Q: Mastering and Recording Enhancement

Marik,

I'm probably repeating what the others have said but I really think you should take your recording to someone else. There are several reasons why -

-Are you sure your monitoring environment is good? When I say good I don't mean nice sounding I mean level matched channels, flat frequency response, room resonances controlled, good speaker positioning and most of all good translation to other systems. A good M.E. will have all of this dialled in (and verified by years of good masters translating well)

-Who did the MS recording? Are you sure they were done correctly? I use MS on almost every acoustic recording session I do and I have often walked into the control room and realised that I left the S mic on cardioid or some other silly mistake. Last week I was recording an upright piano and on the 2 tune the piano was sounding wrong - it turned out the piano player had hit the mics sitting down and moved them slightly (1 inch or so). What mics, pres and converters were used? If the delay between the M and S is real, how did it happen?

-I'm not calling into question your credentials but the way you are speaking about matrixing the MS makes zero sense. I'd be shocked if someone with your musical experience finds the solution in some crazy random solution like those you are suggesting. MS is MS - there's really only one way to decode a properly recorded signal (obviously you can use as much S channel as floats your boat). Are you sure you are decoding the MS correctly (I've made this mistake too).


Get some help or if you like put up a link to a minute of the audio and those of us with some experience in this area can offer real advice.

Best of luck,
Ruairi

[quote author="ruairioflaherty"]Marik,

-Are you sure your monitoring environment is good? When I say good I don't mean nice sounding I mean level matched channels, flat frequency response, room resonances controlled, good speaker positioning and most of all good translation to other systems. A good M.E. will have all of this dialled in (and verified by years of good masters translating well)[/quote]

I do with equipment and in the place where I made about 15 commercial classical piano CDs, as well as hundreds of demos/lives, etc., so here all is good.

-Who did the MS recording? Are you sure they were done correctly? I use MS on almost every acoustic recording session I do and I have often walked into the control room and realised that I left the S mic on cardioid or some other silly mistake. Last week I was recording an upright piano and on the 2 tune the piano was sounding wrong - it turned out the piano player had hit the mics sitting down and moved them slightly (1 inch or so). What mics, pres and converters were used? If the delay between the M and S is real, how did it happen?

Click to expand...

I did the recording and for S used a Fig.8 ribbon, so there is no chance to leave it in cardioid. The mics were in closest possible proximity to each other and never were out of my sight during the session. Once again, when I look at the waveform the channels seem to have about 3ms difference at some places, but dead spot on in others.

-I'm not calling into question your credentials but the way you are speaking about matrixing the MS makes zero sense. I'd be shocked if someone with your musical experience finds the solution in some crazy random solution like those you are suggesting. MS is MS - there's really only one way to decode a properly recorded signal (obviously you can use as much S channel as floats your boat). Are you sure you are decoding the MS correctly (I've made this mistake too).

Click to expand...

Well yes, I realize this solution is quite crazy. I have made quite a few MS recordings and have passive transformer and active decoders, as well as tried it in ProTools by duplicating S channel and then flipping its polarity and mixing M in the center. That's when I payed attention the M-S gives all those problems and this is when I had the idea of moving M towards added S channel. I realize this might even does not make any sense, but that's how it worked. It gives an adequate stereo width, but the piano image is out of center.

Later tonight I will post some links.

Best, M

[quote author="Marik"]Well yes, I realize this solution is quite crazy. I have made quite a few MS recordings and have passive transformer and active decoders, as well as tried it in ProTools by duplicating S channel and then flipping its polarity and mixing M in the center. That's when I payed attention the M-S gives all those problems and this is when I had the idea of moving M towards added S channel. I realize this might even does not make any sense, but that's how it worked. It gives an adequate stereo width, but the piano image is out of center.

Later tonight I will post some links.

Best, M[/quote]

Hey Marik,

I can't be sure from your description that you are decoding correctly in Protools - you probably are but here goes just in case -

1 - the M channel is panned dead centre
2 - the mono S channel is duplicated so you have 2 faders
3 - one of the S channels is panned hard left and the other hard right
4 - the phase is flipped on one of the S channels (if you are using the trim plugin to flip the phase you should obviously insert one on each channel)
5 - it is vital that the pair of S channels are always at the same level, I use a mix group to achieve this.

Often only a little of the S channel is required to give the stereo effect.

When you post the links make sure you give us the M and S channel separately. Ideally a section of music that illustrates the bad midrange you described. Look forward to hearing it.

All the best,
Ruairi

If you're in ProTools, make DAMNED sure that your +S and -S cancel correctly and ABSOLUTELY completely...

If the processing delay compensation is set incorrectly, it will do nothing except mess up your imaging... even a simple polarity-flip will cause incorrect cancelation and make life hell.

Keith

Thank you Gentlemen for your input!
I think I figured out the buzzing matrixing problems. I was trying to EQ the M and S spearately in Protools which completely screwed things up. Using outboard EQ cured at least this particular problem. In any case, the "very flat and uninteresting, like balalayka", as well as "lifeless" sound is still there at the forte playing (as can be heard especcially in the third example).

So here are a few matrixed tracks. The only EQ applied is to tame down somewhat shrill top--a -2db shelf from 2K.
This first one is when the playing is nice and easy:

http://home.comcast.net/~markfuksman/01-AudioTrack_01.mp3

Another one. Pay attention on the difference in the sound when played on piano and then in the same range but the playing becomes more expressive:

http://home.comcast.net/~markfuksman/02-AudioTrack_02.mp3

The third one is the top midrange becomes quite ugly:

http://home.comcast.net/~markfuksman/03-AudioTrack_03.mp3

Hi Marik,

can you give us the M and S tracks separately? Even a 1 minute clip as a WAV file would be very useful. Ideally I'd like to hear the raw M and S tracks with no processing. I'll download the clips you put up and take a listen after 6pm today.

All the best,
Ruairi

Thanks Ruairi,

I have only very limited webspace and will be able to free some more space in a couple days. Will post then.

Best, M

Hi Marik,

that's no problem at all. I didn't get a chance to listen to the MP3s you uploaded yesterday as I worked on my day job until 1am :shock: . I'll get into the studio later today and have a listen.

All the best,
Ruairi