raidboss: tweak sidewise spark callouts #471
Conversation
github-actions
bot
added
💬ko
💬cn
raidboss
|
This is a rework of #354 I made it a separate PR because while some of the code is re-used, this is a different solution, and I think it is too dissimilar to the old approach. |
jacob-keller
force-pushed
the
jk-r4n-clone-callout-improvement-v2
branch
2 times, most recently
from
b1c19ec to
059e293
Compare
|
This changes the callout from something like:
to the following:
|
|
I'm also open to making this a strategy option if you don't want to risk changing the existing callouts at all. |
|
Thanks for the re-work on this -- on a quick look, this seems like a clean approach and output. Will give this a run tonight/tomorrow! |
There was a problem hiding this comment.
Had a chance to do some quick testing, and looks pretty good. I think this is close enough to what already exists that it makes sense to implement this as a change rather than a strat option, so your approach seems good to me.
If I'm understanding the implementation correctly, the example you provided above is what will appear if the final 3 cleaves can be collapsed to a single intercard. If the final 3 cleaves cannot be collapsed to a single intercard, R4N Sidewise Spark will just return the full sequence of the three remaining cleaves like so:
My comments below are based on that assumed behavior, so if I've misunderstood or if it's not working as intended, disregard!
ui/raidboss/data/07-dt/raid/r4n.ts
Outdated
| 3: output.num3!(), | ||
| 4: output.num4!(), | ||
| 5: output.num5!(), |
There was a problem hiding this comment.
It looks like the condition block only returns true if it's a single-cleave scenario or if data.storedCleaves.length === 2. So in the 5-cleave mechanic, this trigger will only fire after the second cleave has been collected, and will not fire again. (Which I think is what you intended, since R4N Sidewise Spark is handling the final 3 cleaves).
If that's the case, do we need outputs for 3, 4, and 5, since those will never be used? Wondering if we can simplify this by removing the lookup thru cleaveNums entirely, including just Output.num2 in the outputstrings, and then calling it explicity rather than doing a number-to-string lookup?
There was a problem hiding this comment.
Yea. I started by copying the previous implementation, so I think we can simplify this a lot more.
There was a problem hiding this comment.
Done. This makes the code a lot simpler too.
ui/raidboss/data/07-dt/raid/r4n.ts
Outdated
|
|
||
| return output.combo!({ dirs: dirs.join(output.separator!()) }); | ||
| const cleaves: number = data.storedCleaves.length; |
There was a problem hiding this comment.
| const cleaves: number = data.storedCleaves.length; | |
| const cleaves = data.storedCleaves.length; |
Typescript should infer the number type automatically from .length.
There was a problem hiding this comment.
At one point I had issues with that, but I think that was a different version of the code. Will check.
There was a problem hiding this comment.
This was simplified so we don't even need to save the value.
Ya, right now it only collapses if all 3 can go the same spot. I think that's better than trying to refactor to do NW => E or what not. |
Yeah, I think that's a totally reasonable approach. |
In m4n, we need to take a list of stored cleave directions and convert them into cardinal safe spots. Similar code is used in two places. A future change is going to refactor this logic. Make StoredCleave an explicit type, and introduce a getCleaveDirs helper function which takes the actors and stored cleave data, and returns an array of DirectionOutputCardinal names. This avoids duplicating the somewhat complex logic of looking up actor data and finding the facing. It also makes it easier to extend this logic in the near future. Signed-off-by: Jacob Keller <[email protected]>
In the trigger for the final hit of sidewise spark from Wicked Thunder, the cleave directions from stored cleaves are obtained, and then the final hit is pushed onto the array. This works for now, but will break when extending the logic of getCleaveDirs to support collapsing the callout for intercardinal safe spots. Instead, push the boss's hit onto the stored cleave array first before calling getCleaveDirs. Signed-off-by: Jacob Keller <[email protected]>
