Both pairs are far too close. What were you monitoring on, this should have been detected at record time. Assuming its a 9ft grand, we would put the main pr 3-4m or so away and 2.8m in the air, and the outriggers would be 4m apart and some 4m in the air.

Do you have access to some convolution reverb. A little bit of a fine hall added to this recording will improve it a lot.

The room is not good though, this is probably the most important thing to fix or at least compensate for. Tone and image are getting much better though, when compared to your first recording.

Keep at it. I think your next recording will be a beauty.

Its a bit dull sounding also, what preamp was used for the ribbons. I sometimes use some high EQ to spark them up a bit, depending on the instrument, room and music being played. Prokofiev needs articulation and clarity, so in this case I would add some EQ to the ribbons only, not the Schoeps.

Thanks a bunch for the advice.

I didn't have a soundcheck, so I had to go with what I set up. I can't wait 'till next time! :twisted:

I do have Waves IR-1, and I have experimented with adding a bit, but chose to leave it off here for evaluation purposes. I also experimented with high end EQ, but again chose to leave it off.

I used the board pres in a Yamaha DM2000.

Any chance I could send you a multitrack snippet so I could see how you would treat this?

No problem.

DavidSpearritt wrote: Both pairs are far too close. ... Its a bit dull sounding also

This seems a contradiction. What am I missing? Please elaborate.

best regards
Lars

Larsfarm, piano mics can definitely be both close an dull. Up high, a piano is not nearly as bright as you might think.

David, thanks. I will put something together for you soon. I'll post it publicly so others can play as well.

Here's the multitrack snippet:

[[url=http://[/URL]="http://dmfrench2.iw…"]Snippet of multitrack from Prokofiev's Piano Sonata No. 6[/]="http://dmfrench2.iw…"]Snippet of multitrack from Prokofiev's Piano Sonata No. 6[/]
[self-extracting RAR archive: 30.9 MB]

So please, David, show me how you would handle this. Thank you so much for doing this for me. I really appreciate your advice.

David and anyone else, after you listen to the pairs alone, I'd love to hear opinions on what I should be looking to get from each pair, and how far off I am.

Thanks guys, this is a great learning opportunity.

David French wrote: piano mics can definitely be both close an dull.

I'm listening and have a lot to learn. Hence the questions. I encounter piano now and then. Grands as well as uprights. Mostly accompaniment, but I still wish to catch them nicely. Occasionally as a solo instrument.

Lars

David French wrote: [self-extracting RAR archive: 30.9 MB

Uhmmm, is RAR really suitable for audio compression? How much did it compress? perhaps one of the lossless compressions flac or apple lossless that reduces to about half would be better? or (dare I propose it) high bitrate mp3? ( BTW exe will not run on Macintoshes )

best regards
Lars

RAR is the best straight data compression I have. It does a pretty good job for audio, much better than zip. I made a self-extractor becuase I doubt most people can use RAR. I may be wrong. I have FLAC, but again, I doubt most people use FLAC; however, they should. I FLAC'ed up these files and it got down to 25MB vs the 31 from RAR.

If others need something else, I will make it.

Here's some more options, representing just about anything anybody could want.

[[url=http://[/URL]="http://dmfrench2.iw…"]WAVE, 16 bit - RAR archive - 15.7 MB[/]="http://dmfrench2.iw…"]WAVE, 16 bit - RAR archive - 15.7 MB[/]

;)

;)

For the curious...

WAVE = 47,078,200 b
[list]

Anyone needing to compress wave-files for transfer of the net should try [[url=http://[/URL]="http://www.monkeysa…"]Monkeys Audio[/]="http://www.monkeysa…"]Monkeys Audio[/] - Lossless and quite effective!

/ptr

Well, David, I have had a good go at improving the multitrack takes but have not succeeded, there are too many basic problems I'm afraid. Here is a summary of my thoughts.

1. The Royer signal is almost unusable, way too far away, and the image is skewed significantly to the right channel. The Schoeps so dominated your stereo mix that I thought both pairs were close, but not so. In the Royer pr, the piano sounds too mono in the middle (right skewed) when soloing this stereo track. This is the exact opposite of what you should do with a Blumlein pr, it should be too close if anything and fill the stereo spread. You should start positioning it so that the sound source fills the 90 degrees, with piano its about 2-3m out and the virtual image has to be in centre of the pair.

