should be corrected now ... thx

Antongiulio Frulio - first I must say great job reproducing that snippet of the JNH score - that is a lot of work and you did a great job. While the WW's and brass really sound terrific - the strings (especially the fast high ones) speak to my problems when scoring using VI or any other string library out there (I have 3 other string libraries).

If I may be so bold - it just says 'samples'. I wish it weren't so but just having come off a large film project - I found myself time and time again substituting upper fast string lines with WW's WW's/mallet instruments.

For me anyway, I can only get the strings sounding right on these 'hollywood' type of cues down low and slow.

Here's to wishing that weren't so [:'(]


Rob

Hi Rob, although your opinion is perfectly valid, if I can share my own, I didn't find that at all to be the case with those strings on this particular piece - sounds great to me - in fact I was very impressed with the realism of the strings for the most part, and even so these kinds of demo's are a giant leap forward in virtual orchestral production. Congrats to Antonio. I also think strings are the hardest to get right, it seems particularly the release notes - if they don't sound right it gives the game away, and also a lot of the strings sounding real is in the mix and that's something I'm myself working very hard on improving all the time. I think it's obvious there's a lot of work in that demo and it's really top notch, great mix too, very powerful all around and good realism, dynamics and energy.

@mpower88 said:

Hi Rob, although your opinion is perfectly valid, if I can share my own, I didn't find that at all to be the case with those strings on this particular piece - sounds great to me - in fact I was very impressed with the realism of the strings for the most part, and even so these kinds of demo's are a giant leap forward in virtual orchestral production. Congrats to Antonio. I also think strings are the hardest to get right, it seems particularly the release notes - if they don't sound right it gives the game away, and also a lot of the strings sounding real is in the mix and that's something I'm myself working very hard on improving all the time. I think it's obvious there's a lot of work in that demo and it's really top notch, great mix too, very powerful all around and good realism, dynamics and energy.

Without a doubt, as I have said, this mock-up is representative of a lot of talent and hard work. Maybe I am just so sensitive to 'that sampled string sound' after literally working around the clock for 8 days to write and orchestrate 35 minutes of music.

What makes me so 'greedy' for better strings (especially the high fast stuff) is VI's WW's, Brass, and percussion. Overall - no other competitive library comes close. And maybe it is the fact that I work alot with live small string ensembles and quartets - spoiled as it were. So many 'human' variables that make live strings so ---- er, alive [8-)] Perhaps MIR will also add a little 'air' around the samples as well - could help.

I can honestly say that when working with live brass and WW's - I don't hear 'that much' of a difference with live and VI - a testament to these samples and programming.

My money (literally) is on VSL. Made this decision 3 years ago and HAVE NOT regretted it since. There are other 'modeling/sample' options on the horizon but even those early demos are not 'knocking me over completely'.

Price for Nirvana Strings (nice title I think [[;)]] ) - For the right solution I would paid dearly.

On a recent 4 min cue the project supported 'real string session'. Cost for copying, studio, musicians - $9,000. And that was a bargain. Cost is always a consideration - but for strings as discussed, I am not sure I wouldn't pay ANYTHING - this side of $2K.

If anyone can make it happen for strings it will be VSL. The next 12-18 months should prove quite interesting.



All the best,

Rob

I agree. I know what you're saying, the ww brass and percussion are really knock out rather than the strings being below par they are simply not *as good* as the others for realism *on their own and dry*, but I think this is because of the mix not VSL. Somebody wrote an interesting thread in the mixing forum about this and I think there was something in it, a modern orchestral recording has the directional mic's close to the instruments, - this is done in VSL so we ignore that stage, but what is left then is the room mics - so we use IR verb for that, but there are always a number of room mics, spillage and delay for sound to travel to different points, and I think it is this affect that often creates more air and warmth and more of a real sounding string. If you put the strings in a dry room and played them - you'd get VSL strings. So what I'm saying is that I don't think there is any inherent flaw in the recording or production of the VSL strings - they're true to their purpose and excellent at that. I think it's now down to performance, programming and more than those two in terms of texture timbre, ambience etc, the mix. I think MIR will go very far in dealing with this issue but I think there is still a lot that can be done with existing tools.

