Spitfire BBC Symphony Orchestra

[Ashermusic]

Gee Piet. I don't doubt you could do better, but "very poor"??!!!

I have to say, I would find it disappointing if I was interested in that library.

[The Saxer]

I think this Rite of Spring mockup shows the weakness of Spitfire libraries in general: the problem of never getting convincing melodic lines. It's obvious in the first bar and doesn't get better... probably... I couldn't listen through the track.
When you are used to VSL's evenness and playability the inconsistent editing of Spitfire libraries is shocking at first. But they merge much easier in full arrangements when details get buried and the sound of the hall shines. There's the weak part of VSL. And I think the Special Edition brass is unusable. Especially because of missing dynamic layers.

I fell into Spitfires hype again and thought "a truly cohesive sound" would be reflected in the samples. It didn't. The BBC brass isn't good. It's more or less boring. But they always get the percussion right.

When you are looking for brass: Soundwise I like Junkie XL Brass but I didn't have time to really work with it. The tree mics sound very natural and the dynamic is hi res. It's not a specialized epic library when ignoring the 6-trombs and 12-horns patches.
And there is also Infinite Brass. I would call it the "classical version of Samplemodeling". Dry samples with custom made impulse responses which sound really natural for my taste. No key switches and very playable. The sound in the new 1.4 update is much better than in the first versions. There is a long thread in vi-land.

[Piet De Ridder]

Gee Piet. I don't doubt you could do better, but "very poor"??!!!

Yes, Guy, very poor.

For starters, Spitifire’s BBC SO prides itself, and I think deservedly so, on its spatial opulence. Very odd and bewildering then, to hear a demo where the entire orchestra and its enfolding space are collapsed into a blurry, smeary, near-mono (!) porridge that never even hints at the spatial depth and dimensional splendour that’s captured in this library (It’s this fact that prompted me to write in the previous post that I’m convinced — and I am because I already heard it elsewhere — that the library is capable of a *much* better sound. The Spitfire people, I’m sure, must be horrified when they hear demo’s such as these. I am horrified, and I'm not even one of its creators.

Furthermore, based on previous demos, I was under the impression that the woodwinds were one of this library’s stronger assets. What I hear here however completely refutes that idea. Uninteresting pseudo-woodwinds that seem incapable of translating even the most basic MIDI-instructions into a more or less believable performance and sound. If you’re familiar with a good performance/recording of “The Sacre” and its score, you can’t but notice, and be annoyed, by the ill-defined sloppy woodwind lines, the absence of essential articulations (in some cases also the absence of notes) and an overal sense of impotence that seems to have incapacitated the entire section. I’m sure some of it is the fault of the samples (or lack thereof), but the maker of this abomination isn’t blame-less either. Far from it. Very, very bad work.

But as bad as these woodwinds are in this demo, they’re still not the worst. The brass and the strings are, and I’ve minced the words for a few minutes before writing them down: ghastly. Atrocious. Horrible. I happen to be a Sacre nut and listening to this offends me. It really does.

So yes, “very poor” is actually a kind verdict. If I were to reply in the thread on VI-C where this demo is posted,, the thing would immediately be moved to the Drama Zone.

_

[ZeeCount]

I only made it about 1 min in, but ouch. That demo has solidified that I made the right call in not picking this library up.

[Ashermusic]

I only made it about 1 min in, but ouch. That demo has solidified that I made the right call in not picking this library up.

I wouldn't necessarily make a conclusion about it based on that video if I were in the market for another orchestra library.

[Guy Rowland]

When I think "very poor" I think of Kirk Hunter or Garritan. Honestly I think these are nowhere near that level. It's not as cohesive as one would hope, but despite the shortcomings, it sounds sorta ok-for-samples to me. That's not a glowing endorsement you'd want on a poster, I realise. But - crucially - when attempting Stravinsky with samples, my expectations are fairly low to begin with.