Currently, during sidewise spark, the clone cleaves and boss cleave are collected. Two callouts are made. As soon as the first two cleaves are collected, a callout is made for the first two safe spots. Then, once the boss cleave is collected, all safe spots are called out. This is problematic, because players will see callouts like: E => S E => S => W => N => W The actual clone hits are about 2 seconds apart. By the time the final callout is made, the first clone will have already hit, and the second clone will be about half a second away from hitting. This requires players to remember that the data for the first 2 spots is effectively useless, and that they really need to pay attention to the 3 final hits. Instead of calling all 5 hits, just call out the last 3. Based on the raid simulator, we are making a call while the 2nd hit is about 0.5 to 1 seconds from hitting. We *could* refactor to delay this call until the second clone has hit. However, this gives the player only ~2 seconds of reaction time from when we make a call like "W => N => W" where they need to be west. I am unsure which of these is better, but at a minimum we should stop repeating information which makes the callout more difficult to follow. Signed-off-by: Jacob Keller <[email protected]>
The way clones and the boss hit for sidewise spark often results in an intercardinal spot being safe for multiple cleaves. When making a call for a safe spot, if an intercard is safe for the entire sequence, just call that intercard. For the collected callouts where there may be a future callout with a new safe spot, include the number of hits. This should be two, though the current code has support for anywhere from 2 to 5. For the final hit after the boss cleave, just call the direction and "stay", so its clear the player can just move to that intercard and stay without needing to keep track of a count. Signed-off-by: Jacob Keller <[email protected]>
jacob-keller
force-pushed
the
jk-r4n-clone-callout-improvement-v2
branch
from
059e293 to
6929855
Compare
Everything else you pointed out should be fixed now! |
|
This looks good to me, if @wexxlee doesn't have any additional comments. |
|
I'm going to go ahead and approve this one, it seems to be working fine. |
|
based on the recent work in m4s, I discovered that we can actually use ActorControlExtra for getting the clone cleaves. I think that is likely more efficient, but will wait for this to merge to do a follow up PR. |
This is another attempt at improving the sidewise spark callouts that hopefully is acceptable to everyone. I dropped the sequence strategy, and just focused on the collapsing of intercardinals during the two callouts, and on avoiding repeat callout of already called or hit spots. - **raidboss: introduce helper function for stored cleaves** - **raidboss: pass final sidewise spark hit through getCleaveDirs** - **raidboss: avoid re-calling previous clone cleave safe spots** - **raidboss: collapse multiple sidewise spark hits to intercards** --------- Signed-off-by: Jacob Keller <[email protected]> f02a1bc
This is another attempt at improving the sidewise spark callouts that hopefully is acceptable to everyone. I dropped the sequence strategy, and just focused on the collapsing of intercardinals during the two callouts, and on avoiding repeat callout of already called or hit spots. - **raidboss: introduce helper function for stored cleaves** - **raidboss: pass final sidewise spark hit through getCleaveDirs** - **raidboss: avoid re-calling previous clone cleave safe spots** - **raidboss: collapse multiple sidewise spark hits to intercards** --------- Signed-off-by: Jacob Keller <[email protected]> f02a1bc
…n#471) This is another attempt at improving the sidewise spark callouts that hopefully is acceptable to everyone. I dropped the sequence strategy, and just focused on the collapsing of intercardinals during the two callouts, and on avoiding repeat callout of already called or hit spots. - **raidboss: introduce helper function for stored cleaves** - **raidboss: pass final sidewise spark hit through getCleaveDirs** - **raidboss: avoid re-calling previous clone cleave safe spots** - **raidboss: collapse multiple sidewise spark hits to intercards** --------- Signed-off-by: Jacob Keller <[email protected]> f02a1bc
…n#471) This is another attempt at improving the sidewise spark callouts that hopefully is acceptable to everyone. I dropped the sequence strategy, and just focused on the collapsing of intercardinals during the two callouts, and on avoiding repeat callout of already called or hit spots. - **raidboss: introduce helper function for stored cleaves** - **raidboss: pass final sidewise spark hit through getCleaveDirs** - **raidboss: avoid re-calling previous clone cleave safe spots** - **raidboss: collapse multiple sidewise spark hits to intercards** --------- Signed-off-by: Jacob Keller <[email protected]> f02a1bc
This is another attempt at improving the sidewise spark callouts that
hopefully is acceptable to everyone.
I dropped the sequence strategy, and just focused on the collapsing of
intercardinals during the two callouts, and on avoiding repeat callout of
already called or hit spots.