Sword Triggers Are Not Continuous

I've been debating reporting this for a while b/c it seems like something that would be more of a pain for you guys than something that would actually make ZC better. Some of what I'm about to say here is just my best guess as to what's happening. There's a bug in there somewhere but I'm not sure exactly what it is or how to fix it? I don't think I've tested if there's been a change in behaviour since original 2.5.

All trigger flags are 'continuous' except for sword triggers (I only tested this with looping secrets and not tiered secrets in general, but I'd guess it does apply more generally). When a sword trigger is used in a way where the continuousness of other triggers would apply, in addition to not behaving in a continuous manner, for some reason screen secrets are not activated. It's especially strange when the strike trigger is involved, because its behaviour differs depending on whether or not you hit it with a sword. I think it's a good bet that this inconsistency with sword triggers was introduced along with the slash and slash/continuous combo types, but I don't know that for sure.

The fix might be to make sword triggers continuous again. Unless all triggers didn't used to be continuous in past versions. If that's the case, then the fix is to make non-continuous triggers actually activate screen secrets and then do the same that was done for slash combo types to all trigger flags (continuous and non-continuous versions of them all).

Maybe I'm just stupid and couldn't solve the room, but there was a room in Lost Isle I cheated my way through. It had looping tiered secrets that, if there is actually a version inconsistency and if the room was implemented in a certain way (I have no way to check how it was actually implemented), might have made the room unsolvable in 2.5? (I vaguely recall a sky dungeon, and I think the room was maybe near the upper-right corner of the map? But maybe plenty of people have played through Lost Isle on 2.5 with no problems and I'm just dumb for not figuring it out. The possibility of a version inconsistency that may or may not exist was in the back of my mind the whole time I was in that room and it was pretty distracting. Me being dumb is the more likely explanation here I think lol.)

Quote:

---

Originally Posted by CSD

I've been debating reporting this for a while b/c it seems like something that would be more of a pain for you guys than something that would actually make ZC better. Some of what I'm about to say here is just my best guess as to what's happening. There's a bug in there somewhere but I'm not sure exactly what it is or how to fix it? I don't think I've tested if there's been a change in behaviour since original 2.5.

All trigger flags are 'continuous' except for sword triggers (I only tested this with looping secrets and not tiered secrets in general, but I'd guess it does apply more generally). When a sword trigger is used in a way where the continuousness of other triggers would apply, in addition to not behaving in a continuous manner, for some reason screen secrets are not activated. It's especially strange when the strike trigger is involved, because its behaviour differs depending on whether or not you hit it with a sword. I think it's a good bet that this inconsistency with sword triggers was introduced along with the slash and slash/continuous combo types, but I don't know that for sure.

The fix might be to make sword triggers continuous again. Unless all triggers didn't used to be continuous in past versions. If that's the case, then the fix is to make non-continuous triggers actually activate screen secrets and then do the same that was done for slash combo types to all trigger flags (continuous and non-continuous versions of them all).

Maybe I'm just stupid and couldn't solve the room, but there was a room in Lost Isle I cheated my way through. It had looping tiered secrets that, if there is actually a version inconsistency and if the room was implemented in a certain way (I have no way to check how it was actually implemented), might have made the room unsolvable in 2.5? (I vaguely recall a sky dungeon, and I think the room was maybe near the upper-right corner of the map? But maybe plenty of people have played through Lost Isle on 2.5 with no problems and I'm just dumb for not figuring it out. The possibility of a version inconsistency that may or may not exist was in the back of my mind the whole time I was in that room and it was pretty distracting. Me being dumb is the more likely explanation here I think lol.)

---

This might be a bug, or it may be an oversight. I do not recall if sword triggers were ever continuous, but are you saying that strike triggers are continuous if hit by something other than a sword; or that sword triggers are continuous without slash?

You can always try running the 'Lost Isle Build' to see if there is a mechanical difference, and if there is, and yo document it (video is best), then we can work to re-implement it.

switch(zc_2.11b) {
case @War Lord :
case @DarkDragon :
case @Gleeok :
case @Saffith :
case @DarkFlameWolf :

Do any of you know if sword triggers were continuous in 2.11 or early 2.5?
break;

default: I don't remember. break; }

I'm saying that the strike trigger, ONLY if a sword is used to trigger it, is NOT continuous. It IS continuous when hit with anything other than a sword.
It's strange to me also that these 'noncontinuous' sword triggers do not activate screen secrets.

Here's an old bug report, coincidentally from the very person who made the room in lost isle that I was talking about earlier. This lost isle stuff isn't relevant, btw.
http://www.armageddongames.net/showt...ehavior-(b816)

It deals with the addition of the slash (continuous) combo types. Apparently, slash combos used to be continuous, they weren't anymore, and so the new combo types were added. Perhaps the same incompatibility bug also applied to sword triggers, but it was never noticed and corrected? That's my guess. Or I suppose the addition of the slash (continuous) combo types somehow changed the way sword triggers function.

