Hello JAB,

Well you mentioned dynamics. I’d say that for orchestral reproduction, dynamics preservation is one of the most important aspects of the mixing/mastering process for achieving the most realism. I do these exact type of midi productions myself and from experience I’ll tell you that I spend maybe more than 70% of my time “working” the instrumentation dynamically with volume automation as one of the the final stages of production.

Secondary to that, I would say that instrument placement within the stereo plane is also crucial. Never use panning automation if your trying to create the sense that we are listening to a “real” orchestra since it would not normally be customary for, say, the brass section of a large ensemble to get up from their seats during a performance and wonder around the stage from left to right :) Carefully balance out each “section” and make a final decision on where they will “be seated” on the stage.

And lastly, as far as signal processing is concerned. In this case, less is definitely more. Remember that a real orchestras players are all in the same room or hall. That's not to say that you should not use different types of reverb for different players though because some reverbs that work well with say a solo flute may not work so well for a French horn section. Instruments played live on a large stage do in fact have different reverb and delay characteristics simply do to the fact that their sounds are emanating from different front to back, left to right placement on that stage affecting absorption and reflection. The spatial acoustics of a large room or hall create different decay and delay rates depending in where the instrument is situated within that space. I also never compress the stereo master (this will totally kill your dynamics) but sometimes do add a VERY small amount of limiting if I’ve got a couple of peaks.
Compression (again, very slight) can be used on a player specific basis (before) volume automation is applied. This may be necessary to bring out solo passages if the instrument does not seem to be cutting through. But now I’m getting into arrangement which is a whole other can of worms. ;)

One last thing regarding the use of equalization. Try to get the mix, during the mixing stage, as close to what you want your final master to sound like before treating the stereo master. EQ’ing individual instruments during the mixing session may be more time consuming but it’s much better than trying to EQ a whole orchestra after the fact. Because there are so many variables and components involved with an orchestra, post EQ’ing almost always snowballs into creating many many more problems leading to potentially disastrous results. To me, “over EQ’ing” anything is still just like borrowing from Peter to pay Paul... Keep it simple and subtle!

Hope that’s what you were looking for.

dk

Thanks dk—

You’ve pretty much hit on everything I’ve been doing along the way, so it’s good to know that I’m on track with others… I guess I was really wondering if anyone out there had any special tricks (or pieces of gear) they use at the final mastering stage. Having said that, I completely agree with you that keeping it simple seems the best way to go with this type of music/mastering… after all we are trying to reproduce the sound of an orchestra in a hall or really great room…

I also use a bit of limiting if need be to get an overall ‘hotter’ level if the dynamic range is wide. I also find that using different amounts of the same type of reverb on particular instruments can work nicely to create the spatial illusion of front to back. Still, I think it’s really hard to get that wide-open true stereo sound of a well-recorded live orchestra. The best verbs and room emulators out there just don’t seem to entirely capture it. Close…but no prize.

Thanks again…

Jim

Jim, you might enjoy this article.
--Rick

http://www.recordingwebsite.com/cgi-bin/Yabb/YaBB.pl?board=dbstips&action=display&num=1002504356

Rick--

Thanks, good article--there was some interesting stuff in there...

Jim

It seems like there are a couple issues at play here... The first is the style of the mastering I'd do on an orchestra has a direct relation to the style of music being played.

For a traditional classical orchestra, I approach the mastering with the idea that as little should be done to it as possible. ie. a more purist approach. It is very dependant upon the room that it is recorded in. Sometimes, that means I need to apply a little overall reverb to the entire stereo mix. As far as processing goes, I might need a bit of EQ and I usually add a touch of peak limiting (using something like a Waves L2). Not much should happen here, though, because the recordings are usually done so that things like EQ really aren't necessary through the choice of mic/preamp, etc...

For a studio, jazz orchestra or a film score, the approach tends to be different. Here, mixes are generally manufactured. While the room is important, because of the large number of other mics involved, it plays less of a role in the recording. In the mix, more of an artificial reverb is used for the orchestral sound. In the mastering of this style of music I tend to take a less purist approach. I find that I may need a bit more EQ (what if the french horns are sounding too heavy in the lower mids and the strings need a bit more sparkle?), perhaps light usage of my multi-band compressor in conjunction with peak limiting. I find that a careful use (but not too much of any one thing) of a number of tools can really help improve a recording.

Now... as to your situation: What style of music are you dealing with? You mentioned an orchestra that is comprised of high-quality samples. Without knowing much about your music, I would guess that the sound you will be getting is closer to the second example I gave. When putting your mix together, I would be careful not to "over-eq" individual instruments. So often with samples what doesn't sound good alone works very well in the context of a large mix. Make sure to check the sound of the individual instruments in the mix as well as soloed.

When you are happy with the mix, then check overall tonality, etc... in your mastering. You may find that you'll need overall tweaks (EQ with a broad Q, a touch of compresson to unify the sound, etc...).

I also sometimes play with a touch of stereo image enhancement. I find that it can really open up the sound of a mix. HOWEVER, you need to be careful because when this is done too much, you can really ruin a recording. I'm not a fan
of the Waves S-1 plugin or stereo widening on the Drawmer Masterflow or TC Finalizer. However, I have found 2 tools that work very well to my ear. The first is a "miracle" box called the DD-2 K-Stereo processor by Digital Domain (Bob Katz). It is referred to as an ambience-recovery processor and it can open up the sound of a recording and maintain all imaging. A truly amazing piece of gear. The other one I like is the multi-band image enhancement in the Sequoia (and Samplitude) workstation. In small amounts it can really open up the sound of a mix.

Hope I am of some help.

--Ben

Thanks Ben for your excellent response. The music I’m working on is film score in style and utilizes Miroslav and Advanced Orchestra samples. So yes, you hit it right on the head, the second example you cited is pretty much on par with what I’m doing. I’ve done some subtitle limiting in the past, but not tried the multi-band as you suggested…I’ll have to pull out the Waves C4 and give that a shot. I’ve also not experimented around with EQ too much as that I’ve felt that the samples are pretty good. Although, with my latest arrangement, I can tell that there is some room for subtitle EQ—particularly with a grand piano sample that seems a bit muddy when mixed with other instruments…

Again, thanks for the great information…it’s really helps a lot…

Jim