seibaby

1.9 Release Bug Reporting Thread

148 posts in this topic

It was inevitable, so here we go. While fixing my broken random encounter patch, I discovered a little bug with Random Number Good: overworld encounters are not using Think's RNG. What should have been a call to the RNG to determine whether an encounter should be triggered, is instead a static number, causing overworld encounters to happen with astounding regularity. I forgot how the vanilla RNG worked; it does in fact return a different number each time, but it's using a tainted version of vanilla's RNG system (since Think's code overwrites part of the random number table vanilla used). It may still be random "enough" for encounter purposes, but it's not what it should be.

The fix is just one byte long, but I'll let think share an updated version of the patch here for inclusion in the next point release. I'll instead share a set of updated patches for the random encounter mod, with the fix baked in, so that it can be properly tested this time.


Patches (updated to v4)

 

Apply directly to your 1.9.

EDIT: With Nowea's help, I re-jiggered some numbers and also went ahead and applied the encounter mod to the overworld as well. This should have the effect of setting the minimum number of steps before an encounter to 9 in dungeons and 5 on the overworld (excepting high encounter areas like forests and WoR deserts). Obviously, this will also fix the faulty overworld encounter logic, so you should notice a difference in overall encounter rate there (increased, according to Mishrak). Dungeon encounter rates should remain the same.

Edited by seibaby

Share this post


Link to post
Share on other sites

Just to clarify for anyone confused by the title, this was a hack that was intended to go into 1.9.0, but was removed because it wasn't working properly.

Share this post


Link to post
Share on other sites

Not really a bug more than it's a typo that existed in the original script. Dialogue #535 in FF3usME if you're still using it. "musn't" should be "mustn't"

Final Fantasy VI - Brave New World000.png

Share this post


Link to post
Share on other sites
5 hours ago, TiredTeddybear said:

Not really a bug more than it's a typo that existed in the original script. Dialogue #535 in FF3usME if you're still using it. "musn't" should be "mustn't"

Final Fantasy VI - Brave New World000.png

@Bishop need your input here.

Share this post


Link to post
Share on other sites

From what I can see, even though I am not Bishop, "musn't" is technically not incorrect, but is an outdated version of "mustn't"?

Share this post


Link to post
Share on other sites

Trooper casts HasteX on himself and his dog.

Dog pounces on the Trooper as a counter.

Trooper counters with a Bayonet to the dogs face.

All the while, Locke is laughing, stealing Dried Meats, and then killing the Trooper with a Boomerang.

Uhh...were those counters supposed to be a thing? (1.9 with Nowea Hard Type on top)

Share this post


Link to post
Share on other sites

It's because the dogs counter magic. I'd consider it a bug, but I'm too amused at the idea of an abused animal turning on its handler to fix it.

Share this post


Link to post
Share on other sites

I like to think it's more a result of overly aggressive training: The dogs were trained to viciously attack anyone who uses magic on them. They just forgot to tell the soldiers that they trained the dogs to do that.

Share this post


Link to post
Share on other sites

Some serious military mismanagement if the rank-and-file soldiers aren't being properly informed about these vicious dogs & properly trained in handling them.

...Oh, wait. I bet it was Kefka who purposely didn't tell them. Probably killed several people in mid-level management to insure that the memo never passed on to the rank-and-file. And I bet he did it all out of petty vengeance for when he cast magic on one of those dogs and got bit in response. It all makes sense now.

Share this post


Link to post
Share on other sites
On 1/19/2018 at 7:01 PM, vonriel said:

I like to think it's more a result of overly aggressive training: The dogs were trained to viciously attack anyone who uses magic on them. They just forgot to tell the soldiers that they trained the dogs to do that.

 

1 hour ago, thzfunnymzn said:

Some serious military mismanagement if the rank-and-file soldiers aren't being properly informed about these vicious dogs & properly trained in handling them.

...Oh, wait. I bet it was Kefka who purposely didn't tell them. Probably killed several people in mid-level management to insure that the memo never passed on to the rank-and-file. And I bet he did it all out of petty vengeance for when he cast magic on one of those dogs and got bit in response. It all makes sense now.

Honestly that part never made much sense in-setting, as the only two groups in the world who are supposed to have Magic are the Empire and some various Monsters. Of course, the fire-spell casting humans (Slamdancer's) in Zozo also raise so many questions as well given that only the Empire is 'known' to have magic-using humans around.

