You are here

PIANO RECORDING

For everything after the recording stage: hardware/software and how you use it.

Re: PIANO RECORDING

Postby Ariosto » Tue Aug 06, 2019 1:03 pm

Hugh Robjohns wrote:
First, you're not a pain -- this is precisely what this forum is about -- helping people to develop their skills and understanding!

H
Many thanks Hugh. Great advice and I think I have learnt a tremendous amount about recording piano. I will try out all your suggestions. Also thanks to Mike, Bob, Tim and all who gave advice.

This is such a great forum!

Peter
Ariosto
Frequent Poster
Posts: 805
Joined: Sun May 04, 2008 12:00 am
Location: LONDON, UK

Re: PIANO RECORDING

Postby hobbyist » Tue Aug 06, 2019 4:33 pm

Bob Bickerton wrote:
hobbyist wrote:It is hard to do a piano in stereo and have it sound natural.

That could apply to anything. Most people record piano with a view to it being played back in stereo and there are some very good recordings out there. Stereo recording with mono-compatibility is still the industry standard.

Bob

Well it *is* stereo but what is the width of the piano viewed from the mikes' location and what is the width the mikes captured?

How would that sound in your LR played on speakers that are farther apart. Especially if the file had been edited to make it sound as wide as possible.
hobbyist
Regular
Posts: 294
Joined: Mon Jul 01, 2019 12:52 am

Re: PIANO RECORDING

Postby Ariosto » Wed Aug 07, 2019 8:35 am

https://soundcloud.com/user-607637178/t ... dth-reverb

This is a new version with Hugh's suggestions - i.e. narrower panning and width and with some small amount of reverb added.

Peter
Ariosto
Frequent Poster
Posts: 805
Joined: Sun May 04, 2008 12:00 am
Location: LONDON, UK

Re: PIANO RECORDING

Postby Hugh Robjohns » Wed Aug 07, 2019 10:55 am

Definitely getting there! :D

I appreciate it's a work in progress, but at the moment the peak level is around -24dBFS, and your R128 loudness is -44 LUFS.... which means I'm having to crank up my monitoring level almost to maximum just to achieve a sensible listening level, let alone getting it comparable to a commercial CD piano recording.

The stereo width and reverb is entirely down to personal preferences, of course, and I don't know what kind of character you're looking for. However, for me, it's now a little bit too narrow -- but much more natural than the fully wide version.

And the reverb makes it sound a little bit like an empty school hall, so I think I'd be looking for something that adds a little more bloom and warmth at the low end, and a softer, wooden room kind of character without too much in the way of distinct early reflections.

H
User avatar
Hugh Robjohns
Moderator
Posts: 24311
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: PIANO RECORDING

Postby Ariosto » Wed Aug 07, 2019 12:08 pm

Hugh Robjohns wrote:Definitely getting there! :D

I appreciate it's a work in progress, but at the moment the peak level is around -24dBFS, and your R128 loudness is -44 LUFS.... which means I'm having to crank up my monitoring level almost to maximum just to achieve a sensible listening level, let alone getting it comparable to a commercial CD piano recording.

The stereo width and reverb is entirely down to personal preferences, of course, and I don't know what kind of character you're looking for. However, for me, it's now a little bit too narrow -- but much more natural than the fully wide version.

And the reverb makes it sound a little bit like an empty school hall, so I think I'd be looking for something that adds a little more bloom and warmth at the low end, and a softer, wooden room kind of character without too much in the way of distinct early reflections.

H
Many thanks Hugh.

My LUFS readings on my meter are -15LU and Peak level -8.3 dBFS and LR range 18.9LU

On my little mixer (PROFX8) out of the computer I have the levels slightly less than half way and headphone output half way or less and that gives plenty of volume. I'm wondering if the Soundcloud output is responsible? I have it on Dropbox -
https://www.dropbox.com/s/h0fxrjkdir3yo ... B.wav?dl=0
and I find that OK volume wise (It's a 24bit file at 96KHz).

The reverb is just a Reaper reverb, so probably not great. I actually prefer it without reverb, but that's probably because I like the clarity of solo instruments and also even orchestra and find a lot of commercials CD's have to much room sound and/or hall reverb which makes them mushy. But many people like plenty of reverb, so maybe I'm not typical.

I could widen the panning a little, I chose 50% as a sort of compromise.

Edit: I'm trying to persuade the performer into recording it again at a slightly faster tempo and have a new version, but not having too much success at the moment. She would rather be on holiday ...

