EWQL/SO vs VSL woodwinds

[michkhol]

Thank you for your effort! I really appreciate that.

Thirdly, it was difficult to work with the project because Apple insists on only providing window resizing hooks in the lower right corner, and that was completely inaccessible to me as you can't drag a window above the Apple menu bar. This is a HUGE window in this project.

Sorry for that. I should have made the window smaller. It fits nicely on my 1680x1050 monitor though.

Finally, I am assuming by glitches, that you are referring to the lack of smoothness when transitioning between notes that are large intervals apart (close to an octave, in a couple of cases). I have experienced this as well, and due to expediency have usually switched to another articulation in those cases.

Not quite. In the oboe part in the second phrase the fifth note sounds much quieter than the rest and with a click. The first phrase is given for the contrast. In the flute II part quite a few transitions have noticeable clicks.
VSL states that it covers transitions up to an octave apart in its legato patches. In the project the notes under suspicion rarely exceed a fifth as far as I remember.

I tried different note lengths, both with gaps and with overlaps (you used neither, and set the notes to begin when the previous note ends, which with most libraries does not guarantee legato but probably doesn't matter so much with Vienna due to the dedicated Legato setups).

I found in some cases that note overlaps sound worse with VSL, especially on legato patches. The algorithm seems to track the interval between notes and engages automatically when it is below some threshold (50ms?). So it makes no sense to overlap them and you have to put the notes significantly apart from each other to avoid triggering the legato.

If I sped up the Conductor track, the glitches went away. I didn't try a lot of numbers, but even at 150% of your very slow tempi, the glitches disappeared. They also became less noticeable if I lowered the Attack velocity of the high note in the large intervals (something a real player would tend to do anyway).

Obviously I cannot change the tempo of the piece but it gives me the idea that maybe legato patches are not meant to be used at slow tempi. I could use a sustain patch but it brings another problem with the VSL oboe: the sustain patch has different timbre and different number of layers so it is tricky to use it along with the legato patch.

So I initially thought maybe the MIDI was non-ideal, as you used precisely the same note lengths and note velocities for every note. I also tried different Note Off velocities vs. the zeroed values in your file. I couldn't discern a difference, so probably VSL doesn't make use of that increasingly irrelevant MIDI CC.

By MIDI specification Note Off velocity specify the release time which is often irrelevant to the the instrument played. I also noticed that legato patch sounds the best if the same velocity is used apparently due to the fact that the layer crossfade stays constant. BTW VSL recommends to automate layer crossfades instead of using velocity.

BTW your French Oboe from SE corresponds to Oboe II in the regular edition. I had to figure that out by ear due to my computer's restrictions on launching PDF's or browsers while running DP. The Oboe II has a much sweeter tone to my ears than the Oboe I, so it's a shame that throwing your nice super-slow melodic line at it isn't a slam dunk.

So it seems like it makes little sense to upgrade to the full woodwinds library then. VSL is obviously having its peculiarities that one has to live with.

Note that the Matrix patch that I used was the Legato-Interval-Speed patch. But it behaved almost the same as the single patch for Legato, which isn't surprising due to the slow tempi of the material.

Again it confirms that perhaps the legato patches are not for slow tempi.

Thanks again for your time and effort, it was a great help! It clearly looks like the difference between SE and full VSL is only in number of articulations and not in the quality of the material and software.

Thank you and Happy New Year!

[mhschmieder]

I'm not the most expert VSL user on this forum by any means, but nevertheless I think the conclusions drawn are fairly certain ones.

I'm sure you realise I wasn't suggesting you change your tempi or anything else; I was simply taking a scientific approach to see how VSL reacts to different factors.

FWIW, the only instruments I have used so far that I have felt have fully smooth legato in almost all contexts, are the ones from Sample Modeling.

That doesn't help you, as so far they have only done trumpet and sax.

I don't have the time for all the stuff that VSL's chief proponents do to make those demos sound so good. I was hesitant to buy into the system as a result, but once I got access to the full articulation selection, I found that MOST of my tracks come out quite well with VSL now.

With the exception of slow legato with seventh, octave, ninth intervals, etc. -- which in my own writing is quite common.

I am sure this problem can be overcome, but at what cost?

I personally feel VSL needs to put more effort into the usability of their libraries. Others are catching up to them, and competition might force the issue. But their main focus right now seems to be on infrastructure, tools, and plug-ins, so who knows?

I suspect they have a large team though, based on what all they are working on. So maybe once VE PRO and MIR settle down, and they get the Choirs library out, they'll think about a V2 of Vienna Instruments.

[jroadrage]

I personally feel VSL needs to put more effort into the usability of their libraries. Others are catching up to them, and competition might force the issue. But their main focus right now seems to be on infrastructure, tools, and plug-ins, so who knows?

I suspect they have a large team though, based on what all they are working on. So maybe once VE PRO and MIR settle down, and they get the Choirs library out, they'll think about a V2 of Vienna Instruments.

I think you're right, Vienna Instruments is by far the most stable piece of software in my studio and there's things that I like about it but it really isn't the most intuitive or efficient way to work. I think the next incarnation of VSL (or I hope) is going to be a hybrid of all of their deep sampling with the sort of real time programmability of something like the Sample Modeling team or even LASS.

[Shooshie]

I tried Wallander, but I didn't like the interface and frankly it didn't sound convincing.

I suspect that those two clauses are related: didn't like the interface and didn't sound convincing. If you use it to its fullest, it's very convincing. I'm a woodwind performer with a pretty high level of skill, I'd say, and Wallander is literally the first and only VI that gives me the feeling of playing the real thing. If timbre is your focus, then all you have to do is change it to your liking. The controls are there.

