This question may have been asked before…I want to purchase Dorico Pro 5, switching from a previous Finale and Sibelius user. The new Dorico Pro 5 with Iconica Sketch sounds nice however I am interested in adding the most realistic orchestral sound library to play my scores. I don’t work in DAW’s…just pure notation, printed out into scores, lead sheets and parts. However for the pleasure of hearing the music, crafting and sculpting the sound in pure orchestrational mode ( as composers of the past would ) and fine tuning and envisioning the final sound, I am in search of a Library of tradtional ( modern) orchestral library that can play in place of the Dorico ( Iconica Sketch) sounds if I wish. Which library has a full range of sounds, techniques, articulations, etc ( for woodwinds, brass, percussion, strings,) that mimic what a large scale orchestra can do and a contemprary composer can write for? In a cursory way, I’ve looked at Vienna Ensemble Pro, Spitfire audio, however i’m unsure of what to look for in terms of critical analysis. My desktop has 48 gb of memory and 1 TB SSD. Is this enough? Is working and trying to load these sounds difficult? Will it even make a vast difference in how I enjoy what I’ve written? Any help, input or advice would be greatly appreciated. Thanks
I think you are in search of the Holy Grail (which is probably always just out of reach).
If you have unlimited funds, and unlimited time to tweak, you could probably buy any of the mainstream libraries and be equally dissatisfied.
If you want a practical and flexible path, I think NotePerformer (with its NPPE extensions) is currently a high quality, and safe starting point.
But there are always interesting alternatives out there to explore. Why restrict your creativity to traditional orchestral sounds? Dorico will work them all.
Janus,
Thank you for the reply and insight. Is iconica sketch and NotePerformer very close in what they do and how they interpret? I’ve heard a lot about NotePerformer, are the sounds and playback techniques far superior to that of standard Dorico with Iconica Sketch?
Sorry. It’s all a matter of personal perception and personal expectation.
You can download a trial of NP and experiment (The NPPE Iconica sketch is a freebie, so you can check the native Dorico version and the NPPE).
I think you’re mostly on your own to figure out what works for you.
I have a 64GB/2TB SSD 8-core laptop that I use with Spitfire samples exclusively (Symphony Complete, Organ, the Hg2O waterphone, etc.) and I’m immensely pleased with it. I also wouldn’t advise anyone to go my route, since it cost me several thousand quid and untold hours of configuration and tweaking. I’ve never liked NotePerformer; I did use the EWQLSO library for a while, and that was OK, but having made the leap (partially of faith, admittedly) to Spitfire, I can’t go back.
The road to satisfaction is long, costly, and arduous; I could be wrong, but I doubt that anyone of an uncompromising bent can eschew it.
VSL libraries are some of of the best options for Dorico, as their libraries are well balanced, very consistent, and come with excellent official expression maps made by VSL themselves.
Noteperformer does two things: it produces pretty good sounds by itself; but more importantly, it also creates intelligent phrasing of connected sounds, rather than just one sample after another.
On top of this, you can now “pass” NP’s intelligent phrasing to other sample libraries for better sounds.
This includes things like Spitfire’s BBCSO, EastWest’s Hollywood ORchestra, VSL’s Synchron, etc, etc.
As a classical orchestra player myself none of the libraries come close enough to replace a real orchestra sound (for my taste Vienna Ensemble came the closest - you would need to spend heaps of money and also tweak things around a lot though).
The best sound you will probably get by imagining it…
If you’re after realism, sounds are only part of it. The element that is also needed is performance: playing those sounds in a way that brings them to life. Dorico is making good steps in that direction but at this stage nobody is doing it better than NotePerformer.
For me (and I suspect many others on here), when working with orchestral scores, NotePerformer is essential. It doesn’t matter how big or expensive sound libraries are, unless you add performance to the playback, it will always sound less than you’d like.
NotePerformer comes shipped with its own library of sounds but in their latest release (v4) you can now use other libraries. That’s a big step forward. The NotePerformer setup is relatively simple and, once you’ve set it up, it loads those sounds automatically.
If you’re after comparisons between the different libraries, at the bottom of this page you’ll hear a bunch of them, including Iconica Sketch, all playing the same piece back through Note Performer. Of course there are better, bigger, more complete libraries but those will come at a cost and inevitably they will require you to upgrade your RAM memory and add another SSD.
Both Dorico 5 and NotePerformer allow users to test them (without any restrictions) before buying.
Replying to the original post, 48gb memory is plenty to begin with. 1TB SSD, especially if that’s your only storage, is on the slim side, though enough to get started. The CPU has a role to play too. I would say that symphonic polyphony with sample libraries requires multi-threaded processing, Intel i9 or equivalent. NotePerformer is more economical in that regard.
Generally, though, I’d say that building an audio library and the competence to use it takes time and patience.
Hi
first off yes it’s the holy grail but you can get pretty close. First you MUST use Vienne Ensemble Pro to host your VST’s
Then Decent libraries are a must. Again Synchron is exceptional
Finally Painstakingly setting up the Expression maps to get the results needed
Best
e
In my opinion, this is critical and the first thing to do is figure out what you expect from playback.
A long time ago someone asked me to listen to two recordings of the same piece, made more than 50 years apart, and I realized that on balance phrasing is more important to me than the modern recording quality. I feel the same way comparing older Boston Symphony recordings of Shostakovich to recent hyper-detailed HD productions, which to my ears sound like exquisitely recorded soulless meanderings.
Playback for the purpose of checking the writing and the sonics - color, chord distribution, timbre can be achieved with whatever sample library gives you the sound you find most pleasing.