I think part of the reason the strings sound different in a real recording is because it is a larger space ensemble in terms of square meters and that is always going to generate a different response at the room mic's than a single instrument, percussion instrument or even a small ensemble like 3 trumpets for example. They cover less physical space and accoustically project a different image than 14 violin players where different players are going to hit different mics at different times.

I agree totally with that assessment. I'm finding similar problems with the trumpets. In most cases the staccato stabs, in ff form, sound wful, just like a synth. And after having heard a few film score mixes and some other dry trumpets I can say that it's totally down to the virtual room sound not being right.

MIR will definately help to solve these problems. Myself I've been trying to get the right mix template for a while. Experimenting with few and many reverbs. Recently I've come to the conclusion that using many reverbs is just not working. I tried to feed my VSL sounds through just one reverb that was setup to give good distance and a liberal amount of 'verb .. and it sounded much more natural.

Bottom line is, is that the room sound needs to be right. If that's wrong then it'll all be wrong. Roll on MIR [:D]

Jay,

I enjoyed your piccolo flute piece, I also enjoyed the piano part, that was fun.
Personally, and I remember mentioning this to you on a piano-brass piece, I think, when the piano plays loud it doesn't make much sense to me that it's that much in the background. A real piano has so much more dynamic range that any f to ff is more dramatic than sampled piano and that's what I missed in the piano part. It's like: "Ok Mr pianist, don't come nearer than 25 feet from my piccolo"! [:D]
But individually I find them great!

Antonguilio,

Very good mock up!

@Guy said:

"Ok Mr pianist, don't come nearer than 25 feet from my piccolo"!

The pianist has a nasty temper when he drinks (which is often) so the piccolo player has a restraining order. [:D]


Guy-- I liked your powerful arrangement of Chopin although it reminded more of Tchaikovsky than Chopin-- which is a good thing since I love both composers. If I had one minor nitpick it's that I personally would like a little more "air" in the mix around the brass. But this is very subjective.

The rate and quality of your output is quite impressive. Did you write some of these pieces in the past or is music just rushing through you?

Antongiulio-- I also liked your dramatic rendition of the Leviathan. Did you work from the score? Or by ear? There is so very little movie music published these days (with the one exception of John Williams).


Best,
Jay

@Guy said:

"Ok Mr pianist, don't come nearer than 25 feet from my piccolo"!

The pianist has a nasty temper when he drinks (which is often) so the piccolo player has a restraining order. [:D]


Jay

lol

Many thanks to all. Grazie.


JBacal:"Did you work from the score? Or by ear?"
By ear because the score is not available.

@mpower88 said:

I agree. I know what you're saying, the ww brass and percussion are really knock out rather than the strings being below par they are simply not *as good* as the others for realism *on their own and dry*, but I think this is because of the mix not VSL. Somebody wrote an interesting thread in the mixing forum about this and I think there was something in it, a modern orchestral recording has the directional mic's close to the instruments, - this is done in VSL so we ignore that stage, but what is left then is the room mics - so we use IR verb for that, but there are always a number of room mics, spillage and delay for sound to travel to different points, and I think it is this affect that often creates more air and warmth and more of a real sounding string. If you put the strings in a dry room and played them - you'd get VSL strings. So what I'm saying is that I don't think there is any inherent flaw in the recording or production of the VSL strings - they're true to their purpose and excellent at that. I think it's now down to performance, programming and more than those two in terms of texture timbre, ambience etc, the mix. I think MIR will go very far in dealing with this issue but I think there is still a lot that can be done with existing tools.

I think part of the reason the strings sound different in a real recording is because it is a larger space ensemble in terms of square meters and that is always going to generate a different response at the room mic's than a single instrument, percussion instrument or even a small ensemble like 3 trumpets for example. They cover less physical space and accoustically project a different image than 14 violin players where different players are going to hit different mics at different times.

I think you make good points here. Dietz - an update on MIR???? [[;)]]


Rob

@hetoreyn said:

I agree totally with that assessment. I'm finding similar problems with the trumpets. In most cases the staccato stabs, in ff form, sound wful, just like a synth. And after having heard a few film score mixes and some other dry trumpets I can say that it's totally down to the virtual room sound not being right.

MIR will definately help to solve these problems. Myself I've been trying to get the right mix template for a while. Experimenting with few and many reverbs. Recently I've come to the conclusion that using many reverbs is just not working. I tried to feed my VSL sounds through just one reverb that was setup to give good distance and a liberal amount of 'verb .. and it sounded much more natural.

