I have thought a lot about this the last few days. We are talking keyswitches, expression maps and negative delay compensation.
For those that do not understand negative delay compensation it’s to be found in Cubase under the volume control in the Inspector. Easily ignored, it’s vital for correct timings of samples. Basically, when samples are recorded, the developer leaves a short gap before the full sound is heard. When a sample is played the gap is the beginning of the playback rather than the sound. This puts out the timing by milleseconds. The exact size of these gaps varies from articulation to arrticulation and developer to developer. Thedelay compensation slider in Cubase is an attempt to compensate for this but…
Getting back to Dorico and Cubase. Many people use Keyswitch instruments. It’s impossible to create the correct delay compensation for keyswitches because the delay is different for each articulation.
The only workaround for this is to have one articulation per track but this is very bad for scoring as the holy grail of one track, per one staff, per instrument is severely breeched. Orchestral Templates are already huge and this multiplies the number of (keyswitch) tracks by a factor of 10.
One might think that one can simply drag the notes back a bit in either Cubase or Dorico and use display quantize to tidy up a score. Firstly this is not good practice, it’s also fiddly and can encounter numerous consequences. in key editors and score editors - its untidy and unsatisfactory.
The Solution is SIMPLE. Steinberg needs to provide a mechanism for providing negative track delay per articulation.
This can be done in various ways. I cannot speak too much for Dorico as I am a Dorico apprentice, but in Cubase, either in the key editor, or in the Expression map editor, provision of delay compensation per articulation would fix this issue. Expression maps are more than ten years without an update, it’s well overdue.
This would be huge for composers and it would be vital for integration between Dorico and Cubase ( which I do accept may not yet be on the horizon). Composers would be able to play back samples in time without the need for proliferation of tracks and the ideal of one track per staff per instrument would be preservable without compromising timing.
I hope Daniel agrees with this and can convey this matter to the development team, it’s really come home to me that these innaccurate timings are frequently unavoidable cvurrently and can be sizable
aurally.
We are well aware of the requirement to allow you to specify an offset on a per-switch basis in the expression map, and plan to implement this in a future version of Dorico.
Am I correct in assuming, that this delay ist meant to adjust the starting position of the resulting sound, rather than the timing of when the switch is sent?
In this case, even though unrelated to the OP, I’d like to also make a case for allowing to set an offset for the timing of keyswitches themselves, if possible.
Hi, yes, that is a control I think for compensating certain latency issues with different libraries and I believe it’s global, i. e. works for all keyswitches.
I was thinking of an adjustment per keyswitch, so that switches can also be sent after the note-on event or a certain amount before the note-off events.
Ah, I see what you mean - a positive or negative offset in ticks for any given base switch. Yes, that would be nice. But I’d also take an additional third Velocity/CC Control option, a copy of “Secondary dynamic” any day.
I’d really like this too. There a quite a few libraries where things like jazz articulations have to be triggered while the note is sounding so note-on and note-off events currently don’t work.
Dorico has gotten negative delay compensation capability in its expression maps since this thread was last updated, it was added in Dorico 5. Cubase does not have it yet.
Do you know if this eventual feature release will affect the way Cinematic Studio Strings can be used? I’m confused by that because I don’t own it. Assuming I don’t play in the MIDI, and set my note velocities using the editor, can I then simply use Cubase articulation set presets and manually edit the notes so as to correct the timing of each affected phrase?
(I guess in a way it’s following Logic’s feature set, though I think Logic can also have articulations set on a per clip basis - so confusing for me to know how to deal with fast legatos in relation to keeping on the beat.)
In Cubase, no, because there’s a KSP multiscript for CSS that someone made that equalizes the delay before all articulations to make them all the same (-350ms). So if you use that script, you can just set the entire track delay to -350ms.
But it can affect the way other libraries can be used, especially when not using such scripts.
The .nkp script & nka presets I have are by kpmuzik and ak70 on vi-control. If I use those I should be able to use that track’s kontakt libraries per those, and it shouldn’t interfere with the negative track delays I set - unless Cubase is still having trouble with automation data messing with that delay processing. Is that what you mean?
It makes all articulations in CSS the same negative track delay so you don’t have to do it on a per articulation basis at all. It also implements single key trills as the method CSS uses to trigger trills isn’t compatible with Dorico expression maps to my knowledge.
It would be interesting if someone tried building an expression map for CSS using dorico’s per switch latency compensation. Sadly I haven’t had the time to try it yet.
I also tried that out back in the day. One or two projects it worked quite well on and saved the bother of having to create an extra hidden staff for trills but there were also a number of issues – particularly a tendency for notes to stick in difficult passages (which may not affect everyone). But as I’ve found NPPE gives better results than the script, my own maps or anything else I’ve tried, I don’t really see why bother using any other method in Dorico unless you have very exacting control requirements and are willing to put in a lot of extra work.
Because NPPE is simply not suitable for cases where the finished product is a sampled one and there will be no performance, and the finished product has to be a nicely finessed rendering. You don’t really have control over the result with NPPE. NPPE is meant to do great sounding demos (key word being demos) and not actually used as a finished product. When using libraries directly in Dorico, I can export the MIDI and bring it into Cubase and load up the same template and have everything sound identical to how it does in Dorico and then be able to further fine tune things in the DAW. This workflow is not possible with NPPE.
When turning in a film score for a director for instance, trying to be like “oh I’m sorry about that note sounding a bit wonky there at 38 seconds in, but NPPE is just being a little strange with that and I can’t do anything about it”. That wouldn’t fly.
ok – if you’re talking about scoring to film and working in both Cubase and Dorico then of course we’re in a completely different situation and of course you’re right. I was thinking in terms of doing a mock-up of a stand-alone work in Dorico only.
Even in a case of not scoring to film, but where I am doing a demo track that is basically a demo of what I would write for film, I still want it to be one where I have full control over the end result, because it’s meant to show what my stuff sounds like for film.
Some people might say to “just write in a DAW then”, but there are two main reasons I want to do this. The first is that I get a nicer view of the big picture (orchestration, doublings) through notation than I do in a sequencer like Cubase, so my orchestration will generally be better and I will consider more things intelligently. The other aspect is that in the past I have done demo tracks in Cubase that I might like to create updated versions of at some point, but replacing all of the samples would be a huge job, similar to writing the track over again. However, if I actually have a notated score, very often I can simply “plug in” newer samples as they become available and get an updated rendering without having to do a lot of work.
I still love NPPE and recommend it for simple demos as it makes life so much easier. But after a lot of work, I basically have a full template that I can load in Dorico or into my DAW that gives me a correctly balanced orchestral sound where I can load up basically any score and play it back with sound that is about as good as NPPE (sometimes better, sometimes worse) but leaves me with full control over all aspects.
Whether I would recommend this avenue to others really depends on what the end goal is. For most people I would just say NPPE is easier.