2. The Schoeps omnis are far too close, again the exact opposite of what you should aim for. Omnis should be further away on error margin if you are going to have a chance. Close omnis sound pretty bad, all woofly, too much bass and an image that is all over the place.

So I am not sure I can fix it in the mix as it were. If you swapped the positions of those two pairs you may have had a chance.

Listening to each pair and swapping between the two, they sound totally and completely different. This should ring loud warning bells. When listening to stereo pairs that you are going to mix together, both the perspective and image should sound the same in both pairs. This is the secret to getting great sound in my experience. Seems obvious I know.

Get the reverb/direct sound balance right in the main pair and then make sure its similar in all other pairs. You have the opposite problem with this recording.

I hope I haven't been too hard on you, but I think your next recording will be much better.

Thank you David.

I have some questions about things that I see as inconsistencies in your replies. I hope you don't take this is as nitpicking, but I'm eally confused and eager to learn from you.

1) Up [="http://recording.org/ftopicp-276522.html#276522"]here[/]="http://recording.or…"]here[/] you said the main pair should be 3-4 m away, but in the last post you said 2-3. My main pair was 8' away, which is 2.44 m, right in the middle of your 2-3 m range.

[[url=http://="http://recording.or…"]Here[/]="http://recording.or…"]Here[/], on the other thread (point 6), you said that you like the Blumlein pair to be behind the omnis such that they get the same direct to reverb balance, but here you said the Blum should be much closer than the omnis.

Now, this has nothing to do with ears, but I decided to check out the angles graphically. See [="http://xs301.xs.to/xs301/06212/PianoBlumDist.jpg"]this scale drawing[/]="http://xs301.xs.to/…"]this scale drawing[/]. One set of lines represents teh extent of the sound source for a 90 degree sweep and the other pair represents the extent of the sound source for a 72 degree sweep (the correct SRA as calculated [[url=http://="http://www.hauptmik…"]here[/]="http://www.hauptmik…"]here[/], which is based on M. Williams's The Stereophonic Zoom. I find that you should be this far away for each case:

90 degrees - 1.168 m
72 degrees - 1.76 m

Like I said, no ears involved, but if the soundboard is the extent of the sound source, then this should be right.

Also, it seems like getting the piano to fill the stereo field in the blum array and getting the correct direct to reverb ratio will usually be mutually exclusive.

Did I mention no sound check? 8)

Monkey's Audio did 2.1% better than FLAC at 24 bit and 4.3% better than FLAC at 16 bit. Thanks ptr!

David French wrote: Monkey's Audio did 2.1% better than FLAC at 24 bit and 4.3% better than FLAC at 16 bit.

You're welcome! - Small procentages but if You're dealing with big files small amounts to big!

Quite an interesting discussion om positioning at the grand you have here...

/ptr

Nice playing!

First, I'd be happy if I could catch a piano like you did. You heard my results in an earlier thread... I hope you don't mind my questions. I make them in order to learn from the dialog. The Schoeps pair sounds quite nice, but it puzzles me a bit. Some sounds in it seems close and others not so close. The softer and lower register sounds full and smooth, but the forte right hand playing sounds has a different character. There is plenty of room in the Schoeps pair too, isn't there? Should one have more? Could you perhaps add your mic positions to the drawing you made earlier?

best regards
Lars

[[url=http://[/URL]="http://xs201.xs.to/…"]Piano mic positioning graphic, JPEG, 52.6 KB[/]="http://xs201.xs.to/…"]Piano mic positioning graphic, JPEG, 52.6 KB[/]

Thanks for the illustration! If you rotated the pairs around the piano, constant radius, how would that alter the recorded sound?

best regards
Lars

Yes, I have been a little inconsistent. Sorry about that. Operating from memory. Lets get the notebook out.

Looking at my notes from the last piano recording I did with the SF24 and Omni outriggers, here are the dimensions.

1. SF24, 2770mm from front leg, piano skewed with tail around so that the tail leg is 440mm further back than the near front leg.

2. SF24 2070mm high

3. 4003's, 2140mm away from near front leg, 740mm out each side of centre line where SF24 was, and 2370mm high.

So in summary, the omnis were a little nearer than the SF24, spaced 1.5m apart, and 2.4m high, whereas the SF24 was further out and lower.

Now I think I said that the omnis should be nearer based on theory, due to the directivity accent from the SF24, but that you should use your ears.

I also said to START with the mics in these sorts of positions but move them as required as I cannot possibly understand what your hall is like. Unfortunately I have been too specific about my case and you have taken me too literally without a sound check. Our rooms are obviously very different and one should always listen and adjust.

