[Boost][GSoC]New algorithm for multiplication

classic Classic list List threaded Threaded
3 messages Options
Reply | Threaded
Open this post in threaded view
|

[Boost][GSoC]New algorithm for multiplication

Boost - Dev mailing list
Sir/Ma'am,
I am highly interested in working on the multiprecision big number types .
I wanted to propose that since In the proposed project there is an increase
in precision from 1000 to 10000 decimal digit , we can implement toom-cook
algorithm instead of the standard karatsuba multiplication.Toom cook
algorithm has a complexity of O(n^1.465) which is quite faster than
karatsuba multiplication O(n^1.58). Please share your views on this.

Thank you

_______________________________________________
Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
Reply | Threaded
Open this post in threaded view
|

Re: [Boost][GSoC]New algorithm for multiplication

Boost - Dev mailing list
 Dear Debajyoti,
Yes. Toom-Cook is one part of the potential plans,in addition to other issues mentioned in the proposal.

Thank you for your interest. There is a process for this.
First Boost must apply for acceptance to the program.
After that, we begin accepting applications for theprojects.

The application process is competitivebecause it is a world-open competition. We canoffer some tips during this competition, butit's mostly up to you to write a good applicationthat realistically reflects project awareness andinterest in and/or similar kinds of experiencein the domain of the project.
The GSoC is for students.
So you are not expected to be a world expert withyears and years of experience. But project awareness,
having basic knowledge about the fundamental technicalissues relevant to the domain, all helps.
For the projects in Multiprecision, I will be directlyinvolved with help from colleagues much more experiencedthan I am. So i will usually be on the board and be ableto answer questions if they are fair in view of thecompetition.
The action tends to really get going toward theend of February if memory recalls.

Good luck.

Kind regards, Chris

    On Wednesday, January 22, 2020, 7:07:27 PM GMT+1, Debajyoti Dasgupta via Boost <[hidden email]> wrote:  
 
 Sir/Ma'am,
I am highly interested in working on the multiprecision big number types .
I wanted to propose that since In the proposed project there is an increase
in precision from 1000 to 10000 decimal digit , we can implement toom-cook
algorithm instead of the standard karatsuba multiplication.Toom cook
algorithm has a complexity of O(n^1.465) which is quite faster than
karatsuba multiplication O(n^1.58). Please share your views on this.

Thank you

_______________________________________________
Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
 

_______________________________________________
Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost
Reply | Threaded
Open this post in threaded view
|

Re: [Boost][GSoC]New algorithm for multiplication

Boost - Dev mailing list
Thank you for the information sir, I am highly interested in this project
and looking forward to work with this project as I believe completing this
project  will have a great impact in many scientific areas and
researches too and I would be very happy to contribute to it.

On Thu, Jan 23, 2020 at 2:59 AM Christopher Kormanyos <[hidden email]>
wrote:

> Dear Debajyoti,
>
> Yes. Toom-Cook is one part of the potential plans,
> in addition to other issues mentioned in the proposal.
>
> Thank you for your interest. There is a process for this.
>
> First Boost must apply for acceptance to the program.
>
> After that, we begin accepting applications for the
> projects.
>
> The application process is competitive
> because it is a world-open competition. We can
> offer some tips during this competition, but
> it's mostly up to you to write a good application
> that realistically reflects project awareness and
> interest in and/or similar kinds of experience
> in the domain of the project.
>
> The GSoC is for students.
>
> So you are not expected to be a world expert with
> years and years of experience. But project awareness,
> having basic knowledge about the fundamental technical
> issues relevant to the domain, all helps.
>
> For the projects in Multiprecision, I will be directly
> involved with help from colleagues much more experienced
> than I am. So i will usually be on the board and be able
> to answer questions if they are fair in view of the
> competition.
>
> The action tends to really get going toward the
> end of February if memory recalls.
>
> Good luck.
>
> Kind regards, Chris
>
>
> On Wednesday, January 22, 2020, 7:07:27 PM GMT+1, Debajyoti Dasgupta via
> Boost <[hidden email]> wrote:
>
>
> Sir/Ma'am,
> I am highly interested in working on the multiprecision big number types .
> I wanted to propose that since In the proposed project there is an increase
> in precision from 1000 to 10000 decimal digit , we can implement toom-cook
> algorithm instead of the standard karatsuba multiplication.Toom cook
> algorithm has a complexity of O(n^1.465) which is quite faster than
> karatsuba multiplication O(n^1.58). Please share your views on this.
>
> Thank you
>
> _______________________________________________
> Unsubscribe & other changes:
> http://lists.boost.org/mailman/listinfo.cgi/boost
>


--
*Debajyoti Dasgupta*
*Student ID: 18CS30051*
*Second Year Undergraduate Student*

*Department Of Computer Science And Engineering*
*Indian Institute of Technology Kharagpur*
*Phone : 8327406555*
*Facebook <https://www.facebook.com/debajyoti.dasgupta.6> | LinkedIn
<https://www.linkedin.com/in/debajyoti-dasgupta-33713a179/>*

_______________________________________________
Unsubscribe & other changes: http://lists.boost.org/mailman/listinfo.cgi/boost