Congratulations by the way on your doctorate at the University! Could not have gone to a more dedicated person.
Ariosto
Frequent Poster
Posts: 805
Joined: Sun May 04, 2008 12:00 am
Location: LONDON, UK

Re: PIANO RECORDING

Postby Hugh Robjohns » Wed Aug 07, 2019 12:42 pm

Ariosto wrote:My LUFS readings on my meter are -15LU and Peak level -8.3 dBFS and LR range 18.9LU

Apologies. Cockup my end! :blush: I don't seem able to download the file from SoundCloud, so I bounced it across to my DAW in real time and hadn't noticed that the SoundCloud volume control wasn't at full level. Mea culpa on that one...

However, I've just downloaded your dropbox file and checked that in RX6 which gave me figures of -31.6LUFS Integrated Loudness, and true peak of -11.53dBTP, and LRA of 19.7.

Anyway... that's all easy to sort out when you've finalised the mix.

I actually prefer it without reverb, but that's probably because I like the clarity of solo instruments and also even orchestra and find a lot of commercials CD's have to much room sound and/or hall reverb which makes them mushy. But many people like plenty of reverb, so maybe I'm not typical.

I'm not a fan of over-done reverb either. It's just that by only part-panning the two mics, the resulting stereo image inevitably won't fill the whole stereo width. To many, that is considered odd...

If you'd recorded with a conventional stereo array, the actual sound stage would fill the whole stereo image width, with the piano itself filling the central portion and the room sound filling out the rest.

Since you've close-miked and part-panned the two tracks, you're missing that outer edge room sound, and that's what a sympathetic reverb could help to replace...

But it's only a suggestion...

She would rather be on holiday ...

Wouldn't we all?! :bouncy:

Congratulations by the way on your doctorate at the University! Could not have gone to a more dedicated person.

Thank you. Most kind! :thumbup: :D
User avatar
Hugh Robjohns
Moderator
Posts: 24311
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: PIANO RECORDING

Postby hobbyist » Wed Aug 07, 2019 5:57 pm

Hugh Robjohns wrote:
I actually prefer it without reverb, but that's probably because I like the clarity of solo instruments and also even orchestra and find a lot of commercials CD's have to much room sound and/or hall reverb which makes them mushy. But many people like plenty of reverb, so maybe I'm not typical.

I'm not a fan of over-done reverb either. It's just that by only part-panning the two mics, the resulting stereo image inevitably won't fill the whole stereo width. To many, that is considered odd...

Why is it odd to have a small sound source not smeared across a wide sound stage?

An orchestra should sound wider than a piano and a piano wider than the announcer.

If you'd recorded with a conventional stereo array, the actual sound stage would fill the whole stereo image width, with the piano itself filling the central portion and the room sound filling out the rest.

What do you call a conventional stereo array?

There is coincident, near coincident, and spaced arrays.
They use sound pressure, both, and time differences respectively.

But the actual apparent stereo source width depends on the distance from the source, the actual mikes, the angle of the mikes, and the room.

So what is the standard for a conventional array?
hobbyist
Regular
Posts: 294
Joined: Mon Jul 01, 2019 12:52 am

Re: PIANO RECORDING

Postby Hugh Robjohns » Wed Aug 07, 2019 6:02 pm

hobbyist wrote:Why is it odd to have a small sound source not smeared across a wide sound stage?

That's not what I said.

An orchestra should sound wider than a piano and a piano wider than the announcer.

Absolutely... and the room acoustic should fill the whole width... but it won't in Peter's recording because he's panned the two channels in from the edges... and, to many, that will sound odd.

What do you call a conventional stereo array?

Harry... or Bill, sometimes... depends what mood I'm in. :lol:
User avatar
Hugh Robjohns
Moderator
Posts: 24311
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: PIANO RECORDING

Postby hobbyist » Thu Aug 08, 2019 1:42 am

Hugh Robjohns wrote:
hobbyist wrote:Why is it odd to have a small sound source not smeared across a wide sound stage?

That's not what I said.

An orchestra should sound wider than a piano and a piano wider than the announcer.

Absolutely... and the room acoustic should fill the whole width... but it won't in Peter's recording because he's panned the two channels in from the edges... and, to many, that will sound odd.

What do you call a conventional stereo array?

Harry... or Bill, sometimes... depends what mood I'm in. :lol:


That was how I read it.

Room acoustic is always everywhere in the playback.

I understand what Peter did. But not technically why it would sound odd. I need to research that. What I do think is that making a modest size source less than full width should not sound odd.

