Vienna Symphonic Library Forum
Forum Statistics

182,299 users have contributed to 42,217 threads and 254,750 posts.

In the past 24 hours, we have 3 new thread(s), 16 new post(s) and 43 new user(s).

  1. Home
  2. Sample Libraries
  3. Vienna Instruments Libraries
  4. "Problems I have with Solo Strings"

  • "Problems I have with Solo Strings"

    L
    L

    A few weeks ago I asked a professional in the commercial music business what I should use for solo strings.

    He mentioned some products he'd had success with. He also said he'd heard that VSL was good but he hadn't tried it.

    So I volunteered to buy the VSL Solo Strings package, and report my findings to him.

    Here's what I will say to him, unless someone can convince me otherwise:

    Do not waste your time and money on VSL Solo Strings. They are not suitable for professional work. Warn your friends.

    1. The "sustain" instrument does not sustain. Specifically, instrument 10 VI_sus_Vib will sustain for the following durations.
    - at velocity 30-50: infinite
    - at velocity 60-80: 6 seconds max
    - at velocity 90-120: 14.5 seconds max

    The above issue was raised by Jurgen (zentrumsounds) long ago, in post http://community.vsl.co.at/forums/p/23498/158878.aspx#158878 and no satisfactory answer has been provided by VSL.

    2. Harmonics cannot be sustained either.
    - at velocity 30-80: 8 seconds max
    - at velocity 90-120: 2.5 seconds max

    The above issue was raised by Jurgen long ago, in post http://community.vsl.co.at/forums/p/21477/147307.aspx#147307 and no satisfactory answer has been provided by VSL.

    3. To be consistent, VSL has ensured that sustained "sul ponticello" notes are inconsistent too:
    - at velocity 30-50: 12 seconds max
    - at velocity 60-100: 8 seconds max
    - at velocity 110-120: 5 seconds max

    4. You cannot play double-stopped harmonics. Specifically, instrument 02 VI_harm-art_sus is monophonic. Double-stopped harmonics have been used in many, many classical scores, but VSL won't support them.

    5. The "legato" instrument 01 VI_perf-legato is also monophonic. Is this a problem? Sort of; I've just skimmed through several string quartets by Beethoven and Bartok, and found many passages where legato lines are interspersed with double-stops. If you want to handle these with VSL, you'll be flipping between articulations often.

    6. The sampling is not as good as VSL would like you to believe. Listen to the note A440 in instrument 03 VI_pon_sus. It fades and seems to restrike after .375 seconds. Nearby notes don't have this annoying quirk. It's just a sloppy sample.

    7. VSL is a CPU hog and a memory hog. My computer can handle 11 MIDI tracks with 17 Kontakt articulations at once, but it can't handle one MIDI track with 2 VSL articulations without frequent interruptions. Kontakt gives me ways to control its resource usage, for example by eliminating reverb or controlling how a sample is loaded from disk. VSL does not give me this kind of control.

    Now I must either buy an expensive new computer that can handle VSL, or spend an awful lot of time working around the limitations, or stop using VSL and ask for my money back.

    Either way, I feel obliged to warn everyone away from this product.
  • H
    H
    last edited
    last edited

     Hi Londoner,

    hope you don't mind that I ignore the offensive tone of you posting and I'll try to be helpful where I can.

    @Another User said:

    6. The sampling is not as good as VSL would like you to believe. Listen to the note A440 in instrument 03 VI_pon_sus.

    ponticello in piano is delicate to perform, the sound has modulating character. I don't here so much difference to other ponticello piano sustains, for example C above A440, or the A one octave higher.

    [quote=Londoner]7. VSL is a CPU hog and a memory hog. My computer can handle 11 MIDI tracks with 17 Kontakt articulations at once, but it can't handle one MIDI track with 2 VSL articulations without frequent interruptions.

    It's hard to tell what's going wrong here, also without knowing any system specs of your computer. If you post your specs, I'm sure other users can tell you what you should expect. I don't want to write here any self promoting, but I'm sure also our users can confirm, that our Vienna Instruments player delivers one of the effectivest performances on the market.

    best

    Herb
  • D
    D

    VSL is most definitely the most efficient sample streaming engine out there.  
  • H
    H
    last edited
    last edited

    @herb said:

     Hi Londoner,

    hope you don't mind that I ignore the offensive tone of you posting and I'll try to be helpful where I can.

