Piano Scale System Diagnostic - closed (now piano video clips)

garylkoh

WBF Technical Expert (Speakers & Audio Equipment)
Sep 6, 2010
5,599
225
1,190
Seattle, WA
www.genesisloudspeakers.com
Here's a track that I've found to be extremely useful in diagnosis of room/speaker interaction and speaker design. It is a recording of every key of a Fazioli grand piano done by Steve McCormack. The piano player is also Steve, and he said that he tried his best to hit each key equally hard.

From Steve - "It was recorded at Soka University (where Jim Merod teaches and runs the "Jazz Monsters" concert series) several years ago. A Fazioli grand piano (a truly magnificent instrument) had been loaned for one of the concerts with Mike Garson. I recorded it to a MOTU 826HD via Firewire to my then Windows laptop. When the concert was over I waited until most of the people had left, then sat down at the piano and made this recording before teardown. Yes, it was 24/96 native, and done in stereo. The conversion to 16/44.1 was done in Wavelab, my standard at the time (these days I have superior tools from Izotope for doing up-or-down sampling)."

Just in case anyone questions Steve's recording "chops" - he was one of the recording engineers behind some of the fabulous M&K Realtime direct-to-disc albums done in the 1970's.

With a pair of well-designed loudspeakers, in a properly set-up room, each note should sound subjectively equally loud. If there is a big difference, especially from one note to the next, you have more work still to do (either in your loudspeaker design or your room set-up).
 
Last edited:

Bruce B

WBF Founding Member, Pro Audio Production Member
Apr 25, 2010
7,006
512
1,740
Snohomish, WA
www.pugetsoundstudios.com
With all due respect Gary, this is far from a good diagnostic tool. As you can see from the 3 images that I've inserted. His "chops" must have not been in piano performance.
As you can see in the first image that the RMS and peak values of each note being played is all over the map, with no consistency at all!
The note being played at 41.7 seconds into the file is actually clipped as opposed to the note played at 2:45 is 6dB down!
I really hope you don't set up Genesis speakers to this file.
 

Attachments

  • Piano..jpg
    Piano..jpg
    251.2 KB · Views: 196
  • -6dB..jpg
    -6dB..jpg
    105 KB · Views: 191
  • Clip..jpg
    Clip..jpg
    131.2 KB · Views: 188

garylkoh

WBF Technical Expert (Speakers & Audio Equipment)
Sep 6, 2010
5,599
225
1,190
Seattle, WA
www.genesisloudspeakers.com
Hi Bruce,

It's not the only tool I use. I found it extremely useful and decided to share it after a discussion at another thread.

Steve said that he tried his best to hit each key equally hard, but may not have been entirely successful.

Since you have a grand piano in your living room, could you do a better job and post it here?

Cheers
Gary
 

Bruce B

WBF Founding Member, Pro Audio Production Member
Apr 25, 2010
7,006
512
1,740
Snohomish, WA
www.pugetsoundstudios.com
Hi Bruce,Since you have a grand piano in your living room, could you do a better job and post it here?

Cheers
Gary

I'm sorry, but I don't see how the file could be useful to anyone.

The only way I'd be able to do it is to take each individual note and adjust it up/down digitally in the workstation. The best thing would be to use a Yamaha Disclavier via midi.
 

garylkoh

WBF Technical Expert (Speakers & Audio Equipment)
Sep 6, 2010
5,599
225
1,190
Seattle, WA
www.genesisloudspeakers.com
The most useful thing I find about this track is that each note from one to the next is subjectively equally loud. Each note should have the same character and timbre like they are from the same piano in the same acoustic. If there are obvious discontinuities, then coherence and integration between drivers is lacking, or the room may be emphasizing some one aspect or harmonics more than another. Further, listening to each note on a piano is instructive.

It also reminds me that measurements aren't everything, and the ear is as important as all my instruments in measurement. Just getting a flat frequency response does not guarantee that a pair of loudspeakers will sound correct.

Since you don't see how the file could be useful to anyone but me, I'll delete it. It was one of Steve's and my "secret tools". We'll continue to keep it secret.
 

JackD201

WBF Founding Member
Apr 20, 2010
12,308
1,425
1,820
Manila, Philippines
I'm sorry, but I don't see how the file could be useful to anyone.

The only way I'd be able to do it is to take each individual note and adjust it up/down digitally in the workstation. The best thing would be to use a Yamaha Disclavier via midi.

In the spirit of WBF, Skip the disclavier and use this



Someday, someday, someday.................
 

Bruce B

WBF Founding Member, Pro Audio Production Member
Apr 25, 2010
7,006
512
1,740
Snohomish, WA
www.pugetsoundstudios.com
Bruce,

Perhaps a Midi-derived file with equal amplitudes could be one track of the WBF test/demo CD?????

Lee

Sounds like a plan!

In the spirit of WBF, Skip the disclavier and use this Someday, someday, someday.................

