For now I worked around this by running the Vista version with Wine. Seems to be working fine for now.

I downloaded the game today and got the ELFCLASS32 error as described in http://vsingleton.blogspot.com/2012/03/wrong-elf-class-elfclass32-possible.html. I followed the instruction and got the error

./mule.sh: 11: ./mule.sh: /bin/java: not found

I ran "whereis java" and got

java: /usr/bin/java /usr/share/java /usr/share/man/man1/java.1.gz

So I simply added the line "JP=/usr/bin/java" to mule.sh. Now running ./mule.sh works up to when I see the message "INFO: Launching game" after which of course the game doesn't launch. What should I do next?

------------------------------------------------

Output of "./mule.sh" below:

Aug 30, 2019 5:56:34 PM com.turborilla.mule.launcher.Launcher run
INFO: Arguments: nohup /usr/bin/java -Xms64m -Xmx256m -Djava.library.path=lib -jar ./data.jar
Aug 30, 2019 5:56:34 PM com.turborilla.mule.launcher.Launcher run
INFO: Classpath: ./launcher.jar
Aug 30, 2019 5:56:34 PM com.turborilla.mule.launcher.Launcher run
INFO: Java VM: 11.0.4+11-post-Ubuntu-1ubuntu218.04.3
Aug 30, 2019 5:56:34 PM com.turborilla.mule.launcher.Launcher run
INFO: Java Runtime: 11.0.4+11-post-Ubuntu-1ubuntu218.04.3
Aug 30, 2019 5:56:34 PM com.turborilla.mule.launcher.Launcher run
INFO: OS: Linux 4.15.0-58-generic amd64
Aug 30, 2019 5:56:34 PM com.turborilla.mule.launcher.Launcher run
INFO: Sun Arch: 64
Aug 30, 2019 5:56:34 PM com.turborilla.mule.launcher.DynamicUpdater isNewBuildNumber
INFO: this build number: 1.3.6
Aug 30, 2019 5:56:34 PM com.turborilla.mule.launcher.Updater isNewerBuildNumberOnServer
INFO: About to download http://www.planetmule.com/updater/build.txt
Aug 30, 2019 5:56:34 PM com.turborilla.mule.launcher.Updater isNewerBuildNumberOnServer
INFO: Could not download build number due to java.lang.NoSuchFieldError: preferIPv6Address
Aug 30, 2019 5:56:34 PM com.turborilla.mule.launcher.DynamicUpdater isNewBuildNumber
INFO: Build number not newer on server
Aug 30, 2019 5:56:34 PM com.turborilla.mule.launcher.NativeFileSelector run
INFO: Renaming native 64 bit files
Aug 30, 2019 5:56:34 PM com.turborilla.mule.launcher.Launcher run
INFO: Launching game

My modified mule.sh file below:

#!/bin/sh
cd data
export LD_LIBRARY_PATH=.:/usr/lib/jvm/java-8-openjdk-amd64/jre/lib/amd64
if [ -f /usr/lib/jvm/java-6-sun/bin/java ]; then
   JP=/usr/lib/jvm/java-6-sun/bin/java
elif [ -n $JAVA_HOME ]; then
   JP=$JAVA_HOME/bin/java
else
   JP=java
fi
JP=/usr/bin/java
$JP -jar ./launcher.jar nohup $JP -Xms64m -Xmx256m -Djava.library.path=lib -jar ./data.jar

I vote for a quick fix :-)

I am for Wampuses.

.... You're wife says "What do you want for dinner" and you answer "5 units of Food".

what?

Melanie?

Wait... are you switching to 3D? Does a strategy, almost turn-based game like this need so much effort? For what?

Or did I misunderstand?

It sure is hard to do, to plan etc, but if you are clever and plan ahead you are sure that this strategy isn't going to have drawbacks. I think a no-brainer, even if it's difficult to achieve, it's a strategy that does you only good.
This is especially true in the very last development turn, when Energy is useless. Usually I convert all Energy plots the last turn so I don't have to think about stockpiling energy.

1. You spend all your free time in the pub.
2. You mumble about base production value when you see a mountain.
3. You ponder about demand and supply at the supermarket.
4. You sell all the jewels of your wife to prevent a pirate attack.

...any other ideas?

I do it, but I hate it because it is a no-brainer.

(No-brainer: choices which aren't really choices, which present little or no disadvantage, and some advantage, so you just make these decisions without thinking because they can only be good. Since a game is a series of decisions to be made, no-brainer strategies narrow the quality of the game and they should be reduced as much as possible)

LTC: Learning Curve Theory.
If you have 3 factories of the same kind you have +1 on all your factories of that kind. In addiction, it stacks, so if you have 6 factories of the same kind, you have +2 on all factories, and so on.
You can see how this is the factor that makes the big production. If you have 9 crystite factories, the bonus would be 3 * factories = 3*9 = 27 LTC bonus. If all your factories were adjacent you would have +9 EOS bonus.
So tell me which one is overpowered 

Yes! No more people converting all energy plots to crystite!

Wait a minute, you say that EOS is preferred over land type, but I actually think LTC is overpowered.
Think about it: EOS doesn't stack: each factory would receive the bonus only once. Now since LTC stacks, it makes land type and EOS a second choice strategy: if you are going to specialize you want to have many factories of the same type.
My solution: reduce the LCT bonus to +1 in total (not for each factory), or don't make it stack.

Thanks for the tips! I bought Race For The Galaxy and can't wait to play. I am also going to check out Dominion at BSP.

Mole: were you successful teaching Kingsburg ?