A GUI program for the CaveFinder project, for searching deep caves. This program works for both Java and Bedrock Edition 1.18+.
This program requires Java 17 or above. Make sure you have downloaded Java.
You can right click the jar file and choose "Java(TM) Platform SE binary" as the open method.
You can also use cmd to run it: open cmd.exe (you can search it in the search tab) , and use the following commands to run it:
Type "java -jar" then type space, drag the jar file to the window and type enter.
Type "java -Xms2048m -Xmx4096m -jar" then type space, drag the jar file to the window and type enter. This is for allocating memory.
You can change the -Xms2048m -Xmx4096m to the memory amount you want. But I suggest to allocate maximum memory 4GB or more for this program. (It's memory-consuming)
You have to wait 10-20 seconds to fully open it. This is because it need to initialize Seedchecker first, which will take a bit long time. In this stage don't do other things.
Use 语言(Language) on top-right to switch language. You can choose Chinese or English.
Use the Start Filtering button to start filtering, and Stop Button to stop filtering.
The top-left part is Parameter Settings, including:
Cave Depth: The cave depth you want, default is -50, range is from -50 to 0.
Thread Count: The thread amount you want to use, range is from 1 to your computer's thread amount.
X/Z coordinates: The coordinate you want to search for caves, range is from -30,000,000 to 30,000,000.
Segment size: Only for incremental mode. Segments if the total seed amount is greater than the value. It depends on your memory allocation. For 4GB memory allocated, the suggested segment size is 5-10 million. If you have more memory, you can try greater segment size. For structureseed search you probably won't care about this, because it searches for 65536 worldseeds for every structureseed and you won't reach the segment size within a few days.
Check Height: Import Seedchecker to check the exact height. This is slower, but much more consistent on giving deep caves. The higher the height, the slower it'll be. (e.g. y=0 is slower than y=-10)
Filter BE impossible seeds: This is a currently hardcoded mode for searching "impossible seeds" on Bedrock Edition. To search for this, you need to set the X/Z coords to (0,0), and I suggest you to enable Check Height for this. (The conditions are strict so it won't be much slower to enable check height) This condition will disable "Entrance1 only" option (Because it's already Entrance1 only) and ALL Biome Climate Parameters options (It has the biome climate conditions itself). This is a fast condition, but it might take you over an hour to find a potential seed, and dozens of hours to find an impossible seed (Spawn in lava and always respawn in lava, which requires no waterfalls and topsolid blocks within the spawn radius)
Entrance1 Only: It's faster, and more likely to give you larger exposed caves (will ignore some small deep caves).
The second part is Height Check Options. This part only available when you enable Check Height.
Height Type: Surface Height means the height of the top of highest non-air block. Choose this if you want to get a non-waterlogged cave (you need a non-waterlogged cave for low spawns and low surface structures). Underwater Height means the height of the highest non-water block in a waterlogged cave (like in a river or an ocean), it corresponds to the depth of water. If you want to have very deep waterlogged cave, you would better to set the AquiferFloodLevelFloodness to Greater Than and set a higher value for it (like 0.5 or 0.7).
Check Height In Range: You can set the range of height checking by editing Range Coordinates. The format of it is "minX minZ maxX maxZ". The coords are relatived to your search coords. For example, if you set it to "-5 -3 5 3" and your searching coordinate is set to (X,Z), it will check height within the coords ranged from (X-5,Z-3) to (X+5,Z+3). The coords are exact position, not block position, so if you need to check height within the block position (X-5,Z-3) to (X+5,Z+3), you should set it to "-5 -3 6 4" instead. The range can only set within ±16 blocks. You can set the Range Height Type to check the lowest, average or highest height in the range.
Next part is Filter Mode.
Incremental: Incrementally searching from a seed range in "Seed Input".
Filter from list : Searching from a seed list. You can load it from file.
StructureSeed(low48bit) : Searches for the seeds in the range/seed list and their sister seeds (low 48 bits same but high 16 bits different, 65536 in total). Sister seeds has same potential structure coords and almost same nether and end dimension. Choosing this is much better if you want to search for a structure in cave (e.g. villages, outposts, igloos) , just use cubiomes-viewer to search for low-48 bit seeds that has these structures and load them as a file into this program.
WorldSeed : Only Search for the seeds in the range/seed list.
The Bottom-left part is Seed Input.
Start Seed & End Seed: For Incremental mode. You don't need this for Filter from List mode.
