Hi George,

thank you for your listening and your very detailed report. I think that the drawbacks of the with take can be more easily corrected (EQ and reverb) than the ones of the without take: I don't know any plugin that would remove the early reflections of a smalll room.
I just have submitted in the audition room a new recording of Chopin's nocturne in Cis performed with the MicThing and the cushion and with such audio editing.
I look forward for listening to your own clips without and with that shall be be much useful to me in my quest for a better sound. Thanks in anticipation,
Didier

Bonsoir Didier. Ah, there is no plug-in that can correct for the early reflections in a small room. I don't like to use EQ too much because it also alters the harmonic balance of the instrument, and that defeats having a fine German piano in the first place. Better than EQ, absorption is the only way to "fool" the mics into believing that the early reflections aren't there in the first place. Low ceilings are the big culprits in many homes. Don't forget the ceiling in your calculations because, if the early reflections from the ceiling is absorbed and delayed; it's as if you have a very high ceiling. For Practical and aesthetic reasons, I am only placing 2in thick panels instead of 4in thick ones on the ceiling, because the fiberglass panels can get heavy. I would hate having these things fall on a Steinway. ZOINKS!

If you don't have room calibration software, in the meantime, make yourself a CD of test tones and do a frequency sweep above 60Hz with good speakers in your music room, and plot which frequencies have dips or peaks. In an untreated room, it can vary as much as 20dB! After treatment, do the same frequency sweep and plot the changes in room response. If you get your room balanced within 5-10dB, you're doing very well.

BTW - I've come to realize that many people love to crank up the reverb, which changes the timbre to an edgy, shrill, and hollow sound as if one is playing in a subway tunnel - hardly appropriate for a classical piano. The problem with reverb software is that the algorithm introduces early type reflections, and this is compounded with the early reflections coming from the room. The resulting sound lacks the attack, decay, and clarity of the original source. I listened to your Nocturne, very nicely played and with good taste - you didn't drown the piece with reverb, just right!

If you ever want to have a reference for acoustic engineering that covers fundamentals, absorption, diffusion, reverberation, studio design, spaces, materials, etc., and one that's not very complicated, here is the bible: Master Handbook of Acoustics, by F. Alton Everest.

Give me some time to post samples of before and after treatment...

Uh oh - I may be one of those who 'crank' up the reverb. I'm glad you said something about this, George. I will do some tests with this in mind next time I record.

Monica, it is very tempting to get carried away with reverb as we've all done it. Upon initial listening, reverb seems to conceal some undesirable effects, but always introduces other artifacts that was never present in the original recording, e.g. loss of clarity, immediacy, texture, changes in timbre, and tonality.

Music has been described as the space or "silence" between the notes. We can all appreciate that something must be happening between these notes, and my belief is that reverb is what predominantly occupies this space. It perpetuates and sustains the inherent energy in the sound over time, much like savoring a fine wine. Our psychoacoustic perception of reverb in music is alluring, hypnotic, evocative, and that is why I think most can easily get carried away in this intoxicating "elixir of sound." Recording a performance involves a 3-way union: The instrument is what makes the sound, the pianist evokes the music, and the acoustics is the communicative pathway. So reverb forms an important part in the communication of the music, and is a physical part of the acoustics that alters our perception of a recording. I feel that it's important to treat the amount and type of reverb very delicately.

How much reverb should one use? For a particular space, it will subjectively depend on the genre of music, dynamics, complexity of sound, rapidity of notes in succession, and individual taste. The characteristics of reverb includes the attack (timbre and tone of sound) and decay (tail, sustain over time). Reverb is like make up (I don't wear any, thanks) - it should be used sparingly to complement and highlight what's already there in the recording. In general, you should only add enough reverb so that you can barely notice it. Sometimes, none at all. You have to be careful so that the reverb tail doesn't drown the next note in succession in a faster piece, otherwise you'll get a mush of notes. When you experiment next time, you'll notice that each piece will need a slightly different amount. Depending on your software, I find the amount can lie between 5-8% of wet reverb. It should always sound realistic, and never encroach on the immediacy and clarity of the performance.

BTW - Whenever editing or adding effects to a recording, never alter the original recording - always archive the original recording onto a gold archival CD or a separate external hard drive. This way you'll you can always refer to the original whenever you change your mind over a particular edit or effect. I've learned the hard way, believe me!

