Demo of C-68 Oboe & Orchestra

[EXer]

Didn't you cheat? It sounds like the real thing...

[Ajbbklyn]

I really dug Dan's chord selection for the string part. It adds some nice color to the melody line. The timbre of the oboe patch is almost incidental to those brooding strings. Sounds like the type of thing you'd hear in film scoring. Y'know, when the melody appears early in the story as a conventional arrangement and then is later reprised as a mood piece with different backing, in this case a lydian-mixolydian-aeolian type progression... really nice.

I guess oboe is one of those instruments that is really difficult to emulate on a synth. I think everyone would agree with that. I've only tried once, using the lowly M50's sequencer. I covered an old Left Banke song ("Pretty Ballerina"). In order to approach the sound on the original recording, I layered a solo violin sound underneath the oboe (on a separate track using the copied MIDI file). I think it came about as close as the M50 could get. It can be heard here. The oboe comes in at 1:09.

[jick]

To be honest I liked chello's track quite better, but I guess that's also because of the arrangement and sound quality..

[danmusician]

Better known as Cor Anglais Have always found that curious - a French Horn is named in English and an English Horn is named in French; go figure!

Okay - this is just for fun - here's a quick Amazing Grace with oboe and more conventional arrangement (don't have time to be as clever as danmusician who produced a very effective ostinato).

Is this a good oboe - or a bad one?

http://soundcloud.com/ngowans/oboe-test

Edited to say to danmusician - just my view, but with a talented oboist son like yours, I'd get him to play (nice father and son duo) - no keyboard will be as good as him!

Absolutely agree, except that he won't play the oboe anymore. Now that he has a wife, child and career, he doesn't have the time to devote to playing well. It is too frustrating for him to play without reaching the quality he once knew. As I said, the oboe is a cruel mistress!

BTW, some have speculated that the French were referring to the English horn as "cor angles," named for the angled crook that holds the double reed. Speculation continues that the English heard the name and mistook it for the homonym "cor anglais." Probably not true, but it's a fun story to tell music students!

[danmusician]

My son happened to come over for dinner this evening. I was telling him about this thread.

He laughed and said, "It probably doesn't sound like a real oboe because it's in tune!"

My wife's response, "I never really thought about it being an oboe sound or not. I just heard you playing and knew I liked the arrangement and I liked the sound."

[Bruce Lychee]

Better known as Cor Anglais Have always found that curious - a French Horn is named in English and an English Horn is named in French; go figure!

Okay - this is just for fun - here's a quick Amazing Grace with oboe and more conventional arrangement (don't have time to be as clever as danmusician who produced a very effective ostinato).

Is this a good oboe - or a bad one?

http://soundcloud.com/ngowans/oboe-test

Edited to say to danmusician - just my view, but with a talented oboist son like yours, I'd get him to play (nice father and son duo) - no keyboard will be as good as him!

Sounds really nice. I'm going to guess Jupiter. Tone sounds very familiar.

[burningbusch]

I would guess the Jupiter as well. The vibrato is too strong (I know it maybe fun to play with it like that but it's not realistic unless you're some kind of Kenny G of oboists) and it's too perfect. I assume you can pull back on the vibrato. Still it's a very good tone and it works.

Here's a very quick example using VSL legato played in real-time. If I was going to create the phrase for real I would spend more time with different articulations for realism. But you get to hear the VSL legato, which IMO, is unrivaled. The vibrato in this example is real.

VSL Oboe

Busch.

[danmusician]

I would guess the Jupiter as well. The vibrato is too strong (I know it maybe fun to play with it like that but it's not realistic unless you're some kind of Kenny G of oboists) and it's too perfect. I assume you can pull back on the vibrato. Still it's a very good tone and it works.

Here's a very quick example using VSL legato played in real-time. If I was going to create the phrase for real I would spend more time with different articulations for realism. But you get to hear the VSL legato, which IMO, is unrivaled. The vibrato in this example is real.

VSL Oboe

Busch.

Nice!

[maestrophillips]

Hi Dan (and all),

The key to effective "realistic" orchestral playing from a sound source (keyboard, computer, etc.) other than the actual instrument, lies in so many areas. Obviously, the samples should be top quality, not garbage GM stuff. Kronos has top quality sound samples. That is always the starting point. If the sounds suck, throw the board away (ha, ha).

Next, articulations. As it was stated in an earlier comment, the embouchure of a wind player does so many things to make a performance real. That has to be accounted for in the articulations for the sounds.

Then, velocities. Numerically, velocity layers range from 0 (silent) to 127 (fff). Kontakt does a good job of showing these layers on one of its screens. The more velocity layers that a program offers, the better. Sidebar, the Kronos pianos has 8 (ish) velocity layers. I forget how many Ivory has, but it is in the same range as the Kronos pianos (I think...I am going for memory here...). For recording, I use the Vienna Imperial Piano, which has to 1,200 samples per key and 100 velocity layers, compared to 8 (ish). The end result: more ability for volumetric nuances.