(That room in lost isle does have a slash (or sword trigger) in it, but the room works fine, it was just a mechanic that wasn't explained at all, and maybe I was out of arrows IDK.)

I'm not sure, unfortunately. But if it breaks Lost Isle we should fix it. I've played through Lost Isle on a recent build and got fairly far without encountering any problems,so if there is an issue it's in the last few dungeons.

It doesn't. That room in question does have sword triggers in a tiered secret but I was randomly supposed to guess to use arrows on it; I thought it might have been a broken continuous sword trigger.

This glitch is a weird inconsistency, especially when the Strike trigger is involved. And also it is very strange that the sword trigger will not activate screen secrets when used in a situation that would normally be continuous. I would call both of these things a bug, regardless of whether it ends up being a version inconsistency, but if it's consistent then I guess there's not much point in mucking around with it.

By coincidence, it was Peteo that reported the bug where slash combos were no longer continuous, in 2008. Nobody seems to know what the behaviour was prior to that inconsistency being introduced, but I guess there are a lot of ways we could have ended up here. Perhaps the same bug was introduced with sword triggers and it was not noticed and corrected. Perhaps the change to fix the slash combos introduced a new bug. Perhaps there always used to be a difference in the way that sword triggers and slash combos worked and an old change unified the code between the two and introduced the version inconsistency, but if that's the case, why do sword triggers work different from other triggers? Just throwing out some random speculation... Nobody seems to know.

Quote:

---

Originally Posted by CSD

It doesn't. That room in question does have sword triggers in a tiered secret but I was randomly supposed to guess to use arrows on it; I thought it might have been a broken continuous sword trigger.

This glitch is a weird inconsistency, especially when the Strike trigger is involved. And also it is very strange that the sword trigger will not activate screen secrets when used in a situation that would normally be continuous. I would call both of these things a bug, regardless of whether it ends up being a version inconsistency, but if it's consistent then I guess there's not much point in mucking around with it.

By coincidence, it was Peteo that reported the bug where slash combos were no longer continuous, in 2008. Nobody seems to know what the behaviour was prior to that inconsistency being introduced, but I guess there are a lot of ways we could have ended up here. Perhaps the same bug was introduced with sword triggers and it was not noticed and corrected. Perhaps the change to fix the slash combos introduced a new bug. Perhaps there always used to be a difference in the way that sword triggers and slash combos worked and an old change unified the code between the two and introduced the version inconsistency, but if that's the case, why do sword triggers work different from other triggers? Just throwing out some random speculation... Nobody seems to know.

---

@_L_ and @jman2050 might remember what was going on at that time. 2008 is that grey period in which I lost hope of 2.5 ever being finished, and went poof for a few years.

I do not have snapshots of the source throughout the 2.5 beta cycle, so I have no way to go back and look at how it used to work.

sword and wand triggers are not continuous, and never can be.
They die after a set amount of frames; no exceptions, do to them updating Link's animation as well.
All the other flags kill the weapon on contact with the exception of strike, which is the exception IIRC.

Quote:

---

Originally Posted by Tamamo

sword and wand triggers are not continuous, and never can be.
They die after a set amount of frames; no exceptions, do to them updating Link's animation as well.
All the other flags kill the weapon on contact with the exception of strike, which is the exception IIRC.

---

This is probably true with the present handling of linkclass weapons. I still do not know, and it would be good to know, if they ever worked as continuous triggers in the past.

Adding this feature would need to occur after a lot of other refactoring happens. If older quests did use them, and they did exist (and thus the quests are unplayable), we need to know about it.

Sword triggers were continuous in 2.10, but they aren't in 2.50.0. I don't know when exactly it changed. Shouldn't be hard to figure it out, I just can't check the logs right now.
In LinkClass::check_slash_block(), comment out the lines

Code:

---

if(!isTouchyType(type)) set_bit(screengrid,i,1);

---

and

Code:

---

if(!isTouchyType(type)) set_bit(ffcgrid, current_ffcombo, 1);

---

and they should become continuous again. screengrid and ffcgrid are used to keep track of which combos have already been checked for the current attack.

And this is why i dont support backwards compatibility, you devs wont make up your minds :s

Quote:

---

Originally Posted by Saffith

Sword triggers were continuous in 2.10, but they aren't in 2.50.0. I don't know when exactly it changed. Shouldn't be hard to figure it out, I just can't check the logs right now.
In LinkClass::check_slash_block(), comment out the lines

Code:

---

if(!isTouchyType(type)) set_bit(screengrid,i,1);

---

and

Code:

---

if(!isTouchyType(type)) set_bit(ffcgrid, current_ffcombo, 1);

---

and they should become continuous again. screengrid and ffcgrid are used to keep track of which combos have already been checked for the current attack.

---

Do you happen to know if this was intentionally blocked, or if it was merely an oversight?

I obviously don;t want to introduce a new and different bug by changing it. i could supply a quest header requirement to these instructions, to fix 2.10 quests; but if continuous triggers are in fact broken, as @Tamamo describes--I accepted her explanation for the present, as it seemed plausible--then they need some kind of reworking to be viable.

