I think it works on the idea of bouncing your tracks seperately.. IE Violins is track 1, Viola's Track 2 etc.

And then you just import the tracks into MIR for mixing.

That's how I understand it. From what has been said, MIR uses to much computer power to allow logic or cubase to operate at the same time to allow realtime sequencer streaming ... what I mean is, keeping your VSL midi as it is and letting MIR handle the mixing .. this seems to be a no no.

SO I think importing rhythms would be easy. In fact any sound at all. But It's made more for orchestral instrument mixing.

In anycase, there's no problem in doing the mix down of the VSL stuff and then overdubbing your beats and what-nots at a later time.

God I hope your wrong [:)]

It would be a very unwanted step for me to first bounce everything, and then import into MIR for yet another bounce. I really hope that you rather just set up MIR as part of your template like you would any other verb. Then you compose your music, and change your MIR parameters as you like them. And when the peice is done you bounce it all into a finished WAV file (or whatever). I dont have time to bounce stuff, import to MIR, bounce again, send to client who wants a few minor tweaks which then require me to do it all again (bounce, import, bounce).

Dietz?

Well .. if you think about it. How would you manage the positions of individual instruments from a stereo bounce!

The way I described, bouncing seperate tracks, is the only way I can think that it would work. From what has been said about MIR on it's uses of computer resources, it doesn't look possible to run this at the same time as a sequencer. Of course I'd rather it be an AU plugin that fits right into logic, but agin from what has been said I don't think we'll see that anytime soon. Perhaps they will think of some way to be able to network two computers (one running sequencer, on running MIR), and do it that way, but how that would interface with logic, or cubase I have no idea. It'd probably work out more costly for software.

In any case, if you had to bounce each track it really wouldn't be that bad. Logic has a facility to bounce all the parts to seperate tracks so long as you name the parts and tracks right you can't go wrong. (And that process only takes 10 mins or so). So importing them to MIR would be actually fairly simple.

Perhaps it will be different, but I can't imagine how else they would be bale to manage what MIR does.

Well. I figured that some data (audio, midi?) was transfered into the machine holding MIR. The MIR machine then takes this clear signal and places it in the room and adds other parameters. In audiotion mode when we play we here the result (although for real time its comparable to other verbs). Then at some point we tell MIR to render - now its offline and slow, but the result is far more detailed.

Thats how I hope it works.

Ok. One reminder: I think it's safe to say, that MIR won't be availible in the next half year. So I don't think you're doing anyone anything good with wanting specific answers to specific questions that may even change until they developers know what the final product will look like. There are many ideas behind this, and some will make it, and some will not for various reasons. That may also be the reason they are not so responsive every next minute you ask, Christian. Take a deep breath, and just enjoy what they're telling you beforehead [[[;)]]] There's still enough time in the 4 weeks between announcement and release to discuss details, hehe [[[;)]]] Otherwise - perhaps VSL staff could make you a sticky MIR thread [[[;)]]]

So far I understood the MIR concept the following way:
You keep your current slaves, you keep your current DAW. You add a powerful additional machine for MIR. you connect your DAW to your slaves, the slaves to the MIR machine, and the MIR machine is your final output (that can be connected back to your DAW also). Some numbers mentioned 4 up to 64 slaves that could be connected (I guess technical limits will be the factor here). I'm sure that when it's called "offline" bouncing that you can bounce multiple tracks at once, coming from all the different slaves.

I'd not even wonder if they'd implement a FXTeleport like engine that is sending the audiotracks via Gigabit network to the MIR machine, as it would make perfect sense to me. Handling different audio streams from various slaves would be quite an easy task then. Only thing that has to be taken into account somehow, that offline bouncing may take longer than online rendering from the slaves - well, buffer the data that comes in, voila.

Once the different tracks are availible in the MIR machine all the magic is happening that Dietz so often described: mixing, balancing and reverbaration are no longer processes seperated from each other, but all controlled via one interface, with the possibility for presets and automation. A multitude of parameters are dealt with here: Position, elevation, stereo width, directivity, filtering, volume (both of the dry and the "wet" signal), and "some less self-explaining parameters" that Dietz wouldn't let out for us.

So once the offline thing is done you got one or maybe multiple audio files for further editing - would make sense for postproduction - that is if MIR will support multi output.

So that's my point of view on this. A user's point of view just reading quite a lot around in the last days (man why do you guys post sooo much in a month where I'm absent).

Hope this helps,
PolarBear

Thanks for your input Polar - Your understanding is somewhat similar to what I have. Regarding asking questions - well I'm only asking so much because I'm not getting any answers... answers I need to decide what to do with my setup now that I'm basing it on the VI Cube. In fact a simple "we can't tell you anything, becuase MIR is atleast ½ a year away" would be immeasureably helpful response.