Now that's what I'm talking about!
 

Jaguar

Well-Known Member
Aug 2, 2010
221
17
925
Bellevue, WA
Hi guys, I would really like to see if we can do a meeting sometime that's focused on measuring and correcting a room. I've been using the XTZ Analyzer, which is somewhat of a beginners' tool and there's still a lot I need to learn about how to use it effectively. I know there are several of us who could learn a lot from a practical, and sophisticated discussion on this. Maybe we can track down that gentleman in Bellevue, who wrote the book on this stuff.
 

The Smokester

Well-Known Member
Jun 7, 2010
347
1
925
N. California
I'm sorry, but I don't see how the file could be useful to anyone.

The only way I'd be able to do it is to take each individual note and adjust it up/down digitally in the workstation. The best thing would be to use a Yamaha Disclavier via midi.

Your posts on this thread have to be among the most vacuous and irrelevant I have seen. Are you actually saying that an experienced piano player cannot play adequately even notes based on your casual perusal of randomly sampled points on a video monitor? What is your criteron? If a Disklavier was used how would it compare? Do you know? If so, show us some results.

If you think you can do better then normalize them yourself, or record them from a player piano if you have the ability (certainly won't be a question of creativity) and give up something constructive. Not destructive. Like your thoughtless posts.
 

Jaguar

Well-Known Member
Aug 2, 2010
221
17
925
Bellevue, WA
Smokester, thankfully Gary won't take a slight criticism of his idea as personally as you seem to. No doubt that someone with his ears can find some usefulness with a file like this, but I think it's a fair point that very precise consistency in the notes is likely to lead to more consistent results. Bruce is one of the few people here who probably does actually know how a Disklavier would compare, so if what he says might sound vacuous or thoughtless, it's not.
 

Orb

New Member
Sep 8, 2010
3,010
2
0
I'm sorry, but I don't see how the file could be useful to anyone.

The only way I'd be able to do it is to take each individual note and adjust it up/down digitally in the workstation. The best thing would be to use a Yamaha Disclavier via midi.

Hi Bruce,
I guess it depends what you want these for.
As an example if comparing different products then this is still very useful as you would possibly take just one segment (one note played) and use this to compare/measure both products or subtle changes you make.
The benefit is that using a real instrument and being able to select where on the FR the note is generated, so can then specifically target-use low/mid/or high notes on the scale to see if they subtly change from the original.
This does require though being able to process the note as a time varying spectral plot using a seperate envelope to show each partial, so we would see around 30-40 partials for up to 20khz presented against time and amplitude.
Alternative for some would be to replay the same note repeatedly until sure subtle changes can be noticed in the system.

In this way it does not matter if the 2nd/3rd/etc note are not all identical, bearing in mind mic setup-position, the frequency,etc I am not sure how or if it is possible that each note will have identical amplitude pattern when looking at it in detail, especially if you consider how the note is played in terms of nuances on the piano that can deliberately add complexity.
Problem (and a reason it is a good one to use including how much of the frequency range a piano covers) is the piano is one of the most dynamic instruments available, especially when using a good concert piano.

Disclavier maybe the way to go for those for some, but then who also has a good hall (and a concert-grand piano version of a Disklavier) and the equipment to do a very good 24/96 recording?

So IMO its great to have this recording made available by Gary and ties in with an aspect we discussed in other threads including measurements, its just depends how you use it.
Thanks Gary and I did not respond in the other thread as we had taken it off topic but want to say it is cool the time was taken to do this and definitely is useful.

Cheers
Orb
 
Last edited:

terryj

New Member
Jul 4, 2010
512
0
0
bathurst NSW
as much as I appreciate the sentiment of this, I am confused a bit why this is in any way better than a quick fr sweep?

maybe it is only for those who cannot measure their systems?

If not, can someone explain what advantages this has over the fr sweep?
 

Jay_S

WBF Founding Member
Apr 20, 2010
309
5
16
San Francisco - East Bay
OK, help me understand this. Bruce, the member who concluded that he doesn't want to use the piano files, had the ability to download them and try them out. But everyone else who wants to try the diagnostic, no luck. Were the files intended for Bruce only?
 

JackD201

WBF Founding Member
Apr 20, 2010
12,308
1,425
1,820
Manila, Philippines
I'd use it because it shows the complete musical envelope comprised of a complex mix of frequencies as opposed to a sweep. I use sweeps for overall FR response pretty much to look for gross anomalies in the speaker/room interface.

Recordings like this could be helpful in checking out what part of the envelope is being emphasized for a particular speaker set up configuration. Like exaggerated decay for example that could be a product of a room with too long a reverberation time. Going up and down the scale could show you what frequencies need damping. So such a disc would also help the user appreciate ratios between direct sound and reflected sound as he or she played with different on and off axis relationships as it pertains to the phantom image. There are other uses but I'm a bit short on time because I'm going out with some buddies.