As always, share your findings, as every recording is a learning experience for me too...

Thank you so much for this information, George. I know what you mean about reverb sometimes screwing things up. I've had problems with using reverb on pieces with a lot of staccatos - the short notes sounded weird. And yes - you can get a tinny sound with the reverb also. Honestly, I don't know why I've been using reverb all the time like I do. With some pieces I play that are slow and dreamy-sounding, there is lots of pedaling going on and I really don't need the added reverb.


I've learned the hard way too. That was a very bad day! Been meaning to get a separate external hard drive for awhile now, but just haven't done it yet. Better put it on my list because I have a sneaky feeling that my computer is getting ready to blow. Zoinks!


I will. Thanks again for all this nice information!

Here are samples of how basic DIY braodband absorption panels can improve the quality of home recordings. I've included samples with and without broadband absorption panels (i.e. bass traps). I only used 3 DIY panels to prove a point that one doesn't have to go crazy with acoustic treatment. I placed two 2ftx6ft 4in thick panels in the corners nearest to the piano, and one 2ftx4ft 4in panel behind me on the wall, triangulated in path of the nearest reflection from the piano-to-wall-to-mic. To simplify the implementation even further, I didn't use a ceiling panel in this test, which would require ceiling bolts. If I used more panels, the sound would improve further yielding more clarity and detail. Here are some observations, you can draw your own conclusions:

Without Absorption Panels:
-Nearby reflections from the wall leads to loss overall clarity and detail across the frequency spectrum from the lows to the highs, regardless if omni or cardiod pattern.
-The bass is uncontrolled, boomy, undamped, and lacks clarity, regardless if omni or cardiod pattern.
-The true tonality of the sound is muffled and whooly.

With Absorption Panels:
-There is an increase in overall clarity and detail across the frequency spectrum.
-The bass is well controlled, not boomy, and well defined with clarity.
-The tonality is natural and you hear the instrument and not the boxy sound of a small room.

Now, if I were to take a recording made in an untreated room, which lacks clarity, detail, and sounds muddy and boomy, and added reverb or EQ, I would be getting more hash and more undesireable artifacts. This is because reverb and EQ add their own artifacts to the sound source and seem to amplify any weaknesses in the room. Whereas, if the recording made in a treated room has enough clarity, detail, and bass control, it will take well to additional reverb or EQ in the mastering stage, and not run the risk of getting a hash result.

TEST: C-sharp minor arpeggio. Steinway B in a 35x14x8.5ft room at vertex of an L-shaped room, with an adjacent room at the other leg. Fixed position throughout test - (2) AKG C414B-XLS spaced pair 15in apart, 1ft from the curve, 5ft high, pointed toward high and low strings. Used omnidirectional and wide-cardiod pattern. NO effects added. I used the AKGs because they are the most neutral and objective mics I own, they don't alter the tone of the instrument.

Listen in the following order:
1. Omni WITHOUT Panels
2. Omni WITH Panels
3. Wide Cardiod WITHOUT Panels
4. Wide Cardiod WITH Panels

Hi George,

In my mind:
WITH better than WITHOUT,
Wide cardioid better than omni.

Thank you very much for the demonstration,

Didier

Hi George,

Thanks for all of this. I listened to the samples and found that between the omni-no traps, and the wide cardiod-no traps, I liked the wide cardiod-no traps sound better. It had a fuller sound - I heard more bass. So then I compared the omni with traps and the wide cardiod with traps but really could not hear much difference between these two. Could that be right?

Hi Monica, The samples "without traps" appear to have more bass because of the uncontrolled boominess and ringing in the 80-150Hz region. It's an acoustical phenomena that affects all small rooms. Generally, the bass accumulates in the corners of the room and imparts an unnatural peaked response in that region. You sometimes here it as a one-note bass with speakers. It's as if you cranked up the bass knob. However, this doesn't translate to an accurate recording. Sure you are getting an unnatural peak in the 80-150Hz range, but at the expense of overall clarity and detail throughout the frequency spectrum. If you listen to the middle registers, the "no traps" sounds muffled and lacks clarity, whereas in the "with traps," you can make out the detail in the middle registers more clearly.

