Hi Francesco

Do you use one and the same convolution reverb for all the instruments. If yes, that's not the way we are used to get such music. My following statement is an universal recipe how to mix band instruments togehter with orchestras.

A) For mixing a band (or band instruments) you should do it in the common way. Just a little bit reverb on the drums guitars etc. without a lot of depth. I would use an algo reverb - one of your DAW. Collect the whole band in a groupe-/bus-channel so that you can mix the whole band with one channel fader. If you also have an E-Bass or a Synthbass you can normally route it directly to the output channel of the project.

B) Mix Orchestra instruments in a typical orchestra way. Use concert room (with a convolution reverb). If possible use different depths for differnt instrument sections. Further: Try to use short fading out rooms or shorten the fading out time of your concert room IR so that you mainly get the depth but not a too long tail. Route the mixed orchestra in another groupe-/bus-channel so that you are able to mix the whole orchestra with one fader as well.

C) Bring now these two different rooms feelings - the one of the band and the other of the orchestra instruments - together which means mix them in the right relation to each other. You also can mix it dynamically and store it as track automation.

Listen to mixes of the London Symphony Orchestra Classic Rock or those pieces of the Vienna Symphonic Orchestra, the Royal Philharmonic Orchestra,... and observe the Band mix and the orchestra mix. They are mostly done as I mentioned above.

----------------------------------------------

This works also with smaller ensembles. Example by Kaufmann with samples (Band mixed as explained above Trumpets in a concert room) or

Happy Music

Beat

Hi Beat, thank you for taking the time to help me. 😊

Yes, the above sample has just one converb.. i see what you say: it's not the right way to achieve depth.. they shown this method in the Special Edition Tutorials so i gave it a try.

I also understand that the sound of the above 8 bars seems a bad combination of band and orchestra..it's not what i meant to do: 1) i accentuated too much the orchestral toms and 2) i tried to add the Overdrive Guitar palm-muted notes because i don't have the staccatissimo articulation for the Cellos and the Double Basses which i wanted to use in order to get a more aggressive feeling at the beginning of each bar. I apologize for such a misleading mix.

I'll go back to a more orthodox use of the orchestral percussions for this piece and i will remove the Overdrive Guitar, i wish to obtain a more natural feeling. Here is the old version of the track which, despite the static sound, gives a better idea of what i had in mind...

again thank you for the useful reply

Francesco

p.s. yes, i will definitely try your method the next time i will do a band/orchestra mix.

Hi Francesco

OK, yes the first mix sounds better over all. But now you could try to bring the orchestra percussion "behind the rest of the orchestra" with a new convolution instance which simulates a bit more depth.

Also reduce the high frequencies of the instruments the farther away they are! ...

...As an example for this matter: Celtic Woman It is not the first time that I publish here this link because we can learn a lot from those mixes. There are playing a huge orchestra, percussions, a choir and several soloists. All those sections are beautifully mixed so that the whole sound comes very transparent. So compare for example the sound of the singers in the front with the one of the choire. The difference is huge concerning the high frequencies. This treatment is applied to all the instruments depending on the distance to the listener. This is one major trick to reach such a transparent mix. Listen also to the sound of other instruments (Strings, Percussion etc.). How do they sound?

Try to copy then all these "cognitions" into your mixes...

Beat

Hi Beat

Thank you for the link. Yes, the Eq difference between the depths is very clear in that mix.. i will try to use this notions to place the percussions behind the other sections.. i'll try to reduce the high frequencies and get the right balance with the reverb tail.. the winds will need a similar treatment to be in the second depth.

Again thank you for helping me 😊

Francesco

I don't particularly agree with 'you should do it the common way' just because it's 'the common way', using an algo 'verb. That's the_universal_way? Not convinced of that and I'm putting it mildly, that sounds more like habit than reason to me.

