To be able to compare the diagrams more easily, I am copy-pasting my previous diagram-post below:

See my answer above. In terms of processing power, the amount of MIR Pro engines makes little to no difference in terms of both CPU and memory use. The amount of running MIR Pro plugins inside VE Pro is what really matters.

Thanks a lot Karel!!! That's great news. It's 22:00 by the way, what a commitment. Very much appreciated!

So I understand now, the DAW runs a single VE Pro server, from which I start the VE 1, VE 2 and VE 3 instances on that same DAW. Each of those VE instances runs a separate MIR Pro engine instance. And all that without requiring extra licenses. That sure proofs I was not optimistic enough.

Thanks again, for your answers and not to forget, for your work that helped produce this wonderful software.

My pleasure :)

In case you're really need to have MIR Pro on more than one machine at the same time, the upcoming "MIR Pro 24" will be a reasonably priced option for additional licenses.

Kind regards,

Thanks for the kind words - just let me add the that "24" was not my idea. ;-)

It works!! I've tested it using the MIR Pro Trial with the 'start venue' and the Mozart Saal trial venue. I am using Cubase 6.0.5 on a PC. The VE Audio Plug-Ins are in FX channels that throw 'instrument groups' into the multiple MIR Pros, which all run on one DAW.

So this is the final (working) diagram:

           ---> DAW VE VST 1 ---> Slave 1 VE Host ==> DAW VE VST 1 ==>                                                     DAW VE 1 ==> MIR 1

DAW                                                                                                              DAW ==> VE Audio Plug-in ==> DAW VE 2 ==> MIR 2 ==> DAW

           ---> DAW VE VST 2 ---> Slave 2 VE Host ==> DAW VE VST 2 ==>                                                     DAW VE 3 ==> MIR 3

Thanks a lot, Karel and Dietz!

I'll post quantitative results when I have them. But for now, with 3 MIR Pros running and five audio input channels each, all playing, my 2600K system has an 18% CPU load in Task Manager, for the whole system that is. Not bad at all, and much less load then I expected. This really sounds promising.

Good to hear that the setup you have in mind works.

Sidenote - do you really plan to mix the sound of different Venues, or was this just for testing purposes?

Kind regards,

@Bas10 said:

It works!! I've tested it using the MIR Pro Trial with the 'start venue' and the Mozart Saal trial venue. I am using Cubase 6.0.5 on a PC. The VE Audio Plug-Ins are in FX channels that throw 'instrument groups' into the multiple MIR Pros, which all run on one DAW.

So this is the final (working) diagram:

           ---> DAW VE VST 1 ---> Slave 1 VE Host ==> DAW VE VST 1 ==>                                                     DAW VE 1 ==> MIR 1

DAW                                                                                                              DAW ==> VE Audio Plug-in ==> DAW VE 2 ==> MIR 2 ==> DAW

           ---> DAW VE VST 2 ---> Slave 2 VE Host ==> DAW VE VST 2 ==>                                                     DAW VE 3 ==> MIR 3

Thanks a lot, Karel and Dietz!

I'll post quantitative results when I have them. But for now, with 3 MIR Pros running and five audio input channels each, all playing, my 2600K system has an 18% CPU load in Task Manager, for the whole system that is. Not bad at all, and much less load then I expected. This really sounds promising.

Hello Karel

Will this work with Logic too?

What is the trick to send my 96 audio tracks from the 1st VE server to the VE server that contains  MIR ?

Best

Cyril

Dietz, It was just for testing purposes, to validate that the different MIR instances indeed have no problem loading different venues. It all works great, as expected from software of your standards, and exceding expectations in the processing load department.

I assume that the Grosser Saal venue probably has a (slightly?) higher load than the Mozart Saal because of the longer tails. Then again, the load of the start venue is probably lower. Based on what I saw and even more on what I heard, I sure am going down the MIR path now. I can now sculpt out beautiful staging, hall presence and finely simulating close micing and combine them into a kaleidoscopical sound field. Although somewhat limited by the choice of venues in the trial, I already got some pretty pleasing sounds out of it.

Although it is not my intention to mix venues, in fact that's what the Berliner Philharmoniker did ages ago. They used to reverberate their recordings in a Berlin church, at night, to later mix it into the to be released recordings, reaping great results. Then who am I to argue with them that mixing venues is a bad idea? ;-)

By the way, that 'start venue' isn't bad at all. In its preset form it gives a nice staging, and if you increase the reverb time the tail isn't bad either, is that the Neuer Saal perhaps?

Cyril, although I am not Karel, having a few years of experience in Logic, I see no problem of getting this to work in Logic. As a matter of fact, the crucial part, the VE Pro Audio Plug-in, is explained in the VE Pro manual on page 14 with screenshots of Logic.

