Messages

76

Bugs and fixes / Re: Replace up, dn, sx, dx with fwd, bkwd, left, right

« on: January 06, 2017, 12:36:39 PM »

It's not either/or; you can have more than one sysvar point to the same location.  Check out DNATokenizing.bas, where the sysvars are set up, there's one for aimleft and aimright to augment aimsx, aimdx.

You can probably just do fw and bw for forwards/backwards, and left/right for left and right.  There are nautical and anatomical terms we could coopt, but I don't think they'd be clearer, even if they're cool (anterior/posterior and forward/aft respectively).

In terms of naming, if you deviate from the 2.XX(.XX) format you're kind of declaring your desire to fork the codebase and do your own thing, for good or ill.  It's fine to do that, just make sure you want to do that.  If you don't want to do that, there's nothing wrong with 2.49.XX.  The numbers don't mean anything except that larger numbers are later revisions, so we can do 2.49.XX.XX.XX etc. all day long.  If you really want to take the 2.50 designation that's probably okay; the C++ version I did 10 years ago isn't exactly in wide distribution   But just be careful you're not going to run out of numbers again.  2.99 is just as much an issue as 2.49, except I'm not giving up the 3.0 designation without a fight

77

Bugs and fixes / Re: EricL's FTP IM + my attempt to save it for what is worth

« on: January 03, 2017, 06:03:58 PM »

I would imagine it's relatively trivial for someone to portsnoop their own computer and find the FTP credentials from a running version of DB, so it's not like making even the entire thing closed source would necessarily mean no one has the FTP login.

Best bet is a separate config file attached to the binary like Peter is saying.  Like this.  I'm not sure if something like that exists in VB6 but there's probably clever things we can do even if you can't do that in VB6.  But it just means people won't see the FTP credentials from casual inspection or web browsing.  It doesn't mean the FTP credentials are secure.

In the long run, like I was saying before, it'd be better to have users login with their forum usernames + passwords.  That would make it easy to revoke access to a specific person if they're abusing things.  A concerted troll would still be able to spoil things, but at least it wouldn't be as easy as getting a single set of FTP credentials.  But that's not a trivial undertaking.

78

Bugs and fixes / Re: EricL's FTP IM + my attempt to save it for what is worth

« on: January 03, 2017, 07:16:35 AM »

At the time my concerns were:

1.  There's no way to limit the size of the directory the FTP dumps to on the server.  If the server runs out of disk space, Bad Things(tm) happen.  And we usually only figure it out after the fact.

2.  It doesn't scale well to lots and lots of users.

3.  It probably was against the terms and conditions of our server.

Now, I was not and am not a webdev, so these things might not be true!  Especially at the time I didn't understand Linux very well at all, which made it frustrating to do anything on the server, and when Eric was around we were using a hosting service (GoDaddy).  Now that we have our own server linuxbox, we can pretty much do whatever we want.  And I've certainly grown as a programmer over the last few years and since I started working at Google I've had to become more adept at linux stuff.

So, thinking about it now, we might be able to limit the size of the FTP folder using this method.  I'll try to dick around with it at some point in the next month maybe.

On the scaling issue, there might be some cleverness the clients can do to make it less of an issue.  And certainly it's not something we have to worry about if the number of simulations is counted in the dozens.  I think the FTP protocol itself is certainly capable of scaling up enough for our needs.

...

I think we can tie in the FTP accounts with forum accounts, so we can be at least be sure if something gets abused who to blame.  That I'm less sure of, as it would require some magic to connect the forum database with the FTP logins, which I think might be tied to system users.

...

If you want to start playing around with setting up a FTP based IM again, I'll see if I can get things kosher for it on the server.

79

Darwinbots3 / Re: How much progress has been made on DB3?

« on: January 01, 2017, 05:39:31 PM »

I'm still working on it pretty regularly.  I have maybe 95% of a very simple single bot simulator done, but my progress recently has been slow because I moved to London in November.  I have some infrastructure stuff I'm working on right now but once that's finished I have another maybe week or two of some UI work/polish and I can release it.  Fingers crossed I don't see why I couldn't release it sometime in the next few months, but I don't like making promises like that, as I'm usually pretty bad at estimating coding time.

Don't expect too much; it's the first executable that ties in a lot of the systems together at once.  It's mostly a way for me to experiment with the glue that holds the different parts of the program together (DNA, rendering, UI thread management, etc.)

80

Off Topic / Re: Wow. This is still a thing. I'm glad

« on: December 25, 2016, 04:52:53 AM »

Welcome to being a kid in the internet era  . Thankfully all my youthful internet posts are probably on defunct sites.

81

Off Topic / Re: My SSD drive broke down. What do you recommend?

