Jump to content

xfs

Community Members
  • Posts

    22
  • Joined

  • Last visited

  • Days Won

    1

Everything posted by xfs

  1. Carthaginians Samnite Skirmisher, Macedonians Heavy Skirmisher, Persians Cardaces Skirmisher are champ javelinists. Looks like they are removed from the game.
  2. I doubt your testing method is scientific. For example, I just did a test of 20 city guards vs 20 immortals (same stats). City guards won by small margin. The ultimate source of data is in-game "Display selection state" where it shows unit stats. The reason nobody uses Ptolemies is because right now everyone uses champ only strategy and Ptolemies don't have good champs, not because its base units are bad. Champ spear cavs have very big differences than other champ cavs, so I don't what you mean by this. Iberian champ cavs are in a weird place. They are very weak for anti-units because units all have very high crush armor. They are also very weak for anti-building because they have low pierce armor (generally worse than elephants for anti-building). They are also weak for scouting because they are not fast enough, and with low armor.
  3. Persians' champions are not "really nice". It is the same spear champions as other civs. But indeed, the low champion production capacity is unbalanced. Ptolemies free houses are penalized by long build time and high cost of barracks, so this is fair. I do use Ptolemies. My army composition is archer cavs and pikemen. This composition is strong in that pikemen are the toughest non-champion tank units in the game and they are prioritized by UnitAI thus protect archer cavs, and archer cavs are a decent DPS with steady output and easily micromanaged. This composition is very effective in defense. Give it a few catapults and it can slowly grind out any fortifications. I managed K/D ratio of 2 during a 1v3 non-champion game using this (lost 450, killed 875). However, this totally does not work against champs, but that is a problem of champs, not the army composition. I don't think champ spear cavs need a buff. Their role doesn't seem to be DPS or tank. They are most effective in scouting, macro, surprise capture, and trade route raiding because of their fastest speed and decent armor. I do think elephants die too fast right now, not sure if it is because of attack prioritization in UnitAt or not enough armor.
  4. @Alekusu do you have replays with Aol_Wonox, aka Benz, aka IdiAmin, aka superd? He does good rush while managing econ equally well.
  5. The secret sause sauce is to build units in 10-batch or 15-batch.
  6. Silver Shield Pike's armor is irrelevant and formation buff will not help with that. Their extremely slow walk speed and low attack make them useless on a macro level. They cannot defend against anything. The enemies simply go around and they will not be able to catch up.
  7. This may be unpopular among Britons players, but for the sake of diversity in strategy, I believe the champion units right now can use some balancing. Right now there is a dominant strategy, that is to use Britons and build sword champions spam. Britons' economy is quite advantageous for quick mass producing of champions starting in mid-game 15 minutes, and in late game once the market trade route is set up, the macro advantage is almost unstoppable. There are several factors that contribute to this optimization situation, and can be considered for potential balancing measures. Sword champions can be trained in barracks, and Britons' barracks are very cheap (300 wood, no stone) for mass training champions. Other civilizations can only train champions from fortresses or champion buildings, with severely limited production capacity. Possible balancing measure: increase the training time for champions from barracks.Swords champions have questionable unit roles. They have deus ex machina level stats, with highest attack, highest armor, highest capture power, fastest walking speed among melee infantry. They are the best at everything and tactically superior to any combinations of other units of equal costs, and they are extremely strategically maneuverable because of their fast walk speed. There is almost no point in making any other units. If the unit roles are properly designed, they should have weaknesses. What is the weakness of sword champions? The consequence of a single unit beating an army composition is greatly reduced diversity in economical and tactical strategy.To discourage creating an army of only champions and balance against the combat effectiveness of champions, a potential balancing measure is to make champion units cost two population.Melee champions should not be both the most combat effective and the most maneuverable at the same time. Their combat effectiveness should cost their maneuverability. This is also realistic, because heavier armor and weapon should make champion infantry move slower than their non-champions counterparts (while they should indeed move a lot faster if under-armored, e.g. Naked Fanatic, or Maiden Guard?). This principle is also represented in the weaker stats of champion cavalries because they have much better maneuverability.Melee champions should have a primary unit role of anti-infantry, not anti-building. Right now a bunch of sword champions can hack down a fortress without significant loss, which is not quite realistic. They should either be more vulnerable to defensive buildings' pierce damage with the same attack strength, or defensive buildings should have higher hack armor. Spear champions have half the attack being pierce damage therefore a lot less effective against buildings. If somehow sword champions are considered to be justifiably more effective against buildings, they should at least gain a weakness compared to spear champions.Finally, is the capture power of champions too high? 5 or 7 if boosted by hero, compared to 2 of normal units. This makes defense towers totally useless. 3, or 4 for champions seems more appropriate. But if champions gain vulnerability in other areas (slower, or lower pierce armor), this advantage may still be appropriate to their unit roles.