Seed List (one per line): For Filter from List mode. It's the seed list you want to search. You can load a seed file into the program (1 seed per line, no other characters or symbols). It's not suggested to load a seed list with over 10 million seeds. You don't need this for Incremental mode.
The top-right part is Biome Climate Parameters.
It includes 8 types. Temperature, Humidity, Erosion, Weirdness, Continentalness determines the biome. It's better to add these climate parameters if you want to search an exposed cave in specific biome. They are searching in order of fastest to slowest in the program (Temperature, Humidity, Erosion and Weirdness are faster than complete cave entrance checking, though erosion and weirdness are still slower than entrance1). You can search on the Minecraft Wiki to look for the biome climate parameter condition for the biome you need. Structures' generation also determines on biomes. Weirdness betweeen -0.05 and 0.05 mostly means rivers, Continentalness below -0.19 means ocean, if you want to search for non-waterlogged cave, I suggest you to exclude weirdness values between -0.16 and 0.16 and continetalness below -0.11, which is the default settings.
Entrance, Cheese and AquiferFloodlevelFloodness aren't Biome Climate Parameters, but they are still here for advanced searching. Normally, Entrance and Cheese <0 means caves, and 99% non-waterlogged exposed caves generates at AquiferFloodlevelFloodness below 0.4. Lower Entrance and Cheese might means larger caves, and lower AquiferFloodlevelFloodness might means less chance to get waterlogged caves.
The bottom-right part is the log, which shows information while searching.
Below these parts, you can see an Export Path, which is the filtered list you want to export to. If that list already exists, the program will show a warning to you: "Result file already exists and will be overwritten. Continue?"
And at the bottom of the GUI there's a progress bar which shows the finished seed amount and searching speed.
https://github.com/KalleStruik/noise-sampler
https://github.com/jellejurre/seed-checker/tree/1.18.1
https://github.com/SunnySlopes
https://github.com/KalleStruik
一个用于搜索很深的露天洞穴的GUI程序。这个程序在Java版和基岩版1.18+基本通用。
此程序需要 Java 17 或更高版本。 请确保您已下载 Java。
您可以右键单击 jar 文件，并选择 "Java(TM) Platform SE binary" 作为打开方式。
您也可以使用 cmd 来运行：打开 cmd.exe（您可以在搜索栏中搜索它），然后使用以下命令运行：
输入 "java -jar"，然后输入空格，将 jar 文件拖到窗口中并按回车。
输入 "java -Xms2048m -Xmx4096m -jar"，然后输入空格，将 jar 文件拖到窗口中并按回车。这是用于分配内存的。
