Does anyone have any experience with this book? By its description it appears to contain many techniques for getting more realistic results when attempting to simulate a full orchestra with MIDI instruments.
I'm not familiar with the book but I can give you a suggestion.
When writing wind, brass or string music the temptation to use an ensemble patch is irresistable. Many people write for a brass section using a "Brass Section" patch. That's fine as long as they 1) play unison and 2) play notes that fall within the practical range of all the sampled instruments.
Say you have a sample of a 5-piece horn section (2 trumpets, alto, tenor, trombone). You play one note and it sounds great. As soon as you play a second note you have effectively DOUBLED the size of the group. The reason being, you're playing two samples of a 5-piece horn section and that equals 10 pieces.
The math is even more unfavoriable with larger ensemble samples. Think of a 60-piece orchestra sample. Now just play two notes and you've gone from 60 to 120 at lightspeed.
One way to deal with smaller ensembles is to use solo instruments. This works great for sections with...say...10 or fewer players. I'll often do this, then I'll make a copy of those tracks, merge them together and feed them into an ensemble sample mixed below the solo instruments.
Of course, this is more difficult with an orchestra. Also, there are other equally important issues like articulation, expression, etc. The doubling ensemble thing was one of my pet peeves while doing sound design.
I think instruments like Gigasampler/studio and Unity that can access HUGE samples and multiple articulations are really the way to go. Especially if you can combine them with a few live players.
I'll post a few more of my experiences later on. I'm sure other folks have some good tips on this subject.
Hi Stuart I'm not familiar with the book but I can give you a s
Hi Stuart
I'm not familiar with the book but I can give you a suggestion.
When writing wind, brass or string music the temptation to use an ensemble patch is irresistable. Many people write for a brass section using a "Brass Section" patch. That's fine as long as they 1) play unison and 2) play notes that fall within the practical range of all the sampled instruments.
Say you have a sample of a 5-piece horn section (2 trumpets, alto, tenor, trombone). You play one note and it sounds great. As soon as you play a second note you have effectively DOUBLED the size of the group. The reason being, you're playing two samples of a 5-piece horn section and that equals 10 pieces.
The math is even more unfavoriable with larger ensemble samples. Think of a 60-piece orchestra sample. Now just play two notes and you've gone from 60 to 120 at lightspeed.
One way to deal with smaller ensembles is to use solo instruments. This works great for sections with...say...10 or fewer players. I'll often do this, then I'll make a copy of those tracks, merge them together and feed them into an ensemble sample mixed below the solo instruments.
Of course, this is more difficult with an orchestra. Also, there are other equally important issues like articulation, expression, etc. The doubling ensemble thing was one of my pet peeves while doing sound design.
I think instruments like Gigasampler/studio and Unity that can access HUGE samples and multiple articulations are really the way to go. Especially if you can combine them with a few live players.
I'll post a few more of my experiences later on. I'm sure other folks have some good tips on this subject.