[Freeciv-Dev] Re: (PR#4376) Slow Motion

["Raimar Falke" <rf13@xxxxxxxxxxxxxxxxx>]

On Thu, Jun 12, 2003 at 11:48:27AM -0700, ChrisK@xxxxxxxx wrote:
> On Thu, Jun 12, 2003 at 06:23:22AM -0700, Raimar Falke wrote:
> > On Thu, Jun 12, 2003 at 04:31:40AM -0700, ChrisK@xxxxxxxx wrote:
> > > On Wed, Jun 11, 2003 at 01:51:00AM -0700, Raimar Falke wrote:
> > > > On Wed, Jun 11, 2003 at 01:11:28AM -0700, ChrisK@xxxxxxxx wrote:
> > > 
> > > > > > > Maybe an interaction with the CMA?
> > > 
> > > > > I think the cma should not be involved as long as the goto is 
> > > > > processed,
> > > > > only after it is finished.
> > > 
> > > > I wasn't able to reproduce this problem here. So please apply the
> > > > patch and post the relevant output.
> > > 
> > > Here are 3 examples:
> > 
> > These are the expected results. In all cases the cities are updated
> > only once after the goto is finished. There should be a delay after
> > the goto is finished but not during the move movement. At least this
> > slowdown isn't caused by the CMA.
> 
> I noticed that the delay is not always present ...
> 
> Now things are getting funny :-) I tried to reproduce it, but failed, until
> I just started to *play*. So here's what I did:
> 
> Start server with attached game
> set saveturns 0
> start client, connect as chris
> start game
> Use GOTO in all actions below
> send cruiser in Kyoto on goto to the open sea
> send all alpine troops from Hiroshima to Yokohama, sentry them
> send transport from yokohama to Ahmadabad
> unload and send alpine troops to bombay, idle them
> make engineers build roads where they are
> send transport back to hiroshima with goto
> buy 8 alpine troops
> turn done
> send alpine troops from bombay to rawalpindi. Now(!) you see a significant
> delay in unit movement when they go through cities, if your fast machines
> allow this.

I don't see a slowdown here and the CMA shows normal behavior:

2: cm_query_result(city='Bombay'(113))
2: cm_query_result: return

open city dialog.

2: A: freeze() current level=0
2: start processing packet 110
2: unit 249 moves from (59,29) to (60,29)
2: unit 249 moves from (60,29) to (61,29)
2: unit 249 moves from (61,29) to (62,29)
2: unit 249 moves from (62,29) to (63,28)
2: unit 249 moves from (63,28) to (63,27)
2: unit 249 moves from (63,27) to (63,26)
2: unit 249 moves from (63,26) to (64,25)
2: unit 249 moves from (64,25) to (65,24)
2: unit 249 moves from (65,24) to (64,23)
2: unit 249 moves from (64,23) to (63,22)
2: unit 249 moves from (63,22) to (62,23)
2: unit 249 moves from (62,23) to (61,24)
2: unit 249 moves from (61,24) to (60,25)
2: unit 249 moves from (60,25) to (59,26)
2: unit 249 moves from (59,26) to (58,26)
2: unit 249 moves from (58,26) to (57,27)
2: unit 249 moves from (57,27) to (56,27)
2: finish processing packet 110

unit moves

2: A: thaw() current level=1
2: cm_query_result(city='Rawalpindi'(208))
2: cm_query_result: return
2: cm_query_result(city='Madurai'(235))
2: cm_query_result: return
2: cm_query_result(city='Karachi'(145))
2: cm_query_result: return
2: cm_query_result(city='Chandrigarh'(230))
2: cm_query_result: return
2: cm_query_result(city='Kanpur'(173))
2: cm_query_result: return
2: cm_query_result(city='Bombay'(113))
2: cm_query_result: return
2: cm_query_result(city='Kumamoto'(211))
2: cm_query_result: return

the cities the unit "touched" are recalculated

> civclient debug log shows no difference for me.

Difference to what?

        Raimar

-- 
 email: rf13@xxxxxxxxxxxxxxxxx
  One nuclear bomb can ruin your whole day.