Cycling power simulator due to GPS data, No need power meters but a GPS logger --

The power-meter has been bringing forth bounty, optimizing training, best conditioning for a competition, optimizing equipment assemble, mathematical knowledge, exercise physiology knowledge and so on. It might be a large difference for users even who are not competitors but just fun-riders or health-conscious riders. Although it might be different with each person to buy a power-meter costed thousands dollars for that.

It's possible to get the power-data almost same as from the power-meter. Due to the software and a GPS logger..

Of course there is a condition. That's preparable arrangement of high accuracy GPS data of cycling. i.e. it's just need a high accuracy GPS logger.
It's desirable that the logger has a high-end chip. A GPS logger could be directly affected by cycling environments -- road width, building coverage, floor space index, roadside trees, satellite constellation by chance, weather, solar flare, and so on.
As for recent high-end chips, There's no need to worry about the geo-coordinates cause which accuracy is very high, but the altitude has been vagabond. It hinges on the chip though, the accuracy of error could be several meters in case of changing number of tracking satelites or the multipass under cluster of high-rise buildings. If the error had been took seriously, power prediction should be impossible.

However, the road is smooth, no doubt it has smooth undulation.
It's only necessary to exclude outliers. It's only necessary to smooth unreal oscillations. That's none other than the real state of the road. And that's the strong point of computing. Of course the data accuracy should be high enough to presume the real state.

Target of prediction is the power. It's also able to predict the cadence though, cadence prediction is one kind of an ideal value due to profile properties.
As for the power prediction, three factors -- speed, acceleration, inclination -- have a profound effect on. Next, air density and wind velocity, road surface and so on have an effect on, which are easy set as real values. Of course human body and equipment are premised on veracious and real values.
The three factors -- speed, acceleration, inclination -- are calculated due to the GPS data, and so I touched about accuracy of GPS logger at this outset.

In the time, a GPS logger within several centimeters accuracy could be get at reasonable cost. If it does, perfect power data could be get by the GPS logger without smoothing.
There is already a map which performs corrections about altitude data. It's worth thinking about usage. It's true, there might be many GPS loggers enough to be used because it's not necessary to be exact altitude but to be exact altitude change.

Please use the software at first.
The software imports a fit file, predicts power and cadence and exports a fit file with predictions. And please try to analyse the exporting fit files by the Golden Cheetah. It's also possible that fit files by cycle computers(e.g. Garmin) without GPS data can be analysed and be predicted as limited to flat course condition.

If the file format of a GPS logger in hand is '.GPX' or '.KML' it's only necessary to convert them to the fit file format by any converter. Incidentally, GPSBabel is famous as a convertor though it can not export to the fit file format.


Other usage, the software has a method. That's a comparison of power predictions between real equipment and suppositional equipment in case of a same cycling records data due to exchange the profile settings.
For an example, that comparison brings mathematical aspects about results of weight saving and aerodynamic improvement, how to optimize a particular coure and cycling. i.e. comparison data of equipment assembles.


Ver.0.16.00 and later, it's able to detect braking at every records, and be able to express braking force(kgf), temperature(Celsius) of brake materials, friction heat(Jule) and radiate heat(Jule) for both of front and rear brakes. Due to setting of "Preference" dialog > "GPS data" tab. In order to get correct prediction values, it's required the GPS logger to be very high accuracy.

Ver.0.20.00 and later, the e-Assist (electric bicycle) function can be set to Profile, and be able to analyse and predict for Simulation (Power and Velocity, Acceleration, Deceleration) and GPS (Analyse and Predict, Create FIT File Analysed).
Profile and Condition are now separated. If you are using ver.0.20.00 or later, you will need to create a new profile file.


powermeter for Android has been developed. It's an application for Android which expresses and records power and other measurements, creates a FIT file, due to GPS function of a smart phone. This is a tutorial page. it will be uploaded to the Google Play in the time.

In a high level, creating puzzles, solving, candidates analysis, hand-build, save and load, and so on --


The software is multi-purpose, easily handled and forceful Number Place software in guessing. Please try it at first.

numberplace for Android has been developed. It's an application for Android which has same functions as this Windows version. Puzzle data files are also compatible.. This is a tutorial page. it will be uploaded to the Google Play in the time.

The menu has been simplified since ver. 0.23.00 for Windows, and the menu structure of the Android version follows this.

