Topic: Barbarian Trainingscamp and no warmill

31 — 40 of 40 Posts ‹‹ previous 1 2 3 4

Adamant Joined: 2012-10-11, 15:21 Posts: 180 Widelands-Forum-Junkie Location: Alemania	Posted at: 2013-05-27, 18:59 teppo wrote: So you put another icon for each type of soldier level next to the generic soldier icon in the window where you set the economy target for all wares. This would solve the problem with soldiers staying in the training facility even if you want them out. Is this what you meant Adamant? I figured out multiple Problems resp like to have more/different Controls about Training and Military Units - generic said. I did work around by blocking Goods/Soldiers in WareHouses EVERYWHERE but those neigbored to a/the TrainingsSite. Thus all Militrary Stuff and Soldiers went there centrally. If all Military Sites are occupied like set additional Soldiers remain in that WareHouse and went out into a TrainingSite if it offer a possible Advancement to that Unit - no Matter what Stuff gets produced. I can avoid to train a Solider to Elite by avoid producing necessary Stuff but thos PreElitary Soldiers cram the TrainSites and I have to kick them out manually. Another Point is that I don't want a Freshie to get expensive Steel by advanced Gear but to archive basic Training first which costs just cheap Food. Last State of Training-Process I archived anytime but just with a Lot Assitance resp Reliability and with full Traning to TopGun. What I want to do is .. may be different resp is different from Player to Player. What I miss is a better Control what means all Kind of new Controls are welcome to me as there are just those via driving the Production. What Controls can I imagine? EG a generic Carreer-Path for Soldiers explaining what Way Soldiers train up, a Definition of aimed Quantities (eg 100 Leet-Warior) or a TrainingsPlan what to train next similar like a ProductionPlan with Items what to produce. Think simple and like a General: you need Gear and Units. What Gear and what Units do you need? Where do you need them? You don't want to handle them always one by one or as Group or as Garnision of a Site - but all that Time by Time. I did several Time rather disappointed that I was not able to specify a Site (selecting it), selecting the Units I want for an Attack and attack with those an EnemySite (or EnemyUnit). Let's assume my best Horse, a medium trained Unit moves out - the other 4 Units move out from other Sites more distant. My best Fighter arrive alone .. well they do Duells anyway .. but due to Calculation the other Soldiers comes to late and my best Soldier did continue as SuicideMission and fight with 33% HP and die tightly at last. Matter of Calculation was that he beat two while other takes the Remain .. failed due to the wrong Units attacked from wrong Sites. Even if I get aware of that right after Instruction of Attack I can not do anything but watch like my Attack starts with the wrong Leg - no Way to retreat, abort or anything. I can't select a Unit or Group but just an empty MilitarySite. Lost of Control if there ever any Control was over the Soldiers aside of Ware-Handling via the Eco-Conrol and specifc Site-Management for Amount of Soldiers. Aside of all Control-Matters there is something I miss extremely in WideLands Military Dimension to support Diversity in Strategies: * XP for Soldiers (like for Worker) eg Recrutee, Corporal, Sergeant, Lieutanant, Captain (0..3 Strives) with better LeaderShip a Soldier * Diversity of Soldier-Type - at least a BowMan as MissileWeapon eg Range 2..3 Fields Imagine that: there is Castle from me and an opponents Castle and my Site takes Territory first. I plant there a Lot of Trees. Now I can observe, that the Opponent can send his Lumberer into my Forest to lumber my Wood and my Soldiers don't do anything to hold him up. The don't protect Borders or Territory but only friendly/own military Sites. They can't patrol, they can't do Recon similar to a Scout, they can't guard a specific other Site (and ignore another Attack). It sounds surely rather hard if I say I did try to ignore the military Dimension of WideLands by a minor Share of Pacifism but the mayor Share about Unsatisfaction about the military Options inside the Game. Nevertheless I like the Game and I like that there are rather many different military Sites which are in my Eyes visually (->Perception!) the BackBone of military Dimension even the TrainLevel of Soldiers is not lesser important than the Site-Size. I won't tell somebody else Soldiers in WideLands can have 16 different TrainStates but in WideLands is per Tribe just a single Type of Soldier but each Tribe have many different Types of MilitarySites - so that's the BackBone. The TrainingSites looks fine but do visually litte aside of consuming Goods and due to the Impact of Training their Meaning is perhaps larger than that of Site-Size. From Aspect of Strategy and Tactic it's rather poor to have as only Option to apply plain Force against Force. I miss any Scissor-Paper-Stone-Matter and other Things which enable more Option to utilize special Attributes and Abilities. Eg Sword-Man is heavier and thus slower but have better protection