     

    OT:

    I think it is a pleasure seeing how professionally you guys, herb & co., handle things here on the forums.

    You're (nearly) always being calm and helpful regardless of whatever way people approach you. 

    Big thumps up & lots of respect from me.

    Looking forward to most likely becoming a MIR-customer later in the year.

    Best regards,

    Henrik
  • C
    C

    I wouldn't agree that the VSL engine is the leanest out there.  I love all of my VSL products.  Some of them take a bit of work to get over the hurdles of them being samples.  Meaning, physical modeling is far superior but at the moment, we don't have that option.  I am sure one day VSL will go this way (or whatever way ultimately becomes the most viable and easy to use).  I do find most of the patches difficult to perfectly blend together when doing certain legato lines but the more I learn the easier it becomes to do so.  I wouldn't say it's an issue with VSL as much as it's an issue of the technology (samples).  And at least they give a crap load of things you can change to attempt to get exactly the sound you're looking for.  It's a learning curve to me.

    I am able to load incredible amounts of sounds in EXS24, Kontakt player etc but when it comes to loading VSL Samples in the Vienna player, the number is cut down exponentially.  I don't know why that is to be honest but, it's consistent on all of my machines so it's not one machine based or a Windows/Macintosh thing.  I think this has been discussed many times before but honestly I find myself spending to much time reading this stuff when I should be using my time with my samples making music and learning how to use the tool to make it happen.  You hear the demos and see the results so it is in fact possible.  It is a matter of learning what to do to get that sound.  I have found that VSL will give me a patch for instance called "legato".  Yet, for a specific legato passage I will find using samples that say nothing to do with "legato" work better.  Trial and error and then some good documentation of your own and I am all set.  I am just glad I have such amazing sounds to work with and not some sloppy looped fake crap recorded samples that have slow ugly wobbling vibrato and a hideous harshness to them.  (will not mention any competitor names).

    MIR I own and love but is another example of this CPU hog you mention.  It requires so much power to run efficiently that the lights have to dim in my home to pull enough juice to power it :).  The nice thing is, i7 processor technology is here to stay and be exceeded.  Soon discussions on computers being to slow to run MIR will become obsolete.  It's the price you pay to go with absolute cutting edge.  I paid it and am happy.  it's far better then any reverb solution I have ever worked with in my life.  But that's another topic in itself.

    I don't think VSL condones people for writing negative posts or things that could possibly be done better to help their business decisions (as customer recommendation should always be an influence in the decision process) but your tone isn't even constructive.  It's combative and quite aggressive with a "bashing" quality behind it.  You could have gotten your point across in a much more professional way but even so, the people who will listen to you and run the other way will be the 15 year old "mp3 generation" iPad users who will go spout what they read in your post to all their friends if VSL ever gets mentioned.  After their moment of glory and ego filling brain puking about how much of an "expert" they are with VSL products, it will actually do two good things!  The person they dumped all the BS on will actually go home and lookup who VSL is (woohoo one for the team!).  Secondly, they will hear the demos and go back to school tomorrow and say to the 15 year old mp3 genius, "you're an idiot" and buy VSL products anyway.

    Good day,

    Maestro2be
  • J
    J

    Wow! Londoner you're pretty astute.  In but, "a few weeks," and 4 forum postings you've crystalized everything there is to know about VSL.  I've been working with VSL now for a little over two years and I'm still learning, but you've got it all figured out.  You're a VSL prodigy.

     

    Seriously, Londoner, don't you think "a few weeks" is such a short period of time to truly grasp the potential of this wonderful tool.  The set backs you've described are easily overcome by the features of the software but you just might have to invest a little more time than "a few weeks" to learn the techniques.  If you have a question about something than ask the forum.  I see you've already got searching the forum figured out so you're off to a good start.  I apologize if I sound condescending but I just don't see how you can form a "professional" opinion about something that, as it appears, you've had so little exposure to.  Are you just venting because you can't get your string quintet project to sound the way you want it?  So, this professional that you've been talking would be foolish not to get a second opinion.      
  • D
    D

    Huh? Unsuitable????