A typical room will have peaks and dips in the frequency response due to nearby reflections that may cancel or add in phase with the direct source. Ever wonder why some notes are louder and some are softer... These peaks and dips can vary as much as 20dB across the frequency spectrum. For any accurate recording, it's not ideal to have a such a varied response curve. The bass traps also work to absorb peaks and dips due to reflections, and help to flatten the frequency response by removing the tubby bass, modal ringing, comb filtering, and flutter echo.

My advice is to have a room that is damped in terms of ringing and flutter echo, and try to get the frequency response as flat as you can. Bass traps provide a great solution to solve these common acoustic problems which plague all small rooms. Have a recording that's clear and detailed. You can always add EQ to get more bass or increase the highs to add more air at the top end, or reverb later without sacrificing clarity or detail. However, there is no way to add clarity or detail to a recording if the room is boomy and has nearby reflections that are competing with the direct piano source.

The panels that I have made are the same material found in all recording studios and concert halls. They are made from OC-703 fiberglass panels. The magic number of panels seems to be 5 for my room: 2 in the corners, 2 in the rear wall, and one on top of the ceiling. However, for simplicity, I chose 3 to illustrate the point of broadband absorption.

What you're saying about getting the most clarity in a recording that you can, and then adding reverb makes sense. I'm glad you explained that. I am curious though - are you married? And if so, what does your wife say about these large panels you are placing around the living room? Unless you have convinced her that it's okay since it's all the name of art.

Pianolady wrote:

With the intention of settling down, and fulfill the dream of having a salon for music, I bought a house with large living spaces. Within the first year, I was lucky to find a gorgeous Steinway B, but still haven't found the wife. Shouldn't everything have happened in reverse order? In any case, the panels are "future wife-friendly" because they match the wall color and are mounted on pedestals for easy removal.

Smart man!

I've been revising my initial thread during the course of the year to reflect changes in technology. I've included additional links and added newer gear on my first thread. I've attached 2 graphs below from thelisteningsessions.com which may help in mic and preamp selection.... Wow, over 34,000 views, I guess this thread has at least been interesting for many.

... I've been away for a while, but haven't forgotten about Piano Society. In many ways, music seems to have been on hold, other than traveling, and spending summer weekends on the Cape. I've been so busy at the office that I wish I had the time to play and record more often.

Happy New Year to all esteemed musicians at Piano Society!

Hi George! I'm so glad you are still around. I know, this time of the year is crazy. But now the new year has begun, so hopefully things can settle down and you can get back to your piano! Happy New Year to you too!!

btw - 34,000 views is a lot! You mean "zoinks!" not "wow," don't you!

Hi George,

happy new year to you too! I look forward to listening to you on Pianosociety!

About your opening post that you have edited, let me stress again that the pianist, the piano, the room and the mic placement are much more important that the selection of the audio gear. You say that this latter one can contribute for 5-10 % to the result: this statement implies that the recording quality could be measured and the relative measurement for two recordings would have the same value for everybody. But such a measurement, if any, is much subjective. I think that your 5-10% is much less for most people and irrelevant to them with respect to the cost of the audio gear that you are proposing: mics above 1000, up to more than 10000 €/$ a pair, preamp above 2000 €/$ (except for DAV), converter above 1000 €/$. I think even that many people might not notice the difference between a take made with 'low-end' microphones and electronics and 'high-end' ones if they could listen to both takes. I'm quite sure that most people (myself first) would not be sensitive to the quality loss encountered in a recording because of the use of low-end audio gear, provided that it is well selected and used, if they could not compare with a high-end reference.

I did a take on this evening with two pairs of mics on the same stand, Samson CO2 and Schoeps CMC6-MK21, which is in your selection and costs up to 30 times more than the Samson. The Schoeps were plugged on a DAV BG1, also in your selection, and the Samson were plugged directly on a PC audio interface to which the DAV was also connected so that both stereo tracks could be recorded on the same PC. Also I used professional cables for the Schoeps, about 10 times more expensive than the cheap ones used for the Samson.

Picture of the mics, Sanmson and Schoeps tracks here attached.

NB Neumann stopped the manufacturing of the M 50, appearing in your list, near 40 years ago. I guess that you intended to put the TLM 50 or the M 150 instead.