oukej
-
Content Count
1063 -
Joined
-
Last visited
-
Medals
-
Medals
-
Everything posted by oukej
-
Campaign Episode 2: ADAPT - Feedback thread ** SPOILER WARNING! **
oukej replied to Wiki's topic in ARMA 3 - GENERAL
Unlike Stratis, which was the main playground in Survive, Altis has much bigger memory footprint. # Could you please take a look at the system pagefile and check whether it is not disabled or set to low size? -
I am sorry to probably disappoint you - it is not the full rag-doll to the ground. Just the flinch - addition to the current state to always visually confirm the hit.
-
In current/upcoming update (not today's) you should see some changes about which I'd like to read your feedback. As said previously. This whole thing is just one thing we can do more or less immediately. That one thing doesn't mean it's all and by far not the end of the story. But for now we should probably stay away from the "could-be's" and "should-be's" in this topic - let's keep it more about tweaking the current status, more about stronger/weaker, better/worse.
-
We'd all like more detailed and better simulated protection using the fire geometry, however, the proper implementation of it has currently been impossible. We are investigating possibilities, but I'd like to stay away from details and promises as long as we can't clearly say "Hey, we've done it and its working" or vice versa. There are more aspects of it - e.g. the injury simulation. Taking me to: U sure you wan't the ragdoll transitions there and back? ;) The changes/tweaks are aimed to simply improve the experience of the game "as it is now" - increase the variety and benefits of wearing vests/helmets (unlike the previous state). A kind of a tradeoff between the possibilities which are currently in the game, realism and enjoyment (liek ...how do you expect to simulate all the things BI, if u no has inkapacitaishun, bl33din' n stuffs?!?! :angryfire: ;))
-
ˇ :( Well, not to let you down - the fight is far from over.
-
Even though we currently have materials and their penetration modeled in a detailed way and this may seem like an off-hand ideal solution without much additional work, we've bumped into some unexpected, currently blocking, issues with it earlier. Btw, meanwhile Solzenicyn has prepared some tweaks that should be already in dev branch. Please, let us know what you think, how do you enjoy the game now and what could - in the given bounds - make it better ;)
-
I've kind of meant both. On the slider in the editor default unit gets 0.6, placed-in-group units get a skill ranging from 0.4 to 0.7 and a spawned unit gets 0.5 - so something like this. 0.2 is a lower extreme value, 0.8 or rather 1.0 is/should be upper extreme. In the (Configure/Game) Difficulty settings it may be best to stick to the default Regular or something let's say ranging from 0.6 to 0.8 But - do it anyway you like. There is never enough feedback. Just please let us always know what configuration you were approx. running. Thanks a lot! Let's wait with thanks for how much you like the change ;)
-
Some tweaks in the AI behavior were introduced in today's dev branch. There might be more details later, but let's put it short - the AI makes mistakes (By that I don't mean the bugs ;)) Generally the firefights and engagements should be more intense, enjoyable and less frustrating. The average AI should no longer give you a head-shot off-hand. (...according to their skill :devil:). And AI gunners shouldn't be flawless top killing machines anymore. It definitely does not solve all the things (e.g. detection, decision making, recoil control), but at least it hopefully softens some of the problems. Together with the average soldier should have become more... average. By default, without a need to change the AI settings unless you wish to. That's anyway a goal we'd like to achieve a we are humbly asking you to help us with it.Please, spam us with your observations and feedback how has the battleground changed for you and whether it has improved the firefights, AI taking a shot, for you. For that I recommend sticking to the average values (Similar to the range mentioned above. AI with skill 0.2 is about the lowest skill you should see/use in the game and so should be its behavior) and rather using default difficulty settings (I'd say global AI skill within the range 0.6-0.8, without changing the precision manually). But if you want to go further and check whether different values work better even with the current conditions - we'd be happy for it as well! ---------- Post added at 16:35 ---------- Previous post was at 16:26 ---------- (and I will try to read thru your last posts asap. I've quickly notices some driving issues...yes, neither I was able to get my driver from one side of Altis to the other, you know, it is a thing with the drivers from the southern countries ;))
-
It should be less resource hungry and save some FPS. Be not afraid... no dead bodies walking around... or... perhaps?
