166
Suggestions & Requests / Re: Some type of addon that can reduce brickcount
« on: June 07, 2017, 04:22:03 AM »-snip-
If you can do that heightmap, you're not far from doing the replacing.
Show Posts
This section allows you to view all posts made by this member. Note that you can only see posts made in areas you currently have access to.
If you can do that heightmap, you're not far from doing the replacing.
167
General Discussion / Re: Blockland Repository
« on: June 05, 2017, 04:18:02 AM »
That is indeed peculiar. On my site, I get SadBlobfish. I'm currently looking through it.
Edit: Located and fixed the issue.
Edit: Located and fixed the issue.
168
General Discussion / Re: Blockland Repository
« on: June 04, 2017, 07:06:04 PM »Thanks! Good to see what's planned.
If I get the motivation later this Summer, I may do a pull request or two. Although, no promises. It's been about 2 or 3 years since I've done anything with PHP, and I have no experience with Laravel.
I've only worked with Laravel for a year. Never have heard of it until Boom mentioned about it and got me going in the right direction. I put an extensive installation guide in the readme.
Also, have you excluded music uploads for the same reasons discussed in this thread?
I actually got two reasons. The first is not copyright, but more about the work that needs to be put through to be able to locate blatant copying music. It's perfectly legal to reuse a piece of content, as long as that piece of content is not copyrighted as well.
The second reason is size. Even if images or models might be big, the music is comparable gigantic. I rather disable such a feature mostly for this reason to avoid having to store so big add-ons on the server. Of course, it should be fully possible to enable music if the host think the drawbacks are not that bad. Currently I'm not sure of how I'll put this into play.
Finally, do you plan on making this project something that anyone can easily download to host their own archive, or something that's intended to be provided as a service? I know that it being on GitHub more-or-less implies the former. However, I wonder how much of this will be reflected in your code. For instance, are you planning on adding options for customization?
My host is only there for a general view of how it'll look like. Sure, I wont take it down nor limit it so it can still be used as a normal host. However, the whole thing with this project is to make others take over it whenever they think they want to be a host.
I currently have the general customization for environment variables, but I tend to add more later on to make it easier to use. For now, one can change some of it through the seed files.
Personally, I would love to run something like this locally, but not necessarily publicly.
Laravel is built so you can do local testing in a closed environment, and then put it in production on the server that is public.
169
General Discussion / Re: Blockland Repository
« on: June 04, 2017, 02:09:34 PM »Glad you resurrected this. I hadn't heard of it yet. I'll give it a bookmark. As long as there's a good deal of file sharing with services like BLG (as you said there will be), I'm in full support. It never hurts to have a little redundancy in the case of an RTB-type failure. I wish this project the best of luck.
It's the plan to add that feature, but I haven't gotten around doing it until I've mangled out the basic functionality.
- Do you plan on adding searching? Having a master list that lists all hosted add-ons together would also be nice.
Yes. It's on my to-do list, but I'm discussing it with my designer.
- Do you have any sort of implementation schedule for features you've said will be implemented (like BLG file sharing)?
Yes, they are all listed here.
- Are you in need of any help with development, or are things just a bit sluggish due to life and whatnot?
Both, really. First, this project was created so anyone that think it's worth something may modify at their hearts content, the very reason I put it out here so early.
Secondly, life is... hard, mildly speaking. I've put the project on lower priority due to things that requires my attention more. I might pick it up this summer again, but I cannot guarantee it.
170
General Discussion / Re: Blockland Repository
« on: June 04, 2017, 08:45:27 AM »
I'm taking back the control of this topic now, now when it has been resurrected from the dead. I've also been busy to bother about it, but I'll eventually continue on it and ignore those pesky features that I currently cannot implement, but sure will in the future.
Thank you for the edit.
Thank you for the edit.
171
General Discussion / Re: Anyone interested in learning how to make addons?
« on: June 02, 2017, 04:46:01 PM »Start with Java coding it's somehow similar to Torque Script.
I can tell that it definitely is not. It's more toward PHP or JavaScript.
172
Modification Help / Re: Why does my GUI only open up while I'm in-game?
« on: May 31, 2017, 07:51:57 AM »No. You shouldn't ever lock a thread here.
Please do not confuse him, I told him the other day to never lock any mod help topics (unless it was a mistake posting)
People may have something important to say but cannot say anything if the topic is locked.
Sure, I was unaware of the unlisted rules, and will adjust me accordingly.
But changing the topic name for a new question is still a bad thing, right?
173
Modification Help / Re: Why does my GUI only open up while I'm in-game?
« on: May 31, 2017, 05:58:13 AM »
I told you yesterday about this. It's because you need to bind it to the global map.
Keep in mind that you need to make sure that you don't override someone else global bind.
Also, please don't change the topic title due to a new question. Instead, create a new thread and lock the old one.
Code: [Select]
GlobalActionMap.bind(keyboard, "x", pushTestgui);Keep in mind that you need to make sure that you don't override someone else global bind.
Also, please don't change the topic title due to a new question. Instead, create a new thread and lock the old one.
174
Modification Help / Re: Please help me with this GUI
« on: May 30, 2017, 12:14:08 AM »Code: (Console) [Select]
Error: cannot change namespace parent linkage for BlocklandLive from GuiControl to GuiWindowCtrl.On line 4 and 16, they have the same name.
Also, BlocklandLiveGUI does not exist.
