Hi folks,
 
Been fiddling around with a few major projects in the last few weeks, all of which are orchestral pieces that have fairly loud sections in them. For two of them, in particular, there are a few fairly intense string passages, and I've begun to notice that for many of the string libraries I own, they tend to sound really grainy when they are layered into a fairly dense arrangement and played at higher dynamics. This is primarily an issue with violins. I am trying to record a few soaring violin section lines, and the sound of the strings is so grainy and harsh that it almost sounds like there is static interference, or something. It was loud enough to be distracting.
 
I think it is caused by two issues: strings can naturally develop a quite harsh, abrasive sound when played hard; and in a dense mix when they are struggling against other instruments, that graininess gets emphasized, since only the louder, harsher elements can be clearly heard. I think that is the case, anyway.
 
My standard libraries for strings are, in descending order of usage: Albion ONE, 8Dio Anthology, Sonokinetic DaCapo, Hollywood Gold, NI Symphony Series. For this one particularly challenging piece, I managed to get a okay-ish recording using Hollywood Gold legato 1st Violins -- the Hollywood ones seem to be a bit smoother and clinical than the 8Dio strings, which are my usual choice for soaring string lines. It also helps to layer in woodwinds, and in this case, I double lines with an ensemble flute patch and that helped a bit too. 
 
Does anyone else experience this? And if so, how do you deal with it?
 
Thanks,
 
Rob
 

I will be told I'm wrong by a few on this forum, but I fixed this problem by using 64bitfp audio files. I had real problems with harsh sounding string until I switched to 64bitfp. I also use various analog emulations like Wave's J37 Tape to 'smooth' string sounds. I feed all of these string sounds into a buss or an aux and then put a tasty compressor along with an EQ - think a Fairchild. Finally, lol...I double all my strings and sometimes brass with a granular version of themselves. I use Wave's codex and take a note snapshot of the instrument - think the lowest and highest note of the instrument. I also use Padshop by Steinberg, but instead of taking a snapshot I will take the whole length of the audio and use that instead.

I hope that helps.

Check harshness around 1-3 kHz. Cut with either a regular, MB eq or de-esser. Try varying on each track, stem or section.

Thanks for the suggestions, folks!

Years ago I read an interview with Hans Zimmer in which he said that if you want to get big-sounding percussion from samples, use lower velocities. Experimentation showed that to be true, if the library was well-recorded. It applies not only to orchestral percussion but also toms and kicks in a rock kit. Low velocities, then use compression and gain to get the volume you want.
 
So I wondered if the same tip might be applicable to other instruments with percussive elements, and found that it was equally true for pianos. Because I naturally play with a heavy hand, I started using the velocity trim slider to back my parts off after recording. That was a breakthrough - especially with extremely well-recorded pianos such as Keyscape, which I was lukewarm to at first but is now my go-to piano. Lowered velocities made the difference.
 
What's any of that got to do with strings? Well, that smooth, sweet, soaring Hollywood string sound had long been an elusive goal for me. Velocity didn't seem like it should be all that relevant to strings, but it turns out that it can be. Again, it comes down to how well-sampled the instrument's lower-velocity layers are. A related factor is microphone position. A close-miked violin will always contain inharmonic components that we're really not meant to hear. The harder the instrument is played, the more of that is in there. I used to fiddle with EQ a lot to combat that, but now with a combination of low velocities and far mics, I rarely use any EQ at all on strings. Or reverb.

In a real string section there are lots of upper end harmonics going on. Especially with violins. My last orchestral mix seemed to be slightly aggressive in the 1-3khz range with that "granularity" almost like I had used an exciter.
 
This isn't an answer. More of an observation.

I’ve started using the lower dynamic layers and just increasing the volume if I want the strings to sound smooth. Those high velocity layers always have a sand paper quality that doesn’t work for flowing melodic lines IMO.

Starise
In a real string section there are lots of upper end harmonics going on. Especially with violins. My last orchestral mix seemed to be slightly aggressive in the 1-3khz range with that "granularity" almost like I had used an exciter.

Very true. However, in a classical setting you don't hear violins directly. They are mostly radiating sound upward, and a classic music hall is very reverberant, so 90% of what the audience hears is indirect, having been acoustically mellowed by absorption and refraction. Natural EQ.
 
Plus violinists usually don't play hard, which is why you need 20 of them to compete with the rest of the orchestra. The exception being when the whole orchestra is rockin' out, in which case those screetchy inharmonics are buried in the mix.
 
Celli and basses, on the other hand, face the audience but the overtones are shifted downward in frequency relative to the violins and thus aren't as annoying.
 
 

Appreciate all the great replies, folks. Thanks. 

Lots of good information in this thread. Thanks everyone. :)