Problem with midi import

[pmj_rit]

Hello,

I've been working on a big project in Sonar for quite a while now (I don't have a lot of spare time).

I've tried to import the midi from it into Finale (having first quantized in Sonar: note durations to 1/32 (that means they all end precisely on a 32 boundary - not that the length is forced to 1/32) note starts to 1/32.

When I import into Finale I set the quantization to 1/32.

Still it Notates this very oddly (see attachment for an example).

Is there is way to prevent this.

[N Grossingink]

In the attachment you pictured, what would be the correct notation you are trying to achieve? Are there actually any 32nds that you want notated, as well as any tuplet values? You know the result you're looking for - anybody else would be guessing. Try to give us some additional information.

N.

[zuill]

The solution may be as simple as "Minimize Number of Rests" in the quantization settings.

Zuill

[pmj_rit]

In the attachment you pictured, what would be the correct notation you are trying to achieve? Are there actually any 32nds that you want notated, as well as any tuplet values? You know the result you're looking for - anybody else would be guessing. Try to give us some additional information.

N.

I've attached what Sonar makes of it (Capture1).

The point about the 32nds.

You're right there are no actually 32nds in the clip. There are in the project.

What this kind of Quantization in Sonar does is the following:

Any note that does not begin precisely on a 32nd is moved to the closest 32nd. Any note that does not end precisely on a 32nd is lengthened or shortened so that it does while keep as close as possible to the orignal.

This being the case I cannot understand Finale's interpretation.

Another attachment might help...but I see they preview in the wrong order.

Anyway in Capture2 The darker vertical lines are the quarter note boundaries, the lighter ones are the 32nd note boundaries.

As you can see it is very precise (and probably would never be played this by a person).

I have spent an inordinate amount of time composing this in Sonar and I would like to avoid, if possible, extensive editing in Finale post import.

Does this make sense now?

I realise that some editing is unavoidable - but Finale literally getting in the way? surely not.

[miker]

If it isn't going to be played by a musician, why bother with Finale at all?

Compose it in Sonar, record it, and move on.

[pmj_rit]

The solution may be as simple as "Minimize Number of Rests" in the quantization settings.

Zuill

Thanks zuill, but this is too crude.

[pmj_rit]

If it isn't going to be played by a musician, why bother with Finale at all?

Compose it in Sonar, record it, and move on.

Why make the assumption that it isn't going to be played by a musician? That wasn't what I said.

If a midi recording was made of two musicians playing the same score and if that recording were displayed with totally fidelity by the sequencer's score editor there would be tiny rests added between some notes or notes would be extended slightly (etc). The end result is unnecessarily detailed and very confusing. Which is what I am trying avoid here.

Unless you have done both midi and score editing this might not make sense.

I prefer working in Sonar because of it's precision (note placement, length, dynamics) and the way it handles audio. i.e.it allows real "human playback".

But it's no score editor compared with Finale.

[miker]

Because you said;
As you can see it is very precise (and probably would never be played this by a person).

If I misinterpreted that, I apologize.

[pmj_rit]

Because you said;
As you can see it is very precise (and probably would never be played this by a person).

If I misinterpreted that, I apologize.

It's OK.

To illustrate further: I've just reverted the midi (in sonar) to the beginning of the session, looked in detail at the midi (not the score). In several places notes don't fall (or end) on the 32nd (or even the 64th or 128th). Much closer to someone playing it. But Finale would have an even harder time interpreting this.

[N Grossingink]

I'd suggest you open your MIDI file in Finale and post an abbreviated Finale file (4 or 5 pages) of the result, making sure that file contains an average representation of the notation insofar as complexity and density go. If we see that, there might be quantization settings that would be helpful. Again, at this point, we can only guess at what you're trying to achieve.

N.

[zuill]

An even better test is for us to see the MIDI file itself and see if we get the same import result.

Zuill

[pmj_rit]

Hello,

I've been working on a big project in Sonar for quite a while now (I don't have a lot of spare time).

I've tried to import the midi from it into Finale (having first quantized in Sonar: note durations to 1/32 (that means they all end precisely on a 32 boundary - not that the length is forced to 1/32) note starts to 1/32.

When I import into Finale I set the quantization to 1/32.

Still it Notates this very oddly (see attachment for an example).

Is there is way to prevent this.

Thanks everyone. I think I'll go the alternative route. Sonar does a passable export to MusicXML with a limited number of staves. This export does not do well with counterpoint within a single stave but fortunately there are not many of those in the file.

I'll create as many separate exports as need then do some copying and pasting. I was just testing the water today ~ the real job of creating a Finale version can't start for a while yet.

Thanks again.