Jump to content

Immersive Citizens - AI Overhaul (by Shurah)


EssArrBee

Recommended Posts

Yeah, I generally tend to stay away from the automatic navmeshing tools as I find them very unreliable and usually they generate some really bad island issues etc as well.

 

However you still haven't answered the main question Shurah: Do navmeshes have a bigger impact on AI other then just allowing them to get places? Does it actually directly affect their behavior?

Answering that one question will help a lot with understanding what you seem to be implying about the compatibility issues with your mod and will clear up a lot of confusion.

Also I addressed the cylindrical patches issue you seem to be suggesting would occur with ETaC and I still don't understand why a single patch cant be made that includes all fixes for the mod and its patches which is what a lot of people do, rather then having to patch the patches?

Cylindrical patches = 'iterative' patches if I understand the problem correctly ... it would seem that shurah is implying that navmesh conflicts cannot be overridden as normal asset conflicts are overridden ... navmesh conflicts behave as a composite? There should only be a single rule governing each navmesh?

Link to comment
Share on other sites

That sounds about right, I'm aware that coc'ing skips startup scripts but it is a quick way to test for FPS, stuttering, and general graphical stability in specific areas. What concerns me is having to install a non-Step mod, Alternate Start, to test a Step install. Is this an endorsed method of stability testing? If so, then I suppose it's not a big deal, even if it does add time to each startup.

This is strictly my personal view, but I've gotten in the habit of installing Alternate Start because I dread going through tediously boring cart ride one more time or help me I'll SCREAM! I also like the option of starting as a variety of characters throughout Skyrim and you can wake up at pretty much any inn in Skyrim so you don't have to coc to a location with unknown results to begin testing.

Link to comment
Share on other sites

Using Alternate Start is a viable option, but dumping yourself into the game from the main menu is not a good idea. When we test, we do it as if we're actually playing the game and set up a couple specific saves for doing so. I have one vanilla save with no mods and I have one save with a full STEP install, nothing extra on top. These are my two testing profiles I use for test.

Link to comment
Share on other sites

A bit offtopic I guess, but the last I did any serious amount of testing the only real reason not to use coc is because all quest flags etc. are not set properly... hence some things will be a bit weird. However if you set those manually the game cannot tell the difference between running the sequence or doing coc. At least to my knowledge. 

 

If there was ever any issues in my testing the good ol, rest a month in a small interior space seemed to work nicely to make sure the scripts etc. had started working properly.  

 

Ofc. just doing coc, and expecting everything to work is hoping for too much. 

Link to comment
Share on other sites

Is there more to the navmeshes than just allowing for the NPCs to be able to navigate the terrain?

This is not a secret for anyone with a minimal amount of skills in navmeshing. For example, the navigation meshes are also used by the AI for general spatial reasoning and spatially localized information. That's the basic of basic, any decent lectures about navigation mesh contain this kind of information. Information which can be found in less than 5 minutes by searching on the net.

If you want to be able to understand a complexe problem in a specific field, it asks a minimal amount of skills in the field and the only way to get skills is through hard work. So don't expect to understand easily why it's not possible to fix some compatibility issues between ICAIO and ETaC with the current tools available.

 

 

Not sure if it is proper to post directly in this thread itself, but after searching through all 17 pages and finding nothing, I simply have to comment on this. Am I the only person who receives an infinite loading screen every time I try to console coc somewhere from the Main Menu with this mod loaded ?

 

I see on the mod's Nexus page, this issue is directly recognized and acknowledged. I'm extremely apprehensive about a mod being admitted to STEP Core which prevents me from quickly testing for stability by coc'ing from the main menu to an area I need to test. Even with Safety Load, SKSE.ini, 6gb of VRAM - infinite loading screen every single time. Is there any recognized fix for this? Any mention of this issue is absent from the STEP notes for the mod, which seems an oversight. I would be grateful for a proposed fix for this, or any word on whether it can be left out entirely without disqualifying the Core patch. Thanks.

I'm aware of the problem. I didn't add a script activation system for the release because the community wasn't ready to believe that it was possible to do so much without using scripts. Besides, a increase duration of 20 secondes (if you load a save or start a new game), for your first loading screen, the first time that you load my plugin, is a minor annoyance. Moreover, as explained in the description, the plugin won't work correctly if you use the coc command.

Anyway, this annoyance will be removed soon, as I'm going to add a MCM menu in order to turn on/off the mod when you are in game. So the activation won't be automatic anymore, but manual.

This will allow me to update the mod more easily.

Edited by shurah
Link to comment
Share on other sites

Is there more to the navmeshes than just allowing for the NPCs to be able to navigate the terrain?
This is not a secret for anyone with a minimal amount of skills in navmeshing. For example, the navigation meshes are also used by the AI for general spatial reasoning and spatially localized information. That's the basic of basic, any decent lectures about navigation mesh contain this kind of information. Information which can be found in less than 5 minutes by searching on the net.
If you want to be able to understand a complexe problem in a specific field, it asks a minimal amount of skills in the field and the only way to get skills is through hard work. So don't expect to understand easily why it's not possible to fix some compatibility issues between ICAIO and ETaC with the current tools available.
 