    Have you listened to the Solo Strings demos?:

    http://vsl.co.at/en/67/4587/4588.vsl#

    Some of them have tutorials included so download them

    and see how they were achieved. The Beethoven piece

    is simply amazing and unbelievable.

    I'd like to know of ANY library that can produce 

    such convincing solo string performances.

    Also check out this Solo Violin video demo by Guy Bacos, its amazing!:

    http://vsl.co.at/en/211/442/344/350/1030/643.htm

    I understand there are issues but there are also workarounds,

    you just have to find them.

    I can't say much about your performance issues, I'm on PC w/ Quad CPU

    and can load dozens of instances.

    Good luck!

    DM
  • L
    L
    DM33, yes, I bought VSL because of the wonderful demos. I will look into the tutorials.

    ddunn, you have provided no facts to back up your opinion, whereas I have provided the facts as I found them (although in retrospect I was wrong about double-stopped harmonics).

    jasensmith, I apologize that you found my post offensive. I did choose an outrageous title. It should have ended with a question mark. I was feeling very frustrated at having spent weeks, indeed, trying to figure out things that I thought should have been obvious even to a newbie. (Why doesn't a long note last for its written duration?) I had also just received an estimate of what it would cost to buy a computer big enough to fully handle VSL. So I wanted a quick response, in order to decide whether to jump ship, and I'm grateful to everyone who responded, even those who are condescending.

    Herb, thank you for your efforts to correct me. Like Henrik, I appreciate what you do. May I be allowed to respond?

    My posting was not intended to have an offensive tone. Any opinions or feelings in it are derived directly from the facts presented. I would love to believe that you can change my mind. Note that I have not yet said anything to my professional friend. First I wanted to post my findings here, where I was sure to draw a barrage of fire, before saying anything negative to my friend.

    I have looked through the PDF manuals to educate myself following your statement that "the sustain patches are mapped with the option of retriggering the sustain sound at all available bow changes". Sadly I have not found the word "retrigger" or "bow" or even "loop" in the manuals. I don't doubt the truth of what you say, but your manuals do not help me to understand it.

    The nearest I have found, perhaps, is something about Release Time. I did this experiment: I created MIDI notes for the sustain articulation (the one which lasts 6 seconds). Each note was 5.5 seconds long. I left only 18 milliseconds between notes. In the VI Perform window I mapped controller 68 to the Master Release. In the MIDI track, I increased controller 68 just before the end of each note, and brought it back to zero after the retrigger. What sound did I get? A bow change after the first 5.5 seconds, then the second note died after only 3 seconds! This seems to indicate that I'd have to write a new note every 3 seconds, and still hear a bow change that sounds nowhere near as good as a real violinist would make. And, there was a long release trail after the last note, even though controller 68 was at zero. This is not good. I look forward to being corrected by you. But wouldn't it be nice if the solution were explained in the manual?

    Regarding my points 2 and 3, I'm not sure what you're saying, but I assume the solution would be similar to the solution for point 1.

    Regarding point 4, I was wrong. I can get double-stopped harmonics, as long as both notes are high enough. My error was to test double-stops where the lower note was outside the range. Sorry.

    What I said in point 5 stands. You are right; that's the way it was designed. And I am right; it is troublesome to use.

    My point 6 stands. The glitch in A440 is very noticeable. It came to my attention because I was working on some real music, and this note stood out. The nearby G# and Bb do not have this problem. Another take with the microphone could have eliminated this problem.

    My point 7 stands, although I don't expect much sympathy in this forum. Everyone will tell me my computer is too old. Yes, maybe it's too old to run VSL. But it's been running Kontakt for a year, with lots more intruments, no problem. Comparatively, VSL is a resource hog, as cgernaey has found, and it may force me to spend a lot more money on hardware I wouldn't otherwise need.

    I'm at a point where I need to radically accelerate my learning curve, or spend a great deal of money, or give up on VSL.
  • L
    L
    Sorry, I meant to say "bow change" is not in the manual. The word "bow" is in the manual.
  • D
    D

    Sorry, I must have missed your "facts" regarding CPU performance.  Oh, thats right, they weren't there.
  • W
    W

    This thread is a bunch of crap.  I wish people here would not give it constant replies, and keep this bullshit in view. But I know you will since it is negative.   