while BowMan is liter and faster and lesser armored and Advantage by RangedAttack but DisAdvantage by CloseCombat. As 3rd I like SpearMan which get its most Value by Mounted Units. I would like to see the Scout to habe a small Soldier-Ability even he is weak but have a good Recon/SightRange. If he get a Horse as Mount he may scout with doubled Speed resp doubled Range. Also for Patrol-Matters in MilitarySites he is useful if he have an increased Sight and Speed to alert the Site and return with Enforcement - he is Recon and not Commando. I don't understand that Players can't attack civilian Sites and Units. I would donate HP to Sites and make them attackable and close Military Sites may send out Troops to defend attacked Sites no Matter if Military or Civilian. Fazit: there exist a Lot of Things where an Improvement can access. At present the Military can't do a Lot and there are just few Methods to control it. If there is a Control for Troops it could serve fine for Fleets as well as we have to deal there same or similar Matter. I don't like the Idea of Military as Crown of Economy but as long as there is no other I want the Military Matter with better Control and Features and Diversity to have more Fun with. Btw: the Icons for Unit-Stats are not really well descriptive to my Eyes (I try to guess the Pixel-Area of the Enemy-Units their Icons as Gauge to guess the Opponents Strength - I can still ever NOT map an Stat-Icon to a Stat-Value. I could train it, yes... but I can't use it well in Battle: I can use it to check my Forces but I can hardly selct them. If an Enemy attack I can't do aLot but wait for Reinforcement resp change Settings to increase Forces. I can not call immediately additional Defenders. Effectively these Icons means to me (only) I can see what Power the Attacker have and compare with own Forces resp guess how or understand why my Forces bashes or get bashed. Brief Point is I would prefer if that Stats are represented different or those of Attacker are invisible as the present Way means to me the Screen is covered by Stat-Icons and I can see almost nothing else. Ivan the Terrible is dead .. Genghis Khan is dead .. and I do not feel well, too. Top Quote

teppo Joined: 2012-01-30, 08:42 Posts: 423 Tribe Member	Posted at: 2013-05-27, 19:54 Adament, why do you put karlrune's words to my mouth? I asked Astuur (and Horatio and others) to playtest and give feedback, which is very different from what you just wrote. Top Quote

Adamant Joined: 2012-10-11, 15:21 Posts: 180 Widelands-Forum-Junkie Location: Alemania	Posted at: 2013-05-29, 14:31 Another Try... You Teppo asked that I put Karlrunes Words into your Mouth. I don't know why exactly but if I read the Thread now that Words origin from Karlrunes. I'm not sure but may be you expect an good Answer and not just a Sorry resp Thanks for Clarification. Sorry and Thanks for Clarification. I did answer a Question of Karlrune - that Question you did identify to origin from Karlrune. Miracle solved? Edited: 2013-05-29, 15:39 Ivan the Terrible is dead .. Genghis Khan is dead .. and I do not feel well, too. Top Quote

Astuur Topic Opener Joined: 2009-02-28, 09:08 Posts: 733 One Elder of Players Location: Frankfurt / Germany	Posted at: 2013-06-02, 22:11 Teppo: >Astuur: Windows build bzr 6573 contains the feature we discuss. Please try it, and share your findings. Sorry I missed that. I have been away some time. I am testing bzr 6576 now. I suppose that also should be okay for testing? I am not quite there yet to understand what else happens, but I can see soldiers coming out of the trainingscamp when they have no chance for further training (high level arms and armour missing). So far so good. Obviously you don't do any preselection of the soldiers but leave that up the trainings site. I have a lot of traffiv now on the road from soldiers of the "untrainable kind" that still stubbornly go to the site, just to be kicked out some time later. But hey! - it works! You can use the wares input control sliders to adjust the trainingslevel that you want. I have a graphical problem in the line immediately under "Trainingscamp". Where "upgrading soldier ......" should be visible I get funny typwrite-like characters, black on a white background -- sometimes. But that may be completely unrelated. That is an awkward testing as WL lacks all capability to show soldier's training levels when they are inside the warehouses.... At some time that ought to be changed, too. For the moment I shift them from one warehouse to the next to see what I have :). I'll do more testing tomorrow and see whether there are other implications... just wanted to let you know that I am at it! Seems good so far and is certainly an improvement from what we had. Being no programmer, I apologize for all my suggestions that imply undue workload and for other misjudgements due to lack of expertise or relevant skills. I am on Win32, have no means to compile, and rely on prefabricated distributions (Thanks to Tino). Top Quote