I know you said you didn't have a sound check, and from your dimensions and sound of your omni pr, your room is smaller than the one I am in. Is this true? I am in a chamber concert hall that seats 1200. Ceiling height at the stage is some 17m. What is your room like.

Sorry to have confused you.

Also, it seems like getting the piano to fill the stereo field in the blum array and getting the correct direct to reverb ratio will usually be mutually exclusive.

Yes a little. Again these are starting positions before repositioning after careful listening and balance.

I find the big problem with Blumlein is that on its own it can sound mono and too far away if a very quick balance is done in headphones without proper listening. So if I am setting up without decent sound check, I start by placing it closer than one would like, to spread the image and I can usually do something with it later with a little reverb. If its too far away there is nothing that can be done.

Interesting.

2770 mm is about 9 feet, and my Blumlein pair was about 8 feet away. Yours was about 6.8 feet high, while mine was only about 4 feet high. So, you were about 1.8 feet (54 cm) farther away. Yours presumably was not too far, but mine was way too far, yet mine was closer. Yes, differeny halls. Mine at 600 seats is smaller. Greater sound density at the same distance? [="http://www.bsu.edu/alumni/media/36164/alumnusnov04mibsursahall.jpg"]Here's[/]="http://www.bsu.edu/…"]Here's[/] a picture of the hall, and [[url=http://="http://dmfrench2.iw…"]here's[/]="http://dmfrench2.iw…"]here's[/] a stereo impulse I made of the hall. The speaker was a Genelec 1037C. It was resting back on a chair aimed up and out into the hall. The mics (4007 spaced 30 cm and 250 cm high) were about 2/3rds of the hall's length back (at FOH). That should give you a very good idea of what I'm dealing with.

Our omnis were the same distance away, yours are twice as wide, and your were just a little bit higher.

We have the same setup except for four major factors: Your SF-24 was twice as high, your omnis are brighter, , your omnis were twice as wide, and your hall is bigger. All in all, It still seems odd to me that my recording is so very much worse. I really want to get to the bottom of this. I woudl be very happy to be able to make piano recordings as good as yours. Could you post a little of this recording you've been speaking of?

David (Spearritt), I'm surprised that you recomend placing the outriggers 4 meters apart, which seems far apart, but I take your word for it based on the high quality of your recordings. Would you recomend a similar
placement as a starting point, if the main mic is a AKG c426b and the
outriggers are Sennheiser 800's ?

Yes omnis at 4m because the centre is filled by the Blumlein pair. Its a great way to start the positioning.

We have the same setup except for four major factors: Your SF-24 was twice as high, your omnis are brighter, , your omnis were twice as wide, and your hall is bigger. All in all, It still seems odd to me that my recording is so very much worse. I really want to get to the bottom of this. I woudl be very happy to be able to make piano recordings as good as yours. Could you post a little of this recording you've been speaking of?

The hall is the big factor of difference here. I will try to post something when and if I get permission from the artist.

When you have your omnis so wide apart, do you usually have
them pointing (downward and) straight ahead or inward?

Outriggers spaced 4m is perfectly acceptable when using such a strong center as blumlein/ortf/XY. For piano, I certainly wouldn't go MUCH further, but 4m is fine. The purpose of the omni outriggers really is to capture the sound of the instrument in the space, not the sound of the instruments mechanics. Leave that to the center pair.

As for direction - aim them how you'd like. With more directional omnis (such as M50s or M296s), I would bring them slightly towards center and angled down slightly towards the instrument. For less directional ones (Schoeps, Earthworks, DPA), it's all a matter of personal preference.

Very interesting. The one thing I don't understand is, if the goal of
the outriggers is to capture the sound of the instrument(s) in the
space, why not have them much further back in the hall. David
Spearritt suggested that having the main pair closer would produce
a wider and dryer stereo image, so why not have the outriggers
much further back to capture more ambience. I guess it would depend
on the acoustics of the particular hall.

aracu wrote: Very interesting. The one thing I don't understand is, if the goal of
the outriggers is to capture the sound of the instrument(s) in the
space, why not have them much further back in the hall. David
Spearritt suggested that having the main pair closer would produce
a wider and dryer stereo image, so why not have the outriggers
much further back to capture more ambience. I guess it would depend
on the acoustics of the particular hall.

Think of it this way (well, I do at least).

Get the mics up on the instrument too close and you capture the instrument virtually devoid of space.