Typically algorithmic reverb presents coloration issues that convolution does not. Conceptually what that kind of statement does for me is prioritize naturalness strictly according to classical music aesthetically, and relegates drums to a ghetto. Why are they not on a stage?  The orchestra percussion is given a convolution in order to get some back to front. Where is the drum kit, or band?

You know, in BFD3 the sends are to 'room' and 'ambience', not to 'reverb'.

I use Hybrid Reverb on drums in every case with little tail; and the return levels are typically not high. Depending on the mix, though I may use quite some room and ambience in BFD3. It depends on my conception of the mix and is a matter of sound design. I'm seeing lack of depth as a good thing. Why is that? What is the actual idea?
People actually try and get more depth mixing the snare drum. Or let's get the kick drum in your face. That may not be what you want with a lot of orchestra, but I'm not convinced by these assertions which suggest things are *the* way to be or "universally _we_ always do this". Give short shrift and a lower priority to more lowly instruments?

There is a much more involved tool available for free, TDR 'Proximity'. Settings are air absorption, depth, width, proximity, reflections; and there are some modulation capabilities. It's quite good for putting things at the back and it's not this simple trick of rolling off the highs.

Zappa.LSO.Nagano

Hi Civilization 3 and thank you for the comment. You made an intresting point about the rock band placement and anyway in the short sample with the overdrive guitar i imagined to place that one guitarist between the Cellos and the D.Basses, there wasn't the complete rock band (although my use of the orchestral toms gave that feeling). I also understand that in most of the cases when mixing band and orchestra they use the Beat universal method.. all the symphonic rock/metal recordings i've listened to seem to be mixed like that..

BTW for this piece i wanted to remove the overdrive as said, i just want to achieve a powerful orchestral mix. I also remember you told me about the hybrid reverb on orchestral drums to get a more cinematic feeling (and i remember you saying it won't suffice to sound as the EW Stormdrum or stuff like that but it works).. i will give it a try on more cinematic tracks and drumkits.

As said the first version of my piece (second attachement of this thread) sounded static to my ears, i wish to fix the score, fix the dynamics (i wrote everything, i will play instead), and get a more fresh mix. We said a lot about depth but what i am after is "dirt" more than depth.

To be honest "Decision Day" from Christian Kardeis sound very good and powerful to my ears, i opened the project file and noticed that (beside the perfect score) more than the half of the sound is done by the room, he used just on reverb as shown in the vids.. and the pre-fader gives the chance to check the room sound separately but also makes difficult to keep track of the dry/wet ratio for the sections... the whole stuff confuses me a bit.

perhaps i just need to keep practising..anyway it was important for me to share the "work in progress" and have feedbacks

@civilization 3 said:

...Typically algorithmic reverb presents coloration issues that convolution does not. ...

I normally accept that different people have different opinions.

But your statement needs - at least - an addition for saving the algorithmic reverbs.

May be that some algorithmic reverbs colour the sound when those echos and refelections have calculated "colours" or when filters shall add such wanted colours. But Calculating an echo or echos does normally not add any colour at the same time...

On the oposit: Convolution reverbs (IRs) always contain the colour of a certain room, of the speakers which delivered the sweep sound and also the colour of the microfones which recorded those sweeps for finally gitting the IR. If you don't believe that listen to the first IRs of Altiverb they sound horrible. But still nowadays... As an exemple 1 the Beethoven-Piece of exister from an other thread (sorry exister for using the example here once more) or as example 2 ... Both examples are results from mixes of MIR... coloured or not?

OK, this topic treats another theme. But I could not resist, to chime 'in...

Go on and happy music

Beat

@Beat Kaufmann said:

OK, this topic treats another theme. But I could not resist, to chime 'in...

I firmly believe that any online thread has to go off topic after a while....