teppo Joined: 2012-01-30, 08:42 Posts: 423 Tribe Member	Posted at: 2013-06-03, 06:29 Obviously you don't do any preselection of the soldiers but leave that up the trainings site. Thanks for the quick response. Quite right. I could, of course, require that the replacement must be less trained than the kicked-out guy. That would Stop the untrainable soldiers from wandering between the warehouse and trainingsite Cause an "overproduction of half-trained soldiers", making it more difficult to get fully trained soldiers. I was feeling that at least the loudest guys here did not want that. I personally like making it even harder to get fully trained soldiers, and surely would not oppose changes towards that direction. I do not see the graphical thing you refer to. Maybe it is a windows related issue? That is an awkward testing as WL lacks all capability to show soldier's training levels when they are inside the warehouses The training matrix is four dimensional, which is difficult to visualize neatly. Maybe that's why. It might be rather straightforward to make the soldier count vs time graph a stacked one, showing own soldiers' general levels as different color shades. Maybe you should make another feature request on that? Top Quote

Astuur Topic Opener Joined: 2009-02-28, 09:08 Posts: 733 One Elder of Players Location: Frankfurt / Germany	Posted at: 2013-06-03, 12:57 Okay Teppo - let's see. Trying to follow your thoughts: Genrally speaking, I would have liked to know some more details about how you achieved the new logic. Now the typical programer's answer to such a querstion is "look at my code" -- but I am code illiterate. Maybe I can talk you into explaining a bit in plain words I am currently playing bzr 6576, Empire and got 2 traingscamps and one Colosseum running. Questions: 1) After the candidates arrive at the training site - how is determined whether they stay or are kicked out again? Kicked whenever training fails? Here's why I am asking this: The trainingsite does indiscriminately kick out soldiers, even when the cause for failed training is not a lack of weapon or armor, but a lack of food. That obviously does not make much sense, except, maybe, if you have other trainingssites that are better supplied in this respect because your road system is inadequate. If you can iron that out perhaps, I think a policy like "send soldier with different training level" (different from the one that was kicked out) may be good. Your code could produce a lot of semi-trained soldiers at times. Be it, that gold or cloth is missing, or that simply not enough iron per time arrives at the weapon/armoursmithies. (some more advanced item need 2 ingots at once, plus coal etc, and will so fail if not all ingredients are present at the time of the attempt) Given the different reasons for failing you cannot guarantee that it will always be wise to try a "cheaper" training level. It may be otherwsise. Hope I make sense. With the current logic, I am a bit worried about the slots in the trainingscamp and the time spent idle, when very many untrainable soldiers keep coming and each is blocking the site for a while. Also one must consider the time it takes to march to the camp and back. This is especially true if you reduce the site's capacity in order to get your trained soldiers sooner (trying to counter the unfortunate effect, that a trainingssite always tries to level all resident soldiers before it applies the next higher training). I often do so, but it could now lead to a trainingscamp that is effectivly not working at all for some time. Hope you get my point and I have no logical errors in my thinking well, this stuff is complicated *Addition:* but there is more: There is also still the problem with the sequential training of unrelated skills. The way I interpret the the conf files there are usually two skills trained in the same facility. As an exmple I use the Empire training camp. The Hitpoints will never get increased as long as the Attack training has is not been completetely done. This is another feature that I never really understood in its true purpose. We could easily rearrange the programs in the conf file, so that the two skills are interwoven, but the real solution seems to me, to have the skills be independent of each other. There however, I am not sure how it must be done. But it seems to me a good moment to tackle this also, as this oddity truly limits the usefulness of the changes you are intending. In the example above, if an economy had no gold, the training of soldiers even with your code would stop after attack level 1(advanced lance needed), because attack levels 2 and 3 each need weapons including gold.(heavy lance, war lance) So the soldiers would be kicked out with no Hitpoint training at all, even though the hitpoint training itsself up to and including level 2 (0=helm, 1= armour, 2= chain armour, 3= plate around) does not not need any gold! Would it not make sense to clear up such irrational interdependencies? Or maybe the original designer had a good reason for this that I have failed to grasp? PS: the graphical glitches seem indeed independent... Edited: 2013-06-03, 17:04 Being no programmer, I apologize for all my suggestions that imply undue workload and for other misjudgements due to lack of expertise or relevant skills. I am on Win32, have no means to compile, and rely on prefabricated distributions (Thanks to Tino). Top Quote