After that comes expression. THIS IS SLIGHTLY DIFFERENT THAN VOLUME...SLIGHTLY. Yes, expression gives you louds and softs like volume. However, the coolest difference is that with expression, you can play a "loud" expression with a piano (p) sample and get a different tone color than a "loud" expression with a forte (f) sample. Both are loud, but both sound tonally different. Give you an example: Imaging a French Horn player playing very hard (volume wise), but sounding muted. Compare that sound with the same French Horn player playing very hard (volume wise), but really opening up the mouth to sound full...and even going to the point of sounding brassy. Different tone color, both loud...that is where expression can lead you.

All of these things come into play when it comes to making orchestral instruments (in this discussion, oboe) sounding realistic. In a recording sense, that would mean multiple articulations/ volume CC, expression CC, etc. layered together in different ways to capture everything. In a live playing setting, it can be done, but it requires so much more work than just pulling up an oboe patch and playing. It requires CC numbers, multiple patches, etc. etc.

I hope this is helpful and was not confusing.

Peace!

[burningbusch]

Then, velocities. Numerically, velocity layers range from 0 (silent) to 127 (fff). Kontakt does a good job of showing these layers on one of its screens. The more velocity layers that a program offers, the better. Sidebar, the Kronos pianos has 8 (ish) velocity layers. I forget how many Ivory has, but it is in the same range as the Kronos pianos (I think...I am going for memory here...). For recording, I use the Vienna Imperial Piano, which has to 1,200 samples per key and 100 velocity layers, compared to 8 (ish). The end result: more ability for volumetric nuances.

I agree with what you posted with the exception of the above, at least as it relates to most orchestral instruments. Of the orchestral libraries I own I don't think any have more than three or four velocities for winds/brass/strings. These libraries give you all you need to create phrases, which can sound phenomenal, but they are not really designed to be played in realtime using the velocity of the keyboard (not if you want to do it right). I disagree with Korg (and Yamaha and Roland for that matter) with regards to many multi-velocity orchestral patches they include. It sounds wrong to me to be playing and have the timbre jump all over the place as you play the phrase, which is pretty much unavoidable with these kinds of patches.

My preferred method is to turn off velocity sensing from the keyboard all together and to use instead a controller for expression (breath, foot, mod wheel). You want to simulate the column of air or bowing of the string as it affects the PHRASE, not just the note. If you have velocity switching going on in conjunction with an expression controller, the two are likely to be fighting each other. It doesn't matter how hard the sax player presses the key, it matters how hard he blows.

I created a batch of wind/brass programs for the Kronos and I believe all of them use a single velocity layer with filtering and other things to create basic expression. They also use mono-legato which definitely helps to make them more playable. The VSL oboe line that I create above uses a single velocity layer with filtering. You apply a controller of your liking for expression.

Busch.

[aron]

Nice sounds guys! What do these patches sound like without all of the wash of reverb? That's one thing I really noticed with the Jupiter 80's patches - they had a lot of reverb.

Both recordings sound very musical!

[Bruce Lychee]

I disagree with Korg (and Yamaha and Roland for that matter) with regards to many multi-velocity orchestral patches they include. It sounds wrong to me to be playing and have the timbre jump all over the place as you play the phrase, which is pretty much unavoidable with these kinds of patches.

My preferred method is to turn off velocity sensing from the keyboard all together and to use instead a controller for expression (breath, foot, mod wheel). You want to simulate the column of air or bowing of the string as it affects the PHRASE, not just the note. If you have velocity switching going on in conjunction with an expression controller, the two are likely to be fighting each other. It doesn't matter how hard the sax player presses the key, it matters how hard he blows.

SN sounds have no velocity switching, making for a very different playing experience than typical sample based libraries. If you have good control over the keyboard, you can create very dynamic phrases without the noticeable layers you might otherwise experience. You won't get the full blown articulations you will with libraries like Hollywood Strings Diamond, but you do get quality sounds that can be used in a very expressive way without using any controllers. That, in turn, allows you to create layered sounds that have mulitple articulations going on at the same time. Very useful for putting ideas together, but also for covering other instruments in live situations.

I really like my software libraries for programming, but I'm really impressed with how Roland has used physical and behavioral modeling to create quality sounds that are highly usable for the keyboard player.

[maestrophillips]

Then, velocities. Numerically, velocity layers range from 0 (silent) to 127 (fff). Kontakt does a good job of showing these layers on one of its screens. The more velocity layers that a program offers, the better. Sidebar, the Kronos pianos has 8 (ish) velocity layers. I forget how many Ivory has, but it is in the same range as the Kronos pianos (I think...I am going for memory here...). For recording, I use the Vienna Imperial Piano, which has to 1,200 samples per key and 100 velocity layers, compared to 8 (ish). The end result: more ability for volumetric nuances.