您可以将 -Xms2048m -Xmx4096m 更改为您想要的内存大小。但建议为此程序分配最大 4GB 或更多的内存。（它消耗内存较多）
您需要等待 10-20 秒才能完全启动这个程序。 这是因为它需要先初始化 Seedchecker，这会花费较长时间。在此阶段请不要进行其他操作。
用右上角的 语言(Language) 切换语言。你可以选择中文或英文。
点击开始筛选按钮启动筛选，点击停止按钮结束筛选。
左上角为参数设置区域，包含：
洞穴深度：期望的洞穴深度，默认值为 -50，有效范围为 -50 至 0。
线程数：使用的线程数量，有效范围为 1 至您计算机的最大线程数。
X/Z 坐标：搜索洞穴的坐标值，有效范围为 -30,000,000 至 30,000,000。
分段大小：仅递增模式适用。当总种子数超过该值时进行分段处理，具体数值取决于内存分配情况：
· 分配 4GB 内存时，建议设置 500 万至 1000 万 · 若内存更大，可尝试更高分段值 · 结构种子搜索通常无需关注此参数（因每个结构种子需遍历 65536 个世界种子，连续运行几天之内筛的种子数量都不会达到分段大小）
筛高度：导入Seed-checker检查精确高度。此模式速度较慢，但能保证稳定发现深层洞穴。检查高度值越大，速度越慢（例如 y=0 比 y=-10 更慢）
筛基岩版无解种子：当前是一个硬编码的筛选模式，用于搜索基岩版“无解种子”。使用时需要：
· 将 X/Z 坐标设为 (0,0) · 建议开启筛高度功能（因筛选条件严格，开启后速度衰减不明显） 注意：此条件将自动禁用只筛Entrance1选项（因为这个模式本来就只筛Entrance1）与所有群系气候参数选项（使用程序内置条件）。这个条件筛得很快，但可能需要一个小时以上才能发现潜在种子，几十个小时才可能筛到真正的“无解种子”（出生点位于岩浆且重生后始终在岩浆中，需满足出生半径内无水流且无露天固体方块）
只筛Entrance1：速度更快，更容易发现大型露天洞穴（会忽略部分小型深层洞穴）
第二部分是高度检查选项。此部分仅在启用检查高度时可用。
高度类型：地表高度指的是该位置最高非空气方块顶部的高度。如果您要筛一个非含水洞穴（非含水洞穴对于筛选洞穴底部的出生点和地表结构来说是必要的），请选择此选项。水下高度指的是含水洞穴（如河流或海洋中）中最高非水方块的高度，它和水深相对应。如果您要筛选一个非常深的含水洞穴，建议将AquiferFloodLevelFloodness设置为大于某值，并为其设置一个较高的值（如0.5或0.7）。
检查高度范围：您可以通过编辑范围坐标来设置高度检查的范围。其格式为 “minX minZ maxX maxZ”。这些坐标是相对于您的筛选坐标的。例如，如果您将其设置为“-5 -3 5 3”，而您的筛选坐标设置为（X，Z），则它将检查从（X-5，Z-3）到（X+5，Z+3）的坐标范围内的高度。这些坐标是精确坐标，而不是方块坐标，因此，如果您需要在方块坐标（X-5，Z-3）到（X+5，Z+3）的范围内筛选高度，则应将其设置为“-5 -3 6 4”。 您只能将范围坐标设置在±16格以内。您可以将范围高度类型设置为检查范围内的最低、平均或最高高度。
接下来是筛选模式区域：
递增模式：根据“种子输入”区的种子范围进行递增搜索
列表筛选：基于种子列表进行搜索，支持从文件加载列表
结构种子（低48位二进制）：在指定范围/列表内搜索种子及其姐妹种子（低48位二进制相同，高16位二进制不同，共65536个）。姐妹种子具有相同的结构坐标和近乎一致的下界/末地维度。若需筛选在洞穴中的结构（如村庄、前哨站、雪屋），推荐使用此模式（可先用 cubiomes-viewer 筛选含目标结构的低48位种子，再导入本程序处理）
世界种子：仅搜索指定范围/列表内的原始种子
左下角为种子输入区域：
起始种子&结束种子：递增模式专用，列表筛选模式无需使用此功能。
种子列表（每行一个）：列表筛选模式专用。支持导入种子文件（每行仅含一个种子，无其他字符）。不建议加载超过1000万种子的列表。递增模式无需使用此功能。
右上角区域是群系气候参数。
它包含8种类型。温度、湿度、侵蚀度、奇异性、大陆性 决定了生物群系的生成。如果你想在特定的生物群系中寻找露天洞穴，最好筛选这些群系气候参数。程序会按照从最快到最慢的顺序进行搜索（温度、湿度、侵蚀度和奇异性的检查速度比完整的洞穴入口检查更快，尽管侵蚀度和奇异性仍然比Entrance1慢）。你可以在Minecraft Wiki上搜索你所需生物群系的气候参数条件。结构的生成也取决于生物群系。Weirdness在-0.05到0.05之间通常代表河流，Continentalness低于-0.19代表海洋，如果你想筛选非含水洞穴，建议像默认设置那样排除奇异性在-0.16到0.16之间的区间，且排除大陆性低于-0.11的区间。
洞穴入口噪声、芝士洞穴噪声和含水层洪水水位噪声 并非生物群系气候参数，但它们仍在这个区域里面，用于高级搜索。通常，洞穴入口噪声和芝士洞穴噪声小于0代表有洞穴，而99%的非含水露天洞穴生成在含水层洪水水位噪声低于 0.4 的位置。较低的洞穴入口噪声和芝士洞穴噪声值可能意味着更大的洞穴，而较低的含水层洪水水位噪声值可能意味着洞穴含水的几率更低。
右下角部分是日志，它在搜索过程中显示信息。
在这些部分下方，你可以看到导出路径，这是你想要将筛选后的列表导出到的路径。如果该列表已存在，程序会向你显示警告："结果文件已存在，将被覆盖，是否继续？"
在 GUI 的底部有一个进度条，它显示已完成的种子数量以及搜索速度。
https://github.com/KalleStruik/noise-sampler
https://github.com/jellejurre/seed-checker/tree/1.18.1