I am wondering what people here think about how VSL strings compare to the other string libraries out there. There is a lot of new stuff being done these days. Is new sampling needed to keep pace with developments in scripting, divisi, etc? Also, which strings are the least "synthy" - this is something that I worry about constantly with sampled strings. It seems that the strings are the hardest of all instruments to sample and not color the sound in a somehow electronic or digital-sounding way even if everything is done with excellent equipment, great musicians, etc. Also, though they may sound great auditioned in isolation it is disturbing how much sampled strings can change in a mix, lose their richness, and start making the dreaded "thin whiny sound" heard a million times with sample performances. What progress has been made in VSL or any other and how do they compare?
-
VSL against other string libraries
-
This is a really complicated subject. To be quite honest, I don't think that any string library is really good. All of them have good sounds; it's the usability that is a problem.
The Appassionata covered a huge hole (for me) that exists with the Orchestral Strings, in that I find the original Strings anemic in sound. Unfortunately the Appassiaonata only covers one sound, and doesn't have the depth of articulation or playing style for me to use it as the primary sound in my music. There are a few things that VSL doesn't do at all well, as far as I'm concerned. However, I don't hear anything better with any of the other developers' libraries.
DG
-
I almost always layer my strings from different librries. Each library has their own strengths, and I usually put that library up front. I enjoy the auto divisi in LASS, and with the introduction of Dimension Brass, I have no doubt that VSL will take this to the next level. What I find missing in the current LASS implementation, is when you only want half of a section playing, there's no way to automate that yet, that I know of. I do wish that VSL had separate 1st & 2nd violin setions.
But I have to admit, that I spend too much of my time listening to sample playback, sometimes I forget what real strings sound like. When I get the opportunity to hear my music played by a live group, I always get goose bumps hearing that organic, imperfect quality of a string section that samples can't quite emulate yet.
-
@William said:
...Also, which strings are the least "synthy" - this is something that I worry about constantly with sampled strings. It seems that the strings are the hardest...People like the Appassionata Strings. So do I. But I had a "bad" experience...
I produced the AIR from Bach with the appassionats last year.
Then I rememberd a similar Arrangement which I've done in 1996 with my Roland String Ensemble M-SE1.
I searched for a sound demo in my "stock"... and I was very surprised how close the sound is/was to the Apassionatas...
Beside the good (real) sound it has more power to get through drums, basses and other instruments of a band... It is/was a Roland...
http://midimanuals.net/manuals/roland/m-se1/owners_manual/m-se1_om.pdf
Beside of these facts I hadn't the nice reverbs and effects of today...
(I still have this special midi-module)
So I agree: It seems that the strings are the hardest of all instruments to sample and not color the sound in...
Best
Beat
- Tips & Tricks while using Samples of VSL.. see at: https://www.beat-kaufmann.com/vitutorials/ - Tutorial "Mixing an Orchestra": https://www.beat-kaufmann.com/mixing-an-orchestra/ -
I wonder, has anyone here had any experience working with the DVZ strings? That's a different school of thought than most. What of its unique usability, auto divisi, and the fact that this company is building a full orchestral library based on the same concept, that is actually going to cost more money than the VSL (when you thought that wasn't possible...)
On Beat's beat, I recently remixed an old track with the VSL, and the only aspect that didn't come out with the same emotional intensity was the strings part (original was done with the god-created Kurzweil K250). I'm far from a great mixer, but I'm a lot better now than 16 years ago and I had used no sophisticated Logic 8 EQ, Space Designer, etc back then; I had just played the parts through... A friend of mine offered me a plausible explanation of how this could have happened but I'm not going to repeat it here...
-
@Errikos said:
I wonder, has anyone here had any experience working with the DVZ strings? That's a different school of thought than most. What of its unique usability, auto divisi, and the fact that this company is building a full orchestral library based on the same concept, that is actually going to cost more money than the VSL (when you thought that wasn't possible...)
I like the idea of how they recorded the strings desk by desk. Unfortunately so far this technique of recording an ensemble has never worked. It just doesn't sound like a section. I also think that of all the major string libraries, this one sounds the worst. It may be because the developers are so focused on playing handfuls of notes to save time, rather than writing proper strings parts, but not a single demo has sounded good to me.
DG
-
@DG said:
I like the idea of how they recorded the strings desk by desk. Unfortunately so far this technique of recording an ensemble has never worked. It just doesn't sound like a section. I also think that of all the major string libraries, this one sounds the worst. It may be because the developers are so focused on playing handfuls of notes to save time, rather than writing proper strings parts, but not a single demo has sounded good to me.
