Bioxx

I'm trying to figure out some ways to make that feasible, we'll have to see.

I don't doubt that it is very buggy in 7dtd. In my rather modest experimentation with it here, it is just too much to handle. So for now I've stripped out all of that code. There are just too many variables when we're dealing with a voxel world.

Actually the way it works is fairly simple. When a collapsible block receives an update it begins a recursive scan to test if it is properly supported. First it checks the block directly beneath itself. If it is solid or marked specifically as a support block, then we exit out and assume that the block is fully supported (this prevents unnecessary scanning for a block which gets ticked in a clearly solid location). If this block is not supported vertically, then it begins the recursive scan horizontally and then vertically. So each block N/S/E/W is added to a list to be scanned. Each of those blocks then performs a vertical test. This time if this block is supported vertically, we add the lower block(y-1) to the list and don't bother checking horizontally, otherwise we check each horizontal block to make sure that it can support this type of block, and then add them to the list. This continues until our Y level reaches reaches a predetermined depth or we run out of blocks to scan. The scan is also limited horizontally to whatever the horizontal support range for a block is. So for example if we have a pillar made of stone bricks that is 3 blocks tall, and we build a platform of stone bricks around the top block, any blocks within a 5 block circular radius would be supported as long as they have a valid path back to the pillar (since stone bricks all have a 5 block range). This satisfies the horizontal range, and the pillar, if it is built on solid ground would most likely fill the vertical requirement. Keeping the ranges fairly small makes sure that we aren't scanning hundreds or thousands of blocks multiple times per tick. As far as the cave stone, I actually thought about doing that, but this works perfectly. The main issues that I'm having have to do with load bearing. If I get rid of that, I can make the current setup work quite well. I'm hoping that I don't have to it as its actually kind of fun to make a support beam or span of horizontal bricks collapse due to too much weight. As far as purpose... I like to add things that I find fun to code. But besides that, the tfc1 system was sloppy and the code was confusing as all hell. This is actually was easier to navigate for when I inevitably have to come back to it in 6 months. Mining collapses are a whole other beast. To offset the obvious nerfing of support beams due to naturally supported ceiling, I'll be implementing mining related caveins in a new way that should hopefully compliment the current system. Also I really hope you guys aren't afraid of Earth Tremors.

Honestly, the more complicated that I make the structural systems, the more I want to simplify it. I've been working on a somewhat simplified version of block weight and structural collapsing due to being overloaded but, the more that I make it work, the more scenarios pop up where it no longer functions correctly, so it may be tossed entirely. This sole feature isn't the entirety of TFC2 so at some point I'm going to have to decide to limit whats there so that I can move on to other areas. I also don't want to release something that I'm not happy with just to say that X feature is crossed off. My goal is to give the player simple rules to follow where they have to think just a bit more when building. I don't want the player to have to be a structural engineer to figure out how to build a house. That said if I can make load bearing work, I would love to have some version of it. Just don't get your hopes up.

Appreciate it

Sorry I don't check this thread often. Can you put that folder in a zip file and re-upload? I'd need a pro account to download all of those files in bulk.

Honestly, the protection meter is quite likely the simplest block that we have in TFC. Looking at that crash report, I would say that it is very highly unlike to be a problem specific to the protection meter. It's likely that placing the block was just the straw that broke the camels back. The real issue is probably related to the entity count honestly. If I were you I would backup the world and then load the world in mcedit and remove all the entities in that chunk (mcedit has this ability right?), and then try again. If it works then do it again and remove just a few entities until it works. That 'Already Tesselating' error is in my experience (when not because I was doing some crazy rendering) usually just related to too many things attempting to render at once. @WillOfStone No it does not. It just reads a simple value that we have stored and picks a side texture.

Thanks for the info gxblt, all good ideas

I might be spending all of my free time in Stardew Valley instead of being productive lately.

Agreed, these are all things that I've been planning to look into at some point.

I love hearing stories like that, thanks for sharing! People from so many different backgrounds and situations like TFC, I always find it interesting.

1.9 is the reason why I'm holding off on creating anything related to combat atm. Once I see how the code changes, I'll have a better idea on what is and is not possible within the confines of MC. As far as the tiers, I like it. Originally, i was just going to have damage/health double every tier, but its easier for everyone if we keep the numbers small and use multipliers based on tiers.

