posted 03-10-17 01:07 PM CT (US)
[This message has been edited by Cataphract887 (edited 06-26-2018 @ 01:18 PM).]
there is some usage of the invisible tiles trick with hundreds of flags on one tile.I do not think that causes the crash, but in general I'd recommend to use an object called DEBRI (which has no graphics) to create invisible tiles.
[This message has been edited by HockeySam18 (edited 08-06-2018 @ 08:31 AM).]
Glad you liked the map design still, I thought it looked much better in earlier versions but I found some issues with mass deployment of shoreless water related to (smooth) unit movement near them so I had to cut back and delete a lot of finished areas with tears streaming down my face as I did so.The achilles heel of shoreless water is the difficulty that transport ships have in unloading on its shores. I usually try and stick some shoreless shallows nearby to mitigate the issue when dealing with an area that is designed to be disembarked on.
Will make a more comprehensive reply later but after a quick skim I would say all the bug reports seem accurate and the balance correct;the unit choices are not so balanced, I have to admit.Balance is no easy task. After nearly two decades, this game still continues to receive balance patches
some terrain changes in the latest WK update meant that all snow grass was replaced by snow road and all dirt 2 with mangrove shallows (should be a simple fix with AoKTS, hopefully).Done! I pushed an update out which should fix this. Fixing bugs and balancing will take a few days at least. Using TS to swap the terrain was a breeze, but I didn't do much testing on the update so that infernal program had better not have done anything evil to my file that I didn't notice...
No way I was touching Khazar territory when they had hordes of light cavalry, buffed cataphracts, cavalry archers with 13 (!) base attack and mangudai with 16 (!!!) base attack running around.The Khazar are absurd! This was by design but maybe I overdid it. My thinking was the scenario should depict how the nordic peoples had the river control, so the player should generally feel in control and safe there. However I wanted the player to not be walking around on land, so the idea was to make the nomads too tough to realistically defeat. This is why their objective involves raiding some treasure rather than killing them all off. The cav archers have insane attack so the huskarls cannot hard counter them, while still performing like huskarls against the townsfolk archers. Much of the scenario was inspired at the time by my recent reading of 'the emergence of rus' by jonathan shepard and it seemed the norse wouldn't dare attempt a land incursion into nomad territory without atleast 10,000 men(IIRC). In order to create a situation where the player is forced to stay to the river an overwhelmingly strong land enemy seemed called for. Hopefully them being an optional enemy you never need to face solves the problem of them being too strong, since you could always avoid them.
I encountered a bug where saving and loading the game makes blacksmith techs vanish from the interface. Fortunately, I was able to win without needing to save and load.The one bug I probably cannot solve. Maybe another trigger array to detect when the player is selecting the blacksmith and refresh which technology he is able to research. Unfortunately custom tech trees via trigger seem to be unready for prime time with multiple huge bugs affecting it. I believe there is some code which checks what techs are valid and it keeps wiping out the trigger changes with updates of its own to the tree, if I recall scripter correctly.
The heroes that spawn from the Khazar wonder are not renamed.Huh, I'm sure I had rename triggers working at some point but maybe I just forgot. Will be fixed.
The Varangian huts seem to have spawned too many units, with the result that some were trapped underneath. I could be mistaking a free placement trick for this, though.Not intended, maybe I will make some patrol triggers so they dont stand around piling up.
Some buildings didn't have the proper ownership changes applied to them; when you conquer Polotsk you can see a house of theirs in Pskov and several farms around the southeast mill of Smolensk. A Khazar feudal age mill remains near Pskov.Whoops, got lazy
Is there a purpose to being able to train trade carts at your market? If not, might as well disable them to avoid player confusion.Somehow I overlooked that completely! Will disable
Sometimes I encounter a bug, after having watched the opening cutscene, where Knut Thorhallson fails to disembark. The ships are then removed, causing him and his company to die. If I skip the cutscene this does not happen, but the condottiero unit representing him in the Varangian camp is not renamed.This threw me into some confusion, but I think I know whats up. I was in the last changes removing shoreless water and I bet I altered the spot he was intended to land at. By this time I was skipping the cutscene always so that explains why I missed it. Not sure why his rename trigger isn't working, one definitely exists for it.
Additionally, some of the tributes do not seem to come through. You don't receive food from Polotsk or lumber from Smolensk, even though the text indicates that you do.Hmm, I thought I had that in the bag. Will have to look into it and also buff the tributes. I did realize later on I had made a situation where its possible to simply farm the villages, or rather its the best option. Thats why I implemented the fires burning buildings down;it used to be possible to kill off the village troop production buildings only, then sit around and farm their undefended economic structures freely. The fires ensure the objective is fulfilled if they cannot make military anymore.