Good intention and malicious intention on the social --

A rational strategy of the prisoner's dilemma is an open-and-shut case. And that had been giving lots of researchers gloom.
Due to folk theorem it's been known that "Cooperation" is the key to create a world of wealth. However, that seems to be realized under purposely designed condition in the real world. For an example, competitor companies realize goal through collaboration in order to get a larger market share and a larger circulation.

"A rotten apple spoils the barrel" as the saying goes, that's used for the human society though, is it always the truth? If it's not, what factor could be effective?

The famous tournament of Robert M. Axelrod polarized opinion.
It seems determinate criticism that the result depends on the initial values. However, Axelrod is a pioneer. Binmore recognized him for creating awareness it's important to select a specified equilibrium among countless equilibriums.



The software introduced at this page is a simulator of the iterated prisonner's dilemma.

Sixteen strategy patterns due to four bits codes do the arms race. Sixteen ones was not arbitrary selection but systematic sampling.

It has couple of observation rules which are reference data for the strategies.
As we will discuss in detail later, each individual player could observe opponent's forpassed cards or own memory of opponents' cards, or could evaluate opponent's strategy codes, opponent's stats of all forepassed cards, opponent's stats of ethical evaluations of all forepassed cards. And each player would decide the card cooperation or denial due to own strategy pattern.
The observation rule is an abilitiy of each individual.

It adopts a geometric rule in order to meet and play with an opponent in random order. Each individual does game with a neighbor opponent selected in random. In case not founded, an individual would move to a random toward. And in case of both sides have a certain vital level they would do cross-reproduction.
In the initial phase, cause the world resource is affluence and plentiful, individuals would get bonus points just due to game and would procreate one after another. In this initial phase a random chance would make a large influence.

It's impossible to predict what equilibrium would be selected. It seems to be the sport of slight haphazards on such a systematic and superficial simulator. However the direction of the equilibrium selected would be exposed to view due what ability(the observation rule) would be given.

For ver.0.86, some additional rules were added. Those were supposed as metaphars of the social world. In each case, it's possible to choice or not.
One is dealing with between two of a kind. Individuals could do game or crossing in case of they have similarity of their genes, due to set a percentage of same bits of the gene.
Another is refusing to deal with, the rich refuses to deal with the weak. Individuals could not do game nor crossing in case of a large disparty, due to set the ratio of vital scores. However, an individual could copy the strategy(gene) of the rich in case of the ratio is in among certain ratios. That's the imitative learning. A reason of limitation of ratios is that the influence is few in case of too much disparty.

For ver.0.87 one observation rule was added. It's evaluation of opponent's strategy codes which is named Rule-X preseted. It's just like an observational ability of opponent's real nature.
As for other observation rules which observe few forpassed cards, those do not include informations of opponents of that moment, it's impossible to know opponent's real nature.
In concrete, it just evaluates opponent's strategy codes which is also opponent's sign expression though, as a social simulation, it might be a kind of the inspiration.

For ver.0.88 further two observation rules were added. One Rule-Y refers opponent's stats of all forepassed cards. Another Rule-Z refers opponent's stats of ethical evaluations of all forpassed cards.
Rule-Y observes the records of opponent's count of forepassed C-cards and count of forepassed D-cards, and evaluates the ratio of those ratio. That's supposed to be an bureaucratic record.
Rule-Z observes the records of opponent's count of ethical behaviors and count of unethical behaviors, and evaluates the ratio of those ratio. That's supposed to be an intuition, although I wonder what metaphor is such an insight.

What ability should be given for what equilibrium could be emerged? -- That's the keynote of the simulation and let us know how to emerge a wealth of the world.

Numerical / Mathematical Rhythm --

There are roughly two types rhythms in mathematically though, for music only one of them is used.
Two types, those are crystal-structural(periodical) rhythm and quasicrystal-structural rhythm. Quasicrystal Structure is also called aperiodic structure though, that's not random structure but ordered and more evolutional structure in some aspects.

Musical rhythms are two beats and three beats, four beats, eight beats, and so on, which are meaning iteration of simple structural unit.
Quasicrystal structural rhythm in a way of saying is a compounding of two beats and three beats. However it's not iteration of simple units just like the compound time as a musical terminology.
Musical structure is based on simple intenger two and three and so on. Quasicrystal structure is based on the Golden Mean.
-- described below



In this page, I would introduce a software which could generate numerical / mathematical rhythms due to set a numeral.

Rhythms are composed with two tones alfa and beta. These two tones make one layer, and eight layers as maximum could be structured. Each layer and multi-layered structure are compliant with the numeral as a concept. Due to two tones and eight layers the rhythm structure would be created, which has sixteen tones(channels) as maximum.
It uses MIDI as a sound source. Software synthesizers of an operating system and utilities of a sound board, external MIDI devices could be selected. Tone and Note, Velocity of each channel(sixteen tones) individually could be setup as MIDI parameters. And, layer count and tempo could be exchange.

