What's new

How Concerned With "Realism" Are You?

Chr!s

Active Member
Something I've always found myself at odds with when it comes to sample mockups, is that, even with some of the "high-end" libraries, I find myself writing something that the instrument in question would have no trouble with in the hands of decent live musician, but the samples (being just snapshots you stitch together) struggle to play convincingly. You can just never have enough samples, both of the notes themselves and the space in between, that result in a truly-convincing sequence of certain passages. This is, without a doubt, most noticeable in trying to create lively, exciting pieces of music with lots of different rhythmic denominations.

So I find that if a passage is sounding more "synthy" these days, despite my best efforts of using different articulations and such, I just say "to hell with it" and leave it that way because the composition just sounds better that way, instead of sacrificing musical ideas because it stops sounding "real".

An example: This lil' ditty was done by Noteperformer according to the composer.



I don't think that too many who are in the know would mistake that for a live recording, but I somehow doubt most listeners really give a shit, and I suspect that if the composer were to present this demo to a client before actually recording it live, it would be enough to convince said client.

How about you? Just stick to writing a good piece, or is sacrificing ideas justified if it's better-suited to the samples?
 
In my work, it's pretty critical as I tend to get hired as the "alternative" to recording a real orchestra. (which kinda makes me want to want to laugh and cry at the same time...) So I tend to put a lot of emphasis on realism. I still end up with only okay results but so far directors are happy. The good news is that not only are libraries getting better, notation software is getting more sophisticated in their playback. This is important to my workflow as I tend to start with pen and paper rather than DAW when it comes to orchestra. So having more sophisticated playback technologies in Sibelius and Dorico (I hate Finale) lets me stay in a composing mode and get pretty realistic using traditional methods of indicating dynamics and playing techniques/articulations. Once all that is done, my composing process is usually done and I can then focus on engineering the cue inside Cubase which includes selecting different/additional libraries, fine tuning notes and MIDI CC's, applying reverbs and panning, etc.
 
Realism -- hmm. Almost forgot what that is....

I don't know what the assignment was for the OP's video, or how much time the composer had or his other constraints, but that recording sounds dated in every conceivable way, including the composition style that harkens to yesteryear.

Leaving aside the style, though (especially since that is very likely exactly what the client wanted), if you have time you can sequence something that sounds much more realistic than that, with fairly basic tools and samples.

When it comes to media music, the demand for hype and excitement has veiled "realism" to the point where it's like a fabled long-ago land. While there are exceptions (notable and excellent ones), the aesthetic has changed in general so that people expect an amazingly synthetic sound, full of electronic pads and pulses, super-human drum and sound design "events," and other bits and bobs that used to be the province only of trailers or maybe sci-fi.

Is This What You're Asking?

I love to work with a real orchestra, so I try to keep musical integrity when I use samples in the hope or expectation that some or all of the sequence will be replaced. When I say "musical integrity," I mean that I make the parts playable, compose material that can be recorded in the breakneck pace that is required, and don't ask for 1,000 string players or something.

But arriving at the point where someone opens a checkbook to pay musicians and an engineer and all that seems to require an astonishingly accurate and painstaking process of an electronic version. And that only sounds good if the samples sound "right" in the musical context you have chosen.

Since, as the OP points out, samples are not always cooperative, I have amassed a huge number of choices so that I can jump from one library to another to get just the right thing at any given time. Otherwise, if you have a meagre palette, you risk being constrained to what that particular string or brass library can execute; since I don't want that constraint, I've ended up with a huge set of resources to get around it.

Not sure if I'm addressing the question so I'll stop there.
 
Last edited:
Do you mean realism as far as how something is composed and orchestrated? If I'm writing for anything other than synthesizers, then I absolutely care about respecting the abilities and limits of the instruments and instrumentalists that would be involved, if I had access to them outside of the virtual world.

A more general sonic realism isn't something I worry a *ton* about though. My biggest concern is that things sound right enough that most people are not distracted from the music itself. Given the tools we have these days, it's not hard to get at least that far.
 
Do you mean realism as far as how something is composed and orchestrated?

I mean realism as in the composition and orchestration is all done well, but the instruments themselves do not sound like a live performance.
 
Why limit yourself to what samples can do? I've started writing in Sibelius before going to DAW, and that opened my eyes/ears to how poor samples still are compared to the real thing.

Game of Thrones has some shocking midi moments (and I wouldn't say it's because the samples aren't up to the task) - if they can, you can.
 
I've started writing in Sibelius before going to DAW, and that opened my eyes/ears to how poor samples still are compared to the real thing.