 

The only info that I am able to extract out of your comment is that everyone can learn what navmeshes do by "google-ing" it.

I highly doubt that all the info on the web about navmeshes is applicable to skyrim.

 

 

All people are asking is to share what you know and have learned so that others can branch off that knowledge.

If you are able to provide specific examples with an image and comment of the -potential- issues then people can understand why and they don't need to pester you anymore.

Link to comment
Share on other sites

Shurah, if you're still planning to work on Guard AI further, would you mind making them more of a nuisance for a thief character? Like, following the player if you're sneaking or simply following you around Jarls' quarters (so that you won't 'accidentally' pocket that gold lying around)?

Link to comment
Share on other sites

Sorry for the slow reply, been taking a bit of a break from skyrim and modding lately
 

Cylindrical patches = 'iterative' patches if I understand the problem correctly ... it would seem that shurah is implying that navmesh conflicts cannot be overridden as normal asset conflicts are overridden ... navmesh conflicts behave as a composite? There should only be a single rule governing each navmesh?

Yeah this is what I'm confused on as well. Shurah seems to be implying that navmeshes are not subject to the rule of one conflict, and by that I mean the actual navmesh record itself, not the NAVI or related records that get merged at run time. Once single patch that incorporates any fixes by Shurah and any expansion edits by ETaC for example should work fine, with no need for patching that patch.

 

This is not a secret for anyone with a minimal amount of skills in navmeshing. For example, the navigation meshes are also used by the AI for general spatial reasoning and spatially localized information. That's the basic of basic, any decent lectures about navigation mesh contain this kind of information. Information which can be found in less than 5 minutes by searching on the net.
If you want to be able to understand a complexe problem in a specific field, it asks a minimal amount of skills in the field and the only way to get skills is through hard work. So don't expect to understand easily why it's not possible to fix some compatibility issues between ICAIO and ETaC with the current tools available.

As hishutup said, how navmeshing is technically implemented and functional often differs dramatically on a per engine basis. In unity for example, you can set up a navmesh like in skyrim but you can set it up so its just thin lines that the NPC can follow rather then having to mesh out the entire space, or you can do it via raycasting which works like an eyeline for the enemies and allows them to dynamically track objects that way although its a little inconsistant. I've even coded my own Ai system in unity that tracks the transform value of an actor and dynamically reposition's them around objects if they stop moving which is more intensive but it can be done. In some of the older engines it was purely done by height mapping where enemies had a certain level of geometry they could traverse per movement and anything above that stopped them, in UDK you can set areas of an existing mesh to be a navmesh as well instead of creating a new one.

 

The one thing that all of these systems have in common is that as long as the AI has the space in the navmesh to get to where they need to go, they figure out the rest. Doesn't matter if the navmesh is a different shape or it has triangles in a different orientation or composition, as long as they have enough space and the navmesh is big enough that they can get there, it works out fine. Of course neater navmeshing allows for the algorithm for pathfinding to work a bit smoother, but even a messy navmesh, which I know JKs and Jennas are certainly not, will still function as long as its got the correct cell connections etc.

Are you saying this is different for gambryo, as you seem to suggest that any mod that makes a new navmesh is automatically incompatible, even if the navmesh still allows them to get to all your new triggers?

 

You state a lot that you don't really have the time for discussions and explanations, but you spend far more time being defensive and dealing with peoples confusion and questions then you would if you simply explained all this info in the first place so that everyone could benefit from you and stop asking you to explain these things.

Edited by Nazenn
  • +1 2
Link to comment
Share on other sites

I'm just going to put this out there and this is not to oppose anything that Shurah is saying but to convey information from the community.

 

I've been in talks with some very good modders who really know their stuff, both on forums and in PMs. From those talks, I'm 98% confident, unless Shurah releases any unknown knowledge, that there is nothing more to navmeshing than what we already know and that is documented. I will recant this statement if any knowledge is given that proves otherwise. This would mean that more mods are compatible with ICAIO than what is led to believe when it specifically comes to navmesh conflicts.

 

Now a disclaimer here. I have NOT personally verified any of the mods that Shurah lists as incompatible. There may be markers hidden by objects, vanilla markers moved in them, or even bad navmesh edits that would definitely make those mods incompatible but I haven't checked this. From what I'm gathering from all these modders is that, as long as the navmeshes allow the NPCs to get to the markers that another mod adds (ICAIO), then it doesn't matter which navmesh from which mod wins the conflicts as long as the edges meet up at the cell borders so NPCs can cross the borders. I said this in another forum, but we could all navmesh the same area and all our navmeshes would conflict and be slightly different, but that doesn't mean one is better than the other. They could all work just the same and get the job done. That's not saying that neat navmeshes don't have their place, they should be neat and properly done.

