Author Topic: Single Bot Simulation  (Read 4072 times)

Offline AZPaul

  • Bot Builder
  • **
  • Posts: 76
    • View Profile
Single Bot Simulation
« on: August 06, 2005, 12:23:42 PM »
Hey y'all.

It's been a while since I said anything, since I've had nothing much to say, but now that's changed.

Oh, I've been lurking in the corners periodically, loading updates, laughing at bunnies and such, looking for strange things like [begin hint] a working .sexrepro [end hint] and [begin hint] some utility to get access to Bot memory locations that I can import to Excel [end hint] and like that, but I just haven't had a lot to say.

Well, now, that's changed.

While patiently awaiting these additional capabilities I really haven't been able to devote much time to this little hobby. Not that I don't have the time to devote, but running my evo sims time and time again without being able to extract the data is pretty useless in my view and, frankly, is boring. So I decided, after feeling that hankering to do some exploring of other aspects of DBII, that I would create another simulation bot. Thus the trouble begins.

I suppose most Bot makers write a few lines of code, cut and paste genes from existing Bots, making a precursor of what they want and throwing the new bot into a small limited sim to see what happens. Well I'm not too different from everybody else so I do that too.

Just a simple little thing Bot. Just walk around, find a veggie and have some lunch. That's all. The grand design can wait till the basics are working, right? Cept the basic wasn't working.

Now, I was somewhat embarrassed by having you helpful people point out missing 'stop', 'end', 'store' and like that in some of my previous code (help for which I am still grateful btw) and was not about to trip over some obvious piece of junk code again if I could help it. So I did some extensive desk checking (I mean, my god, how screwed up could 15 short genes be) but no problem there.  Honest.

After some exhaustive, and, if I may say, brilliant, troubleshooting, this is what I've done and what I see.

A single simple little pre-bot goes into a simple small field with plenty of veggies upon which to feed. She snuggles up to a veggie, zaps it with plenty of -1 shots, gets no energy in return, exhausts her energy and pops like an overripe tomato. Not good.

I have my other simulation setups and I know my PA Alpha bot really works, feeds, survives, prospers. Thinking that maybe I have somehow screwed up this new small field by fiddling with veggie energy, veggie numbers and the like I put a single proven Alpha into the new field and let her loose. She snuggles up to a veggie, zaps it with plenty of -1 shots, gets no energy in return, exhausts her energy and pops like an overripe tomato.

Obviously, I've f#<&%@up the sim field. So I abandon the new field for the larger ones where my Alphas and Deltas live in veggie engorged splendor and I know all is well with the universe. A single simple little pre-bot goes into my well established and working-fine sim field with plenty of veggies upon which to feed. She snuggles up to a veggie, zaps it with plenty of -1 shots, gets no energy in return, exhausts her energy and pops like an overripe tomato. Shit.

Can't be. Run the main evo sim with my Alphas, everything works. Veggies get shot, give up energy, everything works.

Wait! A single pre-bot vs multiple Alphas. Naw, couldn't be.

A single known working Alpha goes into my well established and working-fine sim field with plenty of veggies upon which to feed. She snuggles up to a veggie, zaps it with plenty of -1 shots, gets no energy in return, exhausts her energy and pops like an overripe tomato. Ah!

Put in 5 Alphas. Some of them are being fed by the veggies, periodically. Seems that if only one Bot is feeding the veggies will not give up their goods, no little white dots appear. If a second Bot starts to feed on the opposite side of the field those little white energy dots stream out of both veggies just fine for both bots. Eventually, however, all succumb to the famine and perish.

Put in 10 of the new little simple pre-bots and everybody feeds just fine and goes about their little daily lives. Nothing is amiss with the little simple pre-bot. Nothing is amiss with the sim field layout or parameters.

But only one new little pre-bot or only one lonely Alpha or only one bot of any type in a simulation will not be looked upon kindly by the veggies and will starve to death.

Something is not right in Bot-land.

Two questions:

1. Why?

B. Anyone else seeing this?

3. Have I lost it? Am I in a Twilight Zone episode? Do I need to re-do my meds, again?

d. Did you know that the Fermi paradox can be easily resolved if we consider the possibility that we may be one of the first intelligent species to arise in this gallaxy? Someone has to be first, yes? There may be plenty of 'others' out there just no one has yet made it to the point where they send messages into the sky and listen for answers like we do. Maybe the ones that are doing so are on the otherside of the gallaxy and are too far away for either of us to yet discern the other.  There are hundreds of resolutions to the Fermi Paradox that do not preclude the existance of intellegent (even young inter-stellar) species in this gallaxy.

-P

Offline Shen

  • Bot Destroyer
  • ***
  • Posts: 111
    • View Profile
Single Bot Simulation
« Reply #1 on: August 06, 2005, 01:04:15 PM »
Nope its not you. Its something to do with the shot collisions. If you check the veggies ? .shflav in the console you will see its -2 when its supposed to be -1, as its collecting its own -2 shots somehow. I believe this is fixed and we are waiting on nums for the next release.

