It's really easy - just add a forkthread and a wait to line 372 of FriendlyAsset: [code="c++"]ForkThread(function() WaitSeconds(5); AIUtils.AIChat(unit, announcement) end)[/code]
I've been trying to come up with a non-destructive fix for the boot shop issue for the UberFix, and it's been irritatingly complicated due to how they structured GetItemsList and and GetItemDesires. There is absolutely no reason whatsoever for ScenarioInfo.AllItemsList to be a numerically-keyed table that has to be fully iterated every single time you want to do anything with it-- say, return an item's desire table, something that happens every time an AI shops. That's just awful,
That's because replays are in observer mode as far as the sim is concerned. Observers are Army -1, which alliance functions don't like. Make sure you're not using IsAlly or IsEnemy checks with GetFocusArmy without first checking that GetFocusArmy() > 0. Add that to the AI chat check and you should be good. Edit: There's another one of these in line 722 of HeroGOAP. Same deal, just make sure GetFocusArmy() > 0. Edit 2: These are both stock Demigod
Also found in UseItemActions: that pesky little AI-can-teleport-to-walls problem. Not added by the mod, of course, but easily fixable. Change the category vectors on lines 602-604 from 'categories.DEFENSE' to 'categories.DEFENSE - categories.WALL'.
What were they doing when this happened? Were they trying to cast, or still moving around? Edit: I just looked at UseItemActions.lua, and the AI is set up to send a message before it knows whether or not the ability is ready. The announcement just needs to be moved down to within the 'if (result)' condition, so it only occurs when they're actually fully able to cast the lock.
What is the achievement problem, exactly? Does it affect single player? For that matter, do custom games even track achievements on Impulse? If not, a mod can't fix the problem. If so, and if the problem also exists in skirmish/single player (I don't pay attention to achievements at all, so I wouldn't notice), then sure, maybe I can do something about it?
Wish I could. That stuff is all in the engine, or with the achievements, Impulse. Nothing the community can do about that.
I initially tried changing the buffs, but I recall that having a few unintended consequences - one of which is breaking any mod that rebalances citadel upgrades. The buff creation fix I used is really clean and non-destructive. You're welcome to include a copy of it it, if you're not already overriding GetUpgradesList in AIShopUtilities. If you are, just remove the check for the buff, and it'll resolve fine. Edit: The copy of .24 I have installed right now has an ove
I've always played every test game with both, so I'm not too worried. I haven't had time to look at the code for Peppe's .24 versions yet, but unless he's radically expanding the scope of it, it shouldn't be a problem. All of the fixes I've used in the UberFix related to AI have specifically avoided overwriting anything, and are meant to be fully compatible with any mod that changes anything in those files (e.g. this mod). Of course, I have no idea what extra files are mod
Yes, both of the Experience-related bugs are fixed in the 1.03 UberFix. The reason it can't buy experience is because one of the functions in AIShopUtilities (GetUpgradesList) ignores any cit upgrades that don't have a buff with the same name. Experience upgrades are named 'CPortalFrequency', while their buffs are 'CTraining'. I fixed this non-destructively by just pre-generating some dummy buffs with the same names as any cit upgrades that don't already one associated with
I'll round this out with a self-deprecating coup-de-grace: How rough of a spot is this game in that I'm the only one maintaining the community fix patch? Haha. Seriously. I have no programming experience at all outside of poking around with Supreme Commander lua for a year or so prior to buying Demigod, and I've never even released a mod for either game. I don't even really play Demigod multiplayer, and my sole interest in the game at this point is light moddin
Let me clarify further: If Demigod had been a fairly high-profile release, and hadn't had a total flop of a launch, AND these kinds of fixes and enhancements and AI improvements had been patched in somehow within the first few months, THEN they would have had an appreciable effect on the game's profitability. However, it's been almost two years since the game was released, it had an awful launch, the post-launch support was non-existent, and the community has been stagnant and d
What I mean is that every beta, beta patch, or 'community tested patch' you've seen has also gone through internal publisher/developer QA as well. The only exception I can think of to this would be Tribes 2, and well, look what happened there. It's not that I think the game will never be patched again, but I'm being extremely realistic about how these things happen. Stardock will have to spend some money and hire someone at GPG (or a third party? I don't know what
Chain Lightning and Brain Storm, mostly. Just kill a bunch of creeps with CL as much as you can, see how often the lightning skips links, and definitely tell me if it ever breaks and sticks. The main change I made is moving the target check from IsDead to BeenDestroyed, so lightning should still happen on dead targets, as long as they still exist as an entity. I also moved the trashbag that the effects are added to on the target to standard Trash, from TrashOnKilled.