I agree with what you posted with the exception of the above, at least as it relates to most orchestral instruments. Of the orchestral libraries I own I don't think any have more than three or four velocities for winds/brass/strings. These libraries give you all you need to create phrases, which can sound phenomenal, but they are not really designed to be played in realtime using the velocity of the keyboard (not if you want to do it right). I disagree with Korg (and Yamaha and Roland for that matter) with regards to many multi-velocity orchestral patches they include. It sounds wrong to me to be playing and have the timbre jump all over the place as you play the phrase, which is pretty much unavoidable with these kinds of patches.

My preferred method is to turn off velocity sensing from the keyboard all together and to use instead a controller for expression (breath, foot, mod wheel). You want to simulate the column of air or bowing of the string as it affects the PHRASE, not just the note. If you have velocity switching going on in conjunction with an expression controller, the two are likely to be fighting each other. It doesn't matter how hard the sax player presses the key, it matters how hard he blows.

I created a batch of wind/brass programs for the Kronos and I believe all of them use a single velocity layer with filtering and other things to create basic expression. They also use mono-legato which definitely helps to make them more playable. The VSL oboe line that I create above uses a single velocity layer with filtering. You apply a controller of your liking for expression.

Busch.

I will have to kindly disagree with you. For grins, I just quickly pulled up Symphonic Orchestra "Play Edition" in my DAW Reaper (I was not in the Pro Tools mood tonight). I placed a single instrument short staccato Oboe sample on a track and recorded a lot of successive notes on the Midi grid. I then adjusted the velocities so that each note would stair step up from 1 to 127. I kept the volume and the expression the same. Conclusion: I could hear the change in velocity on EVERY note. Obviously, the closer note changes would not be as dramatic, but you could clearly hear the change. You are right, there are some libraries that are slack in this area....but not all sample libraries. Later, I may do this same thing in my Vienna stuff. That is where the great adage comes in: You get what you pay for (ha, ha).

You are also correct when you talked about the fact that these sample libraries are not really designed for realtime playback. I simply brought that aspect into the discussion because the original point of this thread was addressing realistic playing of an oboe. Since musicians use workstations, such as the Kronos, for live play and composition, they have to be able to distinguish between the two areas and learn how to flow in those areas for their individual situation. If one wants to use the Oboe in a live setting, they can either, 1) expect it not to sound realistic or 2) make it sound realistic. If they choose #2 (which they better - ha, ha), then they have to understand what they have at their disposal within the unit. That is were the multi-velocity orchestral patches can come into play.

When you said that you disagree with Korg (and Yamaha and Roland for that matter) with regards to many multi-velocity orchestral patches they include, they are recognizing that many of their users do not use big sample libraries, but solely use their workstation. If they are going to create units that will assist in making the music sound realistic, then they have to have those velocities. The problem becomes that users are clueless on how to effectively use them to build orchestrations. The tools are there (although, in my opinion, a bit limited), but people just have to know what they have.

If they do not want to go this route and prefer to use just a single patch and just play away, that is cool too. Just know that something will always be missing when we listen.

Also, I do happen to like the "concept" of key swtiching. I have not done it much on the Korg. I also have a Open Labs Neko and I use key switching in my VST's all the time and it works well for me. Like you mentioned, breath controllers are cool too. All of the methods can work....IF the end user knows what they are doing.

Okay, enough rambling from me...

Peace.

[burningbusch]

Pal,

Here's some screen shots from various libraries.


This is the full Vienna Symphonic Library Solo Strings I DVD ($940). You're looking at the solo violin staccato. See where it says 4 velocity layers?


This is the full Appassionata Strings I DVD ($1020). You're looking at the violin sustain. See where it says 4 velocity layers? Notice the number of samples.


This is the LASS Strings. You're looking at the violin section sustain. It says 3 layers at the top. Below you see four layers but the ppp does not have samples (for whatever reason)


This is the LASS Strings. You're looking at the violin section sordino. It has two layers p and f plus release.


This is the main Trumpet Section Sustain from Project SAM Orchestral Brass Classic. You'll see four velocity layers 0-59, 60-89, 90-112 and 113-127 plus release (RT).

I could go on and on with these. I don't think I found any over four velocity layers.

Finally here's the East West oboe stac. You can clearly hear three velocity layers (advancing 5 MIDI velocity per note)

EW Oboe Stac

Busch.

[burningbusch]

I then adjusted the velocities so that each note would stair step up from 1 to 127. I kept the volume and the expression the same. Conclusion: I could hear the change in velocity on EVERY note.

That's how MIDI velocity works. The synth/sample engine increases amplitude with each increase in velocity received. It has nothing to do with sampled velocity layers. A single sample would respond the same way.

Busch.