But if you want to get a specific and exact performance of every nuance of your music then you will quickly realize you’re spending a huge amount of time tweaking in the MIDI editor to get the phrasing, accents, vibrato, micro dynamics and you will become aware of the limitations of any library you have when nothing you try works. Also, certain well known libraries were released with a number of recorded mistakes that have never been corrected or took a long time to fix, and people have been forced to write to the strengths of the sample library instead.
I’m a lot happier with phrasing using modelled instruments and recording the CC data live in Dorico. With practice, I’m also spending less time tweaking things and I’m finding this more creative and therefore enjoyable than drawing curves. To my ears the result sounds like an actual performance (because it is), but there is of course a big trade off here too - you might dislike the available timbre. And it does take a big effort to set up and spatialize at first, especially if combining several libraries such as Sample Modeling, Acoustic Samples, SWAM, AV, etc
My suggestion would be to trial everything you can get your hands on, starting with Note Performer, and listening to the music you like critically, so you can understand what’s important to you before you spend a ton of money you will never get back. Whatever path you choose, it will be a compromise, so find the one you’re more willing to make.
I think the Vienna Ensemble Pro is a bit of a red herring here. It is a platform – a very useful one – but, apart from the bundled Epic Orchestra, it isn’t a sound library.
I agree. VSL libraries are consistent and the provided Expression Maps are good places to start. Using VEPro and MIR will also help. Note performer is very popular on this forum, and the recent advancements which allow it to host other libraries certainly make it more appealing, but you still cannot change an articulation/sampled sound, if you are not happy with the one that Note Performer chooses for you. Therefore, for complete flexibility it doesn’t work so well. It’s advantage is it’s ease of use, but as others have noted, if you are wanting the best possible result, I think you are better off going with individual libraries, Expresssion Maps and a lot of fiddling around in the Dorico CC lanes. VSL also offers a trial period. Spitfire, EWQL, CSS, Berlin etc are all excellent libraries, but I personally think VSL is a good choice. Which specific VSL library is another question, but you can start with an all round collection and upgrade as needed once you start hearing the results.
Another warning/clarification. A lot of libraries outside of VSL (most in fact) aren’t necessarily recorded and edited very consistently. This is because there is the assumption that the composer is doing their work in a DAW and is playing everything in and will be able to hear any balance issues and other problems as they are going along and playing all the lines in. So sample library vendors don’t necessarily feel like they have to make staccato notes match the level of sustained legato notes and things like that. But in notation software, you are often taking a completed score and loading it up and hitting play and hearing what it sounds like. If the staccato notes are twice as loud as they should be and the pizzicato half as loud as they should be, and you take these inconsistencies and multiply them across the entire orchestra, then you have a lot of manual overrides to do and manual editing of passages just to get to hear what a piece sounds like. Out of the box it will often sound so horrible and unbalanced and everything wrong all over the place that it is hard to know where to start, and most people at this point will just give up and throw their hands up in the air and use something else for playback.
These aren’t just some small vendors either that have these inconsistencies. Most big vendors have these inconsistencies and don’t care about them because it would take time and money to correct them by doing further editing of samples (which is a meticulous but boring process that has to be done by a human being). And, that would be money wasted when 99% of their target market is using a DAW and can simply hear the deficiency and correct for it as they are playing the lines in, so the library manufacturer does not feel responsible to address those issues. And if you try to raise with most of them “but in notation software, these issues become horrendously problematic” they immediately say things like “we don’t care about notation software, our products are not for notation software, use a DAW instead.” To have the time to do nice consistent editing, they would often have to raise their prices due to the amount of work all that entails, and everybody expects these libraries for cheap.
About 80% of what NotePerformer does is actually correcting this sloppy or incomplete editing from the vendors, plus simulating “agile” patches in libraries that don’t have them (agile legato patches are frequently needed to perform runs and things like that with individual notes, libraries without them can fall apart in fast passages). NotePerformer often becomes the only way to use most of these libraries in notation software well and have them sound half decent.
Other than VSL, modeled libraries are usually pretty good about this sort of thing too and I’ve also had good luck with them in Dorico, the problem is that the tone is not always there. Acoustic Samples is a newer option for modeled that seems to be a bit better about this, but I haven’t tried those yet.
This is a big reason why I went away from using key switched patches back to using an articulation on each track. Got tired of pp sustains and mf staccatos at in the same velocity range.
It’s “easy work” to balance them out, relatively speaking, but it’s a massive time drain.
Well, some libraries like Hollywood Orchestra Opus (which is way out of balance in this regard) provide gain controls independently for each articulation, so you can customize things that way without having to move away from a keyswitched patch.
But even with that, as you say, it is a lot of work, and it doesn’t always completely solve the problem. Not all articulations are necessarily recorded at the same velocity layers, so even if you balance one velocity, a different one can then be out of balance. It’s also kinda stupid that everybody who uses the library basically has to do this independently - they can’t really share their efforts with other people in a meaningful way, so everybody who has to do this ends up reinventing the same wheel over again. Really, the library vendor should be handling this better. MIDI 2.0 may give us some help though, when it is more widely adopted.
For now I’m trying to stick to libraries that are already super consistent in this way so I can do as little of this as possible.
Yes, when you mentioned this, EW was the company that immediately came to mind.
Not so, you’ll need it to run a big collection of samples
e
Ed, your example sounds ‘realistic’, but not real. I instinctively know that it’s not an orchestra playing, but a computer rendering. What it is that gives it away, I don’t know.
I dare say the time will come, though a combination of better sampling and better ‘intelligence’, along the lines of NotePerformer; but until then, I’m actually OK accepting the imperfections, and accepting that it’s a mock-up, with the emphasis on ‘mock’.
An impression of what an orchestra would sound like is, I think, what most people are happy with, rather than a ‘facsimile’.