"Dr A. N. Walker" wrote in message
...
In article 03n0h.10597$gZ2.8423@trndny07,
Chess One wrote:
[...] I think there are world records for such
things - but I wonder if its true that 10,000 correctly sequenced
iterations
of pi can be remembered, rather than calculated?

"http://pi-world-ranking-list.com/lists/memo" records 13 people
who have recited pi from memory to at least 10000 places; and there is
a recent report that Akira Haraguchi set a world record of 100000 just
over three weeks ago. I wonder what the longest ever [in terms of
number of notes rather than mere time!]

remarkable! my wild guess might be something from Rachmaninoff

correct piano recital has been?
100000 notes would be at any rate several hours, even of quite complex
concertos or sonatas, unless you just played "Islamey" over and over.

although we should continue to note [!] that the pianist has more to do than
recite a linear sequence, since each and every note is of certain and
specific duration, weight, and speed, etc.

but not to finish on that note - it is extraordinary to me that 100k
sequenced datum, which are non-musical could be undertaken - I must suppose
that he calculated the next in sequence [rather than any feat of memory] and
wonder how long it took him to do this overall as well as the typical
interval from one digit to the next?

phil

--
Andy Walker, School of MathSci., Univ. of Nott'm, UK.

In article _Bq0h.22$gf5.1@trndny01, Chess One wrote:
[...] - it is extraordinary to me that 100k
sequenced datum, which are non-musical could be undertaken - I must suppose
that he calculated the next in sequence [rather than any feat of memory] and
wonder how long it took him to do this overall as well as the typical
interval from one digit to the next?

No, not a chance. There are humanly-computable ways of finding
specific digits in a hexadecimal expansion [and so in binary or octal]
of pi, so you could plausibly ask a skilled mathematician to tell you [eg]
the 1234567th such digit; but no known way of doing it in decimal. If
there were, I'd guess it would take hours, even with a calculator, to
get each digit. Calculating pi to even a few hundred places of decimals
without computer assistance is years of work.

We can safely assume that Akira Haraguchi simply has a somewhat
unusual memory.

--
Andy Walker, School of MathSci., Univ. of Nott'm, UK.

On 27 Oct 2006 17:00:51 GMT, (Dr A. N. Walker)
wrote:

In article _Bq0h.22$gf5.1@trndny01, Chess One wrote:
[...] - it is extraordinary to me that 100k
sequenced datum, which are non-musical could be undertaken - I must suppose
that he calculated the next in sequence [rather than any feat of memory] and
wonder how long it took him to do this overall as well as the typical
interval from one digit to the next?

No, not a chance. There are humanly-computable ways of finding
specific digits in a hexadecimal expansion [and so in binary or octal]
of pi, so you could plausibly ask a skilled mathematician to tell you [eg]
the 1234567th such digit; but no known way of doing it in decimal. If
there were, I'd guess it would take hours, even with a calculator, to
get each digit. Calculating pi to even a few hundred places of decimals
without computer assistance is years of work.

There is no easy way of computing the n-th place of pi in any base, be
it binary, hexadecimal or decimal. There is also no repeating pattern
in this number, regardless od base.

We can safely assume that Akira Haraguchi simply has a somewhat
unusual memory.
Yes, that we can.

Johannes

Johannes Seppi wrote:
On 27 Oct 2006 17:00:51 GMT, (Dr A. N. Walker)
wrote:

In article _Bq0h.22$gf5.1@trndny01, Chess One wrote:
[...] - it is extraordinary to me that 100k
sequenced datum, which are non-musical could be undertaken - I must suppose
that he calculated the next in sequence [rather than any feat of memory] and
wonder how long it took him to do this overall as well as the typical
interval from one digit to the next?

No, not a chance. There are humanly-computable ways of finding
specific digits in a hexadecimal expansion [and so in binary or octal]
of pi, so you could plausibly ask a skilled mathematician to tell you [eg]
the 1234567th such digit; but no known way of doing it in decimal. If
there were, I'd guess it would take hours, even with a calculator, to
get each digit. Calculating pi to even a few hundred places of decimals
without computer assistance is years of work.

There is no easy way of computing the n-th place of pi in any base, be
it binary, hexadecimal or decimal. There is also no repeating pattern
in this number, regardless od base.

Amazingly there is a specific expansion for Pi in base hexadecimal
found by Simon Plouffe in 1995 see for example:

http://mathworld.wolfram.com/BBPFormula.html

"Easy calculation" here is a relative term, but it is a pretty
astonishing result.

We can safely assume that Akira Haraguchi simply has a somewhat
unusual memory.

Yes, that we can.

Have any of the very deep Pi memorisers explained how they do it?
(I know of various poems for modest lengths but 100k digits is
astonishing)

The only other time I have seen it discussed the person had
synaesthesia and saw number patterns as colours. He could also do some
pretty amazing mental airthmetic too.

Regards,
Martin Brown

Martin Brown wrote:
Have any of the very deep Pi memorisers explained how they do it?
(I know of various poems for modest lengths but 100k digits is
astonishing)

A common technique is, as you say, to memorise some text where the
lengths of the words corresponds to the digits in the sequence.
Another is to remember a story where the nouns and other significant
words correspond to the digits. So, for example, you could associate
five nouns with each digit to give enough variation.


The only other time I have seen it discussed the person had
synaesthesia and saw number patterns as colours.

Sometimes, the human brain is broken in really interesting ways. :-)


Dave.

--
David Richerby Natural Bulb (TM): it's like a light
www.chiark.greenend.org.uk/~davidr/ bulb but it's completely natural!

On 30 Oct 2006 00:27:12 -0800, "Martin Brown"
wrote:

[...]

Amazingly there is a specific expansion for Pi in base hexadecimal
found by Simon Plouffe in 1995 see for example:

http://mathworld.wolfram.com/BBPFormula.html

"Easy calculation" here is a relative term, but it is a pretty
astonishing result.


Thanks, I didn't know that one!

Johannes