Vienna Instrument MIR Pro, thoughts and tweaks?

Hey Paul!

I don't own MIR Pro so can't answer there, but I can maybe help in making your own piece sound a bit clearer. Let's take a look at the average frequency response of your piece compared to one of the recent John Williams pieces.



I referenced Prayer For Peace from the JW & Spielberg Ultimate Collection album. Just from the visual, it's clear that your piece is missing almost an entire band (high freqs) and is underperforming in low end. Adjusting for that with an EQ and some saturation for the low end, this is what I have. The cymbal is super loud, though. I had to duck the whole 5 KHz band during that hit (normally you'd want that only on the cymbal, no need to compromise the entire mix).

WAV Stream

 

You don't have to cut them, really, they're never quite gone. They're just less audible. A low shelf EQ od 2-3 db will do the job. As for the air absorption, I think MIR already does that (I know Altiverb does), simply by being a convolution reverb. In a nutshell, impulse responses are captured by playing a sweep or a hit (or whatever) in a room on a calibrated speaker and recording it with a mic setup. Then the algorithm compares the two and deduces what was done to the sweep to make it sound like that - it deconvolves it and you get an impulse response. So the natural air absorption of the room should already be there.

 

HI Paul, I do own MIR and all of the add on venues. I share your frustration. At first I though, wow, this is a great idea. You can actually place each player on the stage and MIR will calculate that placement in relationship to the rest of the players in the room. I think it probably works best with VSL's own anachoic instruments that don't add any additional baked in reverb. I think where the problem accures is when you add Berlin Brass with it's own baked in reverb to a MIR treated environment. It doesn't know how to take that into account. Aaron's EQ graph really tells the story. When I used MIR I heard the top and bottom being chopped off. I stopped using MIR because it colored the instruments is a way that I didn't like. Then if you combine VSL instruments with Berlin and Spitfire etc., you end up chasing your tail because all of the different libraries when their built in ambience don't end up blending well together.

I also attended Mike's "Template Balancing" class and spent the next two days balancing my template using his method. I think that Mike's method works so well is that first of all, he is not trying to mix Berlin, Spitfire, VSL, with Cinesamples Woods. All of which have a difference baked in ambience. He's just using Berlin so all of his woods work as a family. Same thing for his Brass and Strings. He was able to combine SM and Adventure Brass because SM brass is bone dry and can be tweaked to match. He can then balance all the sections together because they all have of a distinctive voice that that can be placed in the room. It's definitely the best solution I've tried so far.

The larger problem is that we're all trying to wear too many hats. Composer, arranger, orchestrator, player, mix engineer, technologist. There are about a zillion different variables with no cookbook. Everyone is approaching all of this differently. Herein lies the challenge and the frustration. I'm just not going to live that long. I've leaned to just enjoy the process. All of this is just great fun.

 

Hey Paul, I use MIR for my AM winds and SM brass, Teldex Wide with omni mics. With the winds, I keep pulling back on the wet/dry mix, trying to find a satisfactory blend of simulated close/tree mic blend. If they're too wet, they don't sound right. The brass, however, I found need to stay at a full 100% wet sound, since the orchestral brass sound relies so much on the space. But at 100% wet, I get a tinny sound. What I've done is put an EQ dip around 10KHz and it helps. Still not completely satisfied with the sound, but it's the best I've been able to do.

 

Yeah, I wasn't sure if I should use that acronym yet or not, considering the controversy with the split up. AM for Audio Modeling, which recently split from Sample Modeling. I guess I should have said SWAM. I'll try the MIR % offset, which I usually leave at 0.0 Also, I've never done anything with the Air Absorption. What's your experience with that?

I'd much prefer to work with baked-in room sound (which is what I have for strings), but I love the musicality of the SWAM/SM instruments combined with a wind controller. So I'm trying to make those work.

 

This is an interesting air absorption calculator.
https://www.sengpielaudio.com/calculator-air.htm
You can see that the effects of air absorption are much stronger at higher frequencies above 10k. Below 5k where most musical sounds are, the effects of air absorption are much more subtle. So in the typical distance of a concert hall, it would only have a slight "dulling" effect (affecting mostly the overtones) compared to hearing the instrument up close.

 

Careful with MIR.

it's useful, but you really have to decide if you want to use the default instrument profiles or just go with the default when you load it up.

for stereo signals(recorded in place already) if you want to use MIR on it, use the basic "mic" when you add it, widen it significantly and simply slide it farther back into the room.

Sometimes I place 3 in the room(one reasonably wide infront of the "listener", one half way back into the hall reaaally wide, and one at the back of the hall less wide) each one gets their own EQ. This is nice if I want to say take something like a spitfire library, and dip a little of each mic into them as a send to get some tails.

I don't suggest using EQ too much before MIR - because it already does a lot of those things on it's own when you run an instrument through one of the presets...

