yade-dev team mailing list archive

[Anton Gladky <gladky.anton@xxxxxxxxx>]

Hi, Vaclav,

anyway, I would like to leave the implemented algorithm for the moment, let
me explain why.
As you know, I usually modeling machines. And sometimes I have special
"sheets of facets", which
cover the machine. I need to put those facets in separate file, because I
need to see, what happens inside
the machine. So, I will have several VTKRecorders in one simulation,
something like "Roller-", "Cover-", "RockMaterial-"....
Later, when I do postprocessing, I can switch off "cover-" to check what is
going on inside the machine, or I can switch it on to
show the general view of the machine.

That is why I implemented the feature.

If we use a bitmask, FOREACH will pass through all bodies, checking, whether
it will output to vtk-file or not.
If I have 50000 bodies and 4 VTKRecorders, I will have 50000*4 = 200 000
iterations instead of 50 000.

Also we have a subscribedBodies algorithm in some other engines and this is
a very clear. If we use mask instead, it will be a little bit
complicated to understand, how it is working (even for me, several month
later).

So, I prefer to stay this algorithm, even it is a little ugly.

What do you think?
______________________________

Anton Gladkyy


2010/5/13 Anton Gladky <gladky.anton@xxxxxxxxx>

> I found some comments in GroupRelationData.hpp
>
> Ok, sorry, I understood already, how it should work.
> Thank you.
> ______________________________
>
> Anton Gladkyy
>
>
> 2010/5/13 Václav Šmilauer <eudoxos@xxxxxxxx>
>
>> > I'm just afraid, will it not conflict with Vincent's  predefined
>> > masks?
>> Which ones? I have no idea about Vincent's masks now.
>>
>>
>>
>> _______________________________________________
>> Mailing list: https://launchpad.net/~yade-dev<https://launchpad.net/%7Eyade-dev>
>> Post to     : yade-dev@xxxxxxxxxxxxxxxxxxx
>> Unsubscribe : https://launchpad.net/~yade-dev<https://launchpad.net/%7Eyade-dev>
>> More help   : https://help.launchpad.net/ListHelp
>>
>
>