Oh, and good luck reading up [:)]

They can't tell you anything more than they already told you, because MIR is at least half a year away from being released. [:P]

Anyway - I'd not recommend to change your setup anyway, regardless what they tell you, because as I said, they might change things they don't even think of theirselves.

Oh, and, I'm almost finished reading up already...

hehe... Well as I'm going from Pro. Ed. to VI my setup has to be changed - no choice [:)]

Well you dump out Giga or EXS for the VSL channels, throw in VI and then what?

Then I need a badass convolution verb. I'm looking at AltiVerb which should be released any time now for XP. Problem is its a one computer license, and I would really want to use the new instrument placement featre on all my instrument sections. I would need an install on each computer - thats A LOT of money. If I go down that route, there is no MIR for me. I think I would prefere MIR though. I could also just get some cheapo verb while I wait for MIR, but i'm not a 'half-solution' kind of guy... and I need to offer my clients the best [:)]

So if I knew the approximate release of MIR I would be able to make a qualified decision on what course to take now. I dont want to throw money out the window, but I dont want some half assed solution either.

Sounds like you're not using a verb yet [:P] [;)]

Actually I'm not [:)] That is not with the VI...

I'm still finishing up some tracks with Pro Ed, and waiting for the next 5 collection until I make the giant leap.

So if you come along without any verb now, you will also in the future... if you use SIR now, you can do that in the future too, if it's W2 now for you, that's also possible with the VI, same goes for the Waves IR-1 or any other convolution plugin or verb. But if you have to have an all-new setup, you're right, choices might have to be made. [;)]

PolarBear

@Christian Marcussen said:

Then I need a badass convolution verb. I'm looking at AltiVerb which should be released any time now for XP. Problem is its a one computer license, and I would really want to use the new instrument placement featre on all my instrument sections. I would need an install on each computer - thats A LOT of money. If I go down that route, there is no MIR for me. I think I would prefere MIR though. I could also just get some cheapo verb while I wait for MIR, but i'm not a 'half-solution' kind of guy... and I need to offer my clients the best [:)]

So if I knew the approximate release of MIR I would be able to make a qualified decision on what course to take now. I dont want to throw money out the window, but I dont want some half assed solution either.

Don't know if these comments are of much use but I use Altiverb on 1 mac and have my different instruments output to about 10 busses in Logic. Each Bus has an Altiverb with a particular Orchestral Position and just the early reflections. This saves on CPU usage. The busses then feed to a main output with another Altiverb supplying the tail.

This works well.

But I am planning a PC farm and one idea is to feed sets of instruments from PCs into Logic where the Reverb will be applied.

The sets of Instruments and thus the Busses are:

1 Oboes Flutes Harp
2 Clarinets Bassoons
3 Horns
4 Trumpets Trombones Tuba
5 Vln 1 Vl 2
6 Viola
7 Celli Basses
8 Timpani
9 Percussion L
10 Percussion R

i have just seen the movie about MIR

Many think look like great but one think disturb me: you have many winds dispatched on the stage but there is just 4 position . One for each : v1 ,v2 alti & vc

Is it not realistic for me.

14 v1 on the same place ? we are speaking about musicians, not equilibrist [[;)]]

So, is it possible to dispatches the v1 (and the other) every where we want ?
and if we can, is it possible to divise or multiplicate the position for the same sound or the only way is to open many time the same sound and put it on many position but like this , what about the ram ?

Thank you

Regards,

Thierry [:D]

Don't worry, Thierry, this is not how you think it is.

As you may know, all our samples are recorded in stereo, so they won't sit all "on the same place". There is a simple mathematical relation between the original width of a recorded signal in the SilentStage, and the witdh that the same signal has to have on a given stage with a give microphone-distcance. This is one of the already mentioned "less self explaining parameters" the MIR takes care for in the background.

So the left and right channel of an ensemble will be sent to the appropriate IR-position on any given stage (... actually it is more than just L/R, as any signal inside the MIR is processed in a threedimensional multichannel format, be it the dry instrument or an IR).

Of course it would be more obvious to have 14 small manikins sitting around on the virtual stage, but my idea was to keep the place as tidy as possible, visually [;)]

[[:|]] Its terrible - the more I hear the more I want

Can you speak a few sentances about bouncing etc. Offline/RealTime yadayada...

In a nutshell - no. [H] ... sorry, don't take it personal. This is one of our final construction sites, so to say.

hehe - of course not [:)]

Just dont take it personal thet I ask [:D]

I'm afraid that I just don't "get" it. I have watched the video a number of times and I think that I understand what the goal of all this is. It's just that when I hear the Great Gate start up it doesn't do anything for me at all. I have gone back to the audio only version and I don't like this either. However, I don't know whether I just don't like the programming or whether I don't like MIR. It would help if I could hear some other demos using this technology, because as it stands I have no intention of even considering buying into it. Sorry to be so negative, but I'm sure that you would rather have honest feedback [:(]

DG