Number Six

Yes, those require a small bit of extra work to split the models since a lot of the shoulders have left and right pieces in a single geo together, but are on the radar for the next phase of the project.

Overcomes KB protection and applies higher downward velocity when used on flying targets. It's not a mechanic that exists in the current live build -- it was a hypothetical question. Wasn't sure how many people used it on the ground (where the KU bounce has more of an effect) versus the fun of swatting down airborne targets.

Haven't seen it mentioned but instead of simply buffing damage for KB magnitude, one thing I've been wanting to experiment with is to apply bonus damage if a knocked target hits something, based on how fast they were going when they slammed into the wall, etc. Kind of like fall damage but not strictly up and down. Obviously this is just an idea and I'm not sure how feasible it is due to balance concerns and figuring out the right amount of damage, but I can't deny it would feel epic. Unrelated to that: How would people feel about Air Superiority's KnockUp being changed to a high-magnitude Knockdown?

Sounds can definitely be changed, they’re just part of the fx script. I think projectiles can vary (they definitely take on tinting) but don’t quote me on that. The big question on the sirens music note effects is if the animation times can be made to match up or not.

The tip-off there was things like the yellow Hasten fx that normally goes on your hands appearing on the ground beneath lightning clouds. I think that’s something that’s actually fixed in I25, but it was a subtle bug that was present for most (all?) the life of the game.

Did they previously last for the duration of the effect? This was a quick fix to prevent it from lasting forever, it needs to be redone as a proper ContinuingFX on the power effect (it was originally an ActivationFX like Kheldian eyes which is all sorts of wrong and caused it to show up even if the power couldn’t be cast).

I think he means relative to the last build, making the curve less steep. Overall it should still do more damage at max endurance.

The one thing that changed from my initial post is that the final version doesn’t have a different radius for Defensive and Aggressive Goto. Both allow pets to move around within a 30’ radius of the Goto point, and both prioritize targets closer to that point.

They're implemented and will probably be part of the next beta publish whenever that happens.

It wasn’t possible at the time. It is now. 😉

This is still preliminary (but proof of concept is working), so it may change before it’s patched in. Instead of always chaining to the nearest eligible target, a priority system was implemented. For all of the chain powers, if the target you cast it on is a player, it will try to jump to other players. It can still jump to pets, but will only do so if there are no eligible players in the jump range. Similarly, if you target a pet first, it will prefer to jump to other pets and allied NPCs (but can still chain to players if there aren’t any other targets). The healing and absorb chains were also given “smart heal” logic and try to jump to the eligible target with the lowest HP first.

@Auroxis The main reason for increasing it was due to the issue with pets eating chain jumps. However we think we have a solution for that, so it’s only being increased a bit in order to test how effective the fix is.

If you’re trying to get powers data, bin files are a more reliable way to do it, since the client already has them in order to make the in-game numbers work. Parsing defs requires a bunch of specific knowledge about how they’re used by the game. For example, you need to know what the defaults for every structure field are, since most are optional and the default isn’t always 0. Or how to fill in various things that are determined by context like whether duration, magnitude, or both are resistable/affected by strength buffs, but are encoded into the flags at load time (and persisted in the bin) if the designer doesn’t override it. Since most of that has already been filled out in the bin version, the data there is much more complete and less ambiguous. You have to keep up with it since the format does sometimes change when new features are added, but unless/until something like what Cipher is talking about becomes a reality, they’re probably the most reliable source.

It’s all redirector powers (this is actually fixed internally so it’ll probably clear up at some point during the next patch cycle).

@Bopper Nice work! Can confirm, there's an extra roll for the mod to attach. I didn't catch it when I checked this out last time (was assuming it was a duplicated Chance but didn't find one), but the numbers clearly indicate that's what's happening so I looked again, and found it hidden in the Requires. Requires target.isFriend? ! @ToHitRoll @ToHit / @ChanceMods < && That @ToHitRoll @ToHit / @ChanceMods < is the hacky way the Paragon devs used to get a consistent roll in order to tie multiple attribmods together. In this case it shouldn't even be there at all, since the DoT attribmod is CancelOnMiss and has a TickChance of 0 (which gets increased by global chance mods based on which ability you have slotted). So the chance is being applied twice, once for the mod to attach, and then again on the first tick. My guess is this was probably copied and pasted from the Reactive -Res debuff -- that effect had an attribmod for each damage type, and would have needed the hacky workaround to consistently apply them instead of doing a separate 75% chance roll for each type. We don't actually need the hack anymore since the I25+ engine has effect groups, but that's a whole different discussion. I'll ping @Captain Powerhouseto remove the unnecessary part of the Requires and get that change in the pipeline.

Well to be fair lots of commercial software houses don’t have dedicated QA teams anymore either and routinely go live with bugs that should have been showstoppers. Microsoft I’m looking at you.

If you’re short on RAM, try running the 32-bit version. The 64-bit client (which is now the default) will use about 25-30% more RAM simply because pointers are twice as large on 64-bit architecture.

