@nektarios said:

The statement went like this:

"Each IR of a venue has a special sound. If you ar going to use a common convolution reverb you will get this "venue colour" up to 3-4 times (for each depth once). MIR calculates for each instruments some IRs so that you have the certain venue finally 100 and more times. MIR also has its advantages it mixes your orchestra just by placing all the instruments grafically."

Hi all

I don't know anymore whether these are my own words but I can support their meaning. Listening to one instrument MIR sounds fantastic. The more instruments I play through/with it the more the colour of the choosen venue appears - this is at least my observation... (Whether this is an advantage or not is probably a matter of taste).

Once I was asked to master a couple of VSL-pieces of a good friend. All the pieces with up to 5, 6 instruments did sound great. But with all the pieces played by lots of instruments (symphonic-size) I had no chance to get out the (bad) colours of the used venue. I had to send those symphonic pieces back without any treatment. We had to mix them with new settings. Example with such a bad venue colour: Real Mix compared with MIR Mix .

We both searched for an explanation for this phenomenon. The obvious answer was in fact that because the certain venue was calculated in each instrument one time for sure (or more times?) the typical colour appears more and more as well - and yes using 3 instances of a normal convolution reverb adds this typical colour only once per instance...

How did we solve the problem? My friend dramatically reduced the wet ratio (fewer venue) just for having a nice depth and then he added one final reverb (algo) over all. Voilà was really more pleasant.

Maybe my theory is wrong - nevertheless, the problem we had and still can have (see the example above).

And also: Because this colour of the venue was obviously calculated deep into the sound of each instrument I probably had no chance to master those pieces in a proper way.

Hope that Dietz can clear up this phenomenon and also how to suppress it successfully.

Best

Beat   

Thank you Beat. This is exactly where I was getting at: what this coloring is and how to effectively deal with it. You had mentioned this to me back in October, and now that my MIR project is growing (using Dimension Strings with lots of other instruments), I think I may be facing the same thing, but I am not sure.

Can you post the same example piece with coloring vs no coloring? That way I can better judge the sound.

 Thank you!

 -Nektarios

@nektarios said:

Can you post the same example piece with coloring vs no coloring? That way I can better judge the sound.

Hi Nektarios

I produced this example for my friend and customer for showing him the problem I had not to be able to master his pieces... uuhh English is not my mothertonge. Unfortunately I don't have an example after mixing it once more. ...This was in 2013... Here is a further example

So, to answer your question: No, I can not post such an example. But: Because I don't use MIR (sometimes MIRx) you can compare your pieces with all my newer ones... not any contains those MIR-colours from the examples above even if I used more instruments (example (playalong)).

So I hope Dietz can give us/you some tips.

Beat

Hi Nektarios,

Now I see were you're coming from. :-)

According to Beat's logic, you would get less disturbing "color" as long as you use the same IR over and over again - instead of using many different IRs from many different spots.

Imagine what would happen in case of a real orchestral recording: Typically, you will have your players seated in the same hall, some closer to your listening position, some farther away, each of them pointing into slightly different directions - thus interacting with the hall in many, many different ways. This is what the MIR concept is all about: There are literally thousands of individual IRs, each of them with it's own frequency spectrum, volume envelope, phase characteristics, early reflections patterns and other specific attributes.

By comparison, the conventional "AUX-send to reverb"-approach puts all instruments in the same position. The same frequency spectrum, ER patterns etc. will get used over and over again, without further distinction of any other characteristics of the dry signal than volume and (maybe) pre-delay. _This_ is where _I_ expect to hear some exaggerated “coloring”. ;-)

Beat is used to the AUX-send reverb scheme and he uses it in his own productions with good success, it seems. But there's so much more to MIR than just reverb (and “simple panning”).

AUX-sends are a valid approximation to “reverb” in a mix (and have been used since the early days of multi-track recording), but this approach won't give you much impression of a halls actual acoustic heart and spirit. Talking about “a MIR Mix” is as meaningful as saying “let's talk about the human voice”. There is not just _one_ voice – in fact there are countless pronounced, yet individual colors. The same is true for halls, be they real or virtual: Each of them is different, and each of them will exhibit sonic peculiarities which may (or may not) fit your musical needs. If you don't like what you hear, then you have to make artistic decisions: You could select another Venue, or you will at least have to make a few adjustments to “play” the hall, like you would do in case of any other instrument (and like you would do in case of any other recording/mixing approach, too).

This can be as simple as reducing the overall “wetness” in MIR Pro by means of the Global Wet/Dry Offset. You could achieve your goal by choosing a different Main Microphone position or microphone setup, or by changing an instrument's position and/or rotation on stage, or by using Room EQs for all or just a few selected instruments, and so on.