Anyway since we are here i wish to add a couple of things.. the 3 depths method seems a logical thing to me.. (and i've read your explanation of how the signal is routed into the related Bus with the ER etc.. without sending anything).. i also remember you saying that post-fader is most convinent (in most of the cases) and i agree with that because you mantain the dry/wet ratio... with this method i can simply decide to make a 0db send for the Cello Ensemble and a 1.5db send for the D.Basses which i want a bit further (for ex.) and then play the whole string section and mix the volume of the Basses 'til they sound where i expect to..

Another thing is that i tried the Teldex converb for this piece and, since i have the 3 mic positions for that place, i tried to route the Strigs/Winds/Percussions to the respective stage depth..but i was not happy with the results (and i don't know why).

At that point i listened better to the Special Edition demos and i found that they sounded convincing and powerful to my ears so i downloaded the project files and discovered that they were mixed with a lot of room...more than i expected... it would be nice to learn some details about the Kardeis method, this topic has already been discussed on this forum but no one seems to know precisely why they used pre-fader and there is no way to contact the composer.

Thank you for reading

Francesco

There might be some improvements..

Hi Francesco

I see you are looking for a bit another mix... probably closer to the "Classic Rock Lonodon Symphonic Orchestra"-style.

Your last example has some more pressure... nevertheless it is still drown in the depth of MIR.

Even if MIR supports you in mixing VSL- and other projects and even if it is an easy to use software - for really good results you need to be a master-user with a lot of experiences. So why not trying the way via a conventional mix as well. Group instrument sections / create 3-4depths / route the sections through those depths and youse for all those thepths the necessary EQs, Compressors, Loudmakers and so on until you have reached your personal sound. Once you have it save the project as a staring point for next projects.

This example shows such different rock and classical mixing possibilities:

http://www.beat-kaufmann.com/VSL_New_VI_U/BK_Melody_On_Tour.mp3 (done the classical mixing way)

Have fun

Beat

Hi Beat and thank you for the reply

despite my bad english i feel that "some more pressure" is a positive thing for a mix.. 😊 

just to check i got it: i create a bus (e.g. Strings) set is as the output for all the String Instruments, insert a reverb through the bus and set the wet/dry ratio. For the other sections i use the same convolution (same mic...for example for Grosser Saal i've just one mic) and set a longer tail (e.g. 2.5 for the strings an 2.8 for the winds) ?

Is it what you meant? Or did you refer to that method with just ER in the bus with tail added in the mastering chain?

Francesco

Dear Beat

First of all thank you for the dedication. Please find attached the new sample with the following setup which should be closer to your criteria:

4 Bus (Strings, Woodwinds, Brass+Harp, Perc) and 4 Reverb Bus. Each instrument is routed to the respective section Bus, part of each instrument is sent to the respective section's Reverb with post-fader send (this way i can have slight differences within a section)..then for each section i have a fader which controls the dry signal and another one to control the wet signal.

Far sections have less high frequencies and their respective reverbs have a bit more tail.. the reverb is from Vienna Suite Hybrid Reverb, it's a Hybrid Grosser Saal and i modified only the amount of tail. No comp on the groups, just a bit on single instruments like Timpani, Bass Drum and Snare.. just to avoid the main peaks.

For the mastering chain there is Eq, a bit of Compression (1.5 ratio), Exciter, multiband compressor (without the integrated limiter), finally the limiter and analyzer. No final reverb in the mastering chain.

Does it make sense?

@civilization 3 said:

...Typically algorithmic reverb presents coloration issues that convolution does not. ...

I normally accept that different people have different opinions.

But your statement needs - at least - an addition for saving the algorithmic reverbs.

May be that some algorithmic reverbs colour the sound when those echos and refelections have calculated "colours" or when filters shall add such wanted colours. But Calculating an echo or echos does normally not add any colour at the same time...

On the oposit: Convolution reverbs (IRs) always contain the colour of a certain room, of the speakers which delivered the sweep sound and also the colour of the microfones which recorded those sweeps for finally gitting the IR. If you don't believe that listen to the first IRs of Altiverb they sound horrible.

