But isn't it all relative? You can transpose a song into other keys, and I would not be able to tell the difference.

To me I could play in any key and be happy. I often transition through several different keys when I practice/improvise. I just honestly cannot hear the difference. I'll play through many melodies and think to myself, would this really sound different in C or C# or B? And then I play it in another key and realize, no it is not; it is the same song.


Perhaps long trained musicians hear it differently than I do. I did not grow up knowing or learning music. I only learned once I became an adult.

krstone, I understand what you are saying, but to me, there is great deal of differences in the sound of different keys. And I don't know how to explain it, except that some sound richer, deeper, lighter or happier than others. I read somewhere that composers don't always think about which key to compose in - they just write what is in their head at the time the piece comes to them. Maybe try playing something simple like Mary Had a Little Lamb. Play it in C-major (don't change it to a minor key). Then play it in something like F# major. The black keys sound different, somehow. Maybe it's the way the sound waves come off the vibrations on those strings? (I don't totally get what I just said there. I get what I'm thinking but I can't express it well.) Any of that make sense to you?

D-Flat Major RULES!!!

But don't rule out B-flat major

Perhaps it has something to do with the tempered tuning?

What is tempered tuning?

Our common tuning is the equal temperated tuning. That means that the difference between every note is the same, the distance is mathematically seen the 12th root of 2. That means, after 12 halftones the frequency is doubled, as it is the next octave.
With this tuning system there are no very bad sounding intervalls at all on piano (but one could say also, no very good sounding intervalls). For instance if you play a quinte, you will hear a slight wobbling, maybe every second or so. If you play the quint a half tone higher, and proceed every half tone, you will realize that this wobbling gets contiously more and more, an octave higher the wobbling is doubled. At least that is the case in a perfect tuned piano.

Beside that there are historical temerated tunings with not equal intervals. Some sound worse, some better. Because some intervals have no wobblings, normally received as very strong and powerful chords. Others have so much wobblings that one needs to avoid those keys better.

To me it is problematic to give in the equal temperated tuning different keys a certain character. Because from the temperament all keys are similar (in opposit to the nonequal temperated tunings).
That's why it is problematic (to me) to e.g. associate d major with a festive character (only to mention an example). Already the fact, that some played 300 years ago d major a half tone lower or higher, so that means d flat major or e flat major were the keys in those days which must be associated with the character some refer with d major today.
The different character of different keys come in my opinion more from how often a key is used in general or for certain moods some personal "reference" pieces have in that key, but not from inherent character of a key. Because with equal temperated tuning, there is no certain character anymore of a certain key. By the way, I very much like historical tunings, there are some organs in Germany at which those tunings are used with success for certain baroque music.

Never mind, Terez. I just looked it up. That kind of thing is way over my head.

Here's the long explanation of tempered tuning, or Equal Temperament (link), or perhaps Well Temperament as opposed to Pythagorean Tuning, or perhaps Meantone Temperament. A great deal of Baroque music is written with Pythagorean tuning or meantone temperament in mind, and the differences are fairly slight for the most part, but enough that a certain harpsichord would only be good for certain keys. In , the leading tone of any key should be at a pitch sharper than a 100-cent half step (not sure exactly how many cents would be between the leading tone and the root (i.e., B to C in the key of C) but it's less than the 100-cent distance of every half step in equal temperament tuning.

I don't know a great deal about the differences in tuning, which is why my comment above was a question. lol...it actually seems counter-intuitive to think that tempered tuning would have anything to do with why certain keys sound better on a keyboard with equal temperament (and we all play on equal temperament keyboards). Ah, well....that's why Bach's Well Tempered Clavier was such a revolutionary work, I think - it was meant for a keyboard on which you could play any key! And I think it was the first major work that was intended for such an instrument, though I could of course be wrong.

EDIT: wow, I must be slow - two people posted while I was working on this!

oops - looks like we posted at the exact same time. Thanks for answering but I still don't get it.

Wow! Three people posting at exactly the same time. Now I'll go read Olaf's response.

Also not from what I wrote? If that's the case, what exactly you don't get (indeed 3 posts with same time stamp)?

It's okay, Monica - I don't really get it either!

You guys are too smart for me. Thank you for the explanations, though.

Olaf, off the top of your head, do you know of any piano recordings on the internet that use 'historical tuning' ? (don't waste any more time if you don't know - it's okay)