Sun Mon Tue Wed Thu Fri Sat
1 2 3 4 5 6
7 8 9 10 11 12 13
14 15 16 17 18 19 20
21 22 23 24 25 26 27
28 29 30
|
Master of Go
I finished reading De meester van het go-spel
(The master of
go) by Yasunari Kawabata. A good discussion about this book, from a Go point of view, see Kawabata's Master of Go in the Sensei's Library. I found it quite an
interesting book to read, but as a literary novel, I did not find it very
good. I especially did not like the chaotic order in which the events are
being told. At one point, I even thought that some pages were swapped due
to the book being a misprint. This was in chapter 15 when the author is
talking about moves 59 and 63, but one page later talks about stopping the
game at move 43. The author reported about the game described in the book
in 66 newspaper columns, and it almost seems as if he had accidently mixed
up the order of these columns when working them into this novel. I also
feel that the style of Kawabata is somewhere in the middle of a factual
report and that of fictional novel. I guess that the book would have been
beter, if he would have chosen one of those styles. I think I am also going
to read "The Meijin's Retirement Game", which gives a more detailed
description of the actual game.
3DF Samantha
Last Monday, I tested the 3DF Samantha by 3D Flow, a program
that implements Structure from Motion. I
tested the program on my own set of plant pictures and it did return some
good results. The program can make use of Compute Unified Device Architecture (CUDA) to utilize the processing
power of GPUs. On the PC I used for the test, CUDA is not working, so, I
cannot make any statement with respect to the performance. 3DF Samantha
is a command line utility and follows the normal steps of a Structure from
Motion algorithm. The output looked kind of familiar, and at first I thought
that it could be the case that the implementation is based on some of the other
free implementations available that I have tested before. But I have been
told that the code was written from scratch with the only exception of the
bundle adjustment routines for which the SSBA library is used. The returned point cloud was not very dense (compared to other free applications and
services), but I guess that it a good input for 3DF Stasia, the high fidelity Multiview stereo Technology by 3D Flow,
which has not been released.
Diagonal mazes: exact results
I worked on a program to calculate the exact occurences of loops with certain
lengths in diagonal mazes.
This program tries all paths
up to a specified length and counts how many of these end at the starting
position. Paths that wander too far away, and never can reach the starting
position, are avoided. The numbers found, match those found with the
averages about the random generated mazes. The results are, with increasing
area of the loop. The area is the half of the number of steps, because
each step occupies half a unit square. The number of possible paths increase
exponential with the length. Because of this, it is only possible to
calculate the numbers for short loops. It is not clear whether it is
possible to say something about the number of infinite paths.
2 12.500000
4 6.250000
6 3.662109
8 2.758789
10 2.106667
12 1.732349
14 1.441716
16 1.242055
18 1.083980
20 0.961059
22 0.861584
24 0.779843
26 0.711380
28 0.653467
30 0.603734
Magnolia
Yesterday afternoon, the first flowers of our
magnolia opened. This afternoon about one third of the buds had
opened. This year flowering is quite late, because in March and the
first week of April temperatures have been almost five degrees below
average. The last weeks temperatures have been slightly above average,
but they will get lower again the coming days, with the possibility
of some freezing during the night. Today, there was quite a strong
wind, and I was afraid that the flowers would have been blown away,
but that did not happen.
DVDs and books
This morning at 11:24, I bought six DVDs from bookshop De Slegte for €19,80 under the three for two rule.
I bought: Hero, Final Fantasy: The Spirits Within,
Final Fantasy VII: Advent Children, Lila dit ç,
One Day, and La Fabrique des Sentiments.
In the afternoon, I visited bookshop
Kruimeltje and discovered that they were now selling books for
one Euro each. When I went inside, I noticed that already many books
were sold. I went to the second floor and looked if there are still
some books by Mulisch available. In this bookshop there were often
more books behind the first row on a shelve. When looking, around
the shelves, my eye caught the name Anaïs Nin. I found a whole stack of her diaries translated
to Dutch. I found a complete series ranging from 1931 to 1966. These
are:
- Anaïs Nin dagboek 1931-1934
(The diary of Anaïs Nin 1931-1934) by
Anaïs Nin. ISBN:9060192397.
- Anaïs Nin dagboek 1934-1939
(The diary of Anaïs Nin 1934-1939) by
Anaïs Nin. ISBN:9060192391.
- Anaïs Nin dagboek 1939-1944
(The diary of Anaïs Nin 1939-1944) by
Anaïs Nin. ISBN:9060193482.
- Anaïs Nin dagboek 1944-1947
(The diary of Anaïs Nin 1944-1947) by
Anaïs Nin. ISBN:9060193946.
- Anaïs Nin dagboek 1947-1955
(The diary of Anaïs Nin 1947-1955) by
Anaïs Nin. ISBN:9060194349.
- Anaïs Nin dagboek 1955-1966
(The diary of Anaïs Nin 1955-1966) by
Anaïs Nin. ISBN:9060195507.
Diagonal mazes: coloured by length
This weekend, I developed a script
for MySample to colour diagonal mazes by length of the loops. Below a part of an image for a
1000 by 1000 simulation. The loops are coloured from red to green with
increasing length. Loops that would be infinite if the the random generated
simulation is used to tile in all directions are coloured blue. The image
is rotated around 45 degrees to allow a compact representation.
I found this quite interesting. I am getting the impression that true
infinite paths are very rare in a completely random pattern in all
directions. This because the space that is available for loops above
a certain length is becoming smaller, and thus resulting in a higher
chance that they connect. I was also surprised how compact most loops
are, or better said at how many places they touch themselves.
The Meijin's Retirement Game
This evening, I finished reading the book
The
Meijin's Retirement Game by John Fairbairn, which I started reading on
the sixteenth, the day I received it in the mail after I ordered it from
Go Show Keima.
This book deals with the Go game Honinbo between Shusai
and Kitani Minoru, which has been fictionalized in The
Master of Go. I read the whole book, but I did not study all the
diagrams in detail. I am afraid that most of the issues in the game are far
above my level, but still I found it an interesting read. Although the book
contains an appendix about The Master of Go, I somehow had hoped that
it would be more integrated with this story. I think it would be interesting
to have a novel from the viewpoint of the players, dealing about their
thoughts and struggles. But who can creep into the minds of these Go masters?
Magnolia
This afternoon, almost all the flowers on our
magnolia were wide open and a heavy scent was hanging around the plant.
Henry Miller
This morning, I visited bookshop Kruimeltje
in Enschede for the last time. At 11:24, I bought some books by Henry Miller
for one Euro each.
- De Steenbokskeerkring (translated by John Vandenbergh
from Tropic of capricorn).
- Zwarte Lente (translated by John Vandenberg from Black Spring).
- Rustige dagen in Clichy (translated by Gerrit Komrij from
Quiet days in Clichy), Grote ABC nr. 108.
This months interesting links
Home
| March 2013
| May 2013
| Random memories