right out of the box, it can work OK but much depends on the music. The strings are often tricky (and all my mockups could do with more polishing in this area in particular) and I’ve found really best to default to “legato” which has much more variety than the rather limited “long”. This way, at their best you can get a wonderful spontaneity which is often lacking elsewhere. But they need work and I don’t think it’s particularly an issue of working in a notation environment though there is a bit of that. We may have different tastes regarding natural shaping up of phrases but I feel the strength with this library is the soundstaging – there is actually more of a feel of a real romantic symphony orchestra than in other libraries which can be more realistic on an individual instrument level but don’t quite gel as an ensemble. Most of the few classical demos I’ve come across seem muddy to the extent that I was initially very sceptical about this library but this low end muddiness in the brass above all can give a real atmosphere in the right places. Also this 9th was actually my first mock-up and it’s possible that the others are in some ways more advanced. I have a friend who hates this 9th but there are others who are impressed with what I’ve done in some of the works anyway.
I wouldn’t recommend it for clean, pure classical or baroque music – there are better options out there – but it has a place for the sort of music I like and write.
When I go from legato to sustain and the sustain is - 6db “softer” than the legato it ends up in “needs medical attention” bin for me…
But for my “digital” standard, which is (more or less) the berlin phil digital concert hall way of balancing, I do think this library has the potetial of getting close(er than VSL actually…)
Hi Pieter (I assume that’s you), thanks. I apologize for my initial comments, which sounded a bit dismissive. I assumed you didn’t do anything to it. What stood out to me right away was the flatness of the phrases. Obviously that could be fixed with a little CC work (“little”). Thanks for sharing.
Dan, maybe you were referring to me, David, as I’m not sure who Pieter is? You’re not the first to find the phrasing a bit flat in this work but it’s actually supposed to be a fairly subdued and intimate piece. Don’t worry about being dismissive – I was very dismissive of this library based on my experience of Discover but I had a gut feeling that there was something to it. If you’re still curious, I could create .mp3’s of more and put them on the cloud. If there were an abundance of competent symphonic work done by others, I wouldn’t bother but there doesn’t seem to be so I feel a bit on my own in trying to produce the mockups which probably do need more work than with most libraries and I’m really only half done so far. With VSL SE, I can work rather faster.
To fratveno – I’m looking forward to seeing what you make of it. I notice it’s on an even lower price than the summer sale when I got it! The legato can sound wonderful but they haven’t made the job of programming very easy imo.
We know John Barron (of Discover Dorico) has done Expression Maps for the BBC SO releases, right?
I realize more specific shaping would be necessary, but “out of the box” does not mean one need create EM’s from scratch.
John’s Expression Maps are the only ones I’ve used rather than creating my own from scratch as in all other libraries.They seem generally fairly sound with the proviso that it’s essential to look at how the legato and long patches relate to each other as I think some modifications will be necessary there according to personal taste and the nature of the music.
no problem at all. By the way, if you have a specific score in Dorico format you’d like to hear a mock up of (up to a few minutes), then I’d be happy to try with current knowledge to do what I can with it in BBC Core. I have no time or interest in the tedious business of producing a score from a scan. And don’t say the impressive Doricoised Rite of Spring we’ve probably all heard as not only are there too many instruments missing but it would challenge my modest PC – I tried it already once and soon gave up.
I have BBCSO and the venerable Hollywood Orchestra as my main two libraries. While I haven’t done apples to apples, out of the box BBCSO sounds better than EWHO, much of that due to it’s a much wetter library. And generally you get a decent sound untweaked, whereas EWHO takes some work to be listenable. I’m using John B’s templates, and am working up my own EWHO so will be able to do a direct comparison, but I’m expecting that BBCSO will still sound better off the bat (I think that’s the point of the wet Spitfire libraries isn’t it?) The sampling quality of BBCSO is higher I believe (East West just announced a grand update to the library though, paid of course, but they also have a replacement for Play so who knows).
But unless you go Noteperformer (which has inferior output to my ears - the samples sound too processed), why worry too much about the basic output? It’s only in the tweaking that brings out the music. Here’s a great exammple in Dorico from VSL of Pavane Pour Une Infante Defunte. You can download the file too. Of note is the extensive tempo tweaking to get rubato, the result is the music really breathes! Hard to tell over YouTube compression, but if you download you get a (OK yeah) MP3 which sounds much better. Looking in Play mode there’s really not a ton of extra work here. Of course this is VSL. I would have brought out the phrasing and expressiveness much more, this is aiming to be a neutral performance.
