temple

I had another idea: Only men can fish. If fishing is twice as good as farming for garrisoned units, then 25 fishermen = 50 farmers = 10 fields, which is quite a lot (and that's not counting the fishing boats themselves). Good players fish, but they can't rely on fish to feed their whole economy like they can with berries or hunting. (Or am I mistaken?) That's what I'm trying to change here. Regenerative fish is a nice idea, although I'm not sure it fits with the current game, since then there should also be regenerative berries and woods (and animals)? There are other threads about this. A better fit would be having infinite fish like with infinite fields, but I'm opposed to this idea too, because the real gather rate depends on how far away the fish are, and unlike storehouses and farmsteads, the dock can't be placed closer to the fish, it's stuck on the shore. Players with fish close to the shore will then have a big advantage throughout the game rather than just until the fish is exhausted.

I thought I'd suggest these changes to make fishing more attractive. Currently it's hard to fish efficiently. The gathering rate is equal or faster than berries (it depends on how far away the fish are from the dock), but only three or four fishing boats can gather at a time (plus a few more travelling to/from the dock), so it's hard to rely on them to support your whole economy. Fishing boats take time to build and after fish are exhausted, they're useless. The fishing boat's tooltip says "Fish the waters for Food. Garrison a support or infantry unit inside to boost fishing rate." I thought that was a nice idea. Military units add arrows to ships, so why can't workers add poles and nets to fishing boats? My tentative proposal is to increase the garrison limit to five units, and have each unit add +80% to the fishing boat's effective gather rate. See D657 for details. There is micro involved in garrisoning units, but since they gather twice as fast as farmers and you can have up to 15 or 20 gathering at a time, I think this should make fishing an attractive alternative to fields (until the fish run out). That in turn should make water control more important, which should lead to more interesting games. People won't have to make as many fishing ships which means they'll have fewer to delete later on. Fishing boats could also be used to transport small groups of units (no cavalry or champions), although that's dangerous since they're slow, have little health and vision, and can't fight back. Let me know what you think.

Trac ticket #4581 is about adding garrison flags to some buildings (like houses). I didn't see a task thread for that so I thought I'd post it here.

There are a few buildings that need rubble art: Civic Center: Gaul, Ptolemy, Seleucid Dock: Maurya, Ptolemy, Seleucid Stone 2x4 Palisades Trac ticket #4644.

I thought I should upload some replays before they're made obsolete by a22. Players: temple (1479) vs. bbleft (1692). Map: Mainland, Medium, 200 pop, 37:06. Summary: 3/1. This was during my Ptolemy phase. I think their free corrals is a big advantage, especially on maps without extra food sources, and I like using pikemen as a meat shield, although I didn't use them particularly well this game. I was nervous, not expecting to win, just hoping to play well enough to not embarrass myself. Players: 9025201729 (1319), sckt (1272), shade11 (1465), Lysander vs. temple (1563), tigas, Emperior (1240), craqqy (1240). Map: Rouen, Medium, 250 pop, 1:17:54. Summary: 3/20. One person on our team dropped early, so it was 4v3. I was feeling claustrophobic this game, being surrounded by enemies, and at one point I send some women out to the corner of the map to farm. There was a lot