For design the tile picture --

Tiling a picture falls roughly into two approaches.
One method is making up a wall painting due to paste general purpose tiles set. Another is making up a wall painting due to paste specified tiles which are already designed as parts of the picture.
Latter is more, and much coloring after pasting.

Icons in medieval Europe and geometric abstractive Arabian tilings take a latter approach. Icon was painted a subject of the Bible scene on one tile. Arabian tiling constructed abstractive pattern due to combine specified tiles already painted specified patterns.
In other words those treated tiles as a canvas, just like a jigsaw puzzle, each tile was specialized as a part of the picture.
As in Japan, talking of a tile picture, that's the Mt.Fuji which are painted on a wall of hot springs and public bathes. Those are almost coloring after pasting tiles.

A right picture is decorated on a wall of a hot spring in Atami. It has a nice ordonnance, water-wheel as near distance, a row of mountains as middle distance and Mt. Fuji as long distance. It's constructed with general-purposed tiles, which method is rare. Incidentally, the right picture is not a painting but a photograph.
This recipe is constructing a tile picture with general-purpose tiles set.

The software introduced in this page is also constructing a tile picture with general purpose tiles set.

Anything will do as an original picture, a photograph and an illustration and so on. The software creates a tile picture of those due to any color palette.
There is a lot of flexibility for setup the color palette. It could pick out any colors which express the original picture faithfully, and could apply an arbitrary color palette made by a user. Joint layout could be expressed.
In a word, it's a simulator of the tiling.

A left picture is a tiling of the Mt. Fuji painted by Hokusai, used the safe color as its color palette.

A color palette less color used sometimes makes few noisy tiles which creates an interesting tiling occasionally.

And the software could express a design data which are tile size and count of horizontal and vertical, color indexes of all tiles, RGB values of all tiles.
It also output a csv file of the design data, which makes any aggregate calculation easy by a spread sheet application.

A multi-variated analysis software borned in the spot of marketting.

It's also a general-purpose analysis tool.
Starting convenience tool for marketting with a certain intention. Concretely convenience for deep analysing individual situations of each sample. And good output for a presentation.
The software is created by such demands.

It has twenty four models and some methods in each model which is plentiful in guessing.
And has some functions which are rarely among general tools. For examples, an identification of divergence rate due to significance of the Maharanobis distance, a range predicted due to significance, expression vectors of variables into a projection plane with factors mapping.

Bicycle Power Simulator, the power-meter brought so much --

There has always been scientific views though, in level of users especially of roadbike, they have been told somesthetic and irresponsible things. For an example, speed on flat would be exchanged few km/h due to increase or decrease one kg of bike weight unjustifiedly without speed level.
And with a prevalence of power-meter, such idiotic assertions have been rided out.

Due to the power-meter, it could be verified theory values, could analyse statistics and predict due to data accumulated, could analyse quantity not only of dynamism but also of aerodynamism.
For an example, how much dynamic loss of equipment could be known due to difference between power of pedaling and crank, rear hub, how much gain of aero-helmet, how much diffence between sisxty spokes and twenty spokes, and so on.

The software is created on such a basis of the quantity analysis.
Due to a human body and abilities, specifications of equipment, environments of a course, the software simulates power and torque, force, accel, time, brake material heating in relation to constant velocity and acceleration, deceleration, braking.

Current version simulates based on a certain constant factor though, due to select so much factors it makes possible to simulate dynamic transitions, acceleration and deceleration, braking and so on. Also, it has enough accuracy to be reflected on precise setting.
At first, manipulate the software please.

This Spring in 2019, I've done throughly refactoring and mainly made settings of parameters to be easy understanding, raised count of gear(chainrings, sprockets) as settings which made it possible to set gear as a pisto(single gear).

The purpose of refactoring is a preparing for next update that makes it possible to predict the power by a GPS data and analyse due to exercise physiology. Specifically, importing a FIT file without power data, exporting a FIT file with power data which could be analysed by a physiology analysis software e.g. Golden Cheetah. This update could be finished till this summer in 2019.
Accuracy of a prediction data could not come up to a data exported by a power-meter though, there could been possibilities by the software, e.g. Comparative evaluation about asseble of equipment on a base GPS data recorded running.

please post any impressions you may have

e_mail to webmaster