Sibelius: Lots of problems with dynamics playback

[cjthibeault]

Hello,

I am using VSL Special Edition and the Solo Strings package with Sibelius 6. The solo strings have been working fine, but I am having a great deal of difficulty with the dynamics in SE. I'm writing for full orchestra. 

The initial problem was that low strings and timpani were completely inaudible at dyamics of pp and ppp. I had to redefine their values in the dictionary to be closer to that of p to hear them. mp is considerably louder than p, and dynamics that are mf or louder are ridiculously loud in comparison, when the entire piece is played at volumes that make p or softer audible. 

To compensate, I have placed a look-ahead limiter as an effect in Sibelius, with the input gain raised 20db to make the quiet parts audible enough, and letting the limiter bring down the volume of the louder parts. This generally sounds like crap, and is certainly not the correct way of doing things. 

I have followed the guidelines for optimizing playback in Sibelius, and have enabled the velocity cross fade, and set the controller for it to 11. I have also tried disabling/enabling controller 11 for expression, which has its own problems. If I disable cc11 for expression, I get the change in timbre during crescendo/decrescendo, but not neccesarily the change in volume. 

It has been incredibly frustrating to get things to play back as I would expect. I would like the quiet parts louder, so that they are audible, and the louder parts quieter, but I still want them to be played back with the correct patches, and to have dynamic changes properly followed.

What am I doing wrong?

Thanks,
Chris 

[andi]

Hello Chris!

I recommend disabling CC11 for expression and just using it for Velocity XFade. This way you avoid double dynamics that cause your problems of all dynamic values below mf being too soft.

Are you using the Vienna Symphonic Library House Style? In this file the values for all dynamic levels have been optimized for our sample collections. If you haven't done so yet, you can download it here:
http://www.vsl.co.at/en/68/141/460/576/655.htm

For further finetunings you can change the Dyn.R. setting of your Vienna Instruments in the Control Edit/Edit Cell menu. The higher this level is, the higher is the dynamic range of the instrument and therefore lower the volume of the low velocities. Please be aware that this level is set for every cell seperately.

I hope this was helpful.

Best regards,
Andi

[cjthibeault]

Hi Andi,

Thank you for replying. I've tried disabling CC11 for expression, but the problem is that things such as descrescendos do not work as expected (the change is not as smooth or as complete as I would have expected). 

I installed the VSL House Style. How can I be sure that my score is correctly making use of it? Is there a particular value I can check? As far as I can tell, it is being used. 

I'll check into the Dyn.R too, if nothing else seems to correct it. I'll be playing around with this over the next few days, and I'll let you know if I continue to have problems.

Thank you,

Chris

[andi]

Hello Chris!

The Vienna Symphonic Library House Style has to be loaded for every score you want to use it with. When you are in doubt, if it is loaded, just load it again. In order to keep your layout and just add the dictionary and instrument entries, check only the boxes for these two categories in the Import House Style menu.

For me it works well, to use CC11 for Velocity XFade only and disable it for Expression.

Best,
Andi

[cjthibeault]

Hi Andi,

I double-checked, and I had correctly loaded the house style. Would I be correct in stating that velocity cross fade is most useful when working with the more complete instruments, and not so much with the Special Edition instruments? As an example, a descrescendo from pp to ppp makes no audible difference at all for the English Horn in the Special Edition. I find similar problems with other dynamic gradients. 

I have settled on a solution that allows me to remove the use of a limter by enabling CC11 for both expression and velocity cross fade, but manually tuning the response curves for each instrument. It provides reasonably acceptable results.

What I intend to do soon to further improve this is to notate a number of examples from the Adler orchestration book in Sibelius, and use the accompanying CD material as a reference while tweaking those curves to ensure that I am getting a good approximation.

Anyways, thank you for the response.

-Chris 

[andi]

Hello Chris!

Our bigger collections offer more velocity layers for most articulations. So you are right, that these will work smoother with Velocity XFade. However, we didn't record samples in ppp. So going from pp to ppp will always make the pp-samples only be played a little softer and there will be no difference in timbre.

Best,
Andi

[Nick_E]

Correct me if I'm wrong but, the problem isn't with how many layers of recorded velocity the SE version offers. Individual layers will be softer/louder until they cross into the zone of another velocity layer, at which point a different sample will be used. If this is not how it does things...umm, why not !?

[umarekawaru]

I have the same problem and I imported the House Style Sib7 v 2.1 ppp and pp is inaudible.

[andi]

Hello umarekawaru!

Did you select another controller than CC11 or none for Expression as suggested eartlier in this thread?

Best,
Andi

[umarekawaru]

Tried that. Didn't work. But I pulled the Expression response curve all the way up. That worked. Thanks Andi.

[thomamd]

This does not answer your problems, but is an oboe and English Horn capable of playing pp or ppp. Seems to me that the instruments cannot play that soft. It really cuts through the sound of an orchestra. If the oboist plays a wrong note, the whole room notices.