Get the mics away from the instrument too far (deep into the hall) and you capture the essence of the space, but nothing much of the instrument.

Put the mics where there is a good blend of the two and you now have a great sense of how the instrument actually interacts with the space. That's the purpose of these outriggers.

J.

Mr. French -

I hope you don't mind, but I took the liberty of doing what I could with your piano sample. I found the tracks to be usable overall. Granted, there are things I'd like to be better, but I think you still have a potentially good recording on your hands.

(Dead Link Removed)

What I did -

Each file to its own track.

Left Schoeps at unity gain and panned hard left and right.
Placed left R122 at -2.0 dB and panned -20 dB left
Placed right R122 at -6.2 dB and panned 20 dB right

EQ on Schoeps - HPF at 29Hz
EQ on R122 - High Shelf at 2700 Hz with max gain at peak of curve at 2.2 dB

EQ on total mix - 1dB cut at 814 Hz spanning 2.5 octaves

SIR Reverb - added in lightly to flavor.

Let me know what you think. Given a little more time, I'm sure there's more to do, but I was satisfied with these results after only 5 minutes of playing.

J.

Oh...I forgot - I delayed the Schoeps by 4 ms. They just seemed to fit better there.

J.

Jeremy has done an excellent job in repairing this recording. 8-)

It's like using a stereo pair and two omnis on three
mic stands to create a very wide decca tree.

aracu wrote: It's like using a stereo pair and two omnis on three
mic stands to create a very wide decca tree.

Not quite.

Decca trees are flown over an ensemble, not out in front of. In fact, Decca Trees rarely capture much of that "you are there" sensation which I referred to in the previous post - you know, the good blend between direct versus reflected.

When I reach for Decca tree it's almost always because I don't like the acoustics of the venue. It allows me to get a very deep representation of the orchestra despite the rather close placement. It's particularly good in overly ambient venues such as churches or cathedrals.

The benefit to DT is the amazing stereo representation of the orchestra and the rejection of bad acoustics.

The drawback is that there are often significant phasing issues (which can actually help the illusion of a "wide" orchestra).

Think of the above mentioned pattern as simply a stereo pair with an ambience retrieval pair.

This is real similar to the majority of the recordings that I make.

For full orchestra, I usually go:

Main pair - Schoeps CMC6 MK2S (If the acoustics are good - MK4 in ORTF if not so good.)

Aux/Flank pair - Schoeps CMC6 MK2S/MK21 or Gefell M296 (all depending upon the hall)

Spot mics.

I could easily run with just the two main mics, but adding those flanks really opens up the sound quite a bit. Obviously mix to taste.

J.

Very helpful comparison... and I'm surprised that there would
be more phasing issues with the decca tree configuration than
with the widely spaced outriggers.

I think it depends mainly on how much of the pairs is in the mix. The Blumlein pr with no phase issues has the principal gain. The outriggers are brought up just to add incoherant room, and should not be balanced so that they start to disturb the image.

That is really interesting, it's like you just add a touch of
them for that extra bit of magic, but just a subtle amount,
being careful not to overdo it. In a way it resolves the debate
between using Blumlein versus spaced omnis. I still haven't
gotten over that mp3 of yours of piano, (arrangement from
Carmen?) great piano sound and performance with perfect
ambience and reverberation, ideal for the genre to my ear.
I suppose that that ambience, if overbalanced, could start to
sound too indistinct.

I have permission from the artist to post these samples. The pianist is Jayson Gillham, http://www.jaysongillham.com

The two samples are made with same pianist, same piano, same hall, same piece. You can't get much better than that for consistency. :)

Etude 1 is the setup as described above. Etude 2 is similar but with Schoeps MK21 outriggers instead of the omnis and with slightly different positions of mics and piano. Both main pairs are the glorious SF24.

http://www.lodestarrecordings.com.au/downloads/Etude1.mp3
http://www.lodestarrecordings.com.au/downloads/Etude2.mp3

In each example, the recording process has become
part of the process of musical interpretation. The first
example draws attention to detail and the beautiful
sound of the piano, seeming to probe inside the
instrument and performer. It brings out an emotional
warmth and gives a good idea of the lovely sound
qualities of the Royer. The second example has a
slightly hazy, blurred, impersonal character. Maybe
the outrigger mics are turned up or maybe the Royer
is further away. This version adds to the trancending
"religious" quality of the music inherent in the composition,
and sounds appropriate to it. The piano sounds unified with
the space of the hall.