for alphanumeric limitation only for phrases and sections. Section names are limited on a predefined list and phrase names are never displayed in the game.
DDC keeps EOF section names and always overrides phrase names.
So it is ok for me ...

For your information, DDC v3.3 generates abridged XML only. So if there was a problem with it, we should have more people with this kind of issue.

Envoyé de mon SM-G920F en utilisant Tapatalk

@@raynebc,

During a GP tab import in EOF, a single note with a slide is terminated with a new gem related to the end fret position. I was expected to have only the first note with the slide and its end fret position, but not the last gem at the end of the sustain.

Is it the normal behaviour ?

Thanks

Envoyé de mon SM-G920F en utilisant Tapatalk

Can you share the psarc file ?

Those two elements doesn't have any kind of mandatory relation to each other and should work independently.

I know that the handShape is here only to indicate which fingers to use and can be fully considered as independent from the notes and chords.

But when DDC has to create the other levels of difficulty, it takes care about how many strings are used in the chords inside an handShape, in order to display only the necessary fingers.

That's why DDC analyzes the handShape to see if it contains chords or only single notes. Its behaviour is not the same in these two different cases.

@@raynebc,

If I remember well, when someone creates an handShape in EOF, the tool adds a chord at the beginning with ghost notes. In this specific case where only a single note is followed by chords this doesn't happened.

Is it a normal behaviour for you ?

Thanks

Envoyé de mon SM-G920F en utilisant Tapatalk

Came across another problem. Generating DD on this arrangement causes some of the handshapes to point to the wrong chordId, with the first one pointing to chordId -1.

 

https://drive.google.com/open?id=0B_0HDnEbZi4PSGo4MDJhbjloSTQ

The reason is because the handShape starts with a single note at time "15.791" and the first chord of this handShape starts at time "16.226".

This specific case is not actually managed by DDC, and I need more explanations of what you exactly want to do in order to improve the code.

 Currently, DDC manages the following cases with handShapes :

 - handShape that starts with a chord

 - handShape that starts with a double single notes (two notes at the same time position but without chord)

 - handShape that starts with a single note and identified as an "arpeggios" sequence

Meanwhile and for your issue, the workaround is simply to synchronize the beginning of the handShape with the time position of its first chord.

@Chlipouni 

 

Have you tried replacing '-' with –

 

Here's a thread with other alternatives for how to use dashes.

@@cozy1,

 Thank you a lot for the link, it gave me the solution I was looking for ;)

@@raynebc,

  In french language, the "-" character is used to link two words together.

  As in EOF, this character has a technical meaning, it is not displayed in the game.

  Do you have an idea about how to generate this character in the XML file with EOF ?

Thanks

I am interested to know how you built the .psarc with these characters in?

 

It seems to me that the toolkit is filtering out some characters during the build process.  Was the toolkit behaviour amemded after your testing, using the results?

 

Thanks :D

@Pc Plum,

  Some characters are not displayed in the game (those with [ND] or [NV] tags in my testing XML file)

  Can you give me an example of a specific character you can't use in RS ?

Thanks

Hi raynebc,

  What is the exact purpose of the "PART VOCALS_RS2_EXT.xml" file ?

  For me, only the "PART VOCALS_RS2.xml" works fine with accented characters (RS2014, RST and text editors).

  When I try to use the "_EXT" file, accented characters are not displayed in the game.

Thanks