teppo Joined: 2012-01-30, 08:42 Posts: 423 Tribe Member	Posted at: 2013-06-03, 19:18 how is determined whether they stay or are kicked out again? Kicked whenever training fails? The whole kick-out thing is now training site -driven. If the soldier is at levels that have repeatedly failed, the soldier is a kick-out candidate. If training works, the relevant counters are reset. There is a logical and between types of training: If any of a soldier's level is one that is considered to work, he is safe. Maybe I can talk you into explaining a bit in plain words The history of the training site is important. The soldier's history is irrelevant. A single success in training clears away old sins. Also time makes them bleach, though slowly. When a certain level has already been declared as non-working, the kick out happens rapidly. The trainingsite does indiscriminately kick out soldiers, even when the cause for failed training is not a lack of weapon or armor, but a lack of food. That obviously does not make much sense, You are right. However, I disagree with your conclusions. Especially in case of evade, where only foodstuff is needed. If weapons are supposed to be difficult to get and food easy to get, then lack of food should not be a reason for repeated training failures. From the player's perspective, the whole kick-out thing is not always beneficial. However, it doublessly reduces the benefits of micromanagement, and emphasis the benefits of a fluent economy. Map size permitting, I usually build a warehouse and providers of relevant food items in the proximity of the trainingsite, to smoothen the flow of precious wares. "send soldier with different training level" In my mind, requesting a less-than would make more sense, but nothing prevents us from making a not-equal request. At least the code looks as if it was possible, I have not tried. I will think about it. If everybody else likes your idea, I might think even harder! Your code could produce a lot of semi-trained soldiers at times. Yes -- there is more training with this patch, but on the other hand, fully trained soldiers are harder to get than they were previously. I see that a good thing. It also makes micromanagement less profitable. I see this to be a good thing. With the current logic, I am a bit worried about the slots in the trainingscamp and the time spent idle I would guess that the sites are less idle now than they were previously. A road system must be damn lousy, if the site kicks off all its tranees before the replacements arrive. Of course, if the economy consumes soldiers faster than new ones are made, then the soldiers end up into half-full military sites and trainingsites are more likely to be empty. I do not see a problem there either: If all soldiers are at battle field, then are probably needed there for a reason. Else the player would retreat, wouldn't he? Hope you get my point and I have no logical errors in my thinking No -- you speak clearly. I just disagree. This is a question of taste. One reason why I would not like to discard kick-off because of lack of food is that I would like to introduce one or two additional evade training steps, which would be very expensive in terms of amount of food (bread meat fish beer whatever) and increase the evade skill only marginally. The motivation being that those steps would be so expensive and the gain so small, that those training steps might not be what the player wants. Foodstuff has some value elsewhere (like in empty mines) too, even if there is no real shortage. The motivation would be to make the "golden path" less clear, which would add some salt to the playing experience. One thing that I have not thought too much is the eagerness of the sites to kick out. Each site type has a configurable parameter saying how many times the training must fail in a row before the whole kick-out starts. That number could be tuned up -- or down. Currently barbarian training sites are least patient. If you wish, you could try looking at [widelands installation directory]/tribes/empire/trainingscamp/conf file (and similar files for other sites) with notepad, and find a line saying trainer_patience=[number] and then changing that number. In the example above, if an economy had no gold, the training of soldiers even with your code would stop after attack level 1(advanced lance needed), because attack levels 2 and 3 each need weapons including gold. I have tried with barbarians and atlanteans, and that did not happen. I have not tried with Empire. Did you really try, or were you speculating? If you really tried, I could take a look. Thanks for the long feedback, I value it also when we disagree with some details. Edited: 2013-06-03, 19:22 Top Quote

