182,279 users have contributed to 42,216 threads and 254,747 posts.
In the past 24 hours, we have 2 new thread(s), 23 new post(s) and 36 new user(s).
Hi,
I've uploaded a newish composition of mine to soundcloud and I'd appreciate any suggestions/feedback on whether the mix/realisation is ok. I'm really struggling with my ears and my computer setup this last while and I'm not sure if it's me or my computer so it would be great to hear from others much more qualified as to whether the sound acheived here sounds ok.
https://soundcloud.com/mcelvogue/slow-air-lament-for-a-lost-friend
Regards,
Tom
Hi Grey,
thanks for listening! I was using this composition to try out (albeit a little belatedly) the Cubase expression maps with Solo Strings. Mostly it is switching between legato/non-legato patches and then also "con vib", "molto vib" and "senza vib" articulations from the expression maps. I tried the "slide" articulation within the expression map on a couple of places too.
I left the first part as solo violin and solo cello to see how my rendition worked when these instruments were very exposed with no hiding place. All of the expression is drawn in and while I tried using the velocity x-fade as suggested on various posts, I didn't really find the sound I was trying for using this method. I also decided to abandon MIR for this as I also struggled to get what I was after with MIR.
I deliberately tried to keep the velocity/expression in the lower ranges in an attempt to emphasize some sense of fragility in the compisition. I found it difficult in a number of places where the volume would un-explainedly jump and really sit out against a particular phrase. I have assumed so far (perhaps wrongly) that this is where a particular patch is switched to the next highest velocity layer. I'd appreciate if anyone with more expertise/experience could confirm or comment on whether this assumption is correct as it would be good to understand if I'm doing something wrong or whether this is by design within the VSL Solo Strings.
I decided to use this piece as a learning exercise at acheiving a more realistic sound for my compositions after my last effort failed and the corrective action was quite considerable. I therefore picked a shorter piece to learn on as I figured this would be more fitting to try with before returning to my earlier composition.
I've tried to acheive the sense of bowing/phrasing and I hope this is improved over previous works however I'm not the expert on this by any means so I'd happily welcome all and any feedback on that front as I'm not a string player and in comparison to many on this forum I am also a rank novice in the use of VSL/Midi.
Thanks again for your post and I hope some of the information above answers your question.
Regards,
Tom
... so I'd happily welcome all and any feedback on that front as I'm not a string player and in comparison to many on this forum I am also a rank novice in the use of VSL/Midi.
Hi Tom
"Slow-Tempo-Pieces" are very difficult to play with samples because our ears (actually our brain) have a lot of time to compare sounds which have been with the current one. So the feedback from my side is about this point.
So even the legato of the solo violin is beautiful it is a bit boring over the time. Further the violin player knows after 20-30 seconds for sure: This piece is played with samples. What could you do? Try to use all! of the possible articulations.
...or what ever you have with a certain instrument.
So start with a 4s Crescendo (even if you only use 2s) and switch then into legato and end the musical bow with a diminuendo of 1.5s (If you have VIPro you also can shorten/lengthen those dims and cresc).
Using Dynamic Articulations will help to increase the feel of "real play" because musicians normally vary the volume within a long played note. You can use the X-Velocity function by the way. Unfortunately it isn't very helpful with solo instruments because of phasing issues while crossing the different layers. But with ensembles it is a must in connection with slow tempos.
I know that reaching a good result will need even more time that way... 😔
But the results are even closer to the reality. Listen to this example (done in 2008) it shows the use of all those mentioned articulations above and in the mean time the solo-, chamber-, orchestra-, appassionata-library - still not quit perfect arranged...
All the best
Beat
Thanks Beat,
I understand your points about more switching of articulations however I'm not sure this is possible using the expression maps in Cubase but I could be wrong. I'll experiment further as I'd really like to see if there's a more human, musical and logical way to acheive good results without what seems to be a random selection of patches for each and every project. It all seems so bizarre that VSL I'm sure go to a huge amount of effort to categorise and name each of the samples in a musical and logical manner and unfortunately, in order to get anything musical, one has to abandon the musically logical choices in favour of selecting random samples until one works. It's so time consuming and perhaps this may just be something I will never grasp. Thanks for your response which is much appreciated - I just wish there were an easier and repeatable way to approach this. At the moment I'm finding it a little futile trying to learn a "way" of acheiving anything close to realism and the frustrating thing for me is that if I were to start another project tomorrow - it seems that whatever I "learned" from the current project is now worthless as it's all really down to random selections until the appropriate sample is found? Am I the only person experiencing this frustration ? Is this really the best approach ?
Regards,
Tom
...Am I the only person experiencing this frustration ? Is this really the best approach ?
Hello gain
Don't worry be happy... Choosing the "right" articulation from the whole pool which is sounding "right" and not because it has the right name is a time-intensive way of producing music with sample.
There is a guy called Jay Bacal who produce(d) such fantastic demos (See demos).
Each of us has to decide for himself when it's time for saying "that's it".
My suggestion:
In addition to Sustain and Legato use also Diminuendo for endings but also for starting phrases... This works well because each of those dims starts with a little swell before it fades out. So instead of the boring sustain articulation which comes with a very static volume over the time use the 3s/4s Diminuendo which swells during the first 1-2s until you switch to the next note. Try it out and you will be happy.
And a last truth: Coming close to the reality with samples means play the samples how in reality. Translated in the world of samples: No note should sound like the one before (even if it theoretically should). So you need to change the tempo, the articultaions, the volume in every little time unit.... This means a lot, lot, lot and even more "lots"of work.
All the best and don't give up.😎
Beat
Hi Grey,
thanks for reassuring me on Beat's credentials! I have been a big supporter of Beat also for a long time and indeed use many of his presets. My frustration and questioning is not directed at Beat (or anyone in particular) as I understand he is really trying to help - which is much appreciated.
I had a quick look this morning at the VSL expression maps in Cubase and they don't allow selection of the diminuendo articulations that Beat suggested so I am guessing that I will need to give up on the expression map route and go back to what I had done previously on other projects whereby I have to build my own matrices and populate them with the particular patches and then add the keyswitches into the midi tracks. I had just hoped that in the four years or more since I first started using the VSL products that there may be a more streamlined way of getting a good finished product without having to add all the keyswitching directly into the midi track.
I quickly looked into customizing the expression map however when the next upgrade of these are released, I will be back to square one so I'm thinking the safest way is to go back and start building custom matrices again.
Thanks again and I'd welcome any other suggestions in terms of workflow that might offer an improvement in terms of not having to re-invent the wheel each time.
Regards,
Tom
P Paul moved this topic from Orchestration & Composition on
