3.1.0 has been released!

[Zi-Chan]

Finally. The long-awaited next stable. Good work everyone! I'm happy that this game didn't die when Pumpkin Studios was bought. And i'm sure Open Source will someday rescue the whole world.

Too bad that other games with unique principals more or less died (more or less, because they are still played, but since there is no source no one can continue developing the game mechanics themselves)... E.g. Populous 3, Dungeon Keeper 2...

[Originway]

By popular request, we have built portable versions of Warzone 2100 3.1.0.

I was wondering can you guys add just the program file from buildbot.wz2100.net/files/master/ builds?
lots of people can't download the whole thing everyday because of data caps.

[Vancer2]

Yay!!! New update. The New AI is great. But I miss the old AI sliders :[

[adosmith]

mine keeps crashing 2 minutes in. It also did this in the previous release. Any ideas?

[vexed]

mine keeps crashing 2 minutes in. It also did this in the previous release. Any ideas?

What hardware you got ?
What does the Warzone2100.RPT file say ?

[ramon_grigio]

Nice Work.

But i noticed slowdowns while playing with various Ai's at the same time. I was playing the 10 players map with all Ai's and after a few minutes thing really slowed down.
I checked the cpu/gpu use and none of them was near 100% (4 cpu cores were being stressed somewhat).

My config: A8 3520M (4c 1.6Ghz) HD 6620G (400c @444mhz 128bit) 6Gb ddr3 @ 1333mhz. Windows 7 64bit (up to date).

Tested with diferent video drivers (since gpu wasn't being stressed, i think it's not the problem).

Am i lacking cpu power (task manager didn't show 100% use on any core) or it is the game engine having problems with the Ai's?


Playing smaller maps with less Ai's and campaing went just fine.

Thanks, loved to see the truck rush of the ai all over the map ;D

[Rommel]

14,549 Downloads (This Week) - Saw this on the SF summary page... man that is hammering, looking forward to lots of players in MP!

[mcdebugger]

Russian announce

[PkK]

Just did my first quick game with this version. The game itself went great. But when in the end I clicked "quit" I just got a black screen, and had to kill -9 (normal kill wouldn't do). I also got lots of these (with the player number and type number varying):

info |09:33:02: [structSetManufacture:993] Wrong template for player 3 factory, type 1.
info |09:33:02: [structSetManufacture:993] Assert in Warzone: structure.cpp:993 (!psTempl || (validTemplateForFactory(psTempl, psStruct) && researchedTemplate(psTempl, psStruct->player, true)) || psStruct->player == scavengerPlayer() || !bMultiPlayer), last script event: 'N/A'

Philipp

[vexed]

Just did my first quick game with this version. The game itself went great. But when in the end I clicked "quit" I just got a black screen, and had to kill -9 (normal kill wouldn't do). I also got lots of these (with the player number and type number varying):

info |09:33:02: [structSetManufacture:993] Wrong template for player 3 factory, type 1.
info |09:33:02: [structSetManufacture:993] Assert in Warzone: structure.cpp:993 (!psTempl || (validTemplateForFactory(psTempl, psStruct) && researchedTemplate(psTempl, psStruct->player, true)) || psStruct->player == scavengerPlayer() || !bMultiPlayer), last script event: 'N/A'

Philipp

What hardware do you have ?
Having a black screen usually means bad video drivers...

[NoQ]

lots of these

These are insignificant troubles with Nexus/Semperfi AI; earlier it used to cheat with templates (produce things that were not yet invented, mostly repair tanks), now it can't. It is unrelated to the video issue.

I was playing the 10 players map with all Ai's and after a few minutes thing really slowed down.

NullBot is the reason; AIs do all their calculations in a single game frame, and his calculations are a bit more advanced, so the game has to wait until he cycles through all units on the board etc. This causes the game to have singlular lags even though the average CPU usage is low. It has a lot of optimisations, but still not perfect, so better use it in smaller games (even in 1x1, on hard difficulty, it can be of interest for newer players). Also, warzone is not multi-threaded (apart from pathfinding in a separate thread), so you are actually running most of the stuff in a single 1.6ghz core. The old DyDo AI used to slow things down quite a bit too.

[PkK]

Just did my first quick game with this version. The game itself went great. But when in the end I clicked "quit" I just got a black screen, and had to kill -9 (normal kill wouldn't do). I also got lots of these (with the player number and type number varying):

info |09:33:02: [structSetManufacture:993] Wrong template for player 3 factory, type 1.
info |09:33:02: [structSetManufacture:993] Assert in Warzone: structure.cpp:993 (!psTempl || (validTemplateForFactory(psTempl, psStruct) && researchedTemplate(psTempl, psStruct->player, true)) || psStruct->player == scavengerPlayer() || !bMultiPlayer), last script event: 'N/A'

Philipp

What hardware do you have ?
Having a black screen usually means bad video drivers...

glxinfo:

Code: Select all

OpenGL renderer string: Mesa DRI Intel(R) 965GM 
OpenGL version string: 2.1 Mesa 8.0.5

cat /proc/version:

Code: Select all

Linux version 3.7.1 (root@notebook3) (gcc version 4.7.2 (Debian 4.7.2-4) ) #1 SMP Thu Dec 27 13:02:26 CET 2012

I'll play a few more games and see if the issue reappears.

Philipp

[PkK]

lots of these

These are insignificant troubles with Nexus/Semperfi AI; earlier it used to cheat with templates (produce things that were not yet invented, mostly repair tanks), now it can't.

Which AI would you recommend to use in 3.1.0, and why?

Philipp

[PkK]

Just did my first quick game with this version. The game itself went great. But when in the end I clicked "quit" I just got a black screen, and had to kill -9 (normal kill wouldn't do).
Philipp

I was able to reproduce the issue once, but after rebooting it no longer appears. I suspect that the problem only occurs after suspending to ram and disk a few times, and then stays. Since there has been some discussion of suspend-related issues on the driver development mailing lists, and I am not using the latest version (instead just the latest my distribution provides), this is likely to be some driver issue that has been fixed in the current version.

Philipp

[NoQ]

Which AI would you recommend to use in 3.1.0, and why?

NullBot for 1x1s and small matches (cause it lags if there are too many of them (too complicated thinking), and doesn't behave well in ntw-style maps) and as placeholder in multiplayer (if having not enough players), Nexus if you're looking for something similar to the original game or default 2.3 AI (cheating slider was removed, assumes its minimum non-zero value (no cheating)), Semperfi is a modification of Nexus, just plays better (you'll probably want it if you're looking for a 5x5 skirmish game or when playing 1x3 AIs).

Also, a bug on the game code side has been just found that affects NullBot's early game behaviour in 3.1 (it sends one of its early trucks to capture oils instead of building base due to a bug in groupSize() script api function), so it may be building the initial base a bit slowly (doesn't affect 3.1_rc3, fixed in 3.1_git).