I can see panning too wide could leave a hole in the middle. Not yet seeing how panning less is a problem unless it is overdone.
hobbyist
Regular
Posts: 294
Joined: Mon Jul 01, 2019 12:52 am

Re: PIANO RECORDING

Postby Hugh Robjohns » Thu Aug 08, 2019 10:07 am

hobbyist wrote:Room acoustic is always everywhere in the playback.

No. It isn't. This is stereo mic basics 101...

Try this mind-experiment: You have a piano recording made with two mics (in any arrangement you like), but you pan them both centrally. When you play that back on stereo speakers the piano sound will form a narrow image mid-way between the speakers.

How wide is the spread of 'room sound' across the image? *1

Now extend that to consider what happens if you pan the two mics out from the centre part way. What happens to the piano image now? *2

And the spread of room sound? *3

Now contrast that to what happens if you have an XY stereo array (let's say crossed cardioids for simplicity) a long way back from the piano in a big concert hall. How wide is the piano image? *4

And the spread of room sound? *5

And finally, move the XY array much closer to the piano and answer the same two questions:

How wide is the piano image? *6
And the spread of room sound? *7

Hopefully, you'll now realise that the width of all elements within a stereo image is dependent on many things, but especially the pan positions, and narrowing the image with pan pots affects the room sound image width just as much as the piano image itself.



ANSWERS:
*1 There is no spread: it's a paper-thin-narrow phantom centre image overlaying the (mono) piano sound.

*2 The piano image is no longer quite such a narrow central phantom image, and it will occupy a small amount of stereo width.

*3 The room sound will be as wide -- but no wider -- than the piano sound.

The visual equivalent is watching an old 4:3 format TV programme on a widescreen TV with black strips down each side.

To achieve full stereo width there needs to be around 16dB of level difference (or around 1.5ms of time difference) between the left and right channels, and that can't be achieved with part-panning. So the image width must always be smaller than the total stage width between the speakers.

*4 Very narrow, because the piano physically occupies only a narrow chord of the mic array's whole stereo acceptance angle.

*5 The room sound spreads across the entire stereo image because it occupies the mic arrays complete stereo acceptance angle.

Again the visual equivalent is a widescreen image on a widescreen TV with a small piano at the centre of the image.

*6 The piano now spreads much more widely across the stereo image because it fills a much larger chord within the mic array's stereo acceptance angle.

*7 The room acoustic still fills the full stereo width because it still occupies the mic arrays complete stereo acceptance angle.
User avatar
Hugh Robjohns
Moderator
Posts: 24311
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: PIANO RECORDING

Postby hobbyist » Thu Aug 08, 2019 6:32 pm

Hugh Robjohns wrote:
hobbyist wrote:Room acoustic is always everywhere in the playback.

No. It isn't. This is stereo mic basics 101...

Try this mind-experiment: You have a piano recording made with two mics (in any arrangement you like), but you pan them both centrally. When you play that back on stereo speakers the piano sound will form a narrow image mid-way between the speakers.

How wide is the spread of 'room sound' across the image? *1

Now extend that to consider what happens if you pan the two mics out from the centre part way. What happens to the piano image now? *2

And the spread of room sound? *3

Now contrast that to what happens if you have an XY stereo array (let's say crossed cardioids for simplicity) a long way back from the piano in a big concert hall. How wide is the piano image? *4

And the spread of room sound? *5

And finally, move the XY array much closer to the piano and answer the same two questions:

How wide is the piano image? *6
And the spread of room sound? *7

Hopefully, you'll now realise that the width of all elements within a stereo image is dependent on many things, but especially the pan positions, and narrowing the image with pan pots affects the room sound image width just as much as the piano image itself.

I believe that I said that before.




ANSWERS:
*1 There is no spread: it's a paper-thin-narrow phantom centre image overlaying the (mono) piano sound.

*2 The piano image is no longer quite such a narrow central phantom image, and it will occupy a small amount of stereo width.

*3 The room sound will be as wide -- but no wider -- than the piano sound.

To achieve full stereo width there needs to be around 16dB of level difference (or around 1.5ms of time difference) between the left and right channels, and that can't be achieved with part-panning. So the image width must always be smaller than the total stage width between the speakers.

*4 Very narrow, because the piano physically occupies only a narrow chord of the mic array's whole stereo acceptance angle.

*5 The room sound spreads across the entire stereo image because it occupies the mic arrays complete stereo acceptance angle.