Astuur Topic Opener Joined: 2009-02-28, 09:08 Posts: 733 One Elder of Players Location: Frankfurt / Germany	Posted at: 2013-06-04, 14:02 Wow .. that has become quite long.... Sorry about that, but I could not manage any shorter. At least I have tried to give it some structure So let's get right into it! teppo wrote: The whole kick-out thing is now training site -driven. If the soldier is at levels that have repeatedly failed, the soldier is a kick-out candidate. If training works, the relevant counters are reset. There is a logical and between types of training: If any of a soldier's level is one that is considered to work, he is safe. Ah .. thanks for the explanation! Sounds very good. Can it be that you have factored into it the training at the respective "other" training site? I am asking this because I have made the observation that soldiers inside the Empire trainingscamp stay there (seemingly for ever) when they still lack the Evade training at the Arena or Colosseum -- even when there is no more training for them at the site they are currently in. Could this be a consequence of that last statement of yours? The history of the training site is important. The soldier's history is irrelevant. A single success in training clears away old sins. Also time makes them bleach, though slowly. When a certain level has already been declared as non-working, the kick out happens rapidly. I won't pretend that I fully understand how it works, but I can see that you have clearly put a lot of thought into the optimization. From the player's perspective, the whole kick-out thing is not always beneficial. However, it doublessly reduces the benefits of micromanagement, and emphasis the benefits of a fluent economy. Map size permitting, I usually build a warehouse and providers of relevant food items in the proximity of the trainingsite, to smoothen the flow of precious wares. Yep I do likewise - and I agree an the effects that you mention. It's one of the strong benefits of your addtion. *Again about the "No food -> kick out" rule and my opinion that it should not happen:* You are right. However, I disagree with your conclusions. Especially in case of evade, where only foodstuff is needed. If weapons are supposed to be difficult to get and food easy to get, then lack of food should not be a reason for repeated training failures. True, but still it happens often enough. The demand for bread and meat/fish rises considerably when you run Training sites, and many players run into food shortages once the map's fish/game resources are gone. The Empire cannot replenish any of it. All said, it does happen - so we had better take this situation into consideration and figure out how to react. Nothing of what you say speaks in favour of making the "no food" condition a reason for shifting solders. If it is painful to distinguish between "no weapon/no armour" on one hand and "No food" on the other, then it does not do much harm to let it stand as it is. But if you easily can take out the "no food" condition, I think you should do it. The reason is simple: It serves no purpose und is not benefitial. Every training step for each and every soldier needs food. No training can happen without. So if this is the only reason for kicking out, there is no cure for the situation by calling other soldiers in. You may just as well keep the soldier you have. Okay, there is an exception: Sometimes you need double rations. So exchanging such a soldier for one that needs less food for his level may be a benefit. But even there, I think it may be easier and more effective to wait a moment.(i.e. keep him in and retry later) *About the "send soldier with different training level"* In my mind, requesting a less-than would make more sense, but nothing prevents us from making a not-equal request. At least the code looks as if it was possible, I have not tried. I will think about it. If everybody else likes your idea, I might think even harder! I have no idea if or when any of the regulars will jump in here. I don't doubt that, if training fails, sending a "cheaper" soldier will be the solution most of the time, but it may be worthwhile to try the "not equal". Your code could produce a lot of semi-trained soldiers at times. Yes -- there is more training with this patch, but on the other hand, fully trained soldiers are harder to get than they were previously. I see that a good thing. It also makes micromanagement less profitable. I see this to be a good thing. I can't really see why fully trained soldier will be harder to get. Can you please explain a bit? And I am also not so sure whether I agree on wanting to make them harder. It depends. If you are saying "make them more expensive" I might agree, If you mean "further delay their creation", or "make their training even more time-consuming" I will not! In my experience, depending on the map (many of them are quite small) currently games are often won by high numbers of relatively untrained soldiers. It's a legitimate playstyle, but doesn't always make the game experience better. I don't like it very much. I agree about micromanagement. *About overall efficiency of the training sites.* I said: With the current logic, I am a bit worried about the slots in the trainingscamp and the time spent idle I would guess that the sites are less idle now than they were previously. In a situation where something is missing (gold, cloth, ...) they will be more efficiant - no doubt! And if that is not the case, the training will go on much the same way as it did without your contribution. So -- bottomline - this was nonsense on my side. Sorry. I sort of jumped to the idea in a situation where there was no cure (see below) and saw the soldiers shifting endlessly. *About introducing training levels that are expensive but not very effective.