@HeliopauseNothing is collected unless the game crashes. If it does, you are presented with a menu allowing you to optionally upload a crash report (with or without a process memory dump), or just leave it on your computer for you to review. You can see a picture of what it looks here: Regardless of which option you pick, you can find the report files (if any have been created) under a "Crash" folder inside your COH install.

I think I see the issue. Look carefully at the quote marks on the third line. Hint: Those aren't quotes. 00000000: 4f70 7469 6f6e 2022 4f75 726f 626f 726f Option "Ouroboro 00000010: 7320 2671 e280 9d20 e280 9c70 6f77 6578 s &q... ...powex 00000020: 6563 5f6c 6f63 6174 696f 6e20 666f 7277 ec_location forw 00000030: 6172 643a 3820 4f75 726f 626f 726f 7320 ard:8 Ouroboros 00000040: 506f 7274 616c e280 9c0a Portal....

@Garble, can you test using yesterday's patch to see if there is any difference? Looks like you posted shortly before the update went out.

Yes, imagine a circle on the ground, centered on the spot that you click, that the pet is allowed to move around within but not cross the line. The code that was added for the AI change allows us to evaluate where the pet wants to move and either allow or reject it based on various rules and conditions (instead of moving out of leash distance and then having to run back towards the owner). There are two dimensions to the new tech: leash distance, which defines where the pet is allowed to move, and target preference. The latter operates as a threat multiplier, so when I say 'preference' it means that all things being equal, the pet will prefer a certain target, but other factors such as a large amount of damage done or taunt can sometimes override that. For example, pets classified as melee whose AI is in Follow mode now have a 100' leash distance, which they absolutely won't try to move outside of, and prefer targets that are within 50', which means they get a large threat multiplier for enemies that they can melee while staying inside Supremacy range (with a small buffer to account for collision distance). If you put the pet on Attack instead, the 100' leash is disabled and the pet can go anywhere to follow the target. So here's what I'm thinking for the revised Goto for Melee AI - critiques very welcome: Goto - Defensive: 30' leash distance from the goto location (for reference, FFG's dispersion field has a 25' radius) Scaling target preference based on the inverse of the enemy's distance from the goto location, so targets closer to the center of the circle are preferred Goto - Aggressive: 60' leash distance from the goto location Flat target preference bump for enemies inside a 30' radius from the goto location Also considering setting Stay - Defensive for melee pets to have a small leash range, maybe 10-15' or so. No, it would not keep that, since that's a function of the AI flag that tells it not to move for any reason.

They are, the demons preference got reverted based on player feedback during beta. The issue here is that both "goto" and "stay" were changed to set an AI flag that prohibits it from moving, period (unless it gets too far away from its assigned spot, from say knockback, in which case it's permitted to run back). For ranged pets that works great, but melee pets typically do need some freedom to move around a little, otherwise they stand there like idiots and occasionally throw a shuriken /rock. Previously, pets could roam all over the place with no real limit, even when set to 'goto'.

Thanks @Midnight Guard, @TheSpiritFox, that's helpful feedback to have. I'm quite familiar with and sympathetic to the difficultly of positioning pets to make use of your secondary, having played a ninja/dark extensively before sunset. I am wary about going back to the previous behavior of Goto, since it wasn't really "Goto" in any sense of the word, more like 'maybe go here if you feel like it and aren't too busy otherwise run all over the place chasing stuff and come back here later'. I completely understand how not having even that limited control would make micro difficult on melee pets, but I think we can do better and come up with something that induces less player aggravation. One thing the AI changes make possible is doing some more intelligent positioning based on the AI state. I'd like to leave the ranged behavior as it is, but it's completely doable to make pets that use a melee preference act differently when those commands are issued. What do you think about turning Goto -- for either melee pets only, or just the pets that weren't flagged as "ranged only" -- into an area defense command? That would be fairly easy to do since most of the logic was already written for the supremacy-range-of-owner behavior on follow. The trick IMO is tuning the area size to be large enough that melee pets can be effective, but small enough that they don't scatter too much, especially keeping the example of the FFG bubble in mind.

I keep hearing people say Doooooooooom, but since concurrent player count during peak times has steadily been going up since the first of the year... 🤷‍♂️ Number of active accounts has held remarkably stable for quite a few months and doesn't correlate strongly with concurrency, which suggests that the conventional wisdom of a somewhat more adult-centered player base who doesn't have as much time to play games around the holiday season was probably correct. As expected, once the initial shiny and excitement wore off, some with shorter attention spans drifted off (mostly by the end of the summer), while a healthy number settled back into pre-sunset routines and play a couple nights a week, but are not online every waking hour or even every day.

Honest question: Are you micromanaging melee pets using Goto because you have to, or are you micromanaging them because you are used to having to? I'm curious if you've tried leaving them on follow and testing out the AI changes -- they should be quite a bit better about staying in supremacy range than they were before. We have some tentative plans for the bodyguard mechanic that should make it more flexible in the future and less dependent on the AI, but they're still in the conceptual stage.