This IS an area that I've been pondering on, in regards to progression. The current scaling used in TFC1 clearly isn't satisfactory.

Lisa, tell us if mc launches with JUST forge installed. no TFC

Actually kitty seems to have forgotten that I HAVE rewritten weather. Each island has its own rain map which causes storms to come and go on a per island and per season basis. The system is fully functional on github but needs a lot of polish. That said, floods would be neat, but won't happen. There are just too many issues with making it work well.

I will be making plenty of things configurable, so if you want to sail around using other mods, you're more than welcome. TFC2 just won't have its own mechanics for it. I'd rather spend that time in other areas where I feel like I can provide more gameplay.

One thing that I really want to refrain from doing this time is writing systems that are not well thought out and end up needing to be rewritten every 6 months. My current plan is to get to a point where I have enough basic systems in place to support gameplay on tier 0 islands and we'll release something at that point. Due to the nature of TFC in general, a lot of different systems are connected so each one needs to be in place. On top of that, I'm not always sure how I want to go about doing certain things yet, so it's taking time. More than I would have wanted. I will say this. I don't expect to release anything until MC 1.9. A number of the combat related changes are interesting and I want to be able to take advantage of much of that. Thankfully 1.9 will not be overhauling anything major like 1.8 so Forge will probably be available for 1.9 in a timely manner after its release.

I'd just like to state that TFC2 really isn't that different than TFC1. Most of the same concepts remain. The only difference is that we are attempting to provide the player with a purpose for each playthrough with much more well thought out progression systems. I don't think that magic was never ruled out in TFC1 by me personally, but with my 'make it up as you go' approach, it was just never something that I was looking hard at. Dungeons as well were always intended with our 'move mobs underground' statements. I would say the biggest difference this time is our re-commitment to providing a non-creative gameplay experience. I started to get sidetracked with things like the chisel which had no 'intended' gameplay purposes but ended up being a huge pain to deal with because of multi-block structures and people using it as a way to exploit around the cave-in system, among other things.

Altering the size of tree trunks is actually a real pain which is why we do not do it. Anything smaller than 1m means that the rendering for branches gets very complicated. That said, I've always intended to add moss and things of that nature in TFC1. So I'll add this to my short list of things I can do when I'm struggling on something bigger.

@yeahminecraft16 I got the zip file but haven't taken a look as of yet. Thanks for the submission. @subarctic_guy take a look at the tree world which I have provided in the OP to get a good idea on what is acceptable. Try to stay away from monster trees if you can help it.

This is entirely correct. There is far more to consider with minecraft terrain generation than 'will it look good?'. And yes tfc2 has a sophisticated backend in regards to the island gen model. The hex technique was chosen because it became apparent that using a more realistic looking terrain caused an immense number of problems for things like rivers which would generate on the side of a cliff.

I've perused that paper as well as others a number of times in my quest to figure out terrain for tfc2. What is an acceptable speed for them is not remotely acceptable for minecraft. Additionally, some of what you are describing for placement of trees etc, just isn't really possible in mc or is at the very least much more issue prone than you would think. Traditional terrain generation techniques assume a finite world first of all. This means that various optimizations and assumptions can be used to speed things up. On top of that, the entire world is editable at once. in mc you only have a 16x16x256 area that you are placing things in at a time. This introduces a whole host of issues where that area isn't aware of the areas around it. These are all trials and tribulations that I've experienced when working on terrain gen for tfc1. If you look closely at tfc1 world gen there are tons of problems present, such as the straight line at a certain Z coord where jungle starts generating. I did my best to mitigate that issue but with a non finite world, you just can't. I'm no master coder, nor master at anything else, but I'll admit that I get a little peeved sometimes when people come in and tell me how I should do all these things and that its not that hard (I don't direct this entirely at you entro), as if I didn't spend a great amount of time looking at all these possibilities myself. As a side note, I would HATE playing on the terrain in the picture. Talk about breaking the space bar. If MC blocks were 1/4 their size then it would be fine, but with 1m blocks, no thanks.

Looks cool but that is hand built terrain, probably based off of real world heightmaps. There is only so much that can be done with procedural generation.

And yet there would be complaints. Shades of colors are an issue. I have an idea in mind to get around this however. We'll see once I get to that point.

This is true. I get bitched at whenever I create new color coded things. Stupid colorblind mutants!