Adagio in Gminor - GPO4 Strings sounding authentic...

[jdsnyderii]

Sounding authentic enough at least in my book. When I first got the GPO4 library, I had a hard time wrapping my head around how to make them sound authentic. So, I set out to convince myself that I could do it. Did lots of reading and with some good advice on the forum, I set out to make this recording using a real recording from an ensemble from the Czech Republic. The sound they got was very similar to this, and this piece convinced me that I could get good strings sounds.

Without further a adieu. Adagio in Gminor (and yes the original had the airy reverb to it too).

Jim

[SysExJohn]

Aha! The baroque composers to the fore.

A lovely piece of Albinoni (or should that be Giazotto?), beautifully realised.

Thank you.
John.

[fabiolcati]

Hi John
You demonstrated that GPO strings can sound better than expected.
Can you explain your techniques for achieving this, what you learned "from reading and [...] some good advice on the forum"?
And thank you for adding a Baroque touch to the festival.

[SysExJohn]

Hi Fabio,

What a compliment, being asked for advice.

Most of what I do I learned from the forum anyway, from others well steeped in getting the best from GPO.

But here's a few:
#1. Think about what the composer intended or what forces he/she may have had available.
e.g. most realisations of baroque music have, IMO, too large string and other sections.
a couple of violins or a small violin section is what Bach would have used in the majority of cases.

#2. Use those forces imaginatively. Build sections from the solo strings and add a soloist on top.
I initially write one MIDI track, then replicate it several times. Then adjust the note timings on each replicated track.
Each goes to a different channel that has been shifted back or forwards a few 'ticks'.
Then finer adjustment can be done to individual notes, both timing and expression, especially for the lead instrument.
I use a 1 or 2 dB shift in fader volume between instruments as well as fractional changes in panning to achieve a small spread. Think exactly where you wish to place them, stereophonically, to achieve good separation from other groups of instruments or soloists.

#3. Think about expression and dynamics.
No real player can play fortissimo, less so forte-fortissimo, for very long without rest. So use these levels rarely.
Even more so as the instrument sounds less 'musical' at these levels.
It's also surprisingly difficult to play pianissimo and more so piano-pianissimo.
The normal range is piano to forte.

But just how to translate these to MIDI velocities or cc#11 values?
We have 127 levels to play with.
If we divide this range into 8 (ppp, pp, p, mp, mf, f, ff, fff), roughly 16 levels per dynamic marking, we have a basis for selection. So, e.g. 'p' could be from 32 to 47, 'mp' from 48 to 63. and so on.

Remember these are just guidelines.
So typically realisations could use note velocities in the 32 to 96 range with occasional forays above 100.
The same too with cc#11 values.

The Garritan instrument samples bear this approach out. The only caveat is that the lower values can sound rather quiet, so I tend to compress this slightly by raising the level of the quieter notes.
Working from 104 for ff, 12 per dynamic group, down to the mid 40s for p.
But I let my ear be the final arbiter.

"Rules are made for the guidance of wise men and the obedience of fools."
I can't remember whom the quotation is from.

I compensate for the quieter dynamic levels by using a higher volume fader level, although I never go above -3dB.
Once set for balance, and to make sure that overall levels stay below -3dB FSD, these faders remain static for the whole piece.

#4. For more modern pieces or larger arrangements, e.g. the B minor Mass, I use a string section overlaid with a soloist, as has been advised before by others of this forum.

If the overall level of the recording is a little quiet I use Audacity to amplify it by a few dB.

Just what comes off the top of my head.
I'm sure other, more experienced people, could add to this.

Any help?
John.

As an afterthought, I always use the auto legato (cc#102) control and adjust my notes lengths to overlap the notes within phrases.
I'm equally happy working in the notation, piano roll and detailed list view, and often add broad brush controls by typing in the codes exactly where I want them. i.e. setting initial dynamic levels for a piece as it moves e.g. from piano to forte.

Finer control movement is often then drawn in in a controller pane.

[sururick]

I love this! This is the kind of music I really enjoy listening to! Great work.

[rbowser-]

Sounding authentic enough at least in my book...

In my book too, Jim!

I think it's a cliche, the way the strings in GPO are routinely criticized, complained about, even dismissed. Like many user complaints about virtual instruments, I swear the problem is when people just load an instrument, push a key, and make their total assessment on how that sound emerges straight from the box without any actual Playing of the instrument involved.

Getting an understanding that software instruments need to be PLAYED seems to be something a lot of people just don't get. And so with the GPO strings, it's not as if it's some gigantic undertaking to make acceptable recordings with them. It just calls for the user to be an actual musician and Do something to play the samples, not just trigger them.

In any case, your brief passage here is a great success, imho - it's a very nice recording, and also demonstrates that the GPO strings can sound great. THANK YOU for all of that.

--one small technical note - this MP3 copy cuts off prematurely.

Randy

[DPDAN]

that IS a nice sound you achieved with GPO, one small hint is that the crescendo was extreme and I had to instantly lower my listening volume.

Getting an understanding that software instruments need to be PLAYED seems to be something a lot of people just don't get. It just calls for the user to be an actual musician and Do something to play the samples, not just trigger them.


Randy

Well said Randy!

[etLux]

What a great sound you've achieved on this, Jim!

And if DPDAN says so -- you can believe it.

I would make one single suggestion (and do remember my ears are older than Mt. Everest, and equally eroded.) To me, this sounded just very slightly wide. Though Dan would holler at me for this as bad technique, one thing I've found effective in combat of that without losing the breadth you want is to mix maybe 10-15% of each channel over to the other, spreading the sound more naturally as it would be while reverberating in a hall.

All the best,



David
-----
David Sosnowski
www.DavidSosnowski.com

[DPDAN]

David speaks the truth

[etLux]

David speaks the truth

Holy cow! You mean I got that one right? Son of a gun.

Thanks, Dan!

Best,



David
-----
David Sosnowski
www.DavidSosnowski.com