Action Piece - Composition Walkthrough Pt.11 - Strings Mockup + Stems + Midi Files

This is the final part of the Walkthrough for WE HAVE TO SHAKE THEM OFF! It has been brought to my attention that some people don't know yet about the Tag filtering system here on Patreon. I try to tag every post properly so you can find it again over the Tag system. For example, if you want to re-read all the individual parts of this Walkthrough series you can do so by clicking on the appropriate Tag above or use this link:

https://www.patreon.com/robinhoffmann?filters[tag]=We%20Have%20to%20Shake%20Them%20Off 

After doing Video Walkthroughs through the Woodwinds and Brass Mockup for this piece, we're going through the strings today again in Text form as there is not that much to show with them that would justify a video.

Strings Mockup

A general approach I often follow as also in this piece is to layer Solo String Samples on top of Section Strings.  In this case as I also wanted to experiment a bit I layered even two sets of Solo String Samples on the section.

The Section Strings used here are Berlin Strings and the added Solo Strings are Berlin Strings First Chairs and Cinematic Studios Solo Strings.

The reason why I do this is to create some depth in the sound of the strings that you can not get with section samples alone. Even though most modern sample packages now come with several mic positions where you (as in the case of Berlin Strings) even sometimes have a close mic for the first chair of the section, it still is tricky to create that multi layered sound we know from real recordings. After all, the edit points of all samples are identical across all microphones and for instance triggering a legato transition in spite of 12 or however many players are present during the sample recording, the individuality of the players gets eliminated due to the edits of the samples that create a more uniform sound. By adding solo samples on top the slight differences in attacks and edits create in my opinion a more realisitic representation of how a real section would behave. Of course it is extra work as in most cases just copypasting the section part to the solo part would not result in a satisfying end result. Particularly when the strings are more exposed, this procedure definitely helps quite a bit in my opinion. Of course you need to take care that the different samples match in ambience and tone but with a bit of tweaking of EQs and Reverbs this is not too tricky. You also don't want to mix the solo instruments too loud into the section. If you start to hear them on their own they are already way too loud.

Another thing that really is one of my pet peeves with strings in pieces like this is the velocity of short samples. Many people think that in order to create epic and impressive strings you need to really go for the loudest available dynamic you have with your samples and then wonder why it doesn't sound real.

You need to be aware that short samples are recorded as individual events, so the players can use a lot of force and movement to create that super loud staccato in such an isolated context. However having a row of quick staccato notes, it would be unrealistic to execute them with such force. The other problem with higher velocities is that the notes automatically get longer which will in such quick succession will noticeable make them overlap into each other which again diminishes the realism.

In fact, even in real orchestra situations it might often be desirable to lower the dynamic to gain more precision.

Here's an example from the piece:

Here is this passage as it ended up in the final mockup. Even the accents don't exceed about 3/4 of the velocity range with the main notes sitting comfortably slightly above middle.

https://soundcloud.com/robin-hoffmann/violas-stacc-lower-dynamics/s-dDMhDtUDTcq

If I increase this to trigger more of the higher dynamic samples, notice how it loses its precision as well as realism:

https://soundcloud.com/robin-hoffmann/violas-stacc-higher-dynamics/s-91acESb3UO5

I mentioned already in my other parts about this mockup that an overuse of high dynamic samples often diminishes realism. Just because you have a ff-layer in your samples doesn't mean that it needs to be triggered all the time when there is a ff passage written in the score.

It might be really enlightening to listen to real orchestral recordings with a focus on these things. Always note that samples are recorded individually which allows the player to focus all their energy on this one sample, however in a line would never ever play this loud.

Another issue with string samples very often is the problem to recreate believable string runs. I'm personally not a big fan of prerecorded runs as they are usually limited in scale, range etc. but of course usually sound the most realistic. 

There are a few tricks from the old times of sampling to get a it better realism with strings which would be to use trill patches (layered with staccato) to create runs but none of these sound really believable.

There are just two of the major sample libraries that I know of that have focused on providing patches that quite believably can create "custom" runs by providing transition samples for runs: Berlin Strings and Spitfire Chamber Strings. With both libraries it is possible to create reasonably articulated and realistic sounding runs. All other big players lack that feature and you need to fall back to the above mentioned principles which personally for me are never satisfying.

Stems and Midi Files

Attached to this post you can download the section stems as audio Files as well as the Score as PDF and two different MIDI Files.

The first Midi Export is from the notation app Dorico which is my general starting point to create a Mockup and which is what I import into DAW to create a mockup from. The other Midi Export is from the DAW after I've finished the Mockup. It includes all Keyswitches and CC. It will probably be tricky to replicate the mockup based on this even if you have the same libraries as I do but due to the fact that I often have reprogrammed keyswitches it will probably only cause chaos. However it might still be interesting to have a look into it.

As you might know the next Walkthrough Series has already started and covers my entire score for HERO which of course has been recorded with real orchestra so it will not have the Mockup Part. However, in the near future there will be screencasts coming up with me actually working on a mockup and not just walking through the final result so stay tuned.