I did like that on Hard the Khazar horse archers would fight back against your longships, making sniping them off from the river a less viable proposition due to their high damage output.This is something I was trying to do, have the horse archers only fight back against boats and not aggressively hunt them down in the first place. I'm not sure how but somehow it sortof works. I don't understand the AI here however, its just luck mostly.
One drawback is that it is possible to kill all enemy heroes before you've pillaged their large yurts, although fortunately I destroyed Jarl Hogni's yurt to loot 10,000 gold before all enemy heroes were dead, as I attacked from west to east.Yeah its annoying, you can get penalized for killing off the leaders first and miss out on the loot.
The use of UP-effect modding was excellent. From unit and building stat changes to training times and custom tech trees it thoroughly improved the experience. Particularly welcome were the edits to longships, which remained effective on the water but were much less oppressive towards land assets than the vanilla ones are. Another good change was the way techs worked. With the high HP on units and higher costs all around it demands a considerable strategic choice between recruiting more units and improving their armor/attack/other abilities......Having the varying options was nice - huskarls as raiders, berserks as hard-hitting tough guys, Nordic swordsmen as a cheap option, woad raiders as a fast one, and throwing axemen as a ranged core for your army....I was quite fond of how I could make a varied army now of only infantry! Three units are normal infantry, one a stand in for archery and the fast unit a stand in for cavalry. Making longboats ineffective against land targets was inspired by Ragnars Raids;they were a bit OP when massed up there. I wanted the longboats to be a unit that control the sea, not the land, too.
Not intended, maybe I will make some patrol triggers so they dont stand around piling up.There is also AI code that you can add to make existing units scatter around somewhat (and control the distance, I believe). You might remember that behavior from the ES standard RM and campaign AIs. It could make the Varangians a bit weaker as their forces will be somewhat more far-flung, though.
This is something I was trying to do, have the horse archers only fight back against boats and not aggressively hunt them down in the first place. I'm not sure how but somehow it sortof works. I don't understand the AI here however, its just luck mostly.From what I understand, if an AI land unit has greater range than a ship, or has equal range and equal or greater attack, it will fight back against a ship when attacked. You may recall that in the vanilla game, AI ranged units will shoot at fire ships that attack them or something near them, but run away from longer-ranged, more powerful ships like galleys and cannon galleons.
Yeah its annoying, you can get penalized for killing off the leaders first and miss out on the loot.Maybe add a trigger system that detects which huts are still standing and distributes the tribute accordingly?
Making longboats ineffective against land targets was inspired by Ragnars Raids;they were a bit OP when massed up there. I wanted the longboats to be a unit that control the sea, not the land, too.Good idea. One of the main flaws of Ragnar's Raids/Ironside was the power of longships - once they could be massed, there was no incentive to produce any other type of unit until your invincible fleet of longships had wiped out everything within range of the shoreline.
[This message has been edited by HockeySam18 (edited 08-08-2018 @ 09:51 AM).]
Another good change was the way techs worked. With the high HP on units and higher costs all around it demands a considerable strategic choice between recruiting more units and improving their armor/attack/other abilitiesThis touches on somthing in vanilla AoC with the blacksmith techs being incredibly cheap. The fletching\padded archer armor upgrades are no brainers, as are defence and ranged upgrades as appropriate to unit choice later on. This is in contrast to wheelbarrow\handcart which are the games best technologies from a strategic perspective;mastering when to research these two with their higher costs takes a good deal of skill and experience to get down. I imagine the game would be significantly better if many more techs offered more strategically difficult choices than just those two. At the moment its mostly timing on when to spend the resources that is the primary consideration.
Re: the Khazars - making them that strong was definitely a a good choice. I saw in the AoEZone thread that one tester was complaining about how the two strongest enemies are the ones nearest you, and found it a bit shortsighted that he essentially ignored the hints/dialogues and didn't bother to explore the river to probe out the softer targets. Not a very Viking way to lead a raidHaha, to be fair the warnings, scout reports and hints didn't exist in that build. Still, Its rather amazing you could play 5-6 times and not change up your tactics...
