Author Topic: Bug in 2.4.1  (Read 18683 times)

Offline Numsgil

  • Administrator
  • Bot God
  • *****
  • Posts: 7742
    • View Profile
Bug in 2.4.1
« Reply #30 on: October 20, 2005, 10:50:56 AM »
Oh, and before I forget:

your bots are lying on the bottom of the screen because you had y gravity not set to 0.

Go to the physics tab, open the pulldown menu for gravity, and set it to none.

Offline Griz

  • Bot Overlord
  • ****
  • Posts: 608
    • View Profile
Bug in 2.4.1
« Reply #31 on: October 20, 2005, 11:28:41 AM »
Quote
Grrr, Numsgil! I'm starting to wonder what the problem really is here. Does it work on your computer? Looks like I'm the only one reporting about the problem. The version 2.4.3 acts the same as the other 2.4 versions. The settings don't stay put!!! ..and I still get the 75 run-time error when trying to run a simulation.
no ... you're not the only one ...
SOME of my setting's return when I upload settings or a sim  ...
General for the most part, but not physics, such as gravity, or mutations ...
and  the 'starting positions' seem to shift on me in Species ...
I like to alter them for aglae along with gravity ...
keeping them in the upper half or confined to one corner ...
which makes for some variation of environment within the window.
 
also, the uploaded species, when I load a saved sim ...
are not updated in the 'loaded species' box ...
so if I want to edit, I have to also upload my saved settings.
and ... attempting to load other setting or sims ...
that may have been saved with another version ...
almost always results in a runtime error of some sort ...
either a type mismatch or path error or overflow ...
perhps from an older bot ... who can say? ...
there are too many varialble for me to pin it down.
but basically, unless I set up and save the settings and sim FROM
2.43 ... I run into problems when loading in a sim or settings.
and also the insert box doesn't function ...
which I like to make use of in testing a new bot's affect on a stable population/sim ...
and gene activations don't show either, although I have no use for them yet at this point.
and as Num says ... you may have to go tweak the old bots.

I wonder if as folks do that ... we might upload the bots we get to work
into the beastery under 2.43 or something?

meanwhile ... I go back to 2.36 and 2.37 to play so I don't get too frustrated
at the crashes and loss of sims. ;)

~griz~
不知
~griz~
[/color]
   "The selection of Random Numbers is too important to be left to Chance"
The Mooj  a friend to all humanity
[/color]

Offline Testlund

  • Bot God
  • *****
  • Posts: 1574
    • View Profile
Bug in 2.4.1
« Reply #32 on: October 20, 2005, 11:57:50 AM »
Ok, here's what I did:

1. Uninstalled Darwinbots 2.1 and then deleted the whole folder.

2. Installed Darwinbots 2.1.

3. Unzipped Darwin2.4.3 exe and sysvars2.21.txt into the    

DarwinBots folder.

4. Copied the text from the Animal Minimalis 2.4 Remix from the

FTP, created a textfile in the Robots folder named'Animal

Minimalis 2.4 Remix.txt, pasted the copied text into the file and

saved.

5. Started the program and clicked 'Start a new simulation' and

the problems with the GUI are there!

6. Deleted the bots from the Loaded species list, added Animal

Minimalis 2.4 Remix and started a simulation WITHOUT doing any

settings in the GUI. Got a warning message that the NRG for

veggies are 50000 something bla-bla. Clicked Yes and the bots are

falling down the screen like raindrops as I've mensioned before.

7. Quit and restarted again, this time adding Alga_Minimalis and

tried to make changes by changing starting points, skin and

color. The program crashes with run-time error '380'.

The problem is that changes in the GUI doesn't stay put.

Sometimes it crashes when trying make changes on the starting

position. If I manage to start a simulation the bots behave

crazy.

Obviously you have done some changes in the 2.4 versions that

doesn't work on my computer. I think I will stay with version

2.36.7 instead. That version works if I run in normal speed,

limit veggies to 100 and veg NRG to 10. Can't load saved

simulations though. To bad I can't limit the bots though. Have

had as much as 3200 objects slowing down the program to 1

cycles/sec!
The internet is corrupt and controlled by criminally minded people.

