Re: [Supertuxkart-devel] [GSOC2013] Race History Verifier

[Francisc B. <f.b...@ja...>]

Hello,

Thank you very much for your answer.
I've started looking into the details and another question came to my mind:
is it also of interest to keep track of which configuration element is
non-standard or on a first occurrence of a non-standard config element, the
submission is rejected?

Best regards,
Francisc


On Sun, Apr 14, 2013 at 3:45 PM, Joerg Henrichs <jo...@lu...> wrote:

> Hi,
>
> welcome to our team, and thanks for the impressive introduction you have
> written.
>
> >           This is my first time I am applying to Google Summer of Code
> > and I am looking forward to a positive experience developing open source
> > code for SuperTuxKart. Given my experience developing C++ code, I am
> > mainly interested in the "Race History Verifier" project. I have already
> > checked out the svn version of the project, compiled it from source code
> > and started looking into the "replay" feature. From there, I have
> Good, first steps done ;)
>
> > followed the code's "path" to the KartControl class, which contains the
> > acceleration, steering, braking, nitro information of the kart. I would
> > like to kindly ask you whether I am on the right track and also, whether
> > there is any place where "standard"/unmodified set of kart parameters is
> > set, so that it can be determined of a kart is in those parameters or
> > has modified ones.
> The default parameter (see karts/kart_properties.?pp) are from the file
> data/stk_config.xml, search for:
>    <general-kart-defaults>
>
> This whole section is the default which each kart gets. Mid term (when
> we have different kart physics parameters) kart might overwrite those
> values (that code is already in place), but for now all karts are using
> those parameters (with the exception of the shape of the collision
> object, which depends on the size of the model and differs from kart to
> kart).
>
>
>
> >
> >           In addition, I would like to ask if the idea of the project is
> > to actually only allow to have online highscores for "stock" karts, or
> > also implement a scheme that would map to "real" values of the
> > highscores, depending on the weight of the modifications?
> By now there are actually slightly different ideas, based on some
> discussions (in an unrelated proposal actually): main idea is that for
> an online highscore you submit a replay file (more compressed than they
> currently are). Those results could then be tested (with the result of
> the verifier project), and invalid ones would be rejected.
>
> There was also some talk to perhaps use this in network play (long term
> goal though), to detect if one of your opponent has what appears to have
> hacked his version ;)
>
> Cheers,
>     Joerg
>
>
>
> ------------------------------------------------------------------------------
> Precog is a next-generation analytics platform capable of advanced
> analytics on semi-structured data. The platform includes APIs for building
> apps and a phenomenal toolset for data science. Developers can use
> our toolset for easy data analysis & visualization. Get a free account!
> http://www2.precog.com/precogplatform/slashdotnewsletter
> _______________________________________________
> Supertuxkart-devel mailing list
> Sup...@li...
> https://lists.sourceforge.net/lists/listinfo/supertuxkart-devel
>