Elliptical Curves, Perceptual Matrices

[yemioyehotmail_com]

hello my maths is not so good which of the sample H files can i use to improve my intergration and differentiationfuntionthreads vml?, im aiming for a perceptual matrix alg which could be improved by vector maths as well as concurrency pls what functions call ABI's for multilanguage code
yemi oyekan

[Ying_H_Intel]

Hello Yemi oyekan,

Youmay take a look at MKL manual first
for example,mklman.pdf=>chapter 13, Partial Differential Equations Support, where mkl provide Trigonometric Transform function and Poisson Library interfaceto solve patial differential equation. There are some math formulas on it. You may see if they are same as your alg.
If same, you can call the functions.
There are C source code sample under mkl install directory.
C:Program FilesIntelMKL10.2.1.019examplespdettcsource

If they are not you need and you can only use functions in VML, like
vssin() to compute vector's sin value. it is easy to call
in C language,

float fA[VEC_LEN],fB1[VEC_LEN],fB2[VEC_LEN];
MKL_INT i=0,vec_len=VEC_LEN;
vsSin(vec_len,fA,fB1);

You can refer tomore sample codes under examplesvmlcsource

Regards,
Ying

[yemioyehotmail_com]

Quoting - Ying Hu (Intel)

Hello Yemi oyekan,

Youmay take a look at MKL manual first
for example,mklman.pdf=>chapter 13, Partial Differential Equations Support, where mkl provide Trigonometric Transform function and Poisson Library interfaceto solve patial differential equation. There are some math formulas on it. You may see if they are same as your alg.
If same, you can call the functions.
There are C source code sample under mkl install directory.
C:Program FilesIntelMKL10.2.1.019examplespdettcsource

If they are not you need and you can only use functions in VML, like
vssin() to compute vector's sin value. it is easy to call
in C language,

float fA[VEC_LEN],fB1[VEC_LEN],fB2[VEC_LEN];
MKL_INT i=0,vec_len=VEC_LEN;
vsSin(vec_len,fA,fB1);

You can refer tomore sample codes under examplesvmlcsource

Regards,
Ying

thanks ying
the header files were very usefull,need to brushup my c# got too used to java strong typing,my c# logic too weak,i hope to get help with the concurrency issues below if mutexes,deadlocks and semaphores can help see beyond the 2 player game matrix/decision theory human limit

yemi


perceptual matrices

matrices are usually in grid forms but the extensive forms are usually beyond human analytic capacities as such beyond human perception, as well game theory states game analysis beyond 2 player games are difficult in matrix forms but in extensive form would this not be illogical to the non mathematical bent, in medicine mental illness is any thing beyond the normal of human experience thus my argument is that perceptual fields can be enhanced by human methods but a limit exists based on limits to human neural computational capacity as opposed to digital computational capacities which are limitless, as such seeing beyond perceptual horizons lies beyond our limited capacities, even computational capacities have concurrency problems beyond which glitches, freezing of resources and other errors limit computational aims.

Consider this extensive form matrix

figure1 (paraphrased from mathematical psychology teversky 1970,prentice hall page 242)

P1=side a, disarms= Dc1, arms= C1.

P2=side b, disarms= DC2, arms= C2.

P1 choosing Dc1,minimises loss and maximizes loses potentially/strategically preempting P2 choices, recall in a 2 player non-informative matrix game, p1 plays first thus chaining P2 strategic choices to his first choice,P1 choosing C1,maximises gain and equalizes losses.

[=a conditioning for the next players choice/state/insight.

P2 choosing Dc2 minimizes losses, choosing C2 maximizes gain recall p2's insight has been chained to the upper row,P2 choosing PC2 maximizes losses, choosing C2 equalizes losses again chained to the lower row by P1's earlier choice.

The best choice for individual players in the above example is [stated in the text quoted] may not be to collective gain eventually, which is the experimenters original strategy of hiding perceptual information ,and deciding who is more reasonable or levelheaded as to reason for which ever bias he has chosen his insight state to hinge on, I .e the learning state.

Using Markov chain theory as an hypothesis for the learning theory, using a matrix each cell is a state S1 ,with trials and errors continuing as a sequential process until the right state occurs S2,insightfull learning, or conditioned state where the right choice has been made.

(xn=Sn:Sn-1=S1) trial and error state.

S2=learning, correct response, conditioning.

thus S1 presolution state/unconditioned state,S2 postsolution state/conditioned state xteristic of a all or non learning model represented as a two state markov process.

Thus perceptual matrices may be a learning process, hidden from human cognition by limitations to analytic capabilities unable to compute beyond certain limits of perceptual capabilities.

But concurrency may with queue manipulation by semaphores help in coordinating 100's of processes may make the study of multiplayer, multistrategy games impossible for humans to analyze become better coordinated with less errors thus hopefully by the same ratio improve human perceptual matrices or cognitive abilities.

i refer the reader to my IPP forum code samples on auditory perception to help correct this impression of mine if wrong.

OYEKAN Adeyemi, Olufemi Dr