Ok, here’s a bit of experimentation with an orchestration I did of “O Holy Night,” BBC demos courtesy of John Barron and dko. A big thank-you to them both for sharing their expression maps and bouncing out a demo for me (I haven’t purchased BBCSO yet, so I can’t try it myself).
Xmap B uses the legato patch for the “natural” PT. Although it’s using Pro, it’s not really any different in that regard. The mic positioning is the same.
Again, this isn’t my work, and I’m appreciative to others for their help. It’s still a work in progress.
I’m on the fence about whether BBCSO is a worthwhile purchase. NP sounds more cohesive overall, with no extra work, but it simply has its limitations. As I understand it, there’s not a ton you can do to improve it. BBC, on the other hand, already sounds much better in a number of spots. I think a small amount of tweaking could really add cohesion and make it superior.
Hi Dan! Here’s some experimentation I did over the weekend. As this was the first Thanksgiving weekend in about 8 years I haven’t had at least 4 Nutcracker performances and 2 theater rehearsals (my daughter dances), I decided to test out woodwinds (and 2 French Horns) and upper strings with the first part of the Nutcracker Overture.
In all of the following, the Dorico file is completely unedited from the original, so obviously lots of articulations and dynamics are not played back the way we are used to hearing it. I’m also curious about which libraries produce the best sound with minimal tweaking. As I’m almost always writing (or copying) music that is intended for live performance, the mockup isn’t that important, but obviously getting better sound for no extra work is clearly preferable. All audio has a tiny bit of compression and if any additional reverb is added, it is with UAD Capitol Chambers. Any panning is done automatically with the library/template and not after the fact by me. Dorico mixer levels have been adjusted for rough levels, although obviously much more work would need to be done with levels.
Obviously more time, or a more experienced MIDI musician, would be able to coax more music out of this, but I was interested in testing what could be accomplished just using the stock music, rough levels, and Playback Templates and Expression Maps that were already available.
Thanks for posting that, Todd. The BBC one seems a bit off… are you using John’s expression map? Quite a few notes seemed to be incorrectly triggered, or don’t sound at all.
Yep, that’s the one I’m using. It seems to have a hard time with fast articulation, like the flute and clarinet solos at the beginning. It can probably be edited, but my goal was to basically compare libraries with no additional work. I did add the Init to my VSL expression maps that enable hairpin playback, as the ones posted on their site lack that Init setting.
I found the VSL version very quiet. I would put a VSL Multiband plugin on the Dorico mixer stereo output. There’s some decent presets in the general purpose category. As far as the one with the best feel and dynamic balancing, I’d go for Noteperformer. There’s something about Noteperformer’s forward looking algorithm that results in a better groove. Also if the dynamic markings are done correctly, Noteperformer can give quite a nice performance. In my use, the main criticism is the phasing sound you get when sections are doing block legato. You get that pumping sound. Hopefully Noteperformer will improve this in the future.
Agreed. I was surprised how good the balance was with no additional work, where I had to move the Mixer sliders around quite a bit with the others. With the winds in the middle of the sliders, I too had to crank the strings way up using VSL.
My feeling is that extreme dynamics are being triggered for some reason which is why you can barely hear some of the notes. Also there is no polyphonic legato with this library so in a passage starting bar 45, for instance, the divided first violins cannot play properly as there is no patch for the seconds to use to separate the legato and short notes. Defaulting to sustain as in John’s map instead of legato will make certain passages sound unnecessarily blurred. Otherwise, it’s not really a world away from the results out of the box with my version of John’s map, as you can hear here https://app.box.com/s/haienb7k1dg4enwgfzz6r20pscq9jbql Given about an hour’s work, something reasonably musical could be made out of it, I think.
From the examples as presented, NotePerformer is the best, not because it the instrumental realism is particularly good – it isn’t – but because it effortlessly manages to deal with things like the mixed legato/staccato writing quite effortlessly whereas both VSL and BBC require manual adjustments to work. I don’t know the source of the original. xml but most of it does seem correctly written for “real” musicians --perhaps Fred can maybe confirm. I find the VSL from around the 1 minute muddies the violins because the official VSL EM hasn’t programmed the NoteLength feature very well, completely leaving out the shortest articulations from the automation.
FredGUnn has certainly done us a service here by showing us what happens when you try things out of the tin. For those who really hate working on mockups, NotePerformer can be a wise choice!
I initially downloaded it off the MuseScore site. The person who input it originally had added quite a few editorial articulations and dynamics to affect playback, so I removed anything that wasn’t in the original score. I caught quite a few note mistakes while proofreading too, but it’s possible some more slipped through. I just wanted it to match the original score and see what was possible without any additional work.
Thanks.