[lofi]

Edit:

Be safe, wash hands.

/Anders

[Tanuj Tiku]

I am investing more time on samplemodeling these days and loving performance samples. I am nearly done with old school sample libraries that pride on sound and cannot even be used to stitch a decent connected musical phrase.

Many of these concepts, I find obsolete now.

There are things I can do with performance samples (single patch!) and samplemodeling that no other library can do. I am just surprised how well performance samples works in situations where it was not even designed to work as such. There is something about the sampling philosophy that is more musical.

[Ashermusic]

Man I had the opposite experience. I had The Trumpet, Trombone, and French Horn and I hated working with them and sold them.

I also hated using a breath controller.

[Tanuj Tiku]

Man I had the opposite experience. I had The Trumpet, Trombone, and French Horn and I hated working with them and sold them.

I also hated using a breath controller.

Jay, I would agree that they don't work out of the box and because they have no performance built in, you have to work in a different way from every other library but some of the results can be stunning. I do think, they need help with some layering at times - just a little bit. But, I think there are so many things that the trumpets, horns and tuba can do that nothing out there can.

My new favourite is combining Performance Samples and Samplemodeling.

also, I am a programmer and not a player unfortunately. I want to get a breath controller at some point and learn to use that. But, if you can play very well and use a breath controller, results can be even better.

Here is a quick short excerpt from Hook - 4 SM Trumpets and on instance of PS Trumpet with some envelope shaper on it to make it tighter. I thought it sounded decent. I still have to work on this and I know that the original is not an ensemble but still it sounds decent to me.



And here is another short track I did as a speed writing test - It has some Samplemodeling as back layers but mostly Performance samples brass and even the short strings front main layers are all performance samples.

Performance samples is also quick to work with and you get a lot of mileage from it.

[Ashermusic]

Man I had the opposite experience. I had The Trumpet, Trombone, and French Horn and I hated working with them and sold them.

I also hated using a breath controller.

Jay, I would agree that they don't work out of the box and because they have no performance built in, you have to work in a different way from every other library but some of the results can be stunning. I do think, they need help with some layering at times - just a little bit. But, I think there are so many things that the trumpets, horns and tuba can do that nothing out there can.

My new favourite is combining Performance Samples and Samplemodeling.

also, I am a programmer and not a player unfortunately. I want to get a breath controller at some point and learn to use that. But, if you can play very well and use a breath controller, results can be even better.

Here is a quick short excerpt from Hook - 4 SM Trumpets and on instance of PS Trumpet with some envelope shaper on it to make it tighter. I thought it sounded decent. I still have to work on this and I know that the original is not an ensemble but still it sounds decent to me.



And here is another short track I did as a speed writing test - It has some Samplemodeling as back layers but mostly Performance samples brass and even the short strings front main layers are all performance samples.

Performance samples is also quick to work with and you get a lot of mileage from it.

I don't question that excellent results are possible. However, as a composer and pianist of ....ahem.... a certain age, If I cannot work with real players, which I love, I want to somewhat enjoy the process of composing with virtual stuff. And I just hated it!

Also, hyper -realism is not even remotely my goal. I decided way back when I did the music for "Zorro" that there were things I added to real players that detracted form the "realism" but made the sound more to my liking.

And you know I have great respect for you Tanuj, truly, but those two examples both sound to me, for want of a better word, plastic, especially "Hook." They are too perfect.

Imperfection is interesting.

[Tanuj Tiku]

Jay, whole heartedly agree on all counts.

Live musicians are something else and nothing comes close. I can understand you not wanting to spend your days programming long hours!

I wish I could only work with live orchestras. Unfortunately, that is not always possible.

Having said that, if one had to spend all day programming, like I must do for a living, then I would prefer these over many other choices which have the mic positions but cannot reliably, even in a very sample like plastic way, produce connected musical phrases.