  8. Under optimal conditions, a corral takes 250 food to produce 5 sheep of 500 food in 81 seconds (3.1 food/s). Cavalry meat gathering speed is 5. Under optimal conditions, a farm with 5 women outputs 5*0.5*(1+0.15+0.15+0.25)=3.875 food/second. The problem with corrals and cavs is once the sheep are all finished before the next batch is out, all cavs become idle and will not proceed gathering the next batch of sheep.
  9. Please provide a replay that shows 130 Romans pop at 10min (while doing this early soldier making).
  10. Okay, and? I was not saying it is hard to have 11 soldiers at 5min. It was an attempt to match Celtic civs' economy at 10min. If you make soldiers early on for defense, the economy obviously suffers, and is more worse off in the long run than Celtic civs. What happens if I make defensive soldiers but you continue booming with women? There will be a bigger economy disparity at 10min. To solve this, obviously one must scout. And one must keep scouting to find out possible discrepancy in military strength and sign of rush. In the meantime the economy must go on without wasting a single second, and bifurcate the build order depending on what the opponent does. Multitasking all these is not easy. Scouting in 0AD takes sharp eyes, as the units are really tiny and obscured. And counting them at a glance is even harder. And now your build order is not fixed, instead it depends on the opponent and unluckily for Hellenic civs, you may not have enough wood to at times to make the decision. A simple answer may be just git gud, but I might as well not have time to play.
  11. The reason I lost is when Alekusu rushed at 5 minutes with 11 soldiers, I have not made any soldiers. So there was a big tactical disparity. The reason I have not made any soldiers at 5 minutes is that Hellenic civs have slower economy. Their houses are slower to build than Celtic civs (60 seconds vs 24 seconds). This makes the economy of Hellenic civs inherently slower, with lots of time wasted on production waiting on houses, or houses waiting on wood. In an attempt to match the economy of Gauls, I had to save wood for houses only in the first 5 minutes, and made women only.
  12. I think it is probably not hard to put the production queue on top of the building. But artistically, you probably want some special effects to represent production, e.g. smoke going up, or people going in and out. Unit AI does not have a problem with looking for the next trees. Mines though, yes, but it only needs a few times of managing. You can just use it as a regular builder. Defense towers are quite weak right now, though. More nerfing would not help. Healers have no AOE, and take up population count. I think they are far less useful than temples, if not totally useless. Run is used as a hack for units to quickly go into position in a formation. It would only make sense after stamina is implemented. You can also use i-left drag to select idle workers. But really, the GUI should show numbers of how many workers are gathering what, and how many are idling, and then you know to look for them. You can use Ctrl-q-right click to move attack units only.
