[seqfan] Re: A030057 graph anomaly

[M. F. Hasler]

On Fri, Apr 4, 2014 at 8:54 AM, David Wilson <davidwwilson at comcast.net> wrote:
> Yes, I knew this, but it is odd to see a graph with tick values that are not round numbers.


Not sure I understand; the tick (and data) values ARE round numbers!?
(Tick values 10, 100, 1000 ; data values 2, 4, 8, 13, 16, 29, 32, ...)

Every 2nd values is 2, this is the "bottom line" ; next most frequent
value is 4, this makes the still almost solid looking 2nd line from
the bottom a little bit above half-way of 10 (I think this is Graeme's
point: 4 >~ sqrt(10)).
Next most frequent value is 8, this makes the 3rd line from the
bottom, just beneath 10 (>~ 8).
According to the fact that 8 / 4 = 4 / 2, these three lines are
equidistant (on the log plot), which is not the case for the line of
13, but again for that of the next most frequent value, 16.

What might be more intriguing is the fact that there are inversions of
the order by frequency vs magnitude: generally, smaller values are
more frequent, but
61 seems more frequent than 57 ; 92 more frequent than 91 ; ...
[but this might be no more true when going beyond the precomputed range 1..1e4]

Maximilian


> Perhaps I am more sensitive to this because my current job involves a lot of graphing.
>
>> -----Original Message-----
>> From: SeqFan [mailto:seqfan-bounces at list.seqfan.eu] On Behalf Of Graeme
>> McRae
>> Sent: Thursday, April 03, 2014 8:04 PM
>> To: Sequence Fanatics Discussion list
>> Subject: [seqfan] Re: A030057 graph anomaly
>>
>> There's nothing wrong with it. 3.16227766017 is the square root of 10.
>>
>> --Graeme McRae
>> Palmdale, CA
>>
>>
>> On Thu, Apr 3, 2014 at 4:48 PM, David Wilson
>> <davidwwilson at comcast.net>wrote:
>>
>> > Look at the left axis of the bottom plot in the graph on A030057.
>> >
>> >
>> > _______________________________________________
>> >
>> > Seqfan Mailing list - http://list.seqfan.eu/
>> >
>>
>> _______________________________________________
>>
>> Seqfan Mailing list - http://list.seqfan.eu/
>
>
> _______________________________________________
>
> Seqfan Mailing list - http://list.seqfan.eu/



-- 
Maximilian