DG
I think you're right; I was superficially enjoying looking at the - what looks to be a - very user-friendly interface where one easily modulates through so many performance details (what string, how much gliss./speed of, etc.), as well as admiring the wonderful 'South by Southwest' demo track (and comparing it to the pre-kindergarten 'Inception' soundtrack which I heard last night / I know, I can't resist...). However, when I focused on the sounds per se as opposed to the flowing composition, the sounds left quite a bit to be desired.
However, could that be because it wasn't Meister Bacal behind the mixing desk? Perhaps all top-shelf libraries should pay J.B. to emulate the same work with the best software each company has to offer, and subsequently post all tracks on every company site for our scrutiny and comparison. How's that for an idea?...[;)]
-
What are DVZ? Could you post us a link?
On the VSL strings: I personally still find them to be the best to be found around. With no other library I could even concievably create string sections sounding as those in f.e. Bruckner, Schumann and Bach examples here. And these are Orchestral, Chamber (Bach Piano Concerto) and Solos, no Appassionatas.
@Errikos said:
However, could that be because it wasn't Meister Bacal behind the mixing desk? Perhaps all top-shelf libraries should pay J.B. to emulate the same work with the best software each company has to offer, and subsequently post all tracks on every company site for our scrutiny and comparison. How's that for an idea?...
That is an excellent idea [8-|]
-
@JET said:
What I find missing in the current LASS implementation, is when you only want half of a section playing, there's no way to automate that yet, that I know of.I'm pretty sure the Auto Arranger feature can handle that pretty easily. I've found a way to have my template in Logic set up so I can switch between individual divis and all together.
I don't think the DVZ full set is going to cost anywhere close to what the full vienna stuff does, they've announced the brass section and it's only $599.
DVZ: http://www.audioimpressions.com/
-
DVZ is a good concept but the actual sound ends up being poor. This is the entire phenomenon with sampled strings in a nutshell. No matter how excellent the technical concept is, there is a mystery and artistry to it that somehow makes the final sound good or bad. In other words, the sampling process is itself a meta-performance that can turn out good or bad.
An example is - Appassionata Strings, which are the best overall sampled strings. You simply HAVE to use them in order to do virtual string recordings. But they actually have fewer articulations than the orchestral strings (which definitely sound thin though that makes them good for divisi sections).
Of the other strings in existence the LASS seem the second best after VSL. And this is because of the use of different-character divisi sections that can be combined into the final single line. This is a monster that keeps rearing its head - DIVISI. It must be incorporated as a basic part of the sampling process, not an afterthought, because what makes sampled strings do the various bad things they do is probably the fact that you are taking 21 players and turning them into a single mass. What may prove this is that if you take any given library and record pure string ensemble writing - no other instruments - they always sound better. In other words certain things are happening to the overall sound of the sampled string ensemble in a larger orchestral context that do not happen in a live orchestra. Why? Probably because of this phenomenon of reducing a whole ensemble of players into one simple mass. It will never work, completely.
Of course VSL know this and I predict they are going to pull a divisi card out of their sleeves soon. I'll bet that the Dimension Brass approach which they have pioneered - separating simultaneously playing instruments - is a smaller scale dry run for the big challenge of the divisi/tutti recording of the strings. This same company used an extremely innovative method for recording humanly impossible velocity differences on piano, which were what was missing from all other sampled pianos. The same kind of innovation is almost certainly in the works right now with the strings.
-
Hey Jet, if you want half of a section playing there are a few methods to go about it:
In your project you can have separate midi tracks for each of the A, B, and C divisions of each string section. The C section equals approximately the same number of players as the A and B sections throughout the library, although I find they have them a bit quieter in the mix (probably because they were recorded farther from the decca tree) so you can just use the A and B sections during a half section part, or just the C division (I kind of the sound of the first because the A divisions are very dry and the B and very expressive, so they handle more flexible lines better).
As for having them all on a single midi track, there is a trick using the auto arranger tool: Set the A and B divisions to play the A part of the chord and the C section to play the B part of the chord (set each track to the same midi channel). Next, when playing any line, just place a note below the playable range of the melodic line that you write, the C division will be routed to that note so you only hear A and B. Flip flop this for the Basses if you are using the CPCkeyswitcher.
-
In my humble opinion, strings are NOT the hardest to emulate, not even those of "medium" ( not mediocre) quality.
The THING that STILL seems to throw most everyone off is timpani!
There is not not ONE single company that I can think of that has a demonstrably, halfway *decent* timpani, let alone a FABULOUS timpani!
(Haven't heard VSL timpani yet, either solo or ensemble, so, If someone could point me in the direction proper?) :)