I have some preliminary work to do on the flag editor components, which amount at present to little more than an experiment, but if they prove, then I will think about adding continuous flags for this type of event.

Thanks to Saffith pointing out this portion of code I had forgotten I decided to investigate things. The screengrid array (and later also the ffcgrid array) was designed to ensure that activating slash and pound combos via the sword or hammer respectively would occur only once per attack. Without it, if you had, say, a series of pound combos that linked into each other requiring the player to use the hammer multiple times, the successive pound combos would simply keep activating on every frame that the initial attack was active. I recall this being an issue at some point before 2.10 but I could be wrong. I believe the isTouchyType refers to the continuous variants of those combos.

Anyhow, as far as I can tell looking over my archive of code, this was the only purpose for that code, and in 2.10 and almost all the 2.11 betas the set_bit statement was nested right where the code would change a slash combo to its undercombo or next combo. I can't look at the svn logs, but the clear change between this behavior and the current behavior seems to be when I implemented freeform combos. What happened is that the set_bit statement now resides outside of the code that checks for the exact type of flag or combo that was struck. Where previously set_bit would only evaluate in the specific instance of a slash or slash->next combo being struck and changed into the undercombo or next combo respectively, now set_bit will evaluate at any time a successful cuttable block or trigger is struck with the sword.

As to whether this was an intentional change or not, I can't say without looking at the svn logs as my memory of ten years ago is understandably spotty. Heck, I can't even be 100% sure I was the one who made this specific change. However, looking at the code now, I/m reasonably sure that the thinking was that the screengrid tracking only prevented continuous slash blocks from occurring in those two specific cases mentioned above. Slash combos that changed into the 16-31 secret flags or the armos secret would still exhibit continuous behavior if they changed into more slashable blocks, and then of course there's the very obvious case of a sword trigger changing into itself to create a toggle trigger which, without this change, would trigger on every frame that the sword was active. I definitely remember that being an issue in some 2.10 quests that attempted such functionality using the trigger flags. And yet in hindsight it makes sense that there might be some use cases where a continuous trigger would be desired, and certainly some 2.10 quests were designed with this limitation in mind. Incidentally everything I just said applies to pound combos and triggers as well.

I'm not sure what Tamamo is talking about in that sword and wand triggers can never be continuous. Unless he means that the sword and wand animations eventually end which obviously they do.

As for what to do about it, if it's really necessary to maintain compatibility with 2.10 quests that expected the continuous behavior, then fixing it should be trivial, so long as we can identify the exact version of the betas the change occurred in. For moving forward having continuous versions of the sword (and hammer) trigger flags to go along with the continuous slash combos currently in use seems like the best way to fit both use cases without breaking any 2.5 functionality.

Quote:

---

Originally Posted by CSD

I've been debating reporting this for a while b/c it seems like something that would be more of a pain for you guys than something that would actually make ZC better. Some of what I'm about to say here is just my best guess as to what's happening. There's a bug in there somewhere but I'm not sure exactly what it is or how to fix it? I don't think I've tested if there's been a change in behaviour since original 2.5.

All trigger flags are 'continuous' except for sword triggers (I only tested this with looping secrets and not tiered secrets in general, but I'd guess it does apply more generally). When a sword trigger is used in a way where the continuousness of other triggers would apply, in addition to not behaving in a continuous manner, for some reason screen secrets are not activated. It's especially strange when the strike trigger is involved, because its behaviour differs depending on whether or not you hit it with a sword. I think it's a good bet that this inconsistency with sword triggers was introduced along with the slash and slash/continuous combo types, but I don't know that for sure.

The fix might be to make sword triggers continuous again. Unless all triggers didn't used to be continuous in past versions. If that's the case, then the fix is to make non-continuous triggers actually activate screen secrets and then do the same that was done for slash combo types to all trigger flags (continuous and non-continuous versions of them all).

Maybe I'm just stupid and couldn't solve the room, but there was a room in Lost Isle I cheated my way through. It had looping tiered secrets that, if there is actually a version inconsistency and if the room was implemented in a certain way (I have no way to check how it was actually implemented), might have made the room unsolvable in 2.5? (I vaguely recall a sky dungeon, and I think the room was maybe near the upper-right corner of the map? But maybe plenty of people have played through Lost Isle on 2.5 with no problems and I'm just dumb for not figuring it out. The possibility of a version inconsistency that may or may not exist was in the back of my mind the whole time I was in that room and it was pretty distracting. Me being dumb is the more likely explanation here I think lol.)

---

I added an Emulation Menu toggle for this to 2.53 Beta 36. Care to test to see if it does what it needs to do?

"I added an Emulation Menu toggle for this to 2.53 Beta 36. Care to test to see if it does what it needs to do?"
Suffice to say I don't check this forum very often... Honestly I reported this so long ago at this point I don't even remember how to set this up.