Looking for answer of questions, which can't be find in the MIR-manual

[doubleattack]

Hi all, hi Dietz

I know, some MIR user including myself are pains in the neck.[*-)] 

Sorry! I got the hint from here http://community.vsl.co.at/forums/t/27882.aspx! 

I'm going through the manual  e v e r y  day - surely! -  b u t  :  Sometimes the unsureness is growing!

Meanwhile I have a couple of questions, 'cause I have some time for experimenting with MIR. Keeping myself calm, I will ask them drop by drop!! ;-)  

For today (or better to night) the following:  

Today I noticed a unexpected result, using the stereo width of an instrumenten icon. I did a "much bigger than life" setting - in this way, that both points of the outer ring were  o u t s i d e   of the hotspots (the instrument-icon itself inside). So I expected it would be to hear nothing - obviously an error. I did hear the direct signal   a n d    the space, but the space were sounding different than in the normal way, a bit thinner I would say

and direct and space information were more "disconnectedly". But sounded quite good in a special way, for some effects much useful (for my taste.)

Now I'm wondering where does this different space information (as useally) are coming from? Is the space information "tripled" (to say Mono (from the icon itself) + Stereo (from the outer ring)? So it seemed to me.

How does it work? (Detected in the manual this only: By pushing the width into the "red" areas (if you do it in the instrument panel) this natural width can be exaggerated to a certain among. But what will happen "beyond" (using the general purpose setting cardoid). 

Sorry for my stupid questions, but it isn't out of sheer curiosity, I think it's quite important to know whats happen in this case.

Thanks in advance for the patience!

Best and kind regards

Frank

    

[Dietz]

Hi Frank,

the answer is straight-forward: The validity of an Icon's position is calculated by the coordinates of it's "body" (i.e. the inner circle), while the actual IR's are triggered by the handles of the outer ellipse (which is also the place where the left and right channels of the dry signal will be situated).

So - if you use an Instrument Profile which allows for highly exaggerated stereo settings (General Purpose Profiles are a good example), then MIR will try to interpolate the IR-paramters for any "off-site" position, as long as the Icon itself sits on a valid HotSpot. Although the Ambisonics-format allows for very good interpolation between individual IR-positions on a MIR-stage, it can't move walls 8-) ... in other words, technically the interpolated IR-sets derived from positions _outside_ th HotSpot-area will simply be wrong. OTOH - as long as you like what you hear, nothing is hindering you from using these settings as an effect!

Kind regards,

[doubleattack]

As always, thank you very much for the quick reply, Dietz!

And very good to know!! It wasn't clear for me before.

@Dietz said:

 OTOH - as long as you like what you hear, nothing is hindering you from using these settings as an effect!

 

I like it to use in some cases! It's not for a natural sound, but it's a good choice to get a clear "direct signal" with a "shoot" space (assuming it's more mono, right) - with the rich reflections of the foyer in a long distance (main mic Beethoven - icon on the wall behind the stairs) it sounds like a very good delay! Really nice!! And it increases the possibilties again many times over. So sometimes I will use the "body" of the icon only, giving the Icons "wide wings" to "switch off" the IRs coming from the left and right channels. Doesn't sound good only, it looks good too, like huge dragonflies!

My daughter has been enthralled. She only deliberates about why not is incoming an audience? But that's another question - later!

Best

Frank

[Dietz]

Vienna Konzerthaus Foyer Mic 4 is a special issue in any case as I've discovered a mapping-error just recently 8-) ... this will be corrected with Version 2 (it will be available for all existing RoomPacks, soon).

[doubleattack]

Thanks for this hint! I've noticed, in comparison to mic 3 (center) I have to increase the amount of dry/wet setting. So I was a bit bemused what's happen there. (Hard to say if it were a much different sound, I didn't made a real A/B comparison.

In these days I'm to reinvolve with MIR. Until now I used it mostly in an intuitiv way - and that works fine. (After my first experiments it seems the workflow will don't get so different.) But some things in MIR are very special and I need for some processes of work I did in early times with different tools a new suggestion how does it work in MIR (so to say a translation). So I read a lot of stuff (not only the manual), sometime it's not easy to understand - even if it's written in german.

But I think to get more and more professional results it's important to know lot of details. The relation between venue, instrument mic-setting, main-mic (output-format) , icon-position, setting of inserts, using of character-presets is quite complex (to find out,  w h i c h  tweaks and            w h e r e  in the signal-chain), and if you want a good control about this things it's important to me to know about these things, leastwise a little bit.

So I want not to abuse these forum and the vsl-support for an inquiry office, but would be nice to get for very special questions some precious hints. Thanks in advance for understanding.

Kind regards

Frank  

       

[Dietz]

Frank - you're very welcome. The worst thing that can happen to a VSL forum member is that I politely point her or him to a certain page of the manual ;-) ... and some people will agree when I say that I'm able and willing to put on the emergency brake when things _really_ get out of hand (but we are far from that point, rest assured).

Everything else is interesting both for fellow users and developers.

Kind regards,

[doubleattack]

 Nice to hear. Even though I will try not to carry coals to Newcastle. :-) 

[doubleattack]

Hi all, hello Dietz,

in my last little experiments -  :-D - I've used for the hotspots in the "Großer Konzertsaal" the main mic wide-cardoid  r e v e r s e  (actually for the Foyer). Is that a wrong choice for this venue? (Didn't notice any dissonance - coming from the reverb...) 

I noticed by the way, that the hot spots in these venue are available only with two main mic positions conductor and 7th row. What's the reason for that? So I'm thinking these nice places on the balconies couldn't used for a great dolby surround set up? (I don't have it (yet), but just thinking about, what could be possible - for some special kind of sound art design.

