[Freeciv-Dev] Re: (PR#11828) patch: partial city update for what-if scenarios

["Vasco Alexandre da Silva Costa" <vasc@xxxxxxxxxxxxxx>]

<URL: http://bugs.freeciv.org/Ticket/Display.html?id=11828 >

On Mon, 10 Jan 2005, Jason Short wrote:

> <URL: http://bugs.freeciv.org/Ticket/Display.html?id=11828 >
>
> Vasco Alexandre da Silva Costa wrote:
>
> > Don't wait for an extension of the design to commit this patch if you
> > think it is good and makes things faster though.
>
> Well, according to Benoit's numbers it definitely makes things faster -
> which you'd expect, since it cuts out the biggest part of the refresh.
> The question is whether this is the best way to do it.
>
> Another way would be by caching the tile output values and using the
> cached values if a certain parameter is given to the refresh.  This
> patch does that.  It decreases the CM time with the 10203 savegame on
> Boston from 2200 ms to 840 ms (this is with profiling enabled, which may
> affect the measurement), at a cost of 150 bytes (5-10%) extra memory

Sounds good. Similar to the reported speedup from Benoit's patch. The
memory cost seems reasonable.

> used in the city struct.
>
> These two methods accomplish more or less the same thing (avoiding
> recalculation of tile output) but with very different mechanisms.  The
> cache patch also avoids recalculation of the bonus[] array, which can be
> done separately with the what-if patch.

---
Vasco Alexandre da Silva Costa @ Instituto Superior Tecnico, Lisboa