Complete.Org: Mailing Lists: Archives: freeciv-dev: December 2001:
[Freeciv-Dev] Re: Several bugs discovered when playing large civilisatio
Home

[Freeciv-Dev] Re: Several bugs discovered when playing large civilisatio

[Top] [All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index] [Thread Index]
To: freeciv-dev@xxxxxxxxxxx
Cc: bugs@xxxxxxxxxxxxxxxxxxx
Subject: [Freeciv-Dev] Re: Several bugs discovered when playing large civilisations (PR#1089)
From: Vasco Alexandre Da Silva Costa <vasc@xxxxxxxxxxxxxx>
Date: Tue, 4 Dec 2001 19:03:49 -0800 (PST)

On Wed, 5 Dec 2001, <s.yeates@xxxxxxxxxxxxxxxx> wrote:

> I enjoy playing large civilisations, but have found a number of bugs:
> 
> * Some mass operations (such as selling the same improvement in 700+ cities
> at once via the economy report) cause the client to crash. The message on

Question: crash or get so slow it doesn't even refresh?

> the server is: "cut connection say1 from localhost (connection
> incomplete) due to huge send buffer (2)". Some of the requested actions
> are performed, but not all.

Well, what that means is that the client isn't responding to the server
fast enough so the server send buffer gets full. Since the client isn't
responding, the server cuts his connection. Which is nice. Better than
like it was before the network changes where it would lockup the server
for 30 minutes.

> * When the client is restarted after such a crash, moves are MUCH faster for
> the first few moves. Is this a bug?

Well, i don't see how moving faster can be a bug :-)
I thought you wanted more performance?
Its probably an interaction from the Linux kernel scheduler, X, or something.

> * When a spy tries to subvert a capital city it names a price (1000000 gold),
>  but if you save that much money the city will not subvert.

Integer overflow maybe? I dunno if we are using short ints for that.

> * When many (hundreds) of cities are selected in the cities report, changing
> their production takes an inordinately long time (the display is updated
> between each?)

I believe you are correct on your assessment.

> I know these bugs probably don't impact on most users games, and are low
> priority.

Well, i have been planning to add slightly better networking to the
client, but i've decided to postpone it until after GTK+ 2.0 comes out.
There is a much worse bug in the client than the ones you mentioned, but i
will save those for after GTK+ 2.0 too.

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




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