Another huge tip would be to use those EQs on a per library basis. For instance, whenever I use it for SM brass - I create an EQ for the SM brass. I've got one for Audio Modelling as well, for instance.

you can have up to 32 room EQ's set up, just as an FYI.

That said, sometimes it sounds better, sometimes it sounds worse, I just use my ears. I will say this though... lately I've rarely been using any real reverb on my sample modeling... though EQ, parallel compression, and a bit of basically early reflections from right inside kontakt and I've been enjoying the sound.

 

can you reduce the wet mix of the plugin in your daw?

In reaper - I find myself having to do this pretty often with MIR. because the dry/wet isn't as simple as truly dry. I find using both the dry/wet slider in the plugin as well as the wet slider outside of it help a lot - because theres a weirdness that goes on with drying on MIR... it keeps a lot of the mono-ness + panning that it adds, the EQing, just reduces the actual tail amount - which almost always ends up being losing the body + volume of the instrument.

So my advice for using it on non-dry libraries is to use your DAW's wet/dry knob. Or maybe it could be worked around with a send - but I'll try to create a quick example of the differences.

 

https://www.dropbox.com/s/3uu4c3x2wu8w76s/Mc Graw Example Dry.mp3?dl=0

dry(this will be the loudest) CH solo cello and berlin contrabass trombone.

https://www.dropbox.com/s/konyx4j290p3c58/Mc Graw Example MIR50.mp3?dl=0

this one is the default "mic" on both, left at 50% on the slider, and -6db. Then moved to their spots in the room.

https://www.dropbox.com/s/g0n19ojgqv03q4i/Mc Graw Example DAW50.mp3?dl=0

this one is the same as above, except 100% wet on mir, but the plugin was reduced to 50% wet from the daw(reaper).

https://www.dropbox.com/s/sgmcq2dym7yzgn1/Mc Graw Example both50.mp3?dl=0

This one is both 50% within MIR and 50% within the DAW. So in theory it's like 25% wet. This is probably a good starting point for most libraries.

https://www.dropbox.com/s/ihb6gq6snxh37k5/Mc Graw Example both50instrument.mp3?dl=0

this last one was 50%/50% again, except I changed the default mic in mir to instrument specific presets(cello and contrabass trombone) and put them back at -6db.

Haven't listened to these yet - and while the examples probably don't sound great - hopefully there will be a noticeable difference between the effects of the dry wet in mir and the dry wet in my daw.

https://www.dropbox.com/s/utig121tag4rmuy/mcgraw.png?dl=0

a sick diagram, hope that helps.

 

off topic. CH solo strings isn't bad(I still prefer SWAM). Berlin Brass on the other hand - I got that expansion just to "touch my toes in" to see if I'd like berlin brass(had high hopes, because I'm a huge fan of the samplemodeling/dimension approach to sections) and I think that library feels and sounds like doo doo butter. One day I'm going to need someone like Noam to hold my hand and show me how to un-suckify capsule.

In regards to mikes template masterclass/EQing/Z space - always follow the golden rule! Trust your ears! I know everyone who makes a guide/lesson on things like EQ always says that - but some people ignore that and think simply monkey see monkey do might get them good results - and all it'll do is make you take a trip out of your way before you inevitably have to just go back and use your ears and fix it.

often times I'll just have Pro Q up, chop a section of something out of an MP3 loop that one bar or so with space in between - play it in on the instrument and just leave it looping while I just nudge things around on pro Q until it sounds similar, then I cut the wetness down to like 80/85% for heavily processed instruments(like SM) or down to like 45-55% for ambient libraries like the spitfire AIR stuff. If you're using libraries like berlin - you probably don't want to EQ the teldex out of your samples! find the sweet spot and then back at off just in case**

 

in my example, I used berlin brass exp - using close, and 2 other mics(I think ortf and surround or tree?)

as far as Cubase goes - can anyone confirm this? Truly I'd imagine they'd have dry/wet knobs for plugin effects - or I'm just severely spoiled by reaper in this regard...

It makes parallel compression 10000x easier, because I just put a compressor on, then Dial it back externally until I'm happy.

That and parallel EQ is my standard practice I suppose because of this functionality. I'll clean the low end, and maybe apply a Z-space style EQ setting and then pull some of the original signal back in, or simply blend the EQ'd signal in until it starts to sound unnatural and then back it back off.

I only used Cubase for a month or two a while back - and was frustrated often because things simply weren't where I thought they would be - kind of like when you're a kid and your mom reorganizes your things when you're sleeping over a friends house and now your favorite toys are missing. That said - I'm basically trying to say that I have NO idea where the feature would be - even if I used Cubase. But others here use Cubase religiously and might be able to help if it exists.

 

Sure they do, a lot of insert effects have a dry/wet knob on the GUI. Some old plugins recently had an overhaul and now feature this.
I think you maybe got confused with the pre/post send?