Complete.Org: Mailing Lists: Archives: freeciv-dev: June 2005:
[Freeciv-Dev] Re: (PR#11068) savegame file compression
Home

[Freeciv-Dev] Re: (PR#11068) savegame file compression

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
To: chrisk@xxxxxxxxx
Subject: [Freeciv-Dev] Re: (PR#11068) savegame file compression
From: "Vasco Alexandre da Silva Costa" <vasc@xxxxxxxxxxxxxx>
Date: Mon, 27 Jun 2005 00:47:53 -0700
Reply-to: bugs@xxxxxxxxxxx

<URL: http://bugs.freeciv.org/Ticket/Display.html?id=11068 >

On Sat, 25 Jun 2005, Jason Short wrote:

> <URL: http://bugs.freeciv.org/Ticket/Display.html?id=11068 >
>
> I ran a single test on an autogame with changed compresstype but default
> compresslevel and saveturns 1.
>
> Compresstype 0: 0m26.347s
> Compresstype 1: 0m28.336s (+1.989) (+7.5%)
> Compresstype 2: 0m30.919s (+4.572) (+17%)
>
> Conclusion: bzip -6 is about 2.25 slower than gzip -6.  This is
> consistent with Chris's findings earlier in the ticket.  However this
> isn't really a good measure I think, since with increasing computer
> speeds and slower freeciv AIs the overhead of compression will drop.

I wouldn't bet on near future CPUs getting much faster. Mostly wider or
more of them. In fact, it will probably be a good idea to revisit
multi-threading the AI sometime in the future, despite the fact it is a
PITA to debug.

> Maybe it would be better to look at these as a % fraction of the total
> runtime.  Note that this is with saveturns 1 so the % is 10x higher than
> the default would be.
>
> A 1.7% overhead is not too much I think.
>
> And the file sizes:
>
> -rw-r--r--  1 jdorje jdorje 289729 2005-06-25 12:44 civgame-2000.sav
> -rw-r--r--  1 jdorje jdorje  18116 2005-06-25 12:46 civgame-2000.sav.bz2
> -rw-r--r--  1 jdorje jdorje  29121 2005-06-25 12:45 civgame-2000.sav.gz

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







[Prev in Thread] Current Thread [Next in Thread]