Darwinbots Forum
Welcome To Darwinbots => Newbie => Topic started by: MonteCristo on January 28, 2014, 04:00:14 PM
-
Hello everyone, new to this stuff, but I learn quickly if i want to, question is... What happened to PurpleYouko, because when i look into his profile, he was last been active in 2008. Also, does anyone know what happened to 3DVCE creator Lee Graham? I'm thankfull for any information, and i hope i will learn something from DB.
Btw, i could transilate wiki to Lithuanian, not that it is a big country or anything, but i'd love to help in any way i can.
-
PY isn't active in the community anymore. He's presumably moved on to other things.
I don't think anyone here knows about 3DVCE to be honest. I hadn't heard of it until you mentioned it.
-
i have it. it is no longer active but there is a vid that gives you a link to it but i will just give you the direct url to download it from.
http://www.mediafire.com/download/s71k1ri6xo28qri/3D_Creature_Evolution.zip
have fun :D
-
Well, I didnt know about DB when i was using 3DVCE :P , but nevermind, its a fitness based evolutionary algorithm, honestly its more interesting to look at then DB, but its code isint as deep, so you wont see any smart things that you can see in DB. Idea just popped up, DB doesnt support multiple cores?then why not run as many .exe as there are cores in your CPU, and then set each .exe to a different core? its that simple, or is there a deeper problem?
-
it uses the first avalible core then doubles up and so on.
-
Ok, I remember Numsgil mentioned making a big server for DB, thinking why not make DB a BOINC project? People would connect their computers to to special server, and would share their PC power. That would be a supercomputer, and i would bet, DB would not only run real time, but also mabie faster, just an idea
-
Well, I didnt know about DB when i was using 3DVCE :P , but nevermind, its a fitness based evolutionary algorithm, honestly its more interesting to look at then DB, but its code isint as deep, so you wont see any smart things that you can see in DB. Idea just popped up, DB doesnt support multiple cores?then why not run as many .exe as there are cores in your CPU, and then set each .exe to a different core? its that simple, or is there a deeper problem?
That's possible and some do it, with the sims combined with bot teleporters.
Ok, I remember Numsgil mentioned making a big server for DB, thinking why not make DB a BOINC project? People would connect their computers to to special server, and would share their PC power. That would be a supercomputer, and i would bet, DB would not only run real time, but also mabie faster, just an idea
Seems like IM(Internet mode) to me, or do you mean something different?
-
BOINC combines thousands of PC into one giant web based supercomputer to run a certain task, e.g DB
-
Btw what is bot teleporter?
-
First a point: DB2 (vb6 based) can not multi-thread. Therefor, you can not scale up the performance of a single simulation by adding more cores.
Teleporters (as the name implies) let you share robots between running simulations. The idea of IM is to be able to share robots across the internet with all other simulations connected through the internet therefor making one huge complex super simulation.
-
Yes, I know you cannot just add cores to a single simulation, but I mean if you started lets say 4 DB simulations at once, and you have 4 cores CPU, why not set every DB exe to their affinity in task manager? every simulation would run normaly, but instead of doing just one you would do 4 different sim, and if i don't understand something, just say it, im still a newb
-
Running 4 times DB indeed means it can use 4 cores.
No need to even set affinity to get it to work. It'll work fine without setting it.
-
Oh, it does? Well thanks for the heads up Peter
-
Yea, you just need a way to share the robots between the simulations...
-
And before i go to sleep, wanted to ask: whats the progress on DB3? how much % of its done?
-
DB3 is slow going, because I get bored along the way and start doing novel, untested things :P
My current plan is to have something (either an early alpha for DB3 or at least a physics playground) by summer 2015. I'm approximately on track for that, unless I find a new interesting problem to grab my attention :D. I'm at the beginning of the process of gluing all the bits of a proper physics simulation together (graphics, collision detection, collision response, etc.), but there's a few open questions I don't have answers for just yet.
The first drop of DB3 proper won't really scale well to lots of bots (it's possible you'll need fewer than 50 bots to get realtime-ish speeds), and may or may not have fluid physics, but should have bots running around using DNA, a simple combat model, and multibot physics of some sort (the ability to create articulated structures from multiple bots). On top of that simple base I'll start adding things and streamlining the performance, and development should be at least more public if not more rapid.
-
What would be the biggest diferences between DB2 and DB3?
-
Multibots should be more stable, so you should be able to build larger and more complicated organisms without it collapsing in to a pile of vibrating goo.