  13. Using archer cavalries against RCF Diamond (top 5 player, Gauls). Diamond has better economy than most of the Britons champ spam users and were able to defeat them. First game, a Romans player and Seleucids (me) with no economy against Diamond and a Britons player with massive economy advantage. Two more games, a Britons player (using champ spam) and Ptolemies (me) against Diamond and a noob. Discussion: Basically, archer cavs are a quite good DPS (9.8) if protected by high armor infantry, which takes the aggro. Among non-champions, pikemen have the highest armor. Egyptian pikemen under hero aura have almost the same HP and the same armor as champ swordsmen, and no other non-champions are even near. So I was testing this army composition against champ spam. The critical weakness is pikemen walk half the speed than champs, making it useless for macro and sometimes totally ignored. Also, champ swordsmen run too fast, so purely archer cavs can't win a micro battle against champs. Javelin cavs are not a good DPS even though they have the highest raw number (22.4, higher than champ swordsmen). Their short range makes them die fast and run around a lot before actually starting to attack. archer-cavs-against-rcf-diamond.zip
  14. Not really off topic though, because "I think you can start from uploading them into this topic. If there will be a significant amount of posts, they can be split/moved into a subforum."
  15. Swords cav has 2 capture points so that's not a great advantage in capturing. Besides, the point of swords champions is their massive armor to allow them to ignore everything else until finishing the capture.
  16. There is indeed something new. A19's capture mechanism changed the whole meta-game. Right now there is one single predominant strategy, that is build tons of swords champions, storm in the enemy's base and capture the CC, and build more champions. All the 1400+ score players I played with use Britons and use this strategy, and anyone who does not use this strategy is defeated very easily. There is no more siege, and nothing else matters. Personally I find this new meta-game unbalanced and lacking in diversity. (Another strategy demonstrated by ffm and Wonox above is fast rush and economy skirmishing for game under 15 minutes. OK, this is pro level build order and economy managing, I don't enjoy it and don't have time to practice for it). The basic fact of this mechanism is that, ordinary soldiers have 2 capture power per second, but champions have 5 capture power per second, which can be further boosted by heroes to at most 7. Normal buildings have 500 capture points, CC's have 2000 capture points, fortresses have 4000 capture points. This means with 50 champions CC would take 6 seconds to capture an undefended CC, or 11 seconds to capture an undefended fortress. What happens after capturing the CC is that the CC now generates the attacker's territory and all buildings in that territory get "captured" by "territory pull" 7 points per second, which means all the infrastructure is very quickly finished. Swords champions are able to do this because they have the highest stats in the game: (when fully upgraded) 12 hack armor, and 12 pierce armor both negate 72% damage, and quite high DPS (24.1 hack, or 29.1 if boosted by a hero), and they walk almost as fast (13 meter/second) as the fastest infantry, skirmishers (14). Britons, Romans, and Iberians have swords champions, but Britons' hero offers an additional 25% walk speed boost than the standard +5 attack +2 capture. This is why Britons is the most popular civ now. Spear champions are strictly inferior to swords champions because of lower attack and slower walk speed. Champions are expensive, but it is not very hard to build 100 champions in under 25 minutes by just gathering. There are several implications from these facts: The high capture power of champions decides that towers and fortresses are easily captured and useless for defense, with any number of normal soldiers in garrison. (Unless the defender has an equal number of champions to negate the capture power.)The high armor stats of champions decide that they can walk in and out of heavily tower defended areas unscathed and finish capturing before dying off. This also means archers and skirmishers civs need a huge number of ranged troops to deter (but still can't kill) swords champions, and this army composition is strategically bad for attacking defense structures.The high attack stats of champions decide that even if the defense structures are garrisoned by champions, they can still kill the buildings by brute force relatively fast. This further adds to the uselessness of defense structures. Capturing the CC is now a straightforward job to destroy the opponent. And the only real defense is to also use Britons