As for the genesis of this, sorry I couldn't resist! :) Gary and I had been having an exchange about following scales when setting up speaker positions to be able to get reasonably consistent image focus and pitch articulation from top to bottom, or in the case of our discussion from the bottom up. The discussion stemmed from a short piece I wrote for newcomers to the hobby, who don't have access to measuring equipment and most probably wouldn't know how to use them or interpret the results even if they did have access via freeware or whatever. I also don't see why more experienced hobbyists can't find some use for it. I don't think Gary ever meant for it to be anything more than a tool and not as a definitive example of what a piano should sound like.

What Bruce pointed out was not an objection to the process but rather to short comings in the recording itself as well as some issues regarding consistency of pressure on the keys which could be easily programmed on MIDI. I didn't see Bruce invalidate or attempt to invalidate the use Gary and Steve have for the recording only that a better set of recordings would be, well......better.

:)

Send me a copy Gary. I'll replace it when someone gets the Bosendorfer folks to program and record one of their own.
 

The Smokester

Well-Known Member
Jun 7, 2010
347
1
925
N. California
Bruce's point about note-to-note variation rending this test recording useless is not proved and probably is a tautology. For a recorded piano listening test, what would the relevant recording be if not from a real piano played by a real person? By logical extension I should throw out all my recordings of Horowitz because the recorded notes aren't up to some spec?

What is the note-to-note variation of a typical piano recording? Is it different than this example? Since when do two points define a peak? If each note was used to measure room response one could use a waterfall plot from which useful information could be derived without reference to the relative level with other notes.

Plus, a repected member has opined that he has found this a useful diagnostic. I would like the opportunity to listen and judge for myself.
 

FrantzM

Member Sponsor & WBF Founding Member
Apr 20, 2010
6,455
29
405
The Smokester

Not wearing a moderator hat in this post .. So ..

I find your criticism of Bruce B post harsh and unwarranted. It was also my first reaction: How can one maintain equal loudness with varying frequency .. Striking the notes with equal force would not do trhe trick so would we have to go with the pianist' perception of equal loudness which while it can be well trained, is likely to be inaccurate? With such protocol, there can't be any guarantee that the loudness would be equal. So I find this of limited usefulness. Your point about a real piano is well taken but would not apply here since we are trying to correct a system with a recording that would be the baseline. This recording , the way it was produced is not to be music but a measurement tool and IMO it is flawed and this is what Bruce B pointed at and to me he (Bruce) is correct.

I hasten to say that I laud Gary effort and generosity. I am currently building his server. Gary has been a treasure on this forum, providing information with humor and deep knowledge and I appreciate his contributions and looking forward for more of his posts (by the way Gary I am huge on Burmester ... and have a friend with Genesis II and Burmester 911 in NY) . I simply think that this particular tool is of limited usefulness.
 

Bruce B

WBF Founding Member, Pro Audio Production Member
Apr 25, 2010
7,006
512
1,740
Snohomish, WA
www.pugetsoundstudios.com
Gary is a dear friend of mine and he has given me and the PNWAS a wealth of knowledge.
The very fundamental of this test is flawed (loudness). If this was about tonality, decay, etc, then no problem.

Gary states:
With a pair of well-designed loudspeakers, in a properly set-up room, each note should sound subjectively equally loud. If there is a big difference, especially from one note to the next, you have more work still to do (either in your loudspeaker design or your room set-up).

So if some notes are 6dB different in Peak and RMS value, then if one should play this on their system, would they gather that they have these descrepencies in their speakers, room or system? Most people can hear even a .5dB difference in a single note!

If one should want a piano library of each note being hit with the same exact amplitude, then they could purchase one of the many Piano Sample libraries. These companies go to great lengths in sampling a single note dozens of times using mechanical actuators and specially prepared pianos.

Synthogy

East/West

Steinberg

Yes, this file can be used as a tool, but like building a house, you need dozens of tools. Let's hope the one tool that a carpenter uses, maybe a level?, isn't used for the whole house!
 
Last edited:

About us

  • What’s Best Forum is THE forum for high end audio, product reviews, advice and sharing experiences on the best of everything else. This is THE place where audiophiles and audio companies discuss vintage, contemporary and new audio products, music servers, music streamers, computer audio, digital-to-analog converters, turntables, phono stages, cartridges, reel-to-reel tape machines, speakers, headphones and tube and solid-state amplification. Founded in 2010 What’s Best Forum invites intelligent and courteous people of all interests and backgrounds to describe and discuss the best of everything. From beginners to life-long hobbyists to industry professionals, we enjoy learning about new things and meeting new people, and participating in spirited debates.

Quick Navigation

User Menu

Steve Williams
Site Founder | Site Owner | Administrator
Ron Resnick
Site Co-Owner | Administrator
Julian (The Fixer)
Website Build | Marketing Managersing