วีดีโอ

dang i missed another one

Nice stream

Someone who is playing the Sky games! Blesss~ I love the music at the start and your avatar btw. Dropped a like for you~

🤩 "PromoSM"

I also started with the Ghost Lancer in my first opening. It has been a pretty big life saver a few times

You couldnt go to main phase 2 because you didnt have a battle phase its just part of the rules

I started a crafting questline with cute, miniature bunny people and ended up shedding tears. That's FFXIV for you.😂

Love that emote!!

LOL!!!!!

Damn I just missed ya

It will be wicked if the mechazord is named Shang-Ira (also that pose sequence is nice. You nailed it👍)

Oh please someone do all the rangers color😭😭😭

Island Sanctuary looks like so much fun. I love Gathering and collecting random things, crafting parts for base building and development, and exploring the wilds. It feels perfect for me.

Haha I'd be too scared to try ranked xD

Hello Rita, sorry for going but here is what you can do to help with the dragon maid deck: -Add impermanence (from bundle pack). Droplet is optional -1or2 dark metal dragon for darkness metal dragon (run 1) -1laundry,2or3 kitchen,3 parlor & chamber (core cards) -Noctovision dragon (2or3), helps with link -most only play one of,ernus,tinkhec and lopar. This is up to you -3hospitality,1or2 changeover (most use only 1, welcome 1or2 and tidying at3. -add 2 striker dragon & hieratic seal Hope this helps

listening to this in the background while I'm working on my own "Tag"-type game in forge (I wanted to follow up on that comment to see how things turned out), sharing a few thoughts: * general thought - You duplicated the collision script and had each operate on half, and that seemed to have improved things? I wonder if each script-brain is potentially running as its own thread, that could account for a speed-up... I don't have any sense of what (if any) Forge's guarantees are in terms of parallel node-graph execution, but my suspicion is that it runs single-threaded because of the sorts of the issues/bugs that users would experience scripting in an implicitly-multi-threaded environment, so if splitting it up resulted in a speed-up, that surprises me * node graph suggestion - I've used of "custom events" as a parallel to "calling a function" in a program, which I found useful for common functionality, that might be helpful with the duplicated collision script - if you pull out the collision logic from both of them (and connect it to an "On custom event (global)" node (and you could pass the players you're computing the collision for as an object list, essentially the function arguments), you can "call" the collision computation/updating code by adding a "trigger custom event (global)" in the collision code; then you only have the actual computing-a-collision stuff in one place and you don't have to remember to change it in both scripts * a testing suggestion - they mentioned in one of the forge scripting videos that if you haven't built nav data, bots won't move. from my own testing (on the Void map), if you have nav mesh data built, you can place a nav-cutter and scale it up to cover all ground on the map, build nav data, and then friendly-bots won't move. Enemy bots will move/attack you if they see you, but if you apply the "prevent firing" trait to them then they'll run up to you when they see you, which can also be useful * for another pretty wild approach for collision detection, you might be able to "pin" a small, phased-physics object inside each player's model (by repeatedly updating those object's positions to their target player's positions), then using a boundary on that object (to create an area-monitor), theoretically you'd be able to use the "On object entered area" events for that object's area monitor to drive collision detection, which would be very efficient. There's a _lot_ of fiddly stuff with that approach, like how you link each monitor-object to each player according to how many players there are, how you actually set up the "event listeners", whether moving an object with an area monitor is possible/re-triggers for objects already inside the area before the move... I was thinking about it for my "Tag" map, but for now I've decided against trying it because I don't need it to be that efficient

I skipped around a lot so you may have made note of this in the video at some point, but if we make the assumption that the result of the "get all players" node returns the same view of the players within the context of one execution of the script (which could be not true, for example in the event of players connecting/disconnecting, I'm not sure how it works in that case), then when your inner-loop's iteration index is less or equal to your outer loop's iteration index, you can skip the check because you already checked the reverse case. For example currently if the input list of players is is "a, b, c", the nodes I see at 1:31:59 make the collision checks for "aa, ab, ac, ba, bb, bc, ca, cb, cc" and you're currently skipping the checks "aa, bb, cc" with the "is the player the same" branch. In this example, after you check "ab, ac, bc" you can skip the corresponding checks "ba, ca, cb", which should speed things up by a factor of ~2.

Thank you for posting this!!! :D

I only realised until the VOD that my mic is blowing out.... I guess that's what happens when you don't properly check your settings after a move.... ah well it happens. I will make sure to fix it up next stream. Such a shame, these quests were so good.

PЯӨMӨƧM

Thx for uploading!! You are late to the party - Promo'SM !