    This guy is a "professional"?  If he is, then he would realize that doing a performance such as the Bacal-Ravel quartet is BEYOND "professional" - far beyond.  And yet, this library is not good enough?  I think the real story is - he is not good enough for it.  What are his qualifications to make a sweeping statement like this thread?  Who is he to place himself above it all and judge it?  What arrogance!

    I swore I would not get irritated by this kind of attitude any more, but can't help it.  This is the whole problem with the internet - everybody's opinion is equal.  Even those who don't have a clue what they are talking about.

  • L
    L
    DM33, I'm trying to learn from those tutorials associated with the demos.

    So far, I've looked at the Beethoven, Bloch, Borodin and Bridge. They all consist of one MID file and one or more fxp files. In none of them do I see a tutorial in the sense of a verbal or visual explanation. Am I looking in the wrong place?

    In the Bridge there's a viola C# that starts 1:17 into the piece, and is meant to last 6.3 seconds. However, it sounds only for 4.3 seconds. So the thing I want most to learn, I can't seem to learn from this example. I await any further pointers.

    ddunn, you are obviously on the warpath. Calm down and read where I said that "My computer can handle 11 MIDI tracks with 17 Kontakt articulations at once, but it can't handle one MIDI track with 2 VSL articulations without frequent interruptions." These are facts. If V is less efficient than K, then V cannot be the most efficient thing.

    William, you too are on the warpath, and it's because I chose a bad title for my thread. I already apologized for that before your post. I'll apologize again if it makes you happy. I do not place myself above anyone, and anyone who knows me will tell you I'm not arrogant. I have been very frustrated by a piece of software that takes a heck of a lot of figuring out just to make a note sound its notated length. I'm still hoping one of you will tell me how to do that.
  • T
    T
    Just wanted to chime in with regards to your fifth point, about legato and double stops.

    I own several libraries from other vendors utilizing legato scripts, and I've demoed a bunch of others. None of them can handle polyphonic legato lines on one midi channel. The reason for this is actually pretty obvious, if you think about it. How could the legato script select the proper samples if several notes are being played simultaneously? Sure, it would be possible to approximate what transitions the player had in mind, but it would produce unpredictable results, at the cost of extra strain to the CPU.

    Furthermore, with most other (reasonably detailed) libraries you need several different midi-channels for each instrument. Take the solo violin in gypsy as an example. You need 3 separate midi-channels to utilize all articulations (the master patch, plus the legato and fast legato patches).

    I'd love to learn how the bow change works, though, so if anyone would care to explain, I'd be thrilled. :)
  • W
    W

     So, change the title.
  • M
    M
    I don't know how you guys can take Londoner's post so seriously as to even spend time writing replies to it. Isn't it obvious we have a troll here? Or no, wait -- I can't even believe that a troll paid by the competition can be *that* dumb to post such obvious nonsense and hope people will take it seriously. Come on, I have used VSL for even shorter time than Londoner claims to have, and even I can see the obvious fallacies. Maybe we have the opposite... a VSL-paid troll whose aim was to seed this discussion? In either case, I am actually grateful, because as a newcomer to VSL, I do find some of the posts in this thread enlightening and useful.
  • M
    M

    Yes, that's exactly it. We are all on the warpath, out to get you. Considering your accusatory thread title, did you not expect a few of us to shove back?  Regardless of your insistence that your computer can handle Kontakt but not VSL, until you post some specifics as to your set-up, your argument is useless.

    VSL runs a bit differently than  your standard audio plug in, sort of running in a "shell" outside your actuall sequencer. Thusly, it should actually run SMOOTHER than something inside your application, using up RAM within your sequencer. There are other things too, such as the 1X and 2X buffer along with your audio driver's latency which also need to be addressed. Check into all these things before saying VSL cant even run two instances.

    To say it is uprofessional is unacceptable. Yes there is quite a learning curve, and to this day  (3+ years using VSL) i STILL learn new things about how to use VSL. It is very deep. But that is a good thing. It will seem frustrating as a new user to figure out how to use many options, but it necessitate getting previous programs' architectures out of your head and learning the VSL way. Once you get it down, it is really quite simple. Ten years ago i was programming Hexadecimal code and sending as MIDI data for parameter changes to my JD-800. I eventually figured out how to do that. VSL is ten times simpler!