Offline Griz

  • Bot Overlord
  • ****
  • Posts: 608
    • View Profile
Bug in 2.4.1
« Reply #33 on: October 20, 2005, 12:01:56 PM »
2.43 ...
apparently 'save settings' and 'save sim' does not work ...
at least as far as saving species. [and physics/mutations]
once begun, I can alter/add species but ...
when clicking on 'start new' ... they are not there ...
just whatever was loaded in on the sim ...
be it new sim or saved sim.

now if I use 2.37 to save settings and/or sims ...
then I can load them in with 2.43 ...
realizing of course that the bots that work there ...
are going to act differently ...
and/or give runtime errors.

~griz~
不知
~griz~
[/color]
   "The selection of Random Numbers is too important to be left to Chance"
The Mooj  a friend to all humanity
[/color]

Offline Numsgil

  • Administrator
  • Bot God
  • *****
  • Posts: 7742
    • View Profile
Bug in 2.4.1
« Reply #34 on: October 20, 2005, 12:13:49 PM »
That's weird, I'll see if I can figure out what's wrong.

Bots falling like raindrops means that you have gravity turned on.  I know it's a minor point to your post, but when that happens, you just have to go to physics->turn of gravity.

Offline Light

  • Bot Destroyer
  • ***
  • Posts: 245
    • View Profile
Bug in 2.4.1
« Reply #35 on: October 20, 2005, 12:21:28 PM »
sorry num but they are right, 2.4.3 is well buggy. I tried running about 10 different sims with different bots and each one crashed with various different errors in less than 30 seconds. The bots do act really crazy, they seem to stop working after a very short while. I got this error report for one of them I  dont know if its usefull.

"Out of Memory. path c:\games\darwinbots\robots\alga_minimalis.txt is not a valid robot"

Offline Numsgil

  • Administrator
  • Bot God
  • *****
  • Posts: 7742
    • View Profile
Bug in 2.4.1
« Reply #36 on: October 20, 2005, 12:23:24 PM »
*grumbles heavily under breath*

lol.

If anyone wants to download VB6, and run the code from there, that would make finding disgusting errors 100% easier for you and for me.

I can't seem to reproduce the sorts of errors other people are getting.  I probably know what not to do, and without even thinking I don't do that.
« Last Edit: October 20, 2005, 12:23:37 PM by Numsgil »

Offline PurpleYouko

  • Bot God
  • *****
  • Posts: 2556
    • View Profile
Bug in 2.4.1
« Reply #37 on: October 20, 2005, 01:12:09 PM »
I finally reached a point in my database project where I can take a break so I downloaded the source for 2.4.3 and tried running it.

Most of the GUI seems to work OK but a few parts do indeed lose their values when changing tabs. Notably the veggie cooldown and repopulate threshold.

Ran the sim and it started just fine with my existing settings file.

The robots then began to fall down the screen like rain, just as described by testlund.

I double checked and gravity is definitely turned off. After a while, th efirst Veggie repopulate event happened and almost immediately I got an overflow error.

Here is the offending code.

Code: [Select]
If power > ((rob(n).BODY * 10) / 0.8 + Shell) And Not rob(n).Corpse Then
    'shot will continue on past this bot since it wasn't all used up.
  
    'with reduced impact
    Shots(t).value = (power - (rob(n).BODY * 10) / 0.8 + Shell) * Shots(t).range / (Shots(t).age * SimOpts.EnergyProp)
    Shots(t).nrg = Shots(t).range * (RobSize \ 3)
    power = (rob(n).BODY * 10) / 0.8 + Shell
    Shots(t).Exist = True
  End If

The error is in the line beginning "Shots(t).value =(" but I can't see why since all the individual variables in the code appear to be OK.

This is in the "Releasebod" subroutine.

I will look into it a little more myself but bear in mind this is the first look I've had at the 2.4 code so it might take a while.
There are 10 kinds of people in the world
Those who understand binary.
and those who don't

:D PY :D

Offline Griz

  • Bot Overlord
  • ****
  • Posts: 608
    • View Profile
Bug in 2.4.1
« Reply #38 on: October 20, 2005, 01:20:30 PM »
Quote
That's weird, I'll see if I can figure out what's wrong.

