Kees, what setting should I use for Tune-lab while using your numbers?

Regards,

Anything you want (perhaps even split-scale mode).

Kees

The same is true for my "offset numbers" (but not the original absolute numbers which lock you into a specific partials scheme). The offsets are partial independent.

Kees

I tried your numbers, Kees, with the 5'8" Knabe. I afraid the numbers didn't make it out of the temperament. F3-A3 was not as fast as G3-B3, and the A#3-D4 third was too wide. The equal beating fifth/fourth checks were fairly close, but the overall temperament was much stronger than I would expect EBVT III.

Sorry to have wasted your time. I was experimenting with stronger versions of EBVT and temporarily made some changes in my program and forgot to revert before I computed your numbers.

I can send you the correct numbers if you are still willing to try.

Kees

Ah! That makes sense. It wasn't a waste of time, just not what I was expecting. I used those numbers to pitch correct the piano, so I got to play it with the unisons pulled in quickly before trying to fine tune. I will say, C-major sounded truly amazing with that temperament!

If you could, could you send me EBVT numbers when you get a chance? If you still do the completely custom method, I would be most interested in that, but please don't feel compelled to change things around on my account.

Thank You!

OK, I sent them again. That was Bill's EBWT you tuned there.

Kees

PS. I sent you the offset data rather than the completely customized data because I believe the latter would be a last resort option if the offset idea (due to you) does not work.

I don't really see a program to compute EBVT3 offsets absolutely with a rigid stretching scheme (my original plan) as practical. But a collection of offset files for a suite of pianos can be distributed just as a bunch of tunelab templates.

Cheers,
Kees

Hi guys,

I've been very busy, and will be for the remainder of the week. I've tuned aurally, mainly concert work for the last week or so, so I haven't had that much time to check the numbers. I will get back on track with that as soon as I can!

It's been a while!

Kees: I finally tried your new numbers on my own piano, the Nordiska 5'5" grand. The scaling in not perfect, but fairly consistent in the treble. I used them for the treble, and they worked very well. There were two small tweaks in the temperament, F#3 was slightly sharp, and A#3 was slightly flat, but these were very small adjustments. A4 ended up being a little flat as well. Instead of being a pure 6:3 octave, it seemed to be a nearly pure 4:2 octave. In this case it was no big deal to adjust it sharp, because I had started by floating pitch about 1 cent flat.

Other than that, the numbers worked just as good, probably even a little better than mine for the treble.

I tuned the bass aurally, as I will explain in a bit.

To set the tuning curve for the treble, I did the following:

Tune F3-A4 with auto 6:3/4:1 settings.

Set Tunelab to E6, and play E4 (4th partial) and A4 (3rd partial), then adjusting so that E4 reads just as sharp as A4 does flat.

Tune A4-E6.

Set Tunelab to C7, and play F5 (3rd partial). Adjust overall treble stretch so that F5 reads as pure.

I chose to set E6, as this note falls very close to the middle between A4 and C8, and I chose C7, based on the comment that Patrick made that he wants wide 12th from C7 onward.

This method is based on the assumption that the offsets are absolute in their accuracy from an "ideal tuning curve", and hopefully this would be a more consistent method than simply relying on the program's inharmonicity model.

When I tried to develop a method of imitating this matching for the bass, I came up empty and just tuned the bass aurally.

Any ideas on using a similar method in the bass? What are the exact parameters from A2 down?