and build an equal or more number of swords champions.Siege weapons are as fragile as paper, easy to capture, very slow, and in general much more inefficient in offence compared to above. So they become pretty much irrelevant to the game.Using temples to heal champions is now considered an economy strategy, because champion casualties mean massive economy loss.Walls and turrets cannot be captured, and they may be used for blocking champions. But they are mostly passive and stone is often hard to come by too.A side meta-game derived from the main meta-game is now to deny the opponent access to metal, first scout the map, locate metal deposits, and expand there and defend it. Stone is pretty much irrelevant because fortresses and towers are irrelevant for defense. I am a long time Romans player since a15. Romans have specialty including siege towers, which costs only wood instead of stone, and can be built in neutral and enemy territory, and Scorpio, bolt shooters, which attack invisibly, have splash damage, and used to have the highest DPS (90) in the game. Unfortunately both were seriously nerfed. Siege towers are much slower to build and much more expensive, Scorpio's have much slower rate of fire (DPS now: 49). I personally prefer Romans' melee infantry core, and kept fighting with Britons' slinger rush before. This replay demonstrated my base defense against champion spam. It's a 3-way FFA with julijan (Athenians), imrobbyg (Britons), and me (Romans). Ignore other twos players who were irrelevant. This one is fairly long, one hour, so you may want to use fast forward setting, and Alt-D change perspective. Outcome (everything below is spoiler): I built 300-400 siege towers, which posed a clear deterrence and attrition to champion spam and forced them to retreat many times. Romans siege towers cost 100 wood and 130 seconds to build, have 2000 HP, 17 hack armor, 37 pierce armor, range 96, shoot 1 arrow (DPS 10) empty or 4 arrows (DPS 40) fully garrisoned. Also, it has much less surface for melee champions to attack. A thought experiment: 30 siege towers fully garrisoned kill 30 swords champions in 18 seconds. Important times and events from my perspective: 17 minute, imrobbyg economy skirmishing against julijan.20 minute, imrobbyg started champion production.25 minute and so on, imrobbyg destroyed julijan with champion spam.~40 minute, I built 100 swords champions (I could earlier, but was building towers), and through some strategic maneuver luring away his champions, I attacked his main base.I captured his CC, but forgot his wonder also generated territory. His infrastructure was not converted, and my 100 champions failed the mission and died. I had no more access to metal.I destroyed two of his outer bases.imrobbyg initiated siege ram and champion combined attack. I lost some towers but as soon as I had melee infantry on-site I was able to defend. The siege rams are very quick to kill off, and the champions garrisoned inside cannot afford to go out to defend the rams and endure massive arrow fire. I was focused on something else and didn't bother.I built 100 more spearmen (twice), and through some maneuver, destroyed one more outer base, and captured one fortress at his main base. I should not have garrisoned the spearmen in that fortress because it was very fast to get captured back by champions while the fortress defense of 20 arrows per second did minimal damage to the champions.I destroyed his wonder with a sneak team covered by maneuver elsewhere to attract attention.At this point, I resigned. I could still defend indefinitely. I could build 300 more siege towers, and I had the reserve to deter him from throwing everything and emptying base defense. But generally siege towers alone lacked the strategic initiative compared to concentrated champion spam, and siege towers are too slow to build offensively. So this was a losing battle. TL;DR: Use Britons, build 50 swords champions in 20 minutes, 100 champions in under 25 minutes, become unstoppable.Scout the map, capture metal deposits, deny metal access to the opponent.If your opponent use champion spam, seal your main CC with walls, move farms elsewhere, and do not build defense towers and fortresses anywhere except your main base, and metal deposit.Don't even build defense towers in your main base, which get captured fast. Build wall turrets only, which is strictly better than defense towers for defense purposes (no distance restrictions, cannot be captured vs 500 capture points, 4x/5x HP, 96 range vs 88, half the arrows though you're not gonna have the garrison in time).Do not build any siege weapons except siege rams garrisoned with 5 champions.Do not garrison any building after capture if you opponent have champions. Just destroy it (except CC, wait for the territory pull).britons-vs-romans-ffa.zip