Bots falling like raindrops means that you have gravity turned on.  I know it's a minor point to your post, but when that happens, you just have to go to physics->turn of gravity.
it happened here too ... initially.
now that I have somehow been able to run and save a sim ...
it now defaults to those settings. apparently. I guess. who knows?
another thing I have to keep checking is that the veggies are "blocked" ...
that box never seems to come up checked.

anyway ...
I had the same problems getting even 2.37 to run until I added
that sample defalult.setting thing you showed me.
after that ... I had a chance!
so ...
where does the prog get it's initial default settings from ...
you know ... the very first time one runs it out of the box?
and does it indeed use default.set on startup ...
or what?

I was going to attach a settings that works for me on 2.43 ...
and might work for Testlund as well ...
using Alga_Minimals.txt and your new animials:
which I named Animal_Minimalis2.4.txt ...
but I can see no way to attach or upload files here.

can you point me to how/where to do that ...
or show me where such setting files might already exist?

thanks.

~griz~
不知
~griz~
[/color]
   "The selection of Random Numbers is too important to be left to Chance"
The Mooj  a friend to all humanity
[/color]

Offline PurpleYouko

  • Bot God
  • *****
  • Posts: 2556
    • View Profile
Bug in 2.4.1
« Reply #39 on: October 20, 2005, 01:21:04 PM »
OK here is a breakdown of the calculation

power   =                3286
rob(n).body =   49.03042
shell         =          0
shots(t).range =   14
shots(t).age =   1
energyprop =   1

Shots(t).value = (power - (rob(n).BODY * 10) / 0.8 + Shell) * Shots(t).range / (Shots(t).age * SimOpts.EnergyProp)

therefore....

shots(t).value =   37423.6765
« Last Edit: October 20, 2005, 01:27:42 PM by PurpleYouko »
There are 10 kinds of people in the world
Those who understand binary.
and those who don't

:D PY :D

Offline PurpleYouko

  • Bot God
  • *****
  • Posts: 2556
    • View Profile
Bug in 2.4.1
« Reply #40 on: October 20, 2005, 01:26:24 PM »
Quote
another thing I have to keep checking is that the veggies are "blocked" ...
that box never seems to come up checked.

It isn't supposed to. veggies are by default Unblocked since the blocked veggies are unrealistic under most simulation conditions.

Quote
where does the prog get it's initial default settings from ...
you know ... the very first time one runs it out of the box?

It loads them in from the "default.set" file which is included with the installation

Quote
but I can see no way to attach or upload files here.

At the botom of the window which you use to type in your messages, (just below the post icons) there is a browse button. Click it, select a file from your computer and it will appear as a url in the textbox to the left. Then you just submit the post
There are 10 kinds of people in the world
Those who understand binary.
and those who don't

:D PY :D

Offline Numsgil

  • Administrator
  • Bot God
  • *****
  • Posts: 7742
    • View Profile
Bug in 2.4.1
« Reply #41 on: October 20, 2005, 01:27:06 PM »
Hmmm, power is obviously too large...

I'll go crack open the body function and take a look inside.

Offline Numsgil

  • Administrator
  • Bot God
  • *****
  • Posts: 7742
    • View Profile
Bug in 2.4.1
« Reply #42 on: October 20, 2005, 01:28:28 PM »
When you first run a 2.4 version, it loads the settings file from default.set.

Whenever you exit the program properly, it saves your current settings as lastexit.set.

When you start up a 2.4 version, it loads up lastexit.set if it exists.

Offline PurpleYouko

  • Bot God
  • *****
  • Posts: 2556
    • View Profile
Bug in 2.4.1
« Reply #43 on: October 20, 2005, 01:35:03 PM »
Power is set in the same subroutine.

power = Shots(t).value * Shots(t).nrg / (Shots(t).range * (RobSize / 3)) * SimOpts.EnergyProp

shots(t).value = 3288
shots(t).nrg = 559.6519
shots(t).range = 14
robsize = 120

from the above formula....

power =3285.956
There are 10 kinds of people in the world
Those who understand binary.
and those who don't

:D PY :D

Offline Testlund

  • Bot God
  • *****
  • Posts: 1574
    • View Profile
Bug in 2.4.1
« Reply #44 on: October 20, 2005, 01:35:33 PM »
I'll stay tuned to see what you guys come up with. Maybe there's light at the end of the tunnel.   ;)
The internet is corrupt and controlled by criminally minded people.