Low-key was hoping someone else would say this ;). I start almost always with notation, and then when I move to the DAW to make a mockup of it, many times I'll look at the sheet music...look back at the DAW and then swear at the screen. :roflmao:
 
@Chr!s what a great question. I am a hobbyist composer, but I have experienced exactly what you are talking about. In the end, I will sacrifice the perfect sound for staying true to the composition. But the only person I have to please is myself. I have hope that someday, perhaps when I am dead, some orchestra might play some of my pieces.

The only samples I have found that can do anything and everything I can write are VSL. But of course, a lot of folks do not like the sound of VSL, although I am getting better at using them effectively.

I like the piece you posted very much. I do not relate at all to the modern synth and drum heavy style. I would have been thrilled to have written "Eastern Skies."
 
In my work, it's pretty critical as I tend to get hired as the "alternative" to recording a real orchestra. (which kinda makes me want to want to laugh and cry at the same time...) So I tend to put a lot of emphasis on realism. I still end up with only okay results but so far directors are happy. The good news is that not only are libraries getting better, notation software is getting more sophisticated in their playback. This is important to my workflow as I tend to start with pen and paper rather than DAW when it comes to orchestra. So having more sophisticated playback technologies in Sibelius and Dorico (I hate Finale) lets me stay in a composing mode and get pretty realistic using traditional methods of indicating dynamics and playing techniques/articulations. Once all that is done, my composing process is usually done and I can then focus on engineering the cue inside Cubase which includes selecting different/additional libraries, fine tuning notes and MIDI CC's, applying reverbs and panning, etc.

I would really like to hear your work. It sounds like you must be really great at midi-performance. Why not post a link to your Soundcloud page or something?
 
Realism like maybe a 3/10

I usually aim for it to sound good first and foremost. To get across what you want to say musically always trumps how realistic it is to play with real instruments.

I will lean on certain sounds like the orchestra to use their contextual built in emotional representations, so if making it more realistic makes the music sound better then I will go there but most of the time taking an instrument out of its range or holding a note longer than possible doesnt hurt the actual music so it doesnt bother me.

-DJ
 
I write to the samples I have, and my ability to make them sound decent and believable, which means I tend to compromise to make that happen. I'm not always happy with that arrangement, but I rarely get involved with real orchestras; although that's starting to change with some of my recent and upcoming work... maybe this will change my perspective!
 
First and foremost, I'm trying to convey something interesting and enjoyable through musical language, imagination, emotions. It has to be an enjoyable piece of music - not necessarily something that does its best to create the impression that it was performed by orchestra XY in the Hall so-and-so. That's absolutely secondary.

I do spend a lot of time on massaging and micro-tuning performances. Samples are terrible for the most part. That's just part of the process. But the idea isn't so much to create a perfect impression of something, but to make sure that the listener never thinks about the whole thing. So it's more avoiding obvious giveaways and things that stick out like a sore thumb. I think that over 90% of all listeners only ever think about "realism" when something really sounds off and bad and makes it jarringly obvious that trickery is at work. I' m 100% with Daniel James here - "To get across what you want to say musically always trumps how realistic it is to play with real instruments."

I try to make it sound cohesive and sonorous - sufficiently so to let the music do the talking.

Which in practice does end up in writing to the samples a lot. It's the old Marshall McLuhan adage - the medium is the message, right? And there's things I'll just stay away from because they just can't be done in a musically pleasing way with samples.
 
I'm not very "concerned", but as a noob I welcome limitations that help "guide" me through the process, so I'm trying to be "mindfull" of realism and do try to avoid writing things that couldn't be performed that way. I'm sure I fail in many places, but I hope on the bottom line it helps me write better music anyway.
 
In coming from decades of orchestra performance to the DAW (and these forums), one of several terms that still don't compute is "realism". There seem to be 2 camps here, one supporting realism of tone colour as in "that sounds like a real violin" and the other supporting realism of performance as in "that sounds like a really fine musician playing the violin"

From my experience so far, libraries that support the "real violin" philosophy produce wonderful sounds at the expense of flexibility of phrasing etc, with the result something like a Strad being played by a gifted technician but a hack musician. The few times I have tried to point this out, it has been met with just a wee but of outrage.

The "real musician" philosophy tries to provide the tools for flexibility of phrasing, with the availability of complex programming of many aspects. This is often attained at the sacrifice of some tone color. When people use this type of library, the tone colour camp will criticize the performance as sounding 'synthy', 'unreal', or 'unnatural' no matter the quality of the actual performance.

We all must play to our audiences, especially those here actually trying to make a living in the field. However your audience is not necessarily my audience.