My approach is to pre-mix the orchestra instrument groups into sections. Percussion, Brass, Woodwinds, Strings (and Piano), and for each a left, middle and right stereo channel, that gives me 12 channels, so still 4 stereo channels, per VE Pro slave that is, for special things, a Harp, a soloist, a special directivity profile needed somewhere, etc. I see it as a stage partitioning. Indeed, it might be somewhat limiting, but it appears to be just what I need at the moment,

Hi Bas10

96 tracks is not a typo, in fact it is 97 VI and 3 audio from Play, K4 and Ominisphere.

Best

Cyril

@Another User said:

By the way, that 'start venue' isn't bad at all. In its preset form it gives a nice staging, and if you increase the reverb time the tail isn't bad either, is that the Neuer Saal perhaps?

No, it's just one IR position from Vienna Konzerthaus - Grosser Saal (... I think I even mentioned this in the manual somewhere).

Kind regards, 

@Dietz said:

...you would run into problems when mixing the wet _and the dry_ signals from different Venues. Consequently, MIR Pro' upcoming "Secondary Microphone"-feature will only output the resulting wet signal components, too.

That's indeed what I thought, so I have set the 2nd MIR Pro to "Wet Solo" (as it is called in the MIR GUI if I remember correctly). The funny thing is, when I forgot to do this for the 3rd MIR Pro instance, no problem arose. Perhaps the explanation is that MIR Pro is working better than you think 😉 No phasing occurred, the dry signal was just playing at a higher level, and this pointed me to the fact that I forgot to cut the dry signal.

Thanks again, Dietz.

@Bas10 said:

My approach is to pre-mix the orchestra instrument groups into sections. Percussion, Brass, Woodwinds, Strings (and Piano), and for each a left, middle and right stereo channel, that gives me 12 channels, so still 4 stereo channels, per VE Pro slave that is, for special things, a Harp, a soloist, a special directivity profile needed somewhere, etc. I see it as a stage partitioning. Indeed, it might be somewhat limiting, but it appears to be just what I need at the moment,

Hi BAS 10

Why does an instrument has a profile in MIR ?

Does this interfere in the calculations of MIR ?

If you want to have a 7.1 surround do you have to send 8 channels per VI instruments to the MIR computer  ? 

I will be very happy if either Dietz or Karel could respond to my previous mail ?

Thanks in advance

Best

Cyril

@Bas10 said:

[...] That's indeed what I thought, so I have set the 2nd MIR Pro to "Wet Solo" (as it is called in the MIR GUI if I remember correctly). The funny thing is, when I forgot to do this for the 3rd MIR Pro instance, no problem arose. Perhaps the explanation is that MIR Pro is working better than you think 😉 No phasing occurred, the dry signal was just playing at a higher level, and this pointed me to the fact that I forgot to cut the dry signal.

Thanks again, Dietz.

You're welcome!

It's not so much about phasing, more about the blurring of the precise positions of the dry signals. Due to the very different dimensions and architectural details of the individual MIR Venues they will hardly ever match.

Kind regards,

@Another User said:

If you want to have a 7.1 surround do you have to send 8 channels per VI instruments to the MIR computer  ?

Like pointed out several times already, you can't send surround signals into MIR. This wouldn't make much sense anyway.

As you may know by now, MIR Pro is able to _create_ most common surround formats, though.

HTH,

@Cyril said:

Why does an instrument has a profile in MIR ? 

Like mentioned in the manual, MIR's "Instrument Profiles" act as a meta-layer of information between a signal and MIR's processing, to allow for the highly realistic interaction of the room and a specific signal source.

Hi Dietz

Thanks for your answer, this confirm my thouhts !

So it is not a very good idea do to what Bas10 does, i.e. doing a pre-mix of the "pupitres" ? desk ? 

Each instruments MUST go into MIR SEPARATLY  and it is the reason you did not answer to my previous post.

I have noticed that in VSL there is a habit not to answer to post when the answer is NO, I would prefer you just say "no it is not possible", than to wait for days for an answer.

Best

Cyril

YES, we are bad, Cyril. And YES, sometimes we don't answer posts around midnight anymore like I do now. (... see, again not a "NO" as an answer. *roll*) To say it quite clearly: It's not my favourite pastime to dedicate _so_ much attention to someone like you who is badmouthing VSL-staff again and again.

Obviously I don't understand what you're asking for (what's "pupitres? desk?" ...?) - or you don't understand what I'm answering.

Of course you can send submixes into MIR (... BTW: every ensemble available as Vienna Instrument is one, actually). Just make sure that you use one of the simplified "General Purpose"-profiles as soon as there is no dedicated Instrument Profile available. Depending on the signal and the chosen settings, the results will be pleasing most of the time, sometimes they will be just so-so, or (in rare cases) simply unusable.

Regards,