[Ashermusic]

Having said that, if one had to spend all day programming, like I must do for a living, then I would prefer these over many other choices which have the mic positions but cannot reliably, even in a very sample like plastic way, produce connected musical phrases.

I think there are a fair number of sample libraries that can produce connected musical phrases in the right hands. Roberto Sogietto seems to be able to do it with every freaking library he touches.

But listen if you prefer working with them, more power to you. I really expected to like them. I just didn't and that may say more about me than the products, but there you go.

[Lawrence]

Jay and I diverge here as well. SM instruments feel and Sound more lifelike and human than many of their sampled competition.

I’m pretty sure Roberto is a big fan, btw!

[Linos]

Ouch indeed. I agree with Piet, I find that rendition of the Rite of Spring badly done. I couldn't listen to more than the first few bars. The question is, what does that tell me about BBC SO? I think not that much at all. Because it is very easy to make the best of sample libraries sound bad. Much harder to make them sound good. Marco, the creator of said mockup, has made a version with the Spitfire Symphonic series as well. Have you heard that? In my opinion it is not very good either:



So I am inclined to disregard it, and blame it on not very apt programming.

@Saxer thanks for pointing me towards Infinite Brass. I followed the release of Aaron's libraries. When I heard the classical demos he posted on his website, I quickly lost interest. I will revisit that decision now, as you say that v1.4 sounds much better.

Generally though I am on the opposite trajectory to Tanuj. I have been using dry libraries for years. In all that time I never managed to create that natural, beautiful concert hall sound. I do appreciate the consistency and playability of VSL. I believe that with Sample Modeling musical performances are possible that no samples can reach. But all that at the cost of a natural, organic, beautiful sound.

So here's my plan: I think that using dry instruments for the foreground, solos and exposed parts, is a good idea. To really have the musicality and expressiveness in these lines. But for the background, I want beautifully recorded ambient libraries. Because there the actual musical content is less important there, but the sound much more. At the moment I am using VSL SE for these background tasks, and that is far from ideal. It just gives a flat, sterile canvas on which I try to paint the expressive lines of the musical foreground.

My question now is: which library can create a rich, beautiful orchestral background? Can give an organic, 3D concert hall sound? I don't believe anymore that dry recordings, and or modeled libraries are the best choice for this. They excel at musical performance and expressiveness, i. e. at foreground tasks. But I feel that they tend to lack in sound somewhat. I am looking for the opposite. A library that only needs to be able to master musical background tasks, but that has a beautiful classical orchestral sound.

If not the BBC SO - and I am still considering it, if not with the same enthusiasm as before, because I do like the recordings (as Piet so eloquently put it, I too have heard some 'dimensional splendour' in this library, and that is exactly what I am looking for) - what are my options? Which libraries out there do have a beautiful 3D classical orchestral sound?

[Guy Rowland]

I think we need a thread split, don't we?

And you know I have great respect for you Tanuj, truly, but those two examples both sound to me, for want of a better word, plastic, especially "Hook." They are too perfect.

Tanuj, of course they sound outstanding, and I'm sure you could make them sound slightly worse if you wanted

I've been bleating now for many years that dry samples or modelled instruments are the ultimate way forward. I remember a little while ago saying that it surely can't be impossible to get algorithms to simulate the effect of space better, by feeding in close and ambient mics to machine learning tools to see the entire effect of a room (the low brass effect being a classic example). This might also be true of artificial ensembes, with an AI engine to emulate all the subtle variances between players playing together. Much as that goes against our experience now, logically I think it must be doable - 11 violins are 11 individual violins, each making sounds from the bow / string interaction and playing with reference to the score and everyone around them. Those are 11 emulatable constituent parts in an emulatable space.

I don't know where else you can go with ambient samples other than endless variations on the same existing products. So I've always maintained that concpetually VSL have gone in entirely the wrong direction of travel with the Syncrhon series. I think they should have sunk that cash into R&D for next level realism from dry.

