Menu

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.

Show posts Menu

Topics - hoochy

#1
General Discussion / Rimworld is the best game ever made
September 09, 2016, 11:37:39 PM
I know it makes me sound like some kind of teenage fanboi (I'm 40 though), but right now this is the best game ever made. I must have put over 2000 hours into playing it now, which is pretty ridiculous when I think about things I could be doing otherwise. Sure the game has issues, but most things can be improved or fixed with mods. You go away from it and then there is always some thing in your head that brings you back to it.

I've been playing games since the 80s and this game is the closest yet to the game I always wanted to play in that time. Again, I'm not saying it is perfect because it's not, but it is the best game ever made.

There I said it. Congrats Tynan, the mod makers and the community here.

#2
If I place too many "rottable" objects in a non freezing room there is a significant performance impact by doing so in my current "Late" game. I had about 40000 excess meat in about ~300-400 piles . I moved it to a non freezing room to destroy it. Doing so dropped my frame rate from about 30 FPS to 2 FPS, until all the meat rotted away and then the performance mostly came back. But after I restarted the game by quitting to menu and coming back performance was back to before this problem happened. DEV mode showed no errors when it was running at 2fps, it was simply all those rotting meat contributing to a really bad performance.




I tried to replicate this on a new map but it was harder to see the performance drop from doing so. However it showed another performance problem and that is to do with room sizes. If you start a new map , go into dev mode and click "build full colony" , for me at least the performance is a rock steady 60fps. If I then build half the map into large rooms the performance drops like a rock. The only difference is the "Rooms", if you remove the rooms the performance comes back. This doesn't seem very related to path finding as adding lots of doors does nothing for the performance.

The current map I am playing on has a lot of large rooms which seems to be making the performance around 30fps now. It seems the Alpha 15c performance improvements did a lot for the pawn stuff but there are some "behind the scenes" performance issues with rooms and rotting objects.
#3
Ideas / Performance problem with too many floors?
August 30, 2016, 09:34:08 PM
I have noticed in late game when I build floors over a lot of the map that it seems to impact the performance noticeably. It could also be correlated to something else late in the game I guess but I figured I would mention it. Maybe it is something to do with cleaners searching for sections to clean?

Also is there any "official" way to profile the games performance? I was thinking I may try decompiling rimworld and then profiling it to see where the issues are.

The changes in Alpha 14 have certainly improved my late game performance by a fair amount, perhaps the upgrade to Unity 5 helped, but still it seems at some point it gets slower and slower even though there aren't significantly more pawns or the map has many changes to it. I only play at map sizes less than 250 due to the performance problems in late game. I have a I7 @ 3.8GHz and Radeon 290.
#4
Mods / Security when it comes to C# assemblies
April 19, 2016, 02:11:57 AM
Does Rimworld do anything in particular to stop malicious stuff in mods? Any restrictions on the C# assemblies when loaded? Or does it run at the same security level as the rimworld.exe itself?
#5
Ideas / Instead of research - knowledge and teachers
April 19, 2016, 12:24:54 AM
Research is a good way to "Scale gameplay". Instead of allowing users to create that nuclear reactor on day 0 you require them to research their way there. Spending valuable pawn time on the research bench instead of getting more resources and defenses is a gameplay dynamic.

However I think it could be done better and in a way which would open up more gameplay opportunities, and fits in better with the Rimworld "World". And that is you have to educate pawns themselves on topics to build more things. And this also means certain pawns would already come with certain knowledge. And if your professor teacher pawn who knows everything dies, well then you lose your ability to build those complex things if he never taught anyone else ! Now there is a real reason to keep those frail, one eye'd seniors around, the older someone is the more they should know, in theory!

Research Changes
* Remove research

UI changes
* All buildings should be visible in the menu, any that can't be built should be greyed/red'd out and have a tooltip or messagebox which says "You have no colonist with the appropriate knowledge to build this"

New tasks
* Learning, which replaces the old "research". Now colonists share and teach themselves things, so you need more than one colonist to "share" with.

Pawn changes
* New base trait, intelligence, which defines how quickly people "learn" and "share" such information.
* Each pawn needs data that defines the things they "know", and hence the things they can allow the colony to build.
* Social interactions may even be "teaching" moments, now more reasons to have pawns socialize. Even if it is much slower than getting pawns on the "Learning" task. Everyone knows that guy who thinks he knows something so well who can't just help always talk about it and eventually gets a slap for not shutting up.

New items
* Certain books you can buy that allows pawns to educate themselves on their own at a slower pace than if you had a pawn that knows it. Now a real use for a library!

Since the game already has things like the "Teacher" backstory, and "too smart" trait, and old people, it seems a perfect fit. It makes the game, at any point a lot more dynamic because you can have that great base that no one can touch until your professor pawn dies and now you can no longer repair or build those turrets. It removes a lot of the "safety" you feel once you have researched everything, and also gives you a feeling there is more in the game to do.  And it finally gives a reason to have those crappy frail useless pawns around if they are actually knowledgeable, instead of finding creative ways to accidentally kill them "go haul that rock chunk on the other side of the map during an ice storm". ;)
#6
Help / Is BuildingBase required?
April 18, 2016, 01:33:05 PM
I see that in many mods they define their own "BuildingBase". However in Alpha 13 at least I can't seem to find the base definition of it. Does anyone know the .XML it is in or is it is just some standard you don't have to worry about? My own mod uses it as a parent and it seems to work, even though I didn't define it.
#7
Help / Help with terrain and Community Library
April 18, 2016, 08:38:43 AM
Hi everyone, I am basically just after some details which will speed up my mod development. Basically I want a callback to be called after the map has been created. I then want to store some extra info for each tile on the map that I will use for a production building and maybe a task. I also want to add some extra "ground" types. So any info you can give to speed up my learning process on that would be fantastic! :)

