← Back to team overview

yade-users team mailing list archive

Re: lack of documentation


Václav Šmilauer said:     (by the date of Sat, 4 Nov 2006 16:16:58 +0100)

> Hey, my usual program set is also term+mc+gvim!

wow, what a nice coincidence :) so I'll add few things:

I modified the alternate file plugin, so that I can switch
between .hpp c.pp and .pro files. This can be useful too:

inside the plugin's source code:

call <SID>AddAlternateExtensionMapping('hpp',"pro")
call <SID>AddAlternateExtensionMapping('pro',"cpp")
call <SID>AddAlternateExtensionMapping('cpp',"hpp")

Also I made a mistake, and written that I use :cnew to jump to
compilation errors in gvim. It is a command :cn , not :cnew :)
and I mapped it to alt-F8 (in ~/.vimrc)

:nnoremap <M-F8> :cn<CR>
:vnoremap <M-F8> <esc>:cn<cr>
:inoremap <M-F8> <esc>:cn<cr>

> I will be grateful for
> the guide for 4hedra, since I would like to have a "preview" of the
> model (=working programatically, but inaccurate physically) before
> Christmas, then to tune interaction laws etc. (Maybe I will try a
> simpler version in Python - in case it turned out that 4hedra are
> useless completely). By then, I will see you in Grenoble. 

An extremly simplified example is almost ready. I should finish it
today evening. In the mean time you can read:


(the bottom of this post)

Sorry, I should give you this URL yesterday, but I simply forgot
about it. I will put this link on the webpage in docs setcion. This
is better than nothing...

> BTW I don't think that game physics is necessarily less accurate. There
> is MUCH more money in the gaming industry (than in engineering
> simulation "business") and the scarce information available suggests
> that their algorithms follow rather closely latest advances in the
> research on collision detection etc. It is a pity that there is no
> open-source version of havok or at least some standardized API analogous
> to openGL but for physics. Not to mention dedicated hw physics processors
> (PhysX from ageia), that kind of thingie academia will hardly ever
> produce.

In this aspect you are right. But this game physics about which you
talk is simply not open. So we are left to packages like ODE or
bullet. And I know that ODE is inaccurrate, the best example here is
that the author of OSS game ultimatestunts was using ODE, but had
many problems with it, and finally stopped using it and written his
own formulas. Riding a stunt car required better physics handling
than "usually provided". With the exception of course of those
commercial (closed) game libraries....

But I know nothing about bullet.

A PPU is of course a different matter, since anybody could buy it,
and use it to calculate stuff. But how many people will buy it? Yade
could not depend solely on such hardware, otherwise only a vary
narrow range of people would be able to use it. However additional
support for such a beast would be nice :) But I don't own an PPU so I
can't write a support for it :>

# Janek Kozicki
Yade-users mailing list