There is also AI code that you can add to make existing units scatter around somewhat (and control the distance, I believe). You might remember that behavior from the ES standard RM and campaign AIs. It could make the Varangians a bit weaker as their forces will be somewhat more far-flung, though.Interesting, didnt know you could control the distances. The varangians used this AI logic earlier on but the way they put their units on the very edge of the landmass constantly didnt look very nice. Would be cool if I could get them to have 1 or 2 tile spacing only, maybe.
Maybe add a trigger system that detects which huts are still standing and distributes the tribute accordingly?Indeed I probably should add it to the todo list
A slightly more minor issue, as it did not affect gameplay, was that I noticed several small mistakes within the scenario text and dialogues.Since you didn't mention this previously, I was starting to get my hopes up that it was all good on this front! Only the objectives\hints\history has been put through spell checker to find the mistakes a computer can spot;I will have to give the other dialogues a manual look through sometime
This is in contrast to wheelbarrow\handcart which are the games best technologies from a strategic perspective;mastering when to research these two with their higher costs takes a good deal of skill and experience to get down. I imagine the game would be significantly better if many more techs offered more strategically difficult choices than just those two.There was another cool feature of the scenario's (albeit limited) tech tree, by the way. As you know, experienced players using melee units usually prioritize armor upgrades before attack upgrades (living longer/actually reaching your target = landing more hits, which is worth more than an extra attack point or two). However, for ranged melee units such as mamelukes, throwing axemen, and gbetos, it is better (unless fighting majority ranged armies) to take the attack upgrades first as it increases their ranged damage output. The central role of throwing axemen in this mission made this an interesting choice - do you take the tech first that is better overall for the various units in your army, or do you focus on the strength of your core of ranged melee units? Great stuff.
Something I forgot about was that the player could research Druzhina in earlier builds. This thematic tech fit in well, but juvenal reported a bug where researching it causes his units to damage themselves with area effect damage, something trirem and I couldn't replicate but I decided to remove it to be safe.Interesting. I've never seen this when playing as Slavs in HD, so I wonder if it's an issue with the way the tech was implemented in WK, or if UP changed something, or if forcing techs can create issues. Far be it from my abilities to pinpoint the technicalities of it
Interesting, didnt know you could control the distances. The varangians used this AI logic earlier on but the way they put their units on the very edge of the landmass constantly didnt look very nice. Would be cool if I could get them to have 1 or 2 tile spacing only, maybe.I'm not 100% on that so don't quote me on it. Could also be related to TC location, which means that without a TC it could go haywire. If it doesn't work out, I suppose a more blue collar solution would just be to have triggers task units a couple tiles away from the exterior of each yurt.
Since you didn't mention this previously, I was starting to get my hopes up that it was all good on this front! Only the objectives\hints\history has been put through spell checker to find the mistakes a computer can spot;I will have to give the other dialogues a manual look through sometimeI think that the only issues that I found in the objectives/hints/history were minor punctuation errors (one of the objectives lacks a period at the ending, for example). Taking a few minutes to give it a manual scan probably couldn't hurt, I guess.
By the way, I wonder if you noticed I implemented the faux economy we once discussed quite some time ago in this thread? The AI villages bustle around gathering resources and building structures, while hunters and fishing ships can be spotted working but they don't actually harvest anything and the resources deplete at the slowest rates possible. Instead all AI units have their cost set to 0 so they can train them freely as population and .per file allows. I thought it helps the immersion a little bit not to see every player with 30,000 score off the bat, and maybe casts an illusion that the villages are actually playing.I knew something was up when I saw deer and farms with 30k food content, haha. Didn't know the rest was more extensive than near-instant training. The visual behavior of the AI towns was top-notch. I guess you had the Khazars train villagers from dropsites?
Didn't know the rest was more extensive than near-instant training.Actually, AI unit training times are far higher than in vanilla
I encountered a bug where saving and loading the game makes blacksmith techs vanish from the interface. Fortunately, I was able to win without needing to save and load.I couldn't replicate this bug with either starting inside the editor or starting from campaign selection. Saved and reloaded multiple times including going back to main menu with various different combinations of techs and they were always available upon loading in. Oh boy...well, I dont know that I could fix it anyhow even if I could replicate it.
One other thing, by the way - I noticed that the AI would always train fishing ships before it trained galleys, leading to an infinite cycle of fishing ships being trained and sunk while no galleys were produced when the player is threatening the enemy docks with their longships. If you want to change that, you can just add (not (town-under-attack)) to the fishing ship training rules in the AI, which will cause them to produce galleys to defend until their docks are no longer threatened, in which case they will then train fishing ships.Thanks, this worked! Although they can still make one fishing ship to scout with, I guess. This was actually working as intended, as the lesser of two evils;when the AI had galley training listed higher in its per file it would train them constantly, which resulted in an annoying battle of attrition.