Is it best to just override the relevant functions as I find them and then call the base one, or is there a better way to do this with the community library? I'm assuming there is some issue with order of loading and hooking into things which that library/mod solves but I haven't been able to find much info about it.
#8
Ideas / Fix mining
April 15, 2016, 10:48:17 PM
It seems currently that if you have a map that has more hills or mountains you are at a massive advantage compared to those that have less. The reason is that you can find valuable resources, for free essentially, in the mountains and hills. This concept of digging away mountains and hills, making tunnels and caves, etc, is fun, do not get me wrong. But there should be a way to get minerals from the ground also.

This isn't a z-level thing, well it kinda is but it is a very cheap z-level thing, it is quite a minor tweak Tyrian could make to the game to make it a lot more enjoyable. Upon map creation each tile on the map is given a resource amount. Much in the same way you can often see "resource veins" in the side of mountains, where there is a decent amount of a resource on the map the ground show some differences. So if there is an abundance of "steel ore" in the ground, it will show you visually. Much like how it would be in real life if you discovered a massive iron ore or bauxite deposit, it often is visually seen from the ground.

To get resources from the ground you would need to build a mining production building. I figure you could have a "fueled" one to start with, later getting to electric, perhaps the "Fueled" one even requires a pawn to be using it like a crafting table. You could also add research, so first generation miners can only "dig so deep", to extract minerals. So instead of there simply being "X steel ore" at location "0,0" on the map there are layers to the amount of resources in a tile, the higher your tech level the better your ability to extract all the ore at a particular location. This also opens up the possibility that there are hidden veins of resources on the map that you will only find out about later once you have that tech.

Another method to get minerals could literally be "mining" a tile, so you send a pawn to a rich deposit that is visible from the ground and tell him to "Mine that" much the same way he mines mountains or hills.

So the stuff that is visually seen on the map, like "oh there is a resource deposit here" could just be the easiest stuff to extract. The harder stuff to extract (which should be more plentiful as it would then scale to the late game requirements of increased resource usage) could either be a lottery, ie you just plant down a building and extract all resources from that area. Or you open up the possibility of a way to mineral search on the map ,perhaps with a new building, much like miners do today. Pawns with high "Mining" skill could perhaps be given the potential to give a bit more info about a tile they are on.

So in conclusion I propose adding "resources" to every tile on the map, that can be accessed by "mining the ground". This adds a a new fun layer to Rimworld that is more realistic, and in turn it makes every map type viable for self sufficient colonies.

1) This mining addition/change would make every map viable, though still vastly different. A new gameplay layer to uncover, resulting in more fun. No more need for hills and mountains on a map to make it playable.
2) All of these things can be tweaked exactly as required for the "Story telling" aspect and for gameplay as it is just numbers, so will not break Tyrian's vision at all. It is simply an inground representation of the "above ground" mining currently in the game, with a bit more fun.
3) It is extremely simple to add this to existing game, I believe it is even possible to mod this in currently with C# hacking. It will not add much, if any overhead to the game, whilst vastly increasing its fun and ability to play any map without relying on "outside" traders to work.
#9
General Discussion / Minable components?
April 14, 2016, 11:19:20 PM
Does it seem weird to anyone else that components can be mined? Its like digging a hole and finding intel CPUs in the dirt? :)
#10
Just wondered if the performance has improved any in Alpha 13 over Alpha 12 for big colonies.
#11
Ideas / Multithread the AI and pathfinding!
March 24, 2016, 11:21:47 PM
Rimworld should allow those of us with many CPU cores to be able to utilize them, which is more important on bigger maps and with more units.

I know the default Unity pathfinding is extremely poor at utilizing a CPU correctly but it is possible within the Unity framework to multithread pathfinding and other aspects of the AI. Though depending on how the game is written it may be a major effort to utilize more threads. Given how the game is currently I have the feeling you may regret having based the game on Unity, like most successful projects do that get to a certain size.

So I acknowledge that it may never be done because of the potential big effort required, rewriting a lot of the game may not pay off in terms of sales. But since this game isn't on steam yet I thought I might put the idea forward as it could open the game up to an even bigger audience and allow you to be more creative in the final game you deliver. At the very least it will allow mods to do more which will also increase sales. A game like Factorio which had a pretty cult following still ended of doubling its sales , in a 4week period, by going to steam, but you may still get that without the effort of rewriting a lot of it too, so I understand its probably not high on the priority list.

I have done a lock step multithreaded RTS within unity 5 so I am aware of quite a lot of its shortcomings and may be able to offer some advice, if wanted, even on multiplayer if that is a direction you are going. Thanks for a great game anyhow!
#12
Mods / Ultimate Overhaul Mod invisible character glitch
October 01, 2015, 01:48:20 AM
Sorry don't know where else to post this. Maybe this happens in the normal game too sometimes. Anyhow sometimes characters become invisible, but they are still actually there doing work. You can usually see them in only in bed but when you click on them they don't have the usual character selection cross hair. Also those little icons in the mod pack aren't shown for the invisible character either. Usually going back a few saves fixes it, but it's quite annoying. Is there another fix for this? Is there a certain mod known for doing this I can disable? Thanks.