Now to actually address the OP: I much prefer realism of performance over realism of tone. I find it incredibly frustrating to have a great sounding instrument on which I cannot turn a phrase as I hear it no matter how deeply I program. My audience are my former symphonic colleagues. I have yet to hear "that (such & such) doesn't sound real", but I have heard "can you do something with that vibrato?" or "would you really want me to phrase it like that?"
 
In coming from decades of orchestra performance to the DAW (and these forums), one of several terms that still don't compute is "realism". There seem to be 2 camps here, one supporting realism of tone colour as in "that sounds like a real violin" and the other supporting realism of performance as in "that sounds like a really fine musician playing the violin"

From my experience so far, libraries that support the "real violin" philosophy produce wonderful sounds at the expense of flexibility of phrasing etc, with the result something like a Strad being played by a gifted technician but a hack musician. The few times I have tried to point this out, it has been met with just a wee but of outrage.

The "real musician" philosophy tries to provide the tools for flexibility of phrasing, with the availability of complex programming of many aspects. This is often attained at the sacrifice of some tone color. When people use this type of library, the tone colour camp will criticize the performance as sounding 'synthy', 'unreal', or 'unnatural' no matter the quality of the actual performance.

We all must play to our audiences, especially those here actually trying to make a living in the field. However your audience is not necessarily my audience.

Now to actually address the OP: I much prefer realism of performance over realism of tone. I find it incredibly frustrating to have a great sounding instrument on which I cannot turn a phrase as I hear it no matter how deeply I program. My audience are my former symphonic colleagues. I have yet to hear "that (such & such) doesn't sound real", but I have heard "can you do something with that vibrato?" or "would you really want me to phrase it like that?"
my experience exactly
 
Most of the time if the orchestra is used it will obviously sound the best when it sounds realistic - so I'm mostly very concerned with making it sound realistic. Unless it sounds the best when it doesn't - which also happens, sometimes IMO in electronic or trailer music for example.

But I'm not concerned with whether it IS "realistic" or not - basically whether it can be played or not, or whether it would sound the way it does in an actual ensemble...
Pizzicato runs probably cannot be played. But if I want one I'll do it. And i'll try to make it sound realistic - even though it could not be played.

And if a sample library can't pull of something I have in mind I'll think deeply and try a lot of mad things with the libraries I have... if it really won't work I'll stop working on that track (if I can) and wait till I have a library that can pull it off well enough. Even if it takes years.
 
It just depends on what you're trying to do. I am one who still believes the "Symphony Orchestra" is an important part of western civilization and culture that needs to be preserved, so I am extremely interested in realism and in ensuring that my own writing will be true to what is possible with a real Symphony Orchestra.

I feel that its not at all trivial to get samples to sound realistic to what is written on a score. So I feel the score should take precedent in that regard, even if the sample playback is not quite as realistic sounding as I know it will or would be when a real symphony orchestra performed it. That being said, obviously I want to try to mockup something that sounds as close to realistic as possible, both for the personal enjoyment of hearing it as well as in some way to ensure that the actual score is a realistic one and that when a real orchestra ever performs, it will in fact sound as I intended. If I ever had to create a professional mockup of a score intended for real orchestra, my expectation would be that it would NOT be quite exactly realistic, and might not even sound "good" compared to the final expected real score from a real orchestra. This could be confusing if you are trying to convey an idea to say, a film director, ahead of time before he allocates $100k to record it with a real orchestra. So being as realistic as possible for that task is important, but truthfully the amount of work required to get a realistic mockup like that is very high. Not as high as it used to be, but still quite a lot of work and time involved to closely match what a real orchestra will do with a given score.

All that being said, we live in an electronic media age. Films and such do not have any such requirement to have realistic symphony orchestra sound. That is only a tradition because 50 years ago it was the only way to get a big sound. But now there are lots of ways with samples and synths and all manner of sound creation to effectively make cool sounding music...music that enhances film and is wonderful. Or if you're not into film and just want to make recorded sound for others to hear, then same thing. In that regard, it doesn't need to be realistic at all. In fact I think when your mindset is to just make it sound big and to hell with realism, it will sound "better", use the tools for what they are and just make it sound "good", without regards to symphonic realism. HZ pretty much paved the way for that. And I think there is absolutely nothing wrong with that either, and in fact that is the direction Hollywood has gone, because its just a lot cheaper to produce that way and the viewing public has accepted that.

But back to my previous point, if you have any affection for the true Symphonic Orchestra tradition, then that is a separate and different matter.
 
Top Bottom