TBlazeWarriorT

Here's my suggested code for clif_parse_UseSkillToPosSub. // SKILLUSE MODIFIED CODE int32 skill_success; int32 skill_src_type = 1; if( sd.skillitem == skill_id ) { if( skill_lv != sd.skillitemlv ) skill_lv = sd.skillitemlv; skill_success = unit_skilluse_pos(&sd.bl, x, y, skill_id, skill_lv); skill_src_type = 2; } else { int32 lv; sd.skillitem = sd.skillitemlv = 0; if( (lv = pc_checkskill(&sd, skill_id)) > 0 ) { if( skill_lv > lv ) skill_lv = lv; skill_success = unit_skilluse_pos(&sd.bl, x, y, skill_id,skill_lv); } } if (skill_success == 1) { pc_setreg(&sd, add_str("@skillused_id"), skill_id); pc_setreg(&sd, add_str("@skillused_lv"), skill_lv); pc_setreg(&sd, add_str("@skillused_src"), skill_src_type); npc_script_event(sd, NPCE_SKILLUSE); } // I'll likely be using this or something similar for myself. I even added @skillused_src to know if it was used through an item, since I needed that for my use case. If you don't need it, you can save a few lines. Note that the single target skill code needs to be changed similarly, it uses a more annoying format but the logic is the same. Also to be noted that the single-target code was moved to clif_parse_skill_toid, it seems. Example with this: announce "You have used level "+@skillused_lv+" skill "+@skillused_id+" via "+@skillused_src,bc_self;

I didn't find this available anywhere I spent about an hour doing semi-complex algorithms to compare similar and median pixel colors in several of the newer card bmps that have slightly higher quality, and then manually tweaking them to have a clear top to write text and have clean PNGs to base my cards off of. To make the title: I recommend "Sylfaen" font, size 14, all uppercase, with a 1px white outline and 192 opacity. On Paint.net, I use "Sharp (Modern)" text rendering mode with Anti-Aliasing on. It's not identical to the real card title font which is unknown, but it's almost impossible to notice the difference. To make the number at the top right: copy it from an existing card.

Still works in 2025, thanks 2 important tips: You can toggle debug mode with this (top line is debug, commented. Bottom line is no debug, server boots 10x faster!) ::MSbuild.exe %SolutionPath% /p:PlatformToolset=v143 /m MSbuild.exe %SolutionPath% /p:PlatformToolset=v143 /p:Configuration=Release /m Also, some of the VS Code Setup components have been updated or renamed, here's more or less the ones I used, I don't think anything else is required:

BIG EDIT: If your client supports it (~2022+), just use data\luafiles514\lua files\mapskydata. I found out about this a bit too late, but better late than never. So, here's the basic concept of how to make RGB skyboxes that don't rely on hardcoded skyboxes. I don't see enough people talking about it, so I decided to bring the concept up, as I just did it for two of my custom maps. You can have flat 3D models ~200 units under the main map with a texture, tiled to form a large square, and they'll effectively work as a customizable skybox, going alongside with "clouds2" (or "clouds") mapflag. Pros: Can use any texture/color Can be animated (in theory) Customizable Can be used in any map Not hardcoded/doesn't need to replace any map or skybox Cons: Horizontal camera angles might expose black edges Minor effort editing the map to add the new models Tips: Works well with clouds2 mapflag greywolf/illusion water rsm models are a good base for this, if you don't know how to make a custom model. You will need to change the bottom square texture to be invisible (or the upper one to be opaque), and preferably hex the path to be a new texture so you don't have to replace their textures. Don't use Scale over 5 on the model size (e.g. in Browedit). The tiles will stop rendering properly in some angles if they're too big. Usually a 5x5 grid of 5 Scale models of retextured "ilusion" water at y=200 is enough and produces some of the best results. It's what was used on the images below: I intentionally exposed a small black corner on the image to show it. 90% of the time or if you lock the camera, you can't see the black corners, though.

Bonus: custom art for 'data\texture\À¯ÀúÀÎÅÍÆäÀ̽º\cardbmp\sorry.bmp' made by me (by modifying other assets) sorry.bmp

This will replace the apple sprite on missing items with a question mark based on the question mark emotion sprite. Also includes a bigger version of the question mark texture made by me. To setup this: 1: Replace the apple's '»ç°ú' ACT, SPR and BMP sprites in the data with the ones inside 'data.zip' from this post 2: Add and reference the new 'apple_' textures from 'data.zip' in your itemInfo (so that the apples use the pseudo-custom apple texture instead of also using the '?') Result: Apples still look like Apples. Unknown Items look like a question mark. Less confusion, easier to spot and report! Includes: - »ç°ú.act / spr / bmp - apple_.act / spr / bmp - Partial 'itemInfo' template data.zipunknown.lua

bonus3 bAddEle,Ele_All,x,BF_NORMAL; This is a not so obvious but clever way to increase only normal attack damage through item bonus or player scripts. I was worried I would need to change the source since I didn't find this anywhere, so I assume not many people know about this, and felt like sharing. I had to dig through all ideas and item bonuses to find this hidden gem. This specific use of bAddEle manages to increase all normal attack damage by X against all enemies without affecting skill damage. Explanation: From the documentation, "bonus3 bAddEle,e,x,bf; +x% physical damage against element e with trigger criteria bf", so this was meant to increase damage against specific element enemies. This is the only damage increase with a bf criteria in rAthena's documentation, so this had to be used. So a friend and I thought of only triggering on Normal Attacks, and using Ele_All to affect all enemies, giving us the equivalent of something like a "bAddNormal" that rAthena doesn't have.