175
Suggestions & Requests / Re: Dynamic Enviroments
« on: May 26, 2017, 12:06:46 PM »
As TableSalt mentioned, this have been a feature a for years.
https://forum.blockland.us/index.php?topic=204914.0
However, if you meant to change weather, like sun, rain and snow, then I'm not entirely sure that it's in there. Rain and snow does exist, but only as a predefined default.
Someone goes through it a bit, though, so you could start there.
https://forum.blockland.us/index.php?topic=69259.0
I still don't know why someone haven't made a custom weather creator.
https://forum.blockland.us/index.php?topic=204914.0
However, if you meant to change weather, like sun, rain and snow, then I'm not entirely sure that it's in there. Rain and snow does exist, but only as a predefined default.
Someone goes through it a bit, though, so you could start there.
https://forum.blockland.us/index.php?topic=69259.0
I still don't know why someone haven't made a custom weather creator.
176
Add-Ons / Re: Blockland Glass 4.0
« on: May 22, 2017, 04:14:54 AM »The cause is most likely your Nvidia graphics driver. You need to roll back to an older version of your card's driver. It's a very common issue right now with Nvidia users.
I haven't updated for months since I rolled back due to my modded Minecraft crashed when I started it. The trust is so fragile when it comes to these kind of things.
177
Modification Help / Re: fast node path algorithm
« on: May 21, 2017, 06:38:57 PM »mctwist your sectioning of the pathing can lead to bot self-looping as it finds optimal nodes within its own network but does not lead to the optimal path. it can also lead to looping if the sub-network it connects to has no way other than backtracking to go to the target node.
That is indeed true and I didn't think about that. Of course it could be solved partly with weights. Like, take all nodes and take the mean value of them and then the weight of the neighboring section weight. I'm not sure if a looping would occur though. Would you care to explain more?
178
Modification Help / Re: fast node path algorithm
« on: May 21, 2017, 03:41:29 AM »
If the speed you're looking for is about "not boggling down the server", then I suggest that you split it up in parts. That is, start a search and do a couple of nodes, then schedule an another round and continue. The bots may move within this time. When it has reached the goal, then it's done.
However, if a brick is modified, it will then check if that is a node(I'm not sure how you're placing nodes, so do your thing here to move/replace it), and then do the A* again. I assume that in a heavy match with many AI's, this should reduce the lag on the server. You may also make so the bots wont move until a valid path has been found.
There is also this thing that you may split the graph up into smaller sections and connect them with their own graph. This may of course be a bit complicated, but it could make the A* to guess in beforehand what path is faster, and then do smaller A* searches each time they search.
That is, first do a fast A* on the overview graph. Then do A* on the section the bot is within right now.
Mix together both of the suggestions and you will have something solid enough.
I believe this technique was used in Rail Tycoon.
No-edit: Man, now you got me hyped to make this myself...
Edit: I made an image to maybe make it a bit easier. The orange circles is the sections and the yellow lines are connections between them. This might look like it's complicated to make, but actually, just bundle together a couple of neighbors that are closest to each other and mark their section connections.
This could also give room to optimizations between sections to have a predefined path while just going through sections.
Sectioning like this can be done within itself as well. Sectionception!
However, if a brick is modified, it will then check if that is a node(I'm not sure how you're placing nodes, so do your thing here to move/replace it), and then do the A* again. I assume that in a heavy match with many AI's, this should reduce the lag on the server. You may also make so the bots wont move until a valid path has been found.
There is also this thing that you may split the graph up into smaller sections and connect them with their own graph. This may of course be a bit complicated, but it could make the A* to guess in beforehand what path is faster, and then do smaller A* searches each time they search.
That is, first do a fast A* on the overview graph. Then do A* on the section the bot is within right now.
Mix together both of the suggestions and you will have something solid enough.
I believe this technique was used in Rail Tycoon.
No-edit: Man, now you got me hyped to make this myself...
Edit: I made an image to maybe make it a bit easier. The orange circles is the sections and the yellow lines are connections between them. This might look like it's complicated to make, but actually, just bundle together a couple of neighbors that are closest to each other and mark their section connections.
This could also give room to optimizations between sections to have a predefined path while just going through sections.
Sectioning like this can be done within itself as well. Sectionception!
179
Modification Help / Re: fast node path algorithm
« on: May 20, 2017, 09:43:17 AM »
It all depends on what you have at the start. Dijkstra's could be an alternative if you're handling weights, distances and graphs.
If you can precalculate the paths, that is a static map, then you could use JPS+ and Goal Bounding.
Video
Git
It could be doable for a map, unless you change it. Then it needs to precalculate the paths once again.
If you can precalculate the paths, that is a static map, then you could use JPS+ and Goal Bounding.
Video
Git
It could be doable for a map, unless you change it. Then it needs to precalculate the paths once again.
180
General Discussion / Re: PSA: Please stop using port 30000 for your servers
« on: May 14, 2017, 05:18:44 PM »Glass Hosting servers are given a range of 1000 ports free to use. There's multiple Blockland servers on each node. The user "blockland1" gets 30000-30999, "blockland2" gets 31000-31999, etc. This was also done by Kalphiter's hosting service, RTB hosting, and others.
Why give ports to people that they probably are not going to use? You could easily make it so you force the user to use one port you designated it to. Whenever a user hosts an another server he will get a port that is free and use that. Who hosts over 10 Blockland servers at the same time?