This is something that's inherent to real recordings as much as to their virtual counterpart in MIR Pro, so I can't see any unwanted “coloring” here, thus there's nothing to “suppress”, like Beat implies. Quite contrary - this is exactly what we hoped to achieve when we started with MIR's development over a decade ago, and it's one of the underlying principles of all VSL products: The colors are all there, it's up to you to use them. :-)

Kind regards,

@Dietz said:

... so I can't see any unwanted “coloring” here, thus there's nothing to “suppress”, like Beat implies. Quite contrary -...

Dear Dietz

On the one hand is the theory and on the other hand are results.

The first reason why MIR (and the venues) "colours" the music: Since VSL produce their demos exclusively trough/with MIR all of them sound more or less the same way - the MIR-way. The huge versatility of the dry sample-demos in the earlier days disappeared completely. This means for me: MIR has its own sound, given by the venues and the way it mixes instruments - always the same way. Funny is, that even if we have lots of parameters we can recognice whether a mix is done by MIR or not. The second and main reason why MIR colours the sound are the venues. Those venues are calculated a lot of times into the samples. Sometimes we have results which are not the ones we wanted to have... Please listen once more to two examples for showing these colouring-mixing facts:

A) Realsound - Orchestra trough MIR So what on earth added this boxy or cheesy sound? The samples itself sound way more natural.

B) DS-Cellos - from an official VSL-Demo...  Now listen to this Cello please. Which one sounds more free from added "colours"? Who or what added the potty sound into the DS-Cellos? It is not me who imply anything, it is MIR and its venues which applies such (sometimes extrem) colours. Each typical "room-colour" is probably OK and the sound we want, but what shall we do when it gets too much? ...as we can make out in the two upper examples?

I always use the EQ within your Convolution Reverbs for correcting (suppress) the main colours of a certain used venue. Maybe you could do this as well a bit for the venues of MIR (by default). Then the "common user" will get even better results with MIR without doing anything and in any case - even he use a 100% wet ratio.

All the best

Beat

[I just wrote a lengthy answer which my computer decided to destroy before I could post it, so I'll make it short now.]

Beat, of course you're entitled to dislike MIR, and you can even call it “cheesy” on this forum (run by audio professionals who devoted over a decade to its development, BTW) – but please stop presenting opinions as a matter of fact, and don't spread misleading information.

@Beat Kaufmann said:

I always use the EQ within your Convolution Reverbs for correcting (suppress) the main colours of a certain used venue. Maybe you could do this as well a bit for the venues of MIR (by default). Then the "common user" will get even better results with MIR without doing anything and in any case - even he use a 100% wet ratio.

 It has been said and written over and over again, both in the manuals as well as in these forums: MIR isn't meant to be used ”100% wet”. Never ever.

 The “common user” is an artist, hopefully, who is able to sculpt his/her music in every aspect. While MIR Pro has a nice, intuitive GUI, it was never meant to be just an easy solution for the heedless, but a deep, rewarding new approach to orchestral mixing. Neverthelss we did “a bit” 🤔 of preset work for MIR Pro, too - we created several thousand of them, strictly speaking, for individual instruments as well as for all available MIR Venues:

Vienna MIR PRO Venue Presets

RoomPack 2 – Herb's Teldex Presets for MIR PRO and VI, and  RoomPack 2 – Herb's Teldex Presets for MIR PRO and VI PRO 2

MIRx Mode – Instrument and Venue Presets for Vienna MIR Pro

 … apart from that I'm working on MIRx presets for two additional MIR Venues, one of them brand new (and again very different sounding). That's another 700 tailor-made presets.

Looking at the countless mixes and productions done all around the globe with MIR and MIR Pro during the years, it's quite daring to call all of them “cheesy” and “potty” (? … did you mean to say “boxy”?). While I know that your taste differs from mine  considerably, I think even you could hardly call the sound of this piece anything else than amazing (at least that's what Alan Meyerson said when he heard it): 

Dimension Strings Complete - Trailer 

… and I'm sure that you would hear dozens of other great examples, too, either on VSL's homepage or elsewhere - if you would just allow yourself to find them. 

Regards,

Thank you Dietz for your detailed response. Things are more clear in my head now for sure. I like MIR PRO a lot, and it's tough to go back to the old way of mixing for me. In terms of colorization, I realized a user can end up adding his own "coloring" by incorrectly applying a room eq to instruments, or a character preset. I think this just shows the great flexibility MIR PRO has when somebody wants to sculpt the sound to the sound they are looking for.

Beat, thank you also for your detailed response. Personally, I like all the examples you showed (MIR ones and non MIR ones), I just don't have the ears you have.

I think to better judge the product, you need to own/use it, and see what attributes you need to tweak to get the sound you are looking for. If you do this, I'd be very interested in getting your MIR template. 