My other experiments to get more and more familiar with MIR I used selfmade main-mic setups. I can do it by try and error only, I noticed you can get with the left-right distance parameter a "more deeper" space-impression (hard to explain what happens, it's more psycho-acoustical). But how to use it in a more controled way?  Probable by experience, I'm afraid. (Seem's to me increasing the left right distance is great to use for bigger ensembles.)

I've tryed out the new teldex venue to use for ERs only (trimed to 1.12 s), and the rendered single tracks to give a vintage plate hall tail.  1.12 s it's quite a lot for the ER's, but what do you think how much of trimming the reverb make sense for these venue?

So thanks in advance for every helping reply!

Frank

     

      

[Dietz]

Whew, lots of questions. 😉

@Another User said:

1.12 s it's quite a lot for the ER's, but what do you think how much of trimming the reverb make sense for these venue?

Errr ... my understanding of _early_ reflections encompasses more the 0.1 - 0.3 s range ... everythin else is just reverb 😉

I know that the idea of using "ERs and adding algorithmic reverb" is floating around the Internet since quite some time now, but I'm still not convinced that the approach is one I would choose. Personally I would rather add some nice Lexicon or t.c. to the whole mix than trying to cut out the tail from the impulse responses (... for me, the reverb tail is as important as the so-called ERs, but that's a different story).

But you know - "If it sounds right, it is right!" 😉 ... as long as _you_ like what you hear, it's ok!

Kind regards,

[doubleattack]

Yes, a lot of questions, I know, thanks again for your kindly reply!

To the last point: Always I've thought that these white spread talking gibberish of "ER" and "Tail" in truth and actually is a blend of two reverbs. Sometimes the predelay (=ER) for the tail is set to zero, but anyway. So I'm using these fogged words too.  

Listen to the really incredible sound of the studio rooms of MIR I got inspired to make a old fashion set up, a virtual studio recording in one of the roompack 2 studios to give on top an old common algo-reverb - (I use a inserted hardware t.c.). It's not what I'm aiming for normally, but the idea behind of those setups for me is to get a special "sounds of common remind". I call this "idioms" to use for quotes, so to speak using a sound for a sign of a period. (I try to compose sometimes in this way, trying to avoid the common crossover, more in a way of rigorously switching, so the sound of the reverb can help a lot to make it clear and convincing, and MIR is inspiring a lot on this way.) Thats why I try to simulate the old recording way.

Thinking about what they did formerly (different in each period for shure), how much of reverb information coming from the original recording they have used in the final mix and how to simulate such things, I thought it could be useful to cut out the reverb tail for a better blend. It's good to know a tenor opinion about now.    

So thanks for your pieces of advice!     

Kind regards

Frank

[Dietz]

@doubleattack said:

[...] predelay (=ER) [...]

No! Predelay is the time _before_ the first and all the other early reflections arrives at the point of the listener. The longer the predelay, the farer away any reflecting surfaces are. - The problematic part about this concept is that it treats a source as flying object, because more often than not the fact that the first reflecting surface is the floor is simply neglected. 8-)

[doubleattack]

So working with MIR makes my day! (the floor shouldn't be the problem anymore? : - )  Yes, I remember, read about making the distant with value of initial reflection etc...absorb behavior of concrete or influence of humidity for sound propagation...or damping high frequencies for farer away sound sources... [8-|] I'm geeKing too. But seems it hasn't any effect on me. LOL  

Didn't care anymore since working with MIR, forgot it and now all it's got into a mess. 

So I will do my lot of absurdism using rather emotions than reasons to express my heartstrings.  

To be serious finally: Thanks for rectification!

Best

Frank   

[mpower88]

Dietz, I was wondering if you could explain more about why you chose the ORTF mic arrangement to record the library, specifically I've posted a couple of years back complaining about finding coherency problems when using Altiverb and large mixes, (specifically that I found that convolution seemed to exacerbate severely coherency issues in the mix to start with) that I couldn't fix. I found the new panning tools greatly fix that, and a test on the MIR trial last week I found that coherency on a medium sized mix was very good (I heard this, and I dug up my Waves Analyser which has a stereo coherency meter) it seemed very good. I remember vaguely this was explained as a future proof method - on the surface ORTF is not a great source for panning and altering of stereo width en masse, but with proper treatment it's actually the best way to do it, leaving maximum flexibility later on - almost as an archival technique, so to speak. But can you speak more about that, I'm really interested in technically why it works so well with MIR, and so badly with other convolutions - not just Altiverb, but all other convolution reverbs as well as simple panners, and even without reverb large mixes with lots of panning and width modification (when not using VSL panning). What is the technical science behind that? (without going into trade secrets!)

[Dietz]

I'm not going into the discussion pro /contra ORTF recordings of instruments again ;-) ... but one of the main reasons why MIR / MIR Pro works _at all_ is that any positioning-relevant processing is done in Ambisonics. All IRs are recorded in 4-channel format A and later transcoded to Format B. All dry signals are encoded to Format B. MIR's Output Format (i.e. the Main Microphone) defines the decoding.

This is also explained in MIR Pro's Manual Addendum "Think MIR!" -> http://dl.vsl.co.at/downloader.aspx?ID=7629

... so - it could that you simply like the way how Ambisonic handles "panning". :-)

Kind regards,

[mpower88]

Hi Dietz, I'm sorry - where was that discussion is there a link? I couldn't find anything... ? Was that discussion with me? [*-)]

Yes I've read the addendum - very interesting everyone should take a few minutes to read it.

[Dietz]

@mpower88 said:

Hi Dietz, I'm sorry - where was that discussion is there a link? I couldn't find anything... ? Was that discussion with me?  [...]

Uh - no, don't think so. 😊 It was more the reminiscence of some fruitless discussions on other forums. [^o)]

("Forums"? Shouldn't that be "fora", actually?)