-
yup
-
Thank you for voting and giving us feedback! Were you willing to provide the .Arma3Profile you use for our internal purposes, we'd be grateful - please send the single file to me (use my nick at the company address)
-
Please use AI Discussion for any notes about the AI behavior itself. (I know it isn't easily separable)
-
(+ other good ideas & notes) I - personally - enjoy this sort of notion (unlike the one who'd configure it and test it;)). Let's put those good ideas aside for now. I'd definitely like to return to them and the possible adjustments and UI of individual AI unit (sub)skills, possible defaults, presets, etc... rather soon. On the individual AI unit level, the more control of specific aspects of the unit's behavior, the (seemingly;)) better. But on the global level - as an option that affects every single AI unit in the game always by the same ratio - what do you think should be affected? What makes the game easier or more challenging (in an enjoyable way)? E.g.: If every single AI is less able to hit you, if it doesn't spot it's target so fast...? Conversely, what aspects shouldn't be affected? E.g.: How fast the AI can turn around,...? Please, let your imagination go free (don't think about how it could be done, just what behavioral aspects you'd like to adjust globally) and let us know. (I will later throw a poll)
-
It has been in the game before. We've reduced the influence by narrowing the range. Removing the skill distribution based on the rank in Groups completely was also an option, but then we wanted to retain some diversity inside the groups regarding the leader as the most skilled one (not necessarily reflecting the reality)
-
There is one rather minor change (with possible major in-game consequences) to AI Skill Configuration that's going to appear on today's dev branch build. Skill of an unit of a group inserted in the editor gets it's initial skill (which you can change via the Skill slider in Edit Unit dialogue) set according to this unit's rank as defined in cfgGroups. Previously the range was 0.2 (Pvt.) to 1.0 (Col.) which caused - if the mission designer used groups and didn't readjust the individual units' skill - majority of the AI soldiers on the battlefield to get low skill values. However the default skill of a single unit you place in the editor is 0.6. We also aim towards providing a symmetrical range for AI adjustments. (Tweaking the median is a constant process based also on your feedback in AI Discussion) We have narrowed the range to 0.4 (Pvt.) to 0.7 (Col.), which should help the average soldier to have an average skill and also it reflects the uneven distribution of ranks (majority of units used in groups are Privates and NCOs, very loosely mirroring the reality) This is a single separated change that relates to per-unit skill configuration. For now we still focus on global settings and we'll return to per-unit skill configuration for some bigger work later. (to how an unit's (sub)skill is adjusted, how you can set it in the editor, standardization/description of values etc.) ;)
-
Your report is valid. It shouldn't. After all - we don't have any vital-signs live-stream of subordinates ;) Afaik it's generally this http://feedback.arma3.com/view.php?id=14905 and interconnected issues - things that reduce the enjoyability or give the impression of cheating AI. (nerfing first, pimpin' later ;) - that to say would be too simplified) The UI work is more about adding the ability to properly test the AI, add standards and some reliability. Not locking down current AI, more about locking down some variables to allow for AI improvements. In: Ezekiel 25:17 ;) Armor simulation is a good example (and I believe it has been improved in A3). Others are sometimes better as "optional" or modular features, because they don't always fit everywhere. The community solutions generally (at least today) present a great inspiration to us and indicate what is missing in the vanilla game. Though it is still valid to ask "why we didn't" this discussion doesn't belong into this thread.
-
I have looked into that issue right after the first vid, the issue is wedded to the building. If you wan't to keep an eye on the progress, please start a FT issue and let me know via PM so I can tell you once it has been fixed. In the latter videos you are dealing with completely different issues. What alarms the AI and what doesn't (based on audible sources). And how is the information shared across the groups. But that is the most complex ;) I'm sorry to disappoint you, but human-like, even monkey-like behavior is still beyond reach even of the academic level AI. And that itself is couple of years, decades, ahead of any AI usable in an open game environment & with home-PC computing power :/ Many things are not (often can't be) simulated in Arma 3 on the same level how player perceives the game (e.g. not everything the player hears, the AI hears as well). I'd like to ask you for your patience. We are constantly working on the AI, trying to improve it and your feedback is of a great value to us! The most helpful is of course a isolated 100% repro about objectively (also present-day game-wise) wrong behavior, as simple as possible, 2-3 AIs... ;) submitted into Feedback Tracker . That is like a golden nugget to us ;) But by far not only - a description of general experience helps us greatly to decide on the course of action we should take, what to prioritize, even if we can't eliminate the variables that can count towards it (subjective matter, skill settings, mission, mods, scripts used...) ---------- Post added at 14:30 ---------- Previous post was at 14:24 ---------- I can assure you we do :) Sadly not every scripted solution can easily make it into the engine as a full-scale game-wide mechanic :/
-
Well, I didn't say they don't ;) Sometimes, something at some speed, just not directly and by far not as much as on the squad level. Good point made by2nd Ranger I agree, but then you have to deal with "when" & "how". The other extreme can easily be AI headshooting you behind a bush. Suppresive fire or fire on estimated location is nice, but has to be dealt with properly (before that, several other measures have to be implemented and features adjusted & tweaked). If you play as a leader you can often go around without noticing the death of your subordinate. The same way it works for the AI. It reports on a regular basis, but it doesn't know about each other all the time or instantly (unless they see each other, hear the noise) How fast the AI reacts to death, hit or bullets flying around is also dependable on the AI's skill. Please note that the radio is currently of quintessential importance to the AI. The radio item itself has no real effect on the communication other than visual - subtitles and audible - radio msgs. Removing the item does not remove the comm functionality. However weird that may look, please understand, that the game doesn't currently have any mechanics to deal with commanding or squad communication without a radio. It would be nice to have, but it's really a complex task with great amount of possible game-breakers to deal with and with no as high payoff value.