  17. Now for the commentary after reviewing the replays. I'm surprised ffm is consistently defeated. In my impression ffm is already a very good player, and I've learned one or two tricks from him. Wonox definitely manages his economy extremely well, with 130 pop at 10th minute (in one of the games), which is an achievement itself (previous discussion on Gauls build order has a claim of 110 pop at 10 min, and myself can only get 100 pop with Romans). But without attacking, I think ffm's economy is almost as good, and Wonox only has a slight economy advantage, leading with 5 more pop at 5th minute. I think the deciding factor is that Wonox brings the fight to ffm early on (in all 3 games), which magnifies the economy disparity. A very simple fact in this game is that one soldier fighting is one soldier not gathering. Another very simple fact is that the defender will have to commit more troops to the fight than the attacker, either to ensure not losing the local fight (the attacker comes in concentrated formation, but the defenders are scattered), or by accidentally selecting more than necessary troops, which is very prone to happen (it's quite hard to estimate the number of troops without selecting). These two facts together imply the defender will suffer more in economy. Another effect of being attacked is it does not goes to plan, and it will screw up the defender's build order. Taking a closer look at the first game. At 5 min, Wonox dispatched 20 soldiers to attack ffm's economy. At 6 min right before the fight commenced, the two almost had the same economy level: ffm: 25 women gathering food, 25 soldiers gathering wood, (5 soldiers in training)Wonox: 15 women for food, 10 women and 5 soldiers for wood, 5 soldiers for stoneThe fight lasted 1 minute and 5 losses on each side. All ffm's troops (25, and 5 later trained) were committed to defense against Wonox's 20 and ffm's wood gathering stopped for 1 minute, while Wonox continued wood (10+5 pop) and stone (5 pop) gathering for 1 minute. The effect on ffm's build order was also very clear: house building was interrupted, and consequently hampered troop training; the barrack went up 1 minutes later than Wonox. The population difference went up to more than 10. And that basically decided the outcome. TL;DR: The first striker enjoys an economy advantage by forcing more troops in the defense to halt production.
  18. It didn't show up in the replay list. After unchecking "Filter compatible replays", it shows up in gray, and shows "You don't have the same mods active as the replay does. Required: mod, public Active: mod, public, user". I don't have any custom mod and I don't know what is a "user" mod. I can still manually edit the replay to change the mod requirements.
  19. OK, let me test this replay sharing method. To replay, extract the attachment and move the extracted folder to under sim_log. The replay shows a high-level player's Britons slinger rush strategy. You can see his economy level. britons-slinger-rush-1v3-match-1.zip britons-slinger-rush-1v3-match-2.zip
  20. Now that the game can visualize replays, how to share replays? I can see a sim_log directory, and there are sub-directories like these: 7483, 7483-1, 7483-2, 7483-3, 7483-4, 7483-5, and commands.txt and metadata.json files in those. I suppose I could pack some of these directories into a zip file which can be shared, but I have no idea how this is structured and which ones to completely pack up.
  21. Their spearmen are not weak. In fact their spearmen are standard, with exactly the same HP, same armour and attack stats, and same cost as others. And their spearmen are 25% faster to build.
  22. Have you considered balancing sword cavalry units? In my testing with Alpha 17, 20 sword cavalry units (with no micro) decisively defeat 30 of any combination of spearmen, skirmishers, and swordsmen (with ideal micro). They also defeat 20 spear cavalry with significant margin. They also defeat 20 of any combination of pikemen, archers, and slingers. They do not defeat 20 pikemen and 10 slingers, but they still have great map maneuverability. They do not properly engage ranged cavalry because of auto-micro but if that's disabled the result would be similar to infantry skirmishers. Sword cavalry units are very cheap: 80 food, 35 wood, and 20 metal. They overwhelmingly defeat swordsmen of similar cost. They have the highest DPS (24.6 hack/second) among non-champion cavalry and infantry units after Babylonian Scythed Chariots (31 pierce/second). It seems nothing in the entire infantry and cavalry roster can properly counter sword cavalry spam. Also a question: my machine doesn't seem capable of building the entire game. Is it possible to test it with only public.zip patched with the updated templates?
×
×
  • Create New...