Well, if you take what I said for what it says rather than reinterpret I think we don't disagree very much. "Coloration that convolution does not" does not say that convolution does not color. The color of a room vs the coloration of a filter, then. I wasn't impressed with the Altiverb so far myself.

My point as to drums is beyond consideration of particular reverbs, in general I treat drums with the same 'ear' as the rest, I said I don't really agree with insistence on algo 'verb for drums just because it's the way we always did it.
For me, algo 'verb particularly for drums [while the rest of the mix is 'real' room?] creates a concern I'd prefer not to spend time on. YMMV. You appear to have such reverbs you love that you're going to stick with. I don't. 😊

In service of a tip, Hybrid Reverb has a setting "X-FEED" which tells the reflections how focused to be.
100% is fully diffuse; 0 % means the reflections focus on where you've placed the thing, as narrowly as possible.
So in conjunction with Power Panning I find this a seriously useful tool.

Hi Francesco

 I believe you have found the way how to use MIR and other Reverbs to get the sound you are looking for. You are now able to place the brass instruments in front or behind another section of instruments. You know how it sounds with MIR without MIR, what algorithmic reverbs can do and also convolution reverbs.

Why do I believe that? Your last version shows it. Congratulation! What a step forward. So you will improve the technique more and more until you have founf your style of mixing your music - I am sure.

@Francesco Pirrone said:

For the mastering chain there is Eq, a bit of Compression (1.5 ratio), Exciter, multiband compressor (without the integrated limiter), finally the limiter and analyzer. No final reverb in the mastering chain.

Does it make sense?

Yes.

 The next mixing step you could take: Where shall I place instruments (left to right)?

I can only make out 3-4 main positions between the speakers (L-R) within your music example but we could make out up to 6-8 of such positions even if we have with bad speakers. So this matter could be done with paying a bit more attention. This PDF contains some thoughts about this matter. See page 12,13,14,15. The PDF summed up: It shows and improves the fact that instruments in the front can cover instruments in the back. This can specially happen when both instruments are using the same frequency range. As an example: Let us assume flutes are placed at the same position in the mix like the violins but "behind" those violins. In this case the flutes will disappear in the mix when they are playing similar notes. What to do... the PDF tells the solution.

All the best

Beat

Hi civilization 3

Thanks for your feedback.

Beat

@Another User said:

In service of a tip, Hybrid Reverb has a setting "X-FEED" which tells the reflections how focused to be.
100% is fully diffuse; 0 % means the reflections focus on where you've placed the thing, as narrowly as possible.
So in conjunction with Power Panning I find this a seriously useful tool.

Wow! I didn't notice it..thank you for sharing this tip, i think it will be very helpful to achieve a clear placement of the instruments and to avoid the overlap issues.

This thread is very useful

@Francesco Pirrone said:

...I am just doubtful about a point: for far instruments i have to increase the amout of tail, its lenght, the wetness, the kind of ER or some misterious combination of these things?

You can do what you like to do as long the result simulats the effect you are looking for.

But also: I would try to simulate the reality in a first step. So increasing the tail is not a natural effect... Reduced to the two main facts which tell us whether an instrument sounds from far away or close to us in a room are

A) the ratio "direct sound"/"reflections (ERs+tail)"

B) the damped frequency in the hights (depending on the distance)

I for having the possibility to place instruments really far away I would select an IR which can serve this purpose. Switch the Convolution Reverb to 100% wet and and compare severa IRs. You will have no problem to find some better ones and some of them which are not very helpful.

Finding a good IR can vary the Distance between close and far in a very different way: Example

If you the also use the EQ for damping the high frequencies (just as seen with the mix of Celtic Woman somewhere above here) you will be successful. BTW: I often use MIRx for Instruments which should be far away (percussion).

Beat

I see what you say, that's perfect... let me try to put these notions into practice..

Thanks a lot

Francesco