-
instrument width and tonal quality
Hi all. I've been using Mir with great satisfaction on classical voice, guitar, cello, etc. Curiously, I never had (until now) great results with the acoustic piano. No amount of tweaking EQs or compressors could get me rid of some kind of muddiness. I thought it were my microphones (A B distant, with omni pattern) that were not proper to retranscribe the sound of the piano. Yet, this evening, I simply enlarged quite a bit the width of the instrument on the mir stage, far beyond the width reccomended by the presets for the Bösendorfer imperial, and voila, I got a much better tonal balance, way closer to the original sound, with much better articulation. Has anyone come through this ? Any idea why the piano sound deteriorates when it gets closer to mono ? Ok it could be frequency cancellations coming from the A B setup, but I have exactly the same problem with close cardioid pair. best regards. Stephane Collin.
-
Hi Stephane,
first of all: If it sounds right, it is right. :-)
If you get better results from changing the default width of an instrument (which is meant to be seen as starting point only anyway), then go for it! There's nothing wrong with it.
A sidenote: The "A/B" microphone setup in MIR Pro is derived from the same coincident Ambisonics recording as all other setups. They rely on the same IRs. Different setups come from different Ambisoncis-decodings. In casde of "A/B" the distance created between the caspsules is purely virtual, by means of clever decorrelation alogorithms. It's unlikely that there is perceptible phase cancellation. :-)
BTW - I did several "Venue Presets" for Vienna Imperial - maybe they can give you some ideas, too.
Kind regards,
/Dietz - Vienna Symphonic Library -
Hello Dietz. Thanks for the input. Sorry that I was not clear. I was inputting in MIR a live recording of a piano done with distant omni AB microphones. While on the goniometer the signal looks like mono compatible, I feel there is, in case of this live piano, quite a bit of signal degradation when the MIR instrument controller is set at a normal width, which disappears when I exagerate the width. Not so with my live (stereo) voice recordings or live (stereo) solo strings. I wondered why. Do you mean that you have some presets other than those available in MIR (which are great, for sure) ? Where can I find them ? Best regards, and again congratulations for the VERY nice work done. Stéphane Collin.
-
Hi Dietz. Thanks for caring. Here are my files. This is the dry recording (1896 Bechstein model I piano, Studio Project B3 microphones A/B omni setup 1,5 m distance between each microphone, 1 m from the piano belly) : https://www.box.com/s/e2eolxvwbwjwjusaqpc1 This is the same put through MIR Mozartsaal v2 with the High Directivity planar preset, 1,5 m instrument width, dry wet offset -43 : I hear some kind of mud at 500 Hz or so, and notice quite a degradation of the articulation and the dynamics : https://www.box.com/s/akewlycivlln044x5vrx Here is the same again but instead of 1,5 m, I extend the instrument width to 11 m. All problemns gone, tonal balance is quite similar to the dry recording, articulation and dynamics are back : https://www.box.com/s/4jgr082mfz43rp0qpw8j And here my end result, with a 24 dB roll off at 100 Hz and the Vienna limiter for optimizing level : https://www.box.com/s/3rj8g9dsy0h65drwld8t What do you think ? Best regards. Stephane Collin. p.s. : how do I get carriage returns in the posts ?
-
Here is my second try to make a valid post. (Thanks for the tip; see how quick I learn [:)])
Hi Dietz.
Thanks for caring. Here are my files.
This is the dry recording (1896 Bechstein model I piano, Studio Project B3 microphones A/B omni setup 1,5 m distance between each microphone, 1 m from the piano belly) :
https://www.box.com/s/e2eolxvwbwjwjusaqpc1
This is the same put through MIR Mozartsaal v2 with the High Directivity planar preset, 1,5 m instrument width, dry wet offset -43 : I hear some kind of mud at 500 Hz or so, and notice quite a degradation of the articulation and the dynamics :
https://www.box.com/s/akewlycivlln044x5vrx
Here is the same again but instead of 1,5 m, I extend the instrument width to 11 m. All problemns gone, tonal balance is quite similar to the dry recording, articulation and dynamics are back :
https://www.box.com/s/4jgr082mfz43rp0qpw8j
And here my end result, with a 24 dB roll off at 100 Hz and the Vienna limiter for optimizing level :
https://www.box.com/s/3rj8g9dsy0h65drwld8t
What do you think ?
Best regards.
Stephane Collin.
-
Hi Stephane,
I finally found the time to listen to your examples on a good monitoring system. Thanks for your patience!
I hear quite clearly what you're describing. To me it seems as you're actually longing for the good old "added reverb" sound rather than the proper spatial positioning MIR Pro is able to achieve. :-)
Please take a look at the following little scribble I made for you:
You see very clearly that in reality the perceived stereo-width of a signal will seem smaller as soon as you move away from the signal source. By increasing the width of your piano recording to 11 m in MIR Pro, you get back the the perceived width of the original recording, although your virtual microphone is not in front of the instrument any more.
... although this wouldn't be possible in reality, I'd say that is is OK to do this as long as you like what you hear! After all, it's the beauty of the virtual domain to do things that go beyond the (physical) limitations of the Real World.
Kind regards,
/Dietz - Vienna Symphonic Library -
Thank you Dietz.
That makes sense of course. In the meanwhile, I've been playing with your viframes, and noticed that it is, to a certain point, way more efficient to tweak the microphones patterns and positions in MIR than to rely on EQs. The reason why I didn't notice this before is that I was working more with the VSL strings libraries, and the differences seemed (to me) less obvious. But with a piano recording, being more percussive, it is very obvious.
So nice to discover Mir !
Best regards.
Stephane.
Forum Statistics
195,474 users have contributed to 42,988 threads and 258,260 posts.
In the past 24 hours, we have 2 new thread(s), 8 new post(s) and 46 new user(s).