« on: December 18, 2016, 03:31:10 PM »

I've been using Samsung SSDs.  I don't have any real criticisms for them, but I don't know that they're wowing me either.  But I'm not sure how a hard drive could wow me.  This is the drive I'm currently using, for what it's worth.

82

RANT / Re: Why was Spike so annoying?

« on: November 14, 2016, 05:44:24 AM »

Cmon man, no personal attacks.  You're better than that.

83

Darwinbots3 / Re: Big berthas in DB3?

« on: November 14, 2016, 05:41:56 AM »

I figure we'll deal with that problem when we get to it.  My preferred solution would be to add some tools to, say, randomly kill some robots from time to time.  That would discourage putting all your eggs in one basket.

I might decide to add shots, I'm not sure.  There's a lot of this that I won't figure out until I get far enough to start experimenting with it.

84

Off Topic / Re: what do you do in life?

« on: November 10, 2016, 09:48:24 AM »

what do you guys do in life? work? study?

As it happens, I'm starting my first week at DeepMind.  Before this I was elsewhere within Google/Alphabet, and before that I worked in the video game industry at Havok.  Basically been a professional programmer for 9 years.  Almost as long as I've been working on DB3

85

Off Topic / Re: <Rant> Twitter hacked for the 3rd time <Rant>

« on: August 30, 2016, 02:02:00 PM »

Since you live in the US you can try putting your number on the do not call list.  Not everyone respects it, and it takes ~month to really take effect, but it doesn't hurt.

Also, is this thread just going to be you accusing different forum people of hacking you?  Maybe it's PY!  He's been quiet.  Too quiet.

86

Darwinbots Program Source Code / Re: Buckets?

« on: August 12, 2016, 04:46:40 PM »

It's basically a way to cut down the number of intersections from O(n^2) to O(n), which helps the simulation scale to large numbers of bots.  It probably would be faster to just do one at a time for, say, a simulation of ~25 bots.  But when you get more like 1000 bots, the bucket system helps cut down the number of checks massively.

Another term for it is spatial hash, or uniform grid.  There's a few articles online that talk about it, eg: this one.

87

Darwinbots3 / Re: About Darwinbots3

« on: August 10, 2016, 02:03:59 PM »

just wandering how's db3 going?

I'm almost done with a simple "bot testbed".  Basically a simulation with just a single bot in it where you can program its DNA.  And the only thing the bot can do right now is change its shape.  But it's a good opportunity to flesh out and test all the different sub systems together in a single app, and figure out the high level multithreading strategy to keep the UI, renderer, and simulation threads separated so if one is slow or crashes it doesn't hurt the others.  Basically it's the first "vertical slice" of lots of the features all working together at the same time.

By almost done I mean the thing I have mostly works, but I'm still polishing some of the UI and making sure when it does crash people can send me useful crash reports.  I'm hoping to release it by the end of the month maybe, but I also haven't been actively working on it the last month or so, so we'll see.

Once that's released, I want to try and add in the fluid model, so we can see if bots can swim with it or not, and I'd like to play around with phagocytosis (eating other smaller bots by engulfing them) and mitosis (splitting in half to produce two daughter bots), since they both involve how the bots change their shape and they're probably the biggest question marks in my mind.  There's also an overhaul of the DNA I'd like to do at some point (collapse the boolean stack and the integer stack in to a single stack) to make it faster and some low level architectural stuff I'd like to finish that should allow the program to scale more easily to new features.

88

Off Topic / Re: Quick favor I could do with :p

« on: August 09, 2016, 01:46:31 PM »

eq1: (x1 - x2)^2 + (y1 - y2)^2 <= distance^2

For positions (x1, y1) and (x2, y2).  It's basically just checking if x2 is within the circle of radius distance centered at x1.

eq2: (x1 - x2)^2 + (y1 - y2)^2 + (z1 - z2)^2 <= distance^2

For positions (x1, y1, z1) and (x2, y2, z2).  Same deal, but now it's a sphere.

This extends in to higher dimensions.  You just keep adding in squared terms.

89

Bugs and fixes / Re: Could not find symbol: DllRegisterServer

« on: August 01, 2016, 04:38:22 PM »

The installer is trying to prevent the issues people had like this.  if you can open up all the UI menus without it complaining about missing dependencies you're fine.

90

Darwinbots Program Source Code / Re: Butched buoyancy in DB2?

« on: June 24, 2016, 07:08:43 PM »

Animals can largely change their buoyancy at will with minimal cost.  eg: swim bladders.  So I'd just let bots change their buoyancy without any cost.  And maybe corpses lose buoyancy and just sink.  Plus, from a gameplay perspective, I'm not sure the ability to sink or float is valuable enough to have to jump through hoops for.