Something that some users may not realize is that most woodwinds have notes that stick out or drop a little, with changes of timbre all over the place. Players strive to eliminate those differences as much as possible, but they're still there. Jazz players often exploit those differences via alternate fingerings, creating yet another expressive element in a solo. Realistic instruments will include these variations.

The VSL oboe I heard in that example sounded like a college student, not an orchestral pro, but that's a very subjective thing. Some people like that oboe sound. Oboists are very polarized over topics like this, as it's very hard to get the flowing French-American sound (from the Marcel Tabuteau / John DeLancie lineage) without the British "duck-call" sound (Evelyn Rothwell [Lady Barbarolli] and followers). Germans and other northern Europeans seem to prefer a darker sound with fewer overtones (Heinz Holliger), which leaves me unfulfilled, but which is easier to accomplish, thus requiring less time on reeds than, say, De Lancie's sound. The VSL sound is closer to that northern sound. Wallander is capable of getting all three sounds.

VSL has the edge on the flutes, but I'd sacrifice the timbre somewhat for the ease of performance with Wallander. Since Wallander revised his flutes (for the 3rd time), that sound difference is minimal. As for bassoons, Wallander knocked a home run out of the park on his first try. It sounds like a Heckel Bassoon with a good reed even with no adjusting. His clarinets are also superb. In fact, I'd say that the original flute was the weakest of Wallander's entire line, but it has been greatly improved since his original.

What you have to remember about Wallander's instruments is that they are not really samples. Thus, it is important to have the controls to make them sound as you want them to sound. They are based on samples, but allow a wide range of possible sounds and response curves. Anyone who judges it only by the sounds in the demo will almost certainly disappointed in one or another sound, while anyone who buys them and adjusts the sounds a bit will be very much rewarded. That is the unfortunate problem with the way Arne has chosen to market his products. He thinks people will see the disabled controls in the demo and know what they can do with them, in spite of not being able to hear them. Very few people can imagine that, first of all because they don't know what kind of programmer he is. (excellent) So, it's taking a chance, but it was the best chance I ever gambled on.

Shooshie

[mhschmieder]

Shooshie, I'm looking forward to receiving my set of Saxophones Volume 2 when it arrives next week. And I didn't realise the flutes are on their third revision by now.

Perhaps as much so as Pianoteq, and more so than Sample Modeling, the Wallander products hold out hope for continual evolution and exponential growth in features and improvements. I just wish he had stuck with his original plan of lower-cost piecemeal purchases.

I'll keep that in mind about his bassoons, when next I return to my French Impressionist inspired suite (which uses LOTS of oboe and bassoon family members). Right now my focus is on Broadway-style orchestrations that are a bit jazz-inspired and don't use the double reeds at all (except occasionally bassoon for comedic effect).

[mhschmieder]

Timing can be funny. A high school friend who now lives in L.A., just came out with her first jazz album (she mostly plays and teaches classical music), which might be the first one to ever feature English Horn and Oboe (filling a similar role to Soprano Sax to my ears). I mentioned to her that I'm using VSL for mock-ups, and after recovering her breath, she gave me an SF Bay Area contact.

I have most of the other instruments covered either on my own or through other contacts, should I have the reason or the budget to move beyond mock-ups. This is a bit of a reality check as well, about how much money and time we put into buying and using these products (and fighting their inadequacies).

[Shooshie]

Timing can be funny. A high school friend who now lives in L.A., just came out with her first jazz album (she mostly plays and teaches classical music), which might be the first one to ever feature English Horn and Oboe (filling a similar role to Soprano Sax to my ears).

In the Bay Area you surely should know Paul McCandless! Last I heard he lives in Bolinas, up the coast northwest of the GG bridge. Bolinas is right over the hill from Fairfax, where one of my best friends lives, and whom I visit occasionally. We always go to Bolinas when I'm there. Anyway, Paul played with Paul Winter Group, but most importantly with Oregon, which is one of the great improv groups of all time. If you ever come across the album "Out of the Woods," give it a listen. You'll hear Paul play a lot of oboe, as well as soprano sax and bass clarinet. I don't remember whether or not he does English Horn on that album, but he does play jazz EH. His oboe chops are fantastic. If your friend doesn't know of Paul, you owe it to her to connect her with some Oregon albums. They're even better live; quite the opposite of some groups.

Shoosh

[michkhol]

To make a fair comparison I decided to program in VSL the Strauss Serenade for 13 winds which is one of the Wallander demos. I have the score in PDF, but unfortunately the PhotoScore Lite that comes with Sibelius does not recognize tuplets (and many other markings), so there's a lot of manual work. If anyone has and could share the MIDI file, it would greatly speed up the whole process. I've done only two parts so far.

[mhschmieder]

I've always thought of Oregon as New Age vs. Jazz. I've seen them in concert a few times back in the day. In later days, Paul seems to have drifted more into the Wyndham Hill camp.

At any rate, you are correct that McCandless probably broke the ground originally for double reeds in jazz and jazz-tinged music. I have several of his recordings but haven't listened to them in so many years now that I had forgotten all about him. Now you're making me feel guilty.

The works in which I have been employing oboe and other double reed instruments these past few years, are somewhere between jazz, classical, folk, celtic, and perhaps new age. I find the double reeds to be the most romantic members of the woodwind family -- though sax can be as well, when played with a classical mentality. I find flute evocative but not terribly romantic. Perhaps I am the only one?