Tab improvements- Presets, playing positions, etc.

• Feb 23, 2015 - 00:12

The new tab feature in Musescore 2.0 beta2 is awesome. It makes writing tabs a lot easier. It would be even more powerful with a few improvements/ additions/ alterations:

It would be nice to be able to pick tuning presets in the Edit Strings panel. I use CGCFAD tuning a lot and it's a little frustrating having to change every string. If I could open a menu and pick Drop D, Open D, Standard, etc., then I could put my guitar in any tuning a lot easier. Then you could add a "key up/down" button that would raise/lower each string by a half step (So for CGCFAD tuning I'd only have to click a few times- once on the tuning preset, and twice on the key-down button.). To that you could add string variations, like presets for 7-string guitars and 8-string guitars. And of course, for other instruments, like lutes, mandolins, etc. It would be REALLY cool if tabs on doubled strings (for mandolins and 12 string guitars) would play BOTH notes, the one on the low octave and the one on the high octave.

Also, I've noticed that the tabbing algorithm is very efficient and always bumps the notes to the highest string and the lowest possible fret (I add tabs after writing sheets, using a linked staff.). This is nice, but if I followed the automatic tab I'd be running up and down the guitar neck a lot. I'd like it if there was a way to set a playing position or a "target fret" so that the tabber would try to put all the notes as close as possible to the selected fret, regardless of strings. The frets are a lot closer higher up on the guitar neck, so I prefer to play up there, around the ninth or tenth fret.

Finally, I've noticed that the automatic tabber writes the tabs for the individual notes, not the chords (meaning that the algorithm runs separately for each note and tries to place them as low on the neck as possible, on the highest string.) For example, for power chords (that is, perfect fifths, and people who use tabs write power chords all the time), I often see the tab write them two strings apart, instead of the usual two. Same thing for fourths (but M3's and m3's are already correct, I think). I think that the tabber should have a "chord-shape-recognizer" that always tries to write recognized intervals (P5, P4, M3, m3, etc.) in their standard chord shape, first. So for a guitar in standard tuning, it would look like this:

P8 P5 P4 M3 m3 tt
x x x x x x
x x x x x x
6 x 5 7 2 x
x 5 5 8 4 6
4 3 x x x 5
x x x x x x

Or something along those lines. I understand that it would be VERY difficult to make the algorithm analyze the whole chord, but I think it's worth a try making it recognize specific chord shapes.

Finally: I get annoyed that the number keys don't pick rhythm in tabs staffs. I understand why that is, but I wonder if it would be possible to keep the number line on rhythm hotkey and put the numpad on tab number-note? Or only use the number-note thing on staffs that are ONLY tabs (for people that write just the tabs, not the standard treble clef sheets) and leave the linked staff-tabs to use the numbers for rhythm hotkeys?

Anyways, I hope this gives some good ideas and isn't just rambling. It's already really useful. It basically writes the tabs for me, all I have to do is drag them onto the right string. Thanks for putting so much work into this and good luck on Beta 2.0! It's great so far.


Comments

Hi Joseph,

Thanks for the appreciation and for the comments which, I think, are not "rambling" at all. There are many points to answer; I'll go with order...

...starting with a general note: as it stands now, this "feature request" is a bit too articulated; while posts in the forum allow a wider margin of conversation style, issues are easier to deal with if they are kept focused. It would have been better to split this one into several, one for each of the proposals and suggestions it contains.

Tuning presets: it is a good idea in theory; in practice, it is not so quick to implement. It is not practicable to hard code them in the programme: currently, there are 53 instruments with string data known to MuseScore, it would be easy to arrive at hundreds of presets and strategies should be devised to filter them appropriately while presenting the preset list to the user.

The only viable alternative is to have the infrastructure only and leave to each user to fill in the preset he actually needs and a way to recall them when he needs them; this would require a good deal of user interface. For this reason, I doubt it will get implemented any soon, for sure not before MuseScore 2.0 is out. It is an interesting feature request though, definitely worth of being filed independently!

"Up"/"Down" buttons in tuning: this is a very interesting suggestion; it would be very useful by itself and would also help to reduce the load on the note select dialogue box, which has been criticized by some as not very intuitive to use. As it would be an addition self-contained in the dlg box code, it could make sense to add it even on the threshold of the Release Candidate.

Double string playback: I am afraid this is not possible now, without major structural changes; just as an example, currently MuseScore does not even know that there are double strings!

Fretting algorithm: this is a topic which comes out often. To begin with, the current algorithm does fret chords as a whole, in fact, it frets all the simultaneous chords (in case of multiple voices) as a whole; otherwise it would be impossible to check a string is not used twice.

As a general answer, I believe this would open not one but many cans of worms. To begin with, as I said, there are currently 53 instruments with string data, implementing a guitar-only chord fretting algorithm would make little sense; even limiting to just one instrument, you yourself already introduced the possible variation of different tunings and of different position targets: getting the fretting you prefer (or each of your colleagues prefers) right in the majority of cases would already be a daunting task.

Note that, if you are after a specific fretting, you can always enter it directly in the TAB staff itself: it will keep whatever you put in there.

So, I do not think anything along these lines will be implemented any soon.

Number short cuts: in the initial stage of TAB implementation, this point has been discussed at length, many suggestions came out, with several 'false starts'. The current implementation is the intersection of several targets and limitations.

Numpad is not equally well supported on all platforms and there are out there notebooks without any numpad. Users with a previous experience on well-known guitar notation programmes expect top-row numbers to input fret numbers. For all these reason (and, I'm sure, other I have forgot in the meantime), we arrived at the current implementation.

About having different short cuts for TAB alone and for TAB linked to standard staves, I believe this would make things even more obscure and impractical.

This aspect, like any other, is not cast in bronze; any part of the user interface can be improved, but I believe an alternative to the current implementation should be discussed at length, weighting the pros and the cons with attention.

Thanks again for the time you spent analysing MuseScore's TABs, I hope further discussion will bring more improvements!

M.

Do you still have an unanswered question? Please log in first to post your question.