And personally I think the lightspeed cage resolution to Fermi is the most satisfying. All species that dont grow exponentialy and are really agressive tend to stay at home. Those that absorb and grow as quickly as they can end up expanding there colonies at lightspeed. Eventually the inner colonies are exausted of resources and since the outer planets are being colonised at lightspeed already they cant expand fast enough to support the inner colonies. Result: Ecocollapse.
« Last Edit: August 06, 2005, 01:05:33 PM by Shen »

Offline Botsareus

  • Society makes it all backwards - there is a good reason for that
  • Bot God
  • *****
  • Posts: 4483
    • View Profile
Single Bot Simulation
« Reply #2 on: August 06, 2005, 03:59:35 PM »
How about all this ufo's and sutch , I mean it can't be all bs. I think a combination of things is true:

A.) We are too primative of a species and no one cares about us.

B.) In this stage of the universe life exsists too spread out for any real contact to happen.



Now about this bug: Sounds like Shen is saying: IT IS RESOVED RIGHT?
« Last Edit: August 06, 2005, 04:02:05 PM by Botsareus »

Offline Numsgil

  • Administrator
  • Bot God
  • *****
  • Posts: 7742
    • View Profile
Single Bot Simulation
« Reply #3 on: August 06, 2005, 10:29:55 PM »
There were some problems with shot collisions.  I had to redo them when I made bot sizes more dynamic, but even now it's only maybe 80% accurate (still some occasional should be hits but aren't.)

You know what, let me release what I have of 2.4 to the FTP, so people can check it out.  I've been so distracted the last month that I haven't hardly done any work.  Some feedback might be useful.

Offline AZPaul

  • Bot Builder
  • **
  • Posts: 76
    • View Profile
Single Bot Simulation
« Reply #4 on: August 07, 2005, 10:18:52 AM »
From Shen:

Quote
Nope its not you. Its something to do with the shot collisions. If you check the veggies ? .shflav in the console you will see its -2 when its supposed to be -1, as its collecting its own -2 shots somehow. I believe this is fixed and we are waiting on nums for the next release.

 :clap:  :clap:

My meds [you]are[/you] working! I [you]am[/you] lucid! Oh, Ode to Joy, thank you, Shen!

 :clap:  :clap:

Of course I continue to work around this thing. It is just a nit after all. It's just nice to know my sanity is somewhat stable.

The hole in this eco-collapse scenario, imo, Shen, is that even as the inner core collapses the outer edge remains vibrant and expanding. Fermi's point was that if such a species existed they would have been here by now, or, we should see some evidence of such in the astronomer's gaze.

From Bots:

Quote
How about all this ufo's and sutch , I mean it can't be all bs.

Maybe not ALL bs. Just the vast majority is bs. There are some anecdotes of experiences that still want for some sort of cogent explanation. The problem is that there is no evidence, no data, no artifacts to support a "they're here" hypothesis.  Unfortunate for the UFO crazies the plural of "anecdote" is not "data." There is nothing to support the hypothesis and it fails. Plus you have the problem of  "belief," as in a religious sense, that has become attached to the UFO phenomenon. Such people believably see things that really aren't there.

Until something physical is presented we can not say either way. Carl Sagan said the absence of evidence is not evidence of absence.  As a scientific proof, this is so. But in the UFO instance the absence of evidence does lean quite heavily in one direction.

OK. I will now resume lurk mode and talk at y'all when...

well, when I have something else to say.

Thanks, folks,

-P

Offline PurpleYouko

  • Bot God
  • *****
  • Posts: 2556
    • View Profile
Single Bot Simulation
« Reply #5 on: August 07, 2005, 06:14:42 PM »
Quote
[begin hint] some utility to get access to Bot memory locations that I can import to Excel [end hint]
Working on it.

I have a version of my snapshot utility that imports and executes a DNA file in Excel with full display of all memory locations and stack values, along with logic states at each stage.
It is working fot the most basic stuff right now but needs to have a lot more stuff added to keep up with modern robots.
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
Single Bot Simulation
« Reply #6 on: August 07, 2005, 06:16:57 PM »
Quote
There were some problems with shot collisions.
Have you added a line of code to the shot collision routine to disable collisions if the target bot is the same as the parent bot? That would stop a bot from picking up its own -2 shots.
Poison already works this way to stop a bot getting poisoned by its own shots.
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
Single Bot Simulation
« Reply #7 on: August 08, 2005, 02:24:00 AM »
Quote
Quote
There were some problems with shot collisions.
Have you added a line of code to the shot collision routine to disable collisions if the target bot is the same as the parent bot? That would stop a bot from picking up its own -2 shots.
Poison already works this way to stop a bot getting poisoned by its own shots.
I believe so.

I'll post the code I have to the FTP later tonight.  The problem is that there's a new DLL that's giving me hella trouble, and if anyone else is ever going to be able to run this thiong, I'll have to figure out some sort of fix.