Note lengths can also be extended using percentage of notated duration which is a nice feature.
Can we have an option for delays to behave the same way? E.g. a “Delay %” field that accepts both positive and negative values relative to the note’s duration? This should allow for varied legato speeds if i’m not misunderstood.
I’m not sure how much a percentage field would even make sense in reality. Most libraries are meant to work with a certain amount of delay independent of note length, which often decreases when using a special “agile legato” option.
Any libraries that did not do this would be next to impossible to use in a DAW - unless they also did read-ahead, but if the library can do read-ahead then there’s no need to manually do a delay like this.
That sounds right. Perhaps I suffered from a case of XY Problem.
Maybe asking about an “Absolute note length” option would have made more sense.
Please take a look at this:
The “delay” option shifts all notes in the direction specified, rightly so. But it creates problems in legatos. Using percentage values in note lengths for compensation would be non-ideal because 105% of a 1/4 note is different from a 1/2 note in a legato phrase but both are shifted back 30ms.
I see you are using CSS. I think the issue is that in CSS the delay can be variable depending on numerous factors. This is not normally the case for most libraries. You can minimize this by using the “low-latency legato” instead of the “expressive legato” - most people working in DAW software use the low-latency legato. Also I believe that key velocity of the note has an impact on the speed of the transition too, so you might find you can minimize this issue if you try forcing notes that are short (via Dorico note conditions) to higher velocities. People have also made Kontakt scripts that try to adjust for this, such as the Variable Delay Compensator for Kontakt.
These all boil down to specific problems for CSS. Even in DAWs people fight with these same things too much. I think CSS is one of those libraries that could really benefit from a read-ahead solution like what Tokyo Scoring Strings uses.
That is one of the primary reasons that I really enjoy using Dorico (Expression Maps). You can setup things based on rules in different scenarios and it will work nice most of the time. If it doesn’t you always have the option of playback overrides (but that I like to minimize my usage of).
Dorico encourages its users to write correct notation first, and separates notation from technical limitations of playback.
Thanks to your suggestion, I tried lowering the note velocity and it somewhat improves the situation but that’s still far from what I originally intended, which in this case is achievable through a double-sided extension of legato notes to work around MIDI’s limitations.
The three alternatives now are
a- manually adjust each phrase
b- not use expressive legato mode
c- use note velocity and conditions to lower velocity of short legatos
Given that CSS is a widely used library in the orchestral composition scene, and implementation of an absolute extension would fix the problem while not being impossible to maintain and implement, I think this is worth looking into. Sorry for my initial wrong title.
The one thing that I think Dorico probably should be able to do is have a different offset for the first note under a slur as very often that needs a different delay. Right now it gets “lumped in” with the rest of the legato notes that need to be shifted earlier.
You can try the Variable Delay Compensator script for Kontakt as you might be able to use scripts like this to improve things in Dorico.
Most professional film composers I know working in DAW’s do not use expressive legato mode due to the timing inconsistencies.
BTW I think it is that higher velocity gives a quicker transition rather than lower velocity, but I might be mistaken, I don’t normally use CSS even in a DAW on account of the delay handling.
I personally think building a system that could perfectly accommodate the weirdness in CSS into Dorico would be too much. There are other libraries like EastWest Hollywood Strings that are even worse in this way, where the delay that is needed can actually even depend on what note it is due to the way it is edited and some pitches have a faster transition than others. If Steinberg had to build a system for Dorico that would factor in every single possible pitch with every single possible CC1 setting with every single possible velocity to accurately map the required predelay it would be kind of insane honestly. At a certain point you have to hold the library vendor accountable for these kinds of workarounds. But thankfully we already have Kontakt scripts like the Variable Delay Compensator that can do this.
I initially wrote a long response as to why expecting Dorico to compensate for negative delays would be different from that EHS case you mentioned since MIDI is real-time and we are all doing different interpretations of “workarounds” anyway, and it hit me. Dorico should have already extended notes albeit not in an absolute manner. But it didn’t
I see something weird with what word Natural too like there are other ghostly letters underneath it that say something else. I’m not sure if that is just an artifact of your screen recording or what.
I think you have something wrong with your map, because normally adding a slur should be all that is necessary to get the playback technique Legato to trigger legato.
EDIT: In your earlier video it said legato there, now it does not anymore.
Please also note that there is a useful tooltip when you mouse over that rectangle (where I have circled) that gives you additional details as to why it has selected a particular technique at a particular spot.
Oh I think I should have explained, sorry again it’s on me:
I prefer using the “playing technique” once in the start of a phrase as it looks much more clean and I know the whole phrase is going to be legato. I learned that in this forum actually.
I was using Independent voice playback for my violas.
If you think this is worthy of transferring into a new issue (slur not behaving the same way as labeled technique) for the sake of posterity, I’ll do that. The amount of subject pivots I caused in this single thread is unhealthy.
Yes, I can reproduce this easily. I never used the playing technique “legato” before - I didn’t even know there was one. It certainly isn’t doing the same thing as using slurs when it comes to the note length adjustments. It is probably best in a new topic because of the number of things here.
I genuinely am amazed as how such simple assumption can turn into unforeseen consequences like this whole thread yet is still unsolved. Hopefully Dorico resolves this. Thank you all for your precious time and please support that linked thread.