the case of the disappearing 250047/250000

From tuning . . .

I was eliminating commas where the greatest common divisor of the
entries in the monzo was greater than 1, since these are merely
powers of other commas. But when I moved to 7-limit, I was
unfortunately still checking only the first three components of the
monzo. Hence I erroneously eliminated 250047/250000 and, I think,
1077 other commas under 600 cents where -20<e3<20, -14<e5<14, and -
12<e7<12.

Both this comma and the most complex one on our recent lists were
mentioned here:

/tuning-math/message/6390

and 250047/250000 is mentioned in many posts, for example:
/tuning-math/message/4542

but even earlier here:
/tuning-math/message/1213

and its earliest mention on this list was here:
/tuning-math/message/200

--- In tuning-math@yahoogroups.com, "Paul Erlich" <perlich@a...>
wrote:
> From tuning . . .
>
> I was eliminating commas where the greatest common divisor of the
> entries in the monzo was greater than 1, since these are merely
> powers of other commas. But when I moved to 7-limit, I was
> unfortunately still checking only the first three components of the
> monzo. Hence I erroneously eliminated 250047/250000 and, I think,
> 1077 other commas under 600 cents where -20<e3<20, -14<e5<14, and -
> 12<e7<12.

And apparently 391 others under 600 with log(n*d)/log(10)<18, which
is what the graphs were showing :(

I've uploaded the 3 corrected graphs.