I'm not sure if its just me or is the AI in game guru really terrible? It looks like the characters don't really move towards you or anything, they just see you and try to shoot from where they are and when they do try to move towards you, it looks like they're half blind and can't quite figure out where you are.
The reason I ask is I want to make a single player shooter with mostly melee type ai creatures and if they're not at least somewhat useful in terms of how quickly they're able to find you, then I probably need to use another engine.

AI is still a work in progress and will be worked on again after the DX11 Upgrade work has been completed.

make sure the AI are on the terrain or if they are inside/on a flat entity/object (needs to be very close to terrain height right now) then the object(s) must have forcesimpleobstacle = 3 inside their .fpe file (check the stock buildings, such as the standstone structures, for the setting.

other things to consider are obstacles in the way will block the AI's pathing, you need to make sure there's adequate area for them to move to the player in (either by setting the above setting on all objects they may walk on or turning off physics for said objects) - such as slabs, blood decals etc... i.e. anything flat on the floor.
and another issue is AI don't respond well to being indoors - doors especially are best left missing for now.

in summary;
make sure they are all around terrain height
set forcesimpleobstalce = 3 to all objects they will/might walk on.

AI will be improved soon but it will likely still require you to do at least the fpe change so best to check now.

Ok, I don't yet own a copy of GameGuru... I'm planning to make a doom-like game so the combat will be all inside so its likely that the engine won't be good enough for me just yet. Good luck improving that AI! Maybe this link will help?

http://www.findbestopensource.com/tagged/pathfinding

also Recast https://github.com/recastnavigation/recastnavigation
and Open Steer http://opensteer.sourceforge.net/

I dont think the current AI allows for using stairs and things like that. So it certainly isn't "there" yet.

Yes, it is! However, that hasn't really been a focus of the developers to date. It will improve, I'm sure, after they finish with the engine upgrades that are being worked on right now.

You mean "ALL is still a work in progress"???
Maybe they should mention on the Steam store page that almost nothing works right because it's not finished....


Ouch! LMAO

lol, yes, must be a typo

Was referring to ai. (AI).

AI seems to have most of the necessary functions and data available in the lua, of course you cant see what that tracks to in a closed source engine, testing the AI is my next mini project in gameguru anyway may as well flex my scripting muscles on that

The AI is okay. Just got to play with the settings a little.

What bugs me, is if I start making a game now, will it be compatible with future releases of GameGuru. Or will I need to stick with a version until game completion.

Since Kickstarter the engine development has been fragmented and that's how it stands. Most things taking a direction and changing as part of the ongoing "evolving" theme to development, which has slowed down all things towards a mature development. This may continue for a long time yet...

At the moment AI is limited partly because of what's hard coded and the reliance on all the other things that need to be in place which the AI may require and which may affect/influence it for better or for worse. Everything from physics to obstacle avoidance, pathfinding and behaviours and most of these things and many more can be of differing qualities (simple or advanced) and can make a massive difference and all which may affect the quality of AI itself during gameplay. Other things like engine Performance and efficiency too can impact on the AI and it's behaviour when a lot of AI activity is happening at once.

Hopefully the road ahead is now or will become a little smoother once Direct X11 is in place. Of course many things which AI (and other things) might best require to be in place are themselves likely to be fragmented in coming along as the road map is still fragmented by development direction and user preferences (voting) so some things may be in place sooner than others. This is the way it has been and may well continue and is why perhaps many see various features as unfinished - because they perhaps are depending upon what is envisaged/expected by the developers and users and their priorities, which may vary widely.

Most main features to date indeed perhaps could be said to be under developed, water, terrain and so on, some of everything perhaps being better than nothing so more people half something of what they want or need. Quantity Vs Quality?

The AI currently is to say the least basic but better than you may get and have seen out of the box elsewhere. That does not necessarily mean it could not be a lot better by the will of both the developers and the users.

Apparently Bots Navigating will be possibly receiving attention "soon". At least that may provide the basics of that but any more may take much longer as good AI requires somewhat more than that.



Some of the default current AI scripts do have a variety of behaviors which at the moment do not seem to work, work at all or work well or correctly and have performed somewhat better previously, e.g. go home or run away if health is low, pathfinding itself is hit and miss over entities or terrain due to engine changes/restrictions currently....

A lot could be potentially improved and or added/extended perhaps if as said the developers and users wish it to be so. e.g. advanced pathfinding/nodes/entities to control/manage AI behaviors, random behaviour choices, character lip sync and so on. Characters themselves have an animation set largely untouched and unused - excluding lip sync of course which may never see the light of day at least any time soon....

Many may be very happy with basic AI and have no need/want of anything more.

Not sure what can or should be expected by users in particular in a product where the cost to them might now be described as a giveaway or so it seems to me in the case of the base product irrelevant of any other engine considerations. That is a developer decision and obviously/presumably impacts upon to some extent what one could realistically expect in return, or perhaps not. I don't know...

Happy 2016 game making all.

Not at all, the engine could be free and still earn tgc a lot of money, the real money maker is (and will be) the store and maybe the dlc.
Look at steam, its free but one of the most profitable pieces of software that really doesn't "sell" anything (made by them at least).
BUT without a good core product it won't be successful, so who's really gaining from the better quality features?

Don't get me wrong, I like gameguru and its been the only game making engine I use (aside from a quick play with AGK), its on a good track but it certainly needs a lot more time and work.

SmallG, I like your video of a character tumbling, jumping through windows, climbing and swimming.

I concur! Very impressive.