Also, on the non-MIR examples you showed, I realized there is a longer reverb tail which adds to the pleasing sounds.

Dietz, maybe as a future feature for MIR PRO, it would be great for the ability to add an algo tail from MIR directly. I see myself adding this almost always (sometimes increasing the tail to 6 seconds). So would be nice if I don't have to do this with MIRacle.

And yes, I almost always tweak the dry/wet ratio, even for individual instruments.

Cheers,

-Nektarios

Kind of a strange thread.

MIR adds coloration to the sound - well that's quite the shock, isn't it. Why else would you want to put your sounds in a "room", if not to add the, uh, coloration of that very room? 😕 And what's the "natural" sound of an instrument anyway? It's gotta be played or recorded somewhere. And even completely dry samples still get "colored" by the used mics ... so, I don't know, it's a strange discussion to me.

I can see the possibility of misusing presets (which I have done myself), which would introduce this notion of "coloring" that is not desired. Which is why, I think, a character eq is not exposed to the user.

How did I misuse presets? I took the room eq preset (i.e. cello) of one venue, and applied it to another.

@nektarios said:

I can see the possibility of misusing presets (which I have done myself), which would introduce this notion of "coloring" that is not desired. Which is why, I think, a character eq is not exposed to the user.

@Another User said:

The “common user” is an artist, hopefully, who is able to sculpt his/her music in every aspect. While MIR Pro has a nice, intuitive GUI, it was never meant to be just an easy solution for the heedless, but a deep, rewarding new approach to orchestral mixing.

By "heedless" here, you mean people who'd rather work on composing scores than mixing and mastering? Heedless people like Mozart, Beethoven, and Tchaikovsky?

Dietz, 

I was thinking, maybe MIR PRO can introduce the ability to increase the reverb tail of a venue given the characteristics of a venue. MIR PRO already does this by means of reducing the tail if you wish. For example, on the Mystic spaces, we can reduce the tail which is 8 seconds long, but as a result, this results in all different sound for this particular venue. I would love this feature reversed, where any venue can have the tail increased to achieve the desired sound. This would be so helpful.

@tek0010 said:

The "natural" sound would be the air-pressure wave emitted by the instrument in an anechoic chamber:

https://en.wikipedia.org/wiki/Anechoic_chamber

It was a rhetorical question. Usually, sample libraries aren't anechoic.

My impression was always that VSL was recorded in an anechoic chamber (or very close to it). Other cheap sample libraries usually are not.

Vienna Instruments aren't recorded in an anechoic chamber; far from it. There is quite a lot of "room" on the samples. It is easy to hear the difference when you put them side by side with samples that are recorded that way. If I get a chance next week, I'll put a little audio example together so that you can hear what I'm talking about (sorry, no time before then)

DG

Thank you DG! Btw, the reason I even started this thread, is because of a sample mix you posted a few months back for one of my Dimension Stings pieces (playing a chord progression). I remember you sent me the EQ settings so I can apply them myself in MIR, but it never quite sounded the same yours. Unfortunately, you didn't use MIR PRO to create it. The version I have now sounds much better (with the use of VSL EQ presets), but I could never quite make it sound like your mix.

@DG said:

Vienna Instruments aren't recorded in an anechoic chamber; far from it. There is quite a lot of "room" on the samples. It is easy to hear the difference when you put them side by side with samples that are recorded that way. If I get a chance next week, I'll put a little audio example together so that you can hear what I'm talking about (sorry, no time before then)

DG

Here is the piece I am currently working on (still in draft mode):

Eastern Dream.mp3

It's MIR PRO/Dimension Strings/Chamber Strings/Horns 8/etc.

(I also use a non VSL instrument, the Ukrainian Bandura).

Here is my VE PRO 5 Project File:

https://dl.dropboxusercontent.com/u/33556625/Eastern_Dream_All_06_27_2015.viframe64

Any suggestions or improvements would be greatly appreciated!

Thank you!

-Nektarios

Hello all

Convolution Reverbs use Impulse Responses. IRs are acoustic copies of rooms. All! rooms have their resonances, room modes... called "colours" in this thread.

Here is a short instruction how to equalize those colours of a certain room - using the Convolution Reverbs of the SUITE EFFECTs of VSL...

http://www.beat-kaufmann.com/vitutorials/vi-tips--tricks-4/index.php (No. 36.)

All the best

Beat

Thank you Beat! Very good article. It's interesting because right before I read your post I decided to do an EQ-Match on the piece I posted earlier. I basically matched the EQ with an orchestral piece I like. I was amazed at the difference! Here it is now with the EQ matched:

Eastern Dream (Matched EQ) mp3.

I can def hear a difference. The matched one sounds cleaner and less muddy for sure.

I also used CurveEQ.

I can certainly hear a difference as well on your example Beat.