    Hang in there, open your mind, be patient. The VSL demos are not fake; they really are possible. But itll take a while to get there. Best of luck.
  • N
    N
    FWIW, regarding point 1: It is true that I wish the sustains were a bit longer - as a classical bassist, myself, it is certainly possible to play slightly longer sustains than were recorded for this library. But, VSL did it the right way - one cannot endlessly loop a solo sustain note - the bow direction has to be changed. What some have asked for on the forum - looped solo strings - runs counter to the physics of solo playing. Regarding some of your other points: I have been working on an original double-bass concerto. All I could afford to purchase were the SE libraries, and the solo double-bass even in SE sounds amazingly good. It is far closer to the tone I produce than one should have any reason to expect from a budget sample library. Yes they are samples, and yes they don't capture all the sutble nuances, but the sounds of the different bowing techniques are actually very close. Overall, if one does not know what one is doing, and choose to use articulations in wrong ways and in wrong places it can indeed sound really bad (as it actually should under those conditions). But, if one has a basic understanding of string playing technique, it is possible to achieve very good results, even with the solo strings in SE, let alone what one could do using the full solo strings package.
  • L
    L
    DEAR MODERATOR, please rename this thread to "Problems I have with Solo Strings".

    I hope in the future the people trying to be helpful will outnumber the people expressing spite. I do appreciate the former.

    I still have not figured out why my attempt to connect two notes using controller 68 to control Master Release was unsuccessful, as described above. But I have switched to using controller 64 (sustain) without mapping it to any VI control, and this works better. The documentation led me to believe that mapping a controller to Master Release should work, whereas I found nothing in the documentation about the Sustain controller. So this behavior is puzzling.

    But it is a step forward. As long as I'm willing to restrike a note every 2.7 seconds, and put the sustain controller on during the transition, I can get a good-enough long note from 10 VI_sus_Vib at velocities over 50. It also works for 03 VI_pon_sus at all velocities, and for 02 VI_harm_art_sus up to velocity 80. Harmonics with velocities over 90 must be restruck every 2.5 seconds because of the shorter sample length. In fact maybe I should make 2.5 seconds the general rule.

    This is a workaround. Not what I'd expected from an expensive product that claims to offer "highly sophisticated and elegant virtual instruments ... to provide our users with tools that guarantee maximum efficiency and an intuitive workflow".

    I have also learned a lot about this forum and the kind of answers one can expect. I wonder whether Jurgen (zentrumsounds) ever got reasonable answers to his (very polite) posts and his very valid questions about sustained notes, viola range, downward harp glissandos, and multi-instrument percussionists. There's no evidence in this forum that he ever did.

    To answer those curious about my system: it exceeds the official VSL requirements for the product. The official requirements are: Apple G4 processor of 1GHz, OS X version 10.4, and 1GB memory. My system has a G4 processor of 1.33GHz, OS X 10.4.10, and 1.5GB memory. All audio to/from Pro Tools is on a 7200 RPM external FireWire disk, and all VSL instruments are on another. But even though I've exceeded the requirements stated by VSL, it looks certain that I'll have to buy a new computer if I really want to use this product, as I am constantly being told "you are running out of CPU power" even though I've muted all other tracks and allowed 95% of the processor to be used by a single VSL instrument with a single matrix in which only 2 articulations are in use.
  • D
    D

    Londoner,

    If you open the MIDI file and import the FXP into the VI you'll have all the same articulations as the demo used.

    Now you can study how the piece was crafted and achieved. I guess its more like a self tutorial.

    If the Beethoven does not include the example you are looking for try some of the others,

    maybe the Borodin, Tchaikovsky, or Sarasate?

    Bottom line is that if you are not happy with the VSL Solo Strings you are out of luck because

    its well known that its the best solo strings library on the market.

    Whatever the workarounds are we all try to learn them because these are truly 

    beautiful and unique samples. 

    You may have to compose to the samples strengths in some areas but mostly,

    with these samples and all the rest of the the VSL libs you can let your imagination

    drive the samples and not the other way around like with other libs.

    DM33
  • N
    N

     Londoner, if I may make a hopefully helpful suggestion:  Although I post rarely to this forum, I have found people here to generally be most helpful and courteous.  Your initial post really did get things off on the wrong foot.  Civility, even when one is upset, helps to foster discussion and the search for possible solutions.  This forum really is a good place.