2. Do not waste time and ressources on trying to overthrow the Golden Horde -- concentrate on defending your city instead.CHALLENGE ACCEPTED. HARD MODE ENGAGED.
Interesting, didnt know you could control the distances. The varangians used this AI logic earlier on but the way they put their units on the very edge of the landmass constantly didnt look very nice. Would be cool if I could get them to have 1 or 2 tile spacing only, maybe.Regarding spreading out soldiers, if your goal is to keep soldiers from clumping around their creation buildings, there are a few AIs that set sn-task-ungrouped-soldiers to 1 every minute or so. After a couple seconds, they'll set sn-task-ungrouped-soldiers back to 0. The effect is that they'll spread out and wander around for a couple sections and then stop where they are when the SN is set back to 0. If this is the SN you're talking about, I don't think there's a way to control the distance. The SN that seems relevant, sn-defense-distance, seems to control something else or doesn't do anything at all. I forget which.
Welp. AI with transport ships...Quoted from Kataphraktoi:
And I thought the vikings were great navigators
The landings are done by AI, not triggers. I am sure others like Leif Ericson know better than me how to maximize/randomize the AI's attacks, but it was the best I could do with my knowledge. Randomization (to a limited extent) may be possible, though, but not through the script directly...Quoted from Kataphraktoi:
I should have added an extra comment there;It was pretty good at landing troops, surprisingly so to the point I wondered if you were using triggers to do it. When the player doesn't build palisades on the beach terrain it works well. I played it 3 or maybe 4 times, it was fairly good but could have used some more randomization in attack patterns. (not sure how you could pull that off in a .per file, way beyond my AI skills)
[This message has been edited by Al_Kharn the Great (edited 09-21-2018 @ 09:31 PM).]
I know you weren't as big of a fan of Last Stand at Orkney but I am pretty proud of what I was able to figure out and get the AI to do with the deliberate nudging.I did give it a 4.2 in my (unpublished) review and Siege of Caffa got a 4.6, so its not like I didnt like it. Its a good scenario.
Another interesting thing I discovered was a solution to what I believe has plagued scenario designers who failed to get functional landings from the AI. At some point (perhaps after a certain number of landings or time), the AI stops using transport ships. I could not identify a reason for this, but it was consistent behavior. The AI seemed to simply no longer recognize the transports as transports and left them parked and empty on the water.That is interesting, I will have to remember that if (when) I make a vikings scenario where the AI has vikings. Luckily in my RiverRaids its the player with this unfortunate task of controlling transport ships...
My work-around was to spawn additional empty transport ships for the AI. Interestingly, the AI immediately loads units into these new transport ships, while continuing to ignore the ones they previously had. I set some triggers to spawn new transports whenever more than a certain number of units were on the AI's starting island
Honestly, once I got a reasonable and consistent pattern of attacks from the AI, I did not want to mess with it for the sake of randomization.I'm just saying this hypothetically for next time, but I think at that point you could mix in scripted attacks from an additional vikings AI to good effect. You could direct these attacks at the places and in the times the other viking AI didnt strike. Have them not share LOS to avoid messing their attack patterns up.
[This message has been edited by Kataphraktoi (edited 09-24-2018 @ 06:15 PM).]
Regarding spreading out soldiers, if your goal is to keep soldiers from clumping around their creation buildings, there are a few AIs that set sn-task-ungrouped-soldiers to 1 every minute or so. After a couple seconds, they'll set sn-task-ungrouped-soldiers back to 0. The effect is that they'll spread out and wander around for a couple sections and then stop where they are when the SN is set back to 0. If this is the SN you're talking about, I don't think there's a way to control the distance. The SN that seems relevant, sn-defense-distance, seems to control something else or doesn't do anything at all. I forget which.Thanks, I just got this figured out and implemented and its much better than permanently leaving it on, though I still wish I could limit it a bit more. Setting it to 1 for a single second every few minutes seems to do well.
Sorry, I just saw this! I noticed there were a lot of recorded games in my folder. I will go through them and see if I can't find the one.Quoted from HockeySam:
@Pitt/Great_Artiste/Mash - Please tell me someone has a rec or screenshot of Mash getting defeated by wolves.
[This message has been edited by kud13 (edited 07-09-2019 @ 12:11 PM).]
Copyright © 1997–2024 HeavenGames LLC. All rights reserved.
v2.5.0