Bottom line is, is that the room sound needs to be right. If that's wrong then it'll all be wrong. Roll on MIR [:D]

I have a FP set to the filter on VI - best thing I could have done in regards to the brass - to my ears, in most cases, playing the filter seems to make for a more realistic performance. It is not a static thing either. As a real brass player plays louder the timbre (or filter) opens up. Breath set to Velxfade and FP set to filter is the trick - at least for me.


Rob

If I may I'd like to say a few words on the Chopin Etude No 23.

We all know that Chopin did not write for orchestra with very little exceptions, in fact most people degrade the little orchestration he did. Chopin simply doesn't work well for the orchestra, one reason is: it works TOO WELL on the piano, so why bother. I chose to orchestrate Etude No 23, a very impressive piece when played on the piano, if you haven't heard it you should: A MASTERPIECE FOR THE PIANO. So then why orchestrate it? Simply as a challenge, an orchestral challenge, and even more with sampling, but I thought with VSL I could do it. Of course the piano version is much superior any day, there's no comparison, especially for the purist, but I never wanted to prove anything else than a personal challenge. It was extremely difficult to orchestrate this work, because if the same energy wasn't there it wouldn't work at all. When you look at the sheet music alone:

http://www.sheetmusicarchive.net/compositions_b/et25_11.pdf">http://www.sheetmusicarchive.net/compositions_b/et25_11.pdf

you can see it's not suitable for orchestration. So I had to use all the resources possible to achieve an effect similar to the piano version, which means beyond the written notes, percussion played an important role, the addition of fast repetition notes to keep the energy through out was essential, crazy runs up and down not the way chopin wrote them but to give the same effect. Of course the piano score is all there and I believe you can not alter a single note with the exception of octaves.
In a way it's a crossover between orchestration, transcription and arrangement.

I hope I was able to illuminate the reason I chose to do this.

Hi,

Your impressions "good" or "bad" would be greatly appreciated, I'm sure I speak for all 3 authors of these demos.

Thanks! [[;)]]

@mpower88 said:

I agree. I know what you're saying, the ww brass and percussion are really knock out rather than the strings being below par they are simply not *as good* as the others for realism *on their own and dry*, but I think this is because of the mix not VSL. Somebody wrote an interesting thread in the mixing forum about this and I think there was something in it, a modern orchestral recording has the directional mic's close to the instruments, - this is done in VSL so we ignore that stage, but what is left then is the room mics - so we use IR verb for that, but there are always a number of room mics, spillage and delay for sound to travel to different points, and I think it is this affect that often creates more air and warmth and more of a real sounding string. If you put the strings in a dry room and played them - you'd get VSL strings. So what I'm saying is that I don't think there is any inherent flaw in the recording or production of the VSL strings - they're true to their purpose and excellent at that. I think it's now down to performance, programming and more than those two in terms of texture timbre, ambience etc, the mix. I think MIR will go very far in dealing with this issue but I think there is still a lot that can be done with existing tools.

I think part of the reason the strings sound different in a real recording is because it is a larger space ensemble in terms of square meters and that is always going to generate a different response at the room mic's than a single instrument, percussion instrument or even a small ensemble like 3 trumpets for example. They cover less physical space and accoustically project a different image than 14 violin players where different players are going to hit different mics at different times.

Hi Miklos,

I agree. I remember doing a recording of Enigma variations at the old CTS Wembley Studios in London. It had a wonderful acoustic in it's number 1 studio.

After I had recorded a large string line up I needed a few solo cello lines. To save time we just did it straight off - a mono mic on the Cello but all the other room mics were live and were recorded on the multitrack. When comparing the close mic with all the other mic tracks (set to the same level as the original full mixes) it turned the Cello into an orchestra - a fantastic sound.

Also perhaps the delay of the sound reaching the room mics creates an "overlap" between notes adding smoothness. The biggest problem I encounter with the samples is a sort of suck out between legato chords and this to me differentiates live and sampled the most. Perhaps with delays introduced by a virtual room mic this would lessen somewhat.

Also let me take this opportunity to say what great work you are doing with Guy - some excellent mixes!

Best regards

Julian