Share this post


Link to post
Share on other sites

Weighing in on "musn't".  Google says:

Mark Twain, Huckleberry Finn: "So I said it wouldn’t be no use to try to play it any longer, and I would just make a clean breast and tell her everything, but she musn't go back on her promise."

So it's a weird form, perhaps allowable.  But if we're going for consistent Shakespeare, he used "mustn't", with two 't's.  Agree with recommended change.

Share this post


Link to post
Share on other sites

Is that RNG patch being tested by anyone else? I actually don't have any problems with how it performs personally, but the 9 steps until encounter in dungeons part doesn't seem to be accurate. Sometimes it'll just be 4 or 5 steps until an encounter.

Share this post


Link to post
Share on other sites
1 hour ago, kjinn22 said:

Is that RNG patch being tested by anyone else? I actually don't have any problems with how it performs personally, but the 9 steps until encounter in dungeons part doesn't seem to be accurate. Sometimes it'll just be 4 or 5 steps until an encounter.

I've been counting my steps so far, and I've yet to observe an encounter triggering too early.

Share this post


Link to post
Share on other sites
9 hours ago, kjinn22 said:

Is that RNG patch being tested by anyone else? I actually don't have any problems with how it performs personally, but the 9 steps until encounter in dungeons part doesn't seem to be accurate. Sometimes it'll just be 4 or 5 steps until an encounter.

Which dungeons have you observed this in?

Share this post


Link to post
Share on other sites

I reapplied the patch, so will try again and make sure to note the location if I run into the same issue. It was early as the cave to South Figaro though

 

Edit: Yeah, reapplied, started a new game, and ran into it again. The last area when you're on the cave from Figaro Castle to South Figaro I only had 4 steps before going into an encounter. Everything else seems fine.

Edited by kjinn22
Updated info

Share this post


Link to post
Share on other sites

Are you counting steps in that room, or steps total? The step counter (essentially) only resets upon a random encounter.

Share this post


Link to post
Share on other sites

Steps between encounters, although I'm going out of my way to get into random encounters to test this

Edited by kjinn22
Little more info

Share this post


Link to post
Share on other sites

Testing the random encounter patch. Counted every single step in between battles. Avoided all forests. The first half before the break is for v3 of the patch. Seibaby then made an adjustment, the second half is for v4.

Basic summary of results: looks like it's working now!

Shit, can't upload Excel, so here's a screen shot of v3 (boxes 1-19) and v4 (boxes 21-45)

Spoiler

Encounters 1.pngEncounters 2.png

Edited by Deschain

Share this post


Link to post
Share on other sites

Wanted to make sure, so I went through FC. Still working!

Spoiler

Encounters3.png

Note, at the end of FC, with running from maybe 5% of total battles, my characters are level: 17, 15, 14, 19, 17, 14, 13, 12, 14, 12, 9

Also, before the patch, the average number of steps between encounters, overall, was 24.9. After the patch it was 33.1

Edited by Deschain

Share this post


Link to post
Share on other sites

X-Zone procs from Doomstick seem to possess the "revives undead" property. Had an undead enemy come back from the dead from a proc. In addition a not... dead... undead 'died' and came back as if it was hit with Doom

EDIT: Also dog block animations sometimes play for random characters blocking. None of them have procced reprisal yet so it appears to be purely visual.

EDIT2: Lastly, the equipment screen is behaving very oddly, not properly displaying stat changes in an inconsistent way (not only when using Y to swap between relics and non-relics). I think I've only seen it in relics so far.

Edited by Nowea

Share this post


Link to post
Share on other sites

When a character with imp status uses a Phoenix Down, it revives the target character with 0 HP (I'm assuming due to the half-effectiveness of items used by imped characters, even though 0 is not half of 1). I don't believe this adversely affects anything else. The revived character can still die again on a subsequent hit, and they can still be cured up to a normal HP level. So it's non-gameplay-impacting as far as I can tell. Just odd. May not even be worth fixing, but I thought it was worth mentioning.

Share this post


Link to post
Share on other sites

Create an account or sign in to comment

You need to be a member in order to leave a comment

Create an account

Sign up for a new account in our community. It's easy!

Register a new account

Sign in

Already have an account? Sign in here.

Sign In Now