* ...This is a question of taste (noFood-kickout). One reason why I would not like to discard kick-off because of lack of food is that I would like to introduce one or two additional evade training steps, which would be very expensive in terms of amount of food (bread meat fish beer whatever) and increase the evade skill only marginally. The motivation being that those steps would be so expensive and the gain so small, that those training steps might not be what the player wants. Foodstuff has some value elsewhere (like in empty mines) too, even if there is no real shortage. The motivation would be to make the "golden path" less clear, which would add some salt to the playing experience. Hehe, are you forgetting that we are not alone here, Teppo ? And even if players don't read this postings, they will soon find out. that these options are not worthwhile. Two things will happen then: 1) We will get a bunch of complaints why these training levels are so ineffective, equally many suggestions on how to change it. 2) The options will simply be ignored. I can clearly see what you are trying to do, and I do agree that different tactics or even strategies are welcome. But actually WL offers already a variety of ways to play the game, and some players find it already rather a complicated game. Still more could be done, and I agree about your "golden path" idea in principle. But the alternatives must be valid in themselves. To illustrate what I mean, let me give you an example: If someone could teach the fighting code in the game to distinguish between "Defending" (some other agressor is about to attack my buildings) and "Attacking" (I am myself the agressor), we could have an option for the player to let his soldiers specialize in either one. It would then open the door for many additional ways to play. Real Alternatives, not options to fool players, is what we need in my opinion. Setting "trainerpatience"_ One thing that I have not thought too much is the eagerness of the sites to kick out. Each site type has a configurable parameter saying how many times the training must fail in a row before the whole kick-out starts. That number could be tuned up -- or down. Currently barbarian training sites are least patient. If you wish, you could try looking at [widelands installation directory]/tribes/empire/trainingscamp/conf file (and similar files for other sites) with notepad, and find a line saying trainer_patience=[number] and then changing that number. Cleverly done; but all in all it seems to be quite good "as is". *About the forced consecutive training of unrelated skills and its consequences* I wrote: "In the example above, if an economy had no gold, the training of soldiers even with your code would stop after attack level 1(advanced lance needed), because attack levels 2 and 3 each need weapons including gold." I have tried with barbarians and atlanteans, and that did not happen. I have not tried with Empire. Did you really try, or were you speculating? If you really tried, I could take a look The moment I wrote this I was merely speculating, well not really "speculating", but deducting from former experience. But your experience has puzzled me that much, that I have meanwhile created that exact situation described and found my "speculations" were wrong! The training for every soldier skill stops where it needs to (for lack of gold in this case) -- so you have managed to solve that? I could swear it always was like that, wasn' it? Or does my memory totally fool me? Wonderful! Great work! *About the Empire Trainingscamp (endlessly?) retaining soldiers that need to go to the Colosseum.* Please take look here Marked in red you find the relevant things. The total number of soldiers is 742 now. and I have a total military stregth of 7797. Deduct from this what I had before I started the test: 564 Soldiers and a total stregth of 6204. 6204/564 = 11; That simply means that the soldiers I had were all maxxed out. (1 point for the mere existence, and 10 for the training levels). So I have for the new 178 soldiers (742-564) a strength of 1593; that is 8.95. It's not an Integer as there are some Soldiers in the Trainingscamp that ought to go to the colloseum - but they are not kicked out. The rest is trained to a current maximum. Strangely enough, even after the kick out of those soldiers, I don't get a clear "9" (8 levelpoints are missing, but I cannot find them!) (Numbers given may differ slightly from the screeshot which is older) Those soldiers cannot profit from their stay at the Traningscamp and use up one slot each. I suspect these are the soldiers that initially did not find a place in the Colosseum and so first went to the Trainingscamp. You can kick them out manually and they will go to the Colosseum and complete their training. So this, after all may be the only point where you might want to look into! Or am I wrong again :)? Thanks for the long feedback, I value it also when we disagree with some details. Disagreeing is the essence of any discussion. So that is perfectly fine with me. I like discussions and many things are quite thoroughly discussed here. I may be right, or wrong at times... or simply differ in preferences --- in any case I can learn. Edit: corrected lots of misspelled and mistyped text. Doesn't pay to type things in haste Edited: 2013-06-05, 05:43 Being no programmer, I apologize for all my suggestions that imply undue workload and for other misjudgements due to lack of expertise or relevant skills. I am on Win32, have no means to compile, and rely on prefabricated distributions (Thanks to Tino). Top Quote