The challenge for dry sample modelled instruments is to make them more instantly playable in terms of ensembles. Tanuj's fanfare I suspect is the product of a lot of hours craft and skill involving chains of effects and meticulous programming, along with great composition / arrangement chops. It should be possible to have a product that gets you most of the way there sound-wise at least out of the box. And before someone cries "cheat" or "do away with talent" - this would essentially be no different to playing an ambient library out of the box right now, except it would play and sound better.

But that's for the future. Right now we have these compromises.

[Jack Weaver]

Hmmm... I rather like both of Tanuj's demos. Well crafted and lively. Perhaps a tad much ambience for me, but that is maybe what he likes to hear. You can't fault a person for what he wishes to envision.

.

[Ashermusic]

I never tell people they should or should not like what they like. I am not that arrogant.

That said, and again, my deep respect to Tanuj, if he would be willing to send me the MIDI file, (and if he isn't I totally understand) I would use Hollywood Brass Diamond with my choice of articulations, and what I think people would hear right away is that there is much more warmth and well, soulfulness, to the sound.

I do come to this with a prejudice. On my list of top 10 things that matter:
1. timbre
2. timbre
3. timbre
4. playability
5. timbre
6. timbre
7. timbre
8. timbre
9. timbre
936. Hyper-realism

Again, let me state unequivocally how much respect I have for the people here and if they disagree with me, the fault probably lies with me.

[Tanuj Tiku]

Jay, I am going to create a new thread so that we do not go off-topic. Check the new thread for more details:

viewtopic.php?f=6&p=57564#p57564

[Tanuj Tiku]

Hmmm... I rather like both of Tanuj's demos. Well crafted and lively. Perhaps a tad much ambience for me, but that is maybe what he likes to hear. You can't fault a person for what he wishes to envision.

.

Thanks Jack! In fact you may be right about the reverb. I am currently working off headphones because of the lock down so old habits return as I cannot hear so clearly! In any case, I think I like a little lush reverb sound so that could be it also.

Anyway, all back to BBC now. Sorry for the off topic adventure.

[mushanga]

Now the dust has settled - and seems to have settled rather quickly, given the panache with which Spitfire introduced the library and their plans to use it as a vehicle for community sharing and learning - are others using it apart from playz and Jaap? What are your impressions? Mostly happy, or disapointed?

I am considering it as a replacement for my VSL Special Edition. It served me well to fill out the gaps. But I am not happy with the timbre of its brass and percussion anymore. Is BBC SO a good choice to fill in the background? The quality of the recordings suggest so. It's about as comprehensive a package as VSL SE 1, so I hope it can serve the same role. Only with a sound that I like a lot better.

Hi Linos - did you ever consider cross-grading to the VSL SYNCHRON-ized Special Edition collections? According to VSL timbre/ambience is addressed by porting the samples to the Synchron Stage:

https://www.youtube.com/watch?v=naoHuOZRZSs
https://www.vsl.co.at/en/Synchronized_S ... ion_Bundle

Alternatively MIRx for your existing VI Special Edition collection:

https://www.vsl.co.at/en/MIRx/MIRx_Bundle

Would be interested to hear everyone's thoughts on how well VSL did with the above approaches.

[Linos]

Hi mushanga,

Yes, I spent an unhealthy amount of time in the past on reverb setups for VSL Special Edition and Dimension Strings. I've tried a myriad of techniques and tools, including VSL MIR/MIRx. MIR does the job, but to my ears not better than other tools/techniques. Dry sources with MIR do not sound like samples that have been recorded in a nice hall. Nor does any other tool/technique I tried. So, unfortunately, it is not the solution I am looking for.

As to the Synchronized libraries, I find that many of them sound actually worse than the original VI libraries. I don't like the added reverb and prefer my own solutions. I also much prefer the workflow of the VI Pro Player over the Synchron Player. Also, some of the Synchron-ized libraries feature less articulations than the original ones. For me, all good reasons to stick with the original VI version.