 

City overhaul mods are likely not compatible due to interfering with the markers and NPC routines added by ICAIO. This is due to added buildings, objects, and structures, as well as, because of moved or removed vanilla objects. If you move a grinding stone over a bit then you'll see an NPC grinding in mid-air with ICAIO installed. These changes should also ensure they're not interfering with any quest related routines. In my opinion, city overhaul mods need to be made from the ground up with ICAIO compatibility in mind. I think fixing them in their current forms will be just a large of a challenge that you might as well start over. That's just my opinion, though.

 

So in short, unless proof is provided, there is no ground-breaking revelations to be had here. It's just business as usual. I am glad to see that Shurah made some changes in the latest release to make the mod more compatible with a few other mods, though. Still waiting to see if he'll be using the script to clear the alias from dead NPCs. This will go a very long way in making ICAIO more streamlined by cleaning up the aliases that aren't used or needed anymore.

Link to comment
Share on other sites

Just my 2 cents - wouldn't it be appropriate for Shurah's Navmessh "bugfixes" to be incorporated into USLEP?  My general direction in using a heavily modded skryim has been one toward using mods that are more singular in purpose.  If the core of ICAIO is an overhaul of AI, it stands to reason that Navmesh edits should be beyond the scope of this mod.  Sure, the two are dependent on one another, but adding these Navmesh bug-fixes to USLEP would (1) generally benefit the entire Skyrim community and (2) enhance the overall compatibility of ICAIO.

 

Then it would be up to mod authors who add world content to ensure navmesh best practices are followed.

 

Finally, we would be left with the issue of marker conflicts with mods that add structures, etc.  In this area, it would be very helpful for Shurah to provide a modder's reference regarding which markers are for what purpose.  This would aid in the creation of compatibility patches, with Navmesh now out of the picture due to the incorporation of Navmesh bug-fixes in USLEP.

 

My observation is that Shurah doesn't get enough credit for the documentation he HAS provided.  ICAIO is one of the most well documented mods on the Nexus - just read through all the content he has provided on the main mod page, the compatibility article, and in the Posts.  That said, I think a little additional documentation provided specifically for the modding community, and not just the user community, will go a long way.

 

I would finally recommend that Shurah stay somewhat focused on AI, as that is what he does so well and that is what adds so much to the game when using his mod.  Adding additional structures that are in conflict with existing very popular mods will only add to the confusion.  

 

In my 20+ years in an enterprise IT leadership role, more important than individual talent is the ability to focus that talent in a way that adds to the overall value of the community or team and not just the individual's efforts.  Like many situations in the professional world or in life, in the case of a well modded Skyrim game, the whole is truly greater than the sum of its parts.  ICAIO is a tremendous part, but it's value is diminished if it doesn't play well with the whole.

 

Ultimately, the end goal of any user is to be able to USE ICAIO along with other mods that have been long enjoyed in the load order.  I really hope the compatibility continues to progress.

Edited by jbvertexx
  • +1 2
Link to comment
Share on other sites

Just my 2 cents - wouldn't it be appropriate for Shurah's Navmessh "bugfixes" to be incorporated into USLEP?  My general direction in using a heavily modded skryim has been one toward using mods that are more singular in purpose.  If the core of ICAIO is an overhaul of AI, it stands to reason that Navmesh edits should be beyond the scope of this mod.  Sure, the two are dependent on one another, but adding these Navmesh bug-fixes to USLEP would (1) generally benefit the entire Skyrim community and (2) enhance the overall compatibility of ICAIO.

 

Then it would be up to mod authors who add world content to ensure navmesh best practices are followed.

 

Finally, we would be left with the issue of marker conflicts with mods that add structures, etc.  In this area, it would be very helpful for Shurah to provide a modder's reference regarding which markers are for what purpose.  This would aid in the creation of compatibility patches, with Navmesh now out of the picture due to the incorporation of Navmesh bug-fixes in USLEP.

 

My observation is that Shurah doesn't get enough credit for the documentation he HAS provided.  ICAIO is one of the most well documented mods on the Nexus - just read through all the content he has provided on the main mod page, the compatibility article, and in the Posts.  That said, I think a little additional documentation provided specifically for the modding community, and not just the user community, will go a long way.

 

I would finally recommend that Shurah stay somewhat focused on AI, as that is what he does so well and that is what adds so much to the game when using his mod.  Adding additional structures that are in conflict with existing very popular mods will only add to the confusion.  

That would be nice, but navmeshes are one thing that simply can not be forwarded from using xEdit. The UP Team would have to do the work themselves. I know I've seen mentioned that Riverwood was already complete covered by USLEEP, but Shurah did his own for his own reasons, I'm guessing. You are right that he has excellent user documentation. Very detailed.

 

Shurah, like myself and others, is also part of the RS Team that will be redoing Relighting Skyrim from the ground up so the focus will be split a little, but my focus is always split as well. I have a list of 7 or 8 projects going at once. When I get tired of one, I which to the other to break up the flow and keep things fresh.

Link to comment
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
×
×
  • Create New...

Important Information

By using this site, you agree to our Guidelines, Privacy Policy, and Terms of Use.