*6 The piano now spreads much more widely across the stereo image because it fills a much larger chord within the mic array's stereo acceptance angle.

*7 The room acoustic still fills the full stereo width because it still occupies the mic arrays complete stereo acceptance angle.
===============================================


I find the room is the width of my speakers. No matter what the actual venue was. You can diddle the source width but the room width will still be where my speakers say it is. You can change the source width within that for sure.

The distance, spacing, and angle, as well as diddling the pan knob all affect where in that space I hear the source and its apparent width.

If you are saying that you can narrow the source so I do not hear it full width on my speakers then I agree.

If you are saying that you can narrow it and I won't hear anything else past the width the mikes captured including any room sound that was wider than the source, then again I agree. Although my amp may give me some white noise on the edges:)

I am not fully sure what we disagree about.
hobbyist
Regular
Posts: 294
Joined: Mon Jul 01, 2019 12:52 am

Re: PIANO RECORDING

Postby Hugh Robjohns » Thu Aug 08, 2019 8:36 pm

hobbyist wrote:Room acoustic is always everywhere in the playback.

hobbyist wrote:If you are saying that you can narrow it and I won't hear anything else past the width the mikes captured including any room sound that was wider than the source, then again I agree.

I am not fully sure what we disagree about.

There seems to have been a reversal of position in your statements.
User avatar
Hugh Robjohns
Moderator
Posts: 24311
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: PIANO RECORDING

Postby hobbyist » Thu Aug 08, 2019 11:58 pm

Hugh Robjohns wrote:
hobbyist wrote:Room acoustic is always everywhere in the playback.

hobbyist wrote:If you are saying that you can narrow it and I won't hear anything else past the width the mikes captured including any room sound that was wider than the source, then again I agree.

I am not fully sure what we disagree about.

There seems to have been a reversal of position in your statements.

Or a lack of communications. Perhaps one of us did not understand what was meant, or one of use said what we meant poorly.
hobbyist
Regular
Posts: 294
Joined: Mon Jul 01, 2019 12:52 am

Re: PIANO RECORDING

Postby Hugh Robjohns » Fri Aug 09, 2019 12:14 am

Glad that's cleared up then! ;-)
User avatar
Hugh Robjohns
Moderator
Posts: 24311
Joined: Fri Jul 25, 2003 12:00 am
Location: Worcestershire, UK
Technical Editor, Sound On Sound

Re: PIANO RECORDING

Postby hobbyist » Fri Aug 09, 2019 1:44 am

Hugh Robjohns wrote:Glad that's cleared up then! ;-)

I hope so. Sometimes I do have problems communicating and being misunderstood.
hobbyist
Regular
Posts: 294
Joined: Mon Jul 01, 2019 12:52 am

Re: PIANO RECORDING

Postby pianowillbebach » Fri Aug 09, 2019 1:52 am

I'm a huge Steinway fan - never played or recorded on a Model O though!
pianowillbebach
Poster
Posts: 11
Joined: Wed Jun 12, 2019 5:46 pm

Re: PIANO RECORDING

Postby Bob Bickerton » Fri Aug 09, 2019 2:19 am

That's a smaller size which enables you to carry it open a model railway O gauge..........

Bob
User avatar
Bob Bickerton
Frequent Poster (Level2)
Posts: 3872
Joined: Fri Dec 20, 2002 1:00 am
Location: Nelson, New Zealand
http://www.bickerton.co.nz

Re: PIANO RECORDING

Postby Rich Hanson » Fri Aug 09, 2019 10:19 am

Bob Bickerton wrote:That's a smaller size which enables you to carry it open a model railway O gauge..........

Bob

You do have to be specially trained though.


Why yes, that is my coat over there, shall I fetch it?
Rich Hanson
Frequent Poster
Posts: 1307
Joined: Sat Sep 06, 2003 12:00 am
Location: UK

Re: PIANO RECORDING

Postby Ariosto » Sat Aug 17, 2019 5:28 pm

This is an update as its still work in progress. Same work - Granados "the Lover and the Nightingale." Slightly different interpretation. No edits and no added reverb or any other processing.

The two mics were approx 6 inches higher but in the same position. (DPA2006C omni mics). recorded on a MixPre-3. A 24bit file and 96KHz Sample Rate.

https://www.dropbox.com/s/8rsr3873w5ooj ... N.wav?dl=0
Ariosto
Frequent Poster
Posts: 805
Joined: Sun May 04, 2008 12:00 am
Location: LONDON, UK

Previous

Who is online

Users browsing this forum: Martin Walker