teppo Joined: 2012-01-30, 08:42 Posts: 423 Tribe Member	Posted at: 2013-06-05, 14:13 Hi again. If it is painful to distinguish between "no weapon/no armour" on one hand and "No food" on the other, then it does not do much harm to let it stand as it is. The piece of code handling the training site knows that it need wares 7,10/11, and 16 for one step. It does not know which of those were done at bakery, which at helmsmith. It would, of course, be possible to add such knowledge. That might be a maintenance issue later, when there is one more place that needs to be adjusted if somebody altered the economies (or introduced a fourth tribe into the game, for example). In short, currently foostuff-wares are not different from other wares. Making a distinction might lead to ugly code somewhere, but is of course doable. Maybe my view is biased since we still disagree on this. Somebody else, cut in and comment. Hehe, are you forgetting that we are not alone here, Teppo I remember seeing a poll here at widelands.org some time ago. I suppose that the webmasters know how to make polls. Would that be an overkill? What should the poll options be? We will get a bunch of complaints why these training levels are so ineffective, equally many suggestions on how to change it. On the other hand, we might also get complains that soldiers are not exchanged fast enough. I have been doing another micromanagement related modification -- namely allowing the military sites to exchange soldiers automatically when a more trained (or less trained, depending on what the player prefers) becomes available. It actually plays quite nicely together with this kickout code, in my opinion. This other modification is now stable enough for playtesting. Windows version not foreseen in the next few days. Anyway, if that is also taken into use, somebody might be irritated whenever a soldiers remains at training! Please take look here Marked in red you find the relevant things. The screenshot was taken when the economy did not consume soldiers. Currently, when kickout happens, it is always the most trained, stalled solder that will go away. Also arts that are not taught in that school count. In your case, lack of Evade levels makes the marked soldiers to be "less trained" than the others. Since your economy does not consume soldiers, the more trained ones go back and forth, which causes the standstill. Thanks for the remark. I will do something. In general, I got the feeling that you are not suggesting that we should remove this kick-out thing from the development branch? Top Quote

Astuur Topic Opener Joined: 2009-02-28, 09:08 Posts: 733 One Elder of Players Location: Frankfurt / Germany	Posted at: 2013-06-05, 15:49 OKay, This time I understoood your explainations All of them! As for polls Yes, there is definitely an option to do polls. If I remember right, there was a time when everybody could do one - seems gone. Also Sirver had initiated regular polls at a time, but he's not so fond of them now. Even though nobody ever promised that the polls would in fact influence the further development, we all felt some urge no to totally ignore what seemed a majority. And democracy really works best when all voters are equally informed and equally intelligent - which definitely is not the case. In any case,, no poll make sense without a thorough discussion prior to voting. Best give them all little time - I am sure they will react. It is strangely quite currently. In general, I got the feeling that you are not suggesting that we should remove this kick-out thing from the development branch? Heaven forbid! It's an excellent improvement. It cures the standstill, it solves the intertwined skills trainings, and it works without killing anything else - as best as I can judge. Who would want to go back? I hope I have already asked most of the silly question, so that others need not bother you again The development crew may have other, not so silly question that they want to have explained, but from a player's perspective, I am very pleased with the solution. Thank you very much! I must stay away from WL (not totally, I hope) for some days soon, but will try to play the other tribes ASAP. Being no programmer, I apologize for all my suggestions that imply undue workload and for other misjudgements due to lack of expertise or relevant skills. I am on Win32, have no means to compile, and rely on prefabricated distributions (Thanks to Tino). Top Quote