Oh, nice catch. Those should probably be zeroed out, as all but one of them is wrong, and will be influencing prices (and possibly even availability) for any function that returns the main shop unit (ubgshop01) instead of the shop tab blueprint (ugbshop10 in this case). Out of curiosity, which function is being used to the shop blueprint? I'll try setting Tree = {} and see if that causes any problems. If not, it's going in the UberFix.
I can post one now, but like I said, I haven't actually tested some of the new Oculus code at all. It's probably fine, and as long as you don't mind if it breaks (and can get me a log), knock yourself out. 1.03 RC4
Community QA is not a thing that has ever happened, and never will. Keep in mind that companies like Stardock are real, actual businesses employing people in suits with business and law degrees to make sure the well-meaning operations side of the company doesn't do something really foolish that costs them a lot of money (insert joke about Demigod's unprofitability). Such as: relying on random people on the internet to determine whether or not an official update that is aut
From Rook's blueprint:[code="c++"]CanFireWhileMoving = false,[/code] So no, it's not in DG 1.3. I have no problem with the change to Permafrost, and agree that 5/10/15 seems to be developer intent, and that it was an overlooked QA issue. If this is the general consensus, and nobody has objections, I can put this into the UberFix. Just so we're clear, it's not the movement speed debuff, but the ROF debuff. Movement speed is 3/7/10, which is inline with all oth
Ok, I went ahead and finished this, and gave it a quick test. I don't have time to really run it through the paces right now, but I have confirmed that it's not generating any errors, and that all the sync variables it adds are in fact available and properly reflected in the UI. I'll release it as a standalone mod for now, keeping the UberFix as clean and reliable as possible until the end of the tournament. Then, assuming it proves stable, I'll merge this into the next po
I'm gonna try to get this sync enhancement stuff done in the next day or two, but you'll probably still want to use the next fix-only version I release in the tournament. If the version you're using now proves solid enough, I'll drop another with a few tweaks to some of the existing fixes (like, for example, I found a much better way to handle the sticky lightning fix that results in far, far fewer broken visual links). Also, I totally missed adding debuff removal to all levels of
I like Demigod, but let's be honest with ourselves - when you can't even scrabble together a dozen 3-player teams for your second publisher-sponsored tournament and have half of them show up to a single match, the game is deader than all hell. I've seen this happen to so many games before, most better-supported than Demigod was. Games don't get more popular with age. There's nowhere to go but down. etc etc. Even if another patch were released, there's no eviden
Alright, here's what I've come up with for a cumulative changelist for 1.02 -> 1.03. I've picked through the code, the google code issues, through previous versions, and through all of the readmes, so hopefully I've got everything. Need to keep better docs in the future. -------------------------------------------------------------------------------------------------------- v1.03 ------------------------------------------------------------------
Sure. One of the best examples is the Rook weapon fix between 1.02 and 1.03 - in 1.02, it's replacing the entire unit blueprint to change one line. In 1.03, it's one line in three tiers of otherwise empty tables. Another example, turning the Electrocution fix from 1.02: [code="c++"]BuffBlueprint { Name = 'HOculusElectrocution01', DisplayName = ' Electrocution', Des
While compiling a full changelog, I found a few more small fixes that I overlooked when doing the conversion of the mod to non-destructive hooking. Mostly effects-related things. They'll be in the next release.
I'm gonna be a downer again (it's what I do best) and say that the game needs fewer favor items rather than more of them, or at least needs most of the existing ones improved. How many of the stock DG items do good players actually use? 4 or 5 out of 30. If anything, make a small mod that tweaks existing weak favor items to be more specialized and desirable, without making them too good. If an item does more for any given demigod than BotF does for UB, it's probably
Demigod Forums