-
I have to confirm that. KnowsAbout tells you that the AI is more or less sure about a threat. But it doesn't tell anything about the AI's estimate of the location and direction of the threat. By audible sources the AI will (should) never get that estimate purrfectly correct.But there are many variables entering the observed issue. Like e.g. I wasn't able to clearly reproduce the issue in other buildings. On the other hand, the geometry seems fine. There could be something with the door (thin aperture around - visually unnoticeable, possibly shuffling AI checks) True. That is completely missing and sadly it ain't an easy thing to introduce. After introduction of 3D chatter that shouldn't be the issue anymore ;) The info is shared directly only inside a group. Even there, the shared information doesn't give exact position (subject of tweaking), the error is on a soldier-level. And unless the AI has clear LoS it won't open fire on a target it learned about only by a report. It will also cease fire after couple of seconds after loosing the LoS, even if another AI in the group has a LoS on the target and keeps giving the reports about it's movement. If you see the AI constantly firing thru a view obscuring object or even wall or door (issue reported by Msy) without supposedly having a clear LoS, it means there's a bug somewhere - either in the object itself (wrong geometries) or how it's perceived.
-
Thanks for the vid. It was quite easy to reproduce. (one has to run immediately after killing the opfor guy) We'll look into that. Just a note: the AI has probably heard you (shooting (even suppressed shot would have been audible outside) & running) - that's a correct behavior. Wrong is that their estimate is too precise and they even fire thru the door only if player happens to be behind them.
-
setPos'ing AI generally isn't recommended, because the AI gets confused and needs to - sort of - "reinitialize" itself. This more of a limitation than a bug :/ (I may bring more information on monday.)
-
:popup:
- 874 replies
-
- 3
-
-
- flight model
- flying
- (and 3 more)
-
You should be now able to find the answers here: https://community.bistudio.com/wiki/Category:AI Skill slider in Editor/Insert Unit allows you to set skill from 0.2 to 1.0, default is 0.6, and the value set is the same as one simply set by setSkill command (last executed prevails). The sub-skills are directly inherited from it unless a sub-skill is set specifically in array using setSkill. This value is interpolated according to definitions in cfgAISkills. In most cases it remains unchanged with the current vanilla settings. And only then, on the start of a mission, it gets multiplied either by skillEnemy/skillFriendly or precisionEnemy/precisionFriendly (which is derived from skillEnemy/skillFriendly and that is exactly the value you set in the Game Options). To visualize it (the picture may not be technically purrfect ;), but should give u a little aid understanding what's going on) That idea came to my mind as well some time ago. It would make things more linear and graspable. But it would mean either narrower hard-limit for the range a mission designer can use to allow for some global "adding/subtracting".
-
The sub skill functions are not linear and expressing the qualitative nature of it in percentage can be misleading. The descriptive words (recruits, professionals) are the way to go even if everyone imagines something different under such word. Correct me if I got it wrong, but what you ask for is to have a "semi-global", mission specific (but not game-wide nor individual unit-specific) set of - let's say - sliders to adjust all the AI units under each side, am I right? It would be definitely be a new feature and I am afraid that it would clutter the settings. Not prio, but it's a good idea that could perhaps deserve it's own ticket? ;) Shouldn't be a big deal though to script such a solution, that would adjust the skill/sub skill values of all the units under a given side. For more feedback about the AI Configuration & Settings, please use http://forums.bistudio.com/showthread.php?169122-AI-Configuration-feedback
-
Does the setskill array scripting command override this? It does not, the value of each of the sub-skill from the array is multiplied by the value of either precisionFriendly/Enemy* (for aimingAccuracy or AimingShake) or skillFriendly/Enemy (for the rest) before actually used for the algorithm behind an action performed by the AI unit. *"precisionFriendly/Enemy" is derived from skillFriendly/Enemy.