My headphones blinded me...

I think I know little about this problematic... Reading this forum I can see that quite many people here want to have a wide, rich sound, which is something you hear when sitting near to a big grand piano. On the other hand this is not the sound you hear typically in the records. Sound is more thin; it leaves musical space for other instruments for example. How do they make it? I am not an expert of recording or mastering piano sound, but I am quite sure that producers cut some frequencies out to make it more thin. Too much reverb can of course destroy it too.

My main personal preference is to have a digital piano sound to live stage situation which cuts trough mix and does not take whole "musical space". For pop or jazz groups (which is my main environment) IMO typical mistake is to have too much bottom with piano sound. Same sound which is nice with your headphones or studio monitors can have too much bass on stage PA. I like piano sounds that are easy to mix in any situation. In fact this was the main reason for me to move from Roland to Clavia as my main piano. Roland was too wide (or rich) and lacking the clarity. It had kind of nice dark sound but it was not that good on stage. My experience is that when you need a lot of EQ this is bad sign. On the other hand a good piano sound is good in many situations (headphones, monitors but on stage as well) and you don't need that much EQ or any processing (of course little bit in some acoustics). PTQ is probably sometimes too wide or rich on stage. Other thing is that it may not have clear enough attack to make it powerful to cut trough...?

I've been taking part on a software forum for about ten years (the company makes Band-In-A-Box and PowerTracks Pro Audio).

Most of the users are singers and instrumentalists but they all say the same thing about mixing and mastering to get the best sound:

1)  Do not judge your sound with headphones.

2)  Try your music sound out on several systems - car, home stereo, boom box.  They will all sound different.

My comments on piano:

Piano solo may require some compression - this is particularly true for classical piano.  Although I suspect that commercially produced classical piano CDs have little or no compression applied, classical piano can have a tremendous range of sound levels.  Without compression, you will turn the volume up so the quiet passages can be heard while driving and when the loud passages come, the sound will be unbearable.  Not enjoyable at all.

I made a CD of jazz piano solo (not my playing, but that of Doug Mackenzie - you can google him - his midis are online).  I used a Close Mic setting and turned the reverb completely off.  I like it.

In addition to no reverb, I added compression with the following settings:
Multiplier:  minus 6 dB
Threshold:  minus 6 dB
Attack:  zero seconds
Release:  zero seconds
Smoother:  OFF

The multiplier and threshold settings could change depending on how "level" a result is wanted.  I use zero for attack and release because piano sounds change level extremely fast (percussion instrument).  I leave the smoother off for the same reason.

I'm not sure that these are the "best" settings for piano and if anyone has differing comments, I'd be very interested to hear them.

Glenn

EDIT:  Two pieces rendered with Pianoteq (neither is me playing).  Both with the settings noted above.  I can't remember which piano was used (K1 has been my favourite for quite a while).  It could even be a setting where I put one mic beneath the piano and two above the  soundboard.

http://www.box.net/shared/z658yum37m
http://www.box.net/shared/mq7651ntsv

Etto:

Please explain a bit further if you will.  I have much to learn about compression.

What settings would you use for solo piano?

Glenn

Hello Glenn,

I have been reading this thread with interest; it wasn't until you posted the Chopin Scherzo opening excerpt that I needed to add commentary:

The second version (maxed and compressed) caused me to hear digital overload distortion.  At first, I thought my audio volume was turned too high and caused my AKG-702 headphones to "crack up" in distortion.  Then, I reduced the loudness on your excerpt to approximately one-third of full scale, and listened again --- the digital overload distortion remained!

I make no claims to be an expert in compression, but offer these observations to your excerpt:

Notice that the break-up occurred both when the low B-flat octave was played -- AND -- when the two-handed E-flat minor chord was played in the top half of the piano.  When the bass and the treble each cause digital overload distortion, this suggests that a single-band (wideband / all-frequency) post-processor was used.  (Alternately, if a multiband compressor was used, its output signal was set sooooo high that all bands were driving the digital signal to distortion.)

As you are probably aware, a good multiband compressor should be used more conservatively to bring up the low level of sound, but not slam the loud signals into distortion.  A quick search of YouTube for tutorials on understanding audio compression should be of help.

Cheers,

Joe

Hi Fab!

What you realized then is what every beginning sound engineer has to go through.
To manage such a situation you'll need some tools.

First of all Headphones are for most things unusable to get what you deserve.

1) You can't judge Reverb or Room Mic relations very well. (unless you've verified them for years)

2) Use a good pair of Speakers (Mastering systems cost tons of money)
http://www.kef.com/gb/loudspeakers/qser.../bookshelf

those are great (pricy)

3) As some mentioned before, crosscheck on different speakers. write down your experience. (to muddy, to sharp etc...)

4) For pianos less compressors are of good use not to damage the sound ( they start to distort the sound very easily). Use a good limiter

http://www.brainworx-music.de/de/plugin...p0lloudab1

The BX XL is great for its price. Also VST Voxengo's Elephant3 and LAWO are good

I don't like the PTQ internal very much but reduce the dynamic with the dynamic-slider a bit.

5) To judge compression on headphones doesn't work for me. I listen to speakers with bad dynamic on a very low level (even the cheapest can work for that. I use some 30€ JBL PC speakers)

6) Check Phase! (especially in the base region)

some useful tools: BX digital V2 EQ (with mono maker for the base) or NugenAudio   Mono filter

So after all that you have to be patient too. Because some of us needed years to get it done, but it's time to begin...

Heinke

Hi Glenn and Fab,
here I am again; meanwhile interesting things has been added to this thread.
I listened to your examples and aside the subtle distortion the compressor seems to work well, this is why is difficult for you to hear the difference.
Basically compression can be approached in a "creative" way but is important to know what ever parameter means, how is translated in your sound and maybe also basic differences between compressor types (fet, opto, vca..) that in today's digital plugins are only simulated.. If you need gentle compression for a piece like Beet "chiaro di luna" first tempo you can easily achieve it starting with an opto compressor. In the last 'tempo' a fet is more suitable.
But the discussion could go on forever, so maybe is better if I post some practical example
The general purpose of a compressor is to level the sound in a continuos way with maximum transparency (when not used as a special effect). Because in the audio field you have a lot of different situations its parameters help you to set it to fulfill every situation. An extreme case could be a peak limiter where you use fast attack and release times, high ratio and high threshold; in this way only the very loud peaks get catch and leveled. The same setting is very good on snare or bass drum where you set a slower attack to preserve or exaggerate the punch. But lets go back to piano solo.
It is very important to listen to the song characteristics and decide where to go; the idea to compress it as light as possible is very good but as you stated before extreme dynamic could be a problem in some listening environments. But it's your choice.. For the audiophile the reproduction should be as true as possible so you have only to be sure that your piano dynamic fits well in your media's dynamic range. A 16 bit digital recording (CD ie) can embed 96 db; enough for a piano; so the only precaution should be a limiter with soft knee to maximize the recording level and avoid alias effect in low passages. Conversely today a lot of people listen mp3s in ipod or similar so if this is my target I'd choose for the same piano as before a soft compressor with long attack, medium release, low threshold and mid ratio. In this way I limit the dynamic in a range that can be heard in headphones in noisy environments (traffic, school and a like) and avoid mp3 artifacts on low dynamic passages.
As you can see there is not a general rule for this..
The important thing is to set attack and release times fast enough to achieve your goal, slow enough to avoid distortion and preserve attack transient if needed, ratio and threshold set to avoid pumping or breathing side effect; all this related to the kind of piano you are mixing, your taste, and the final media you will use for that.
It can help to exaggerate threshold to check what compressor is effectively doing and put it back to a level where you achieve a 2 to 6 db reduction without any side effect.
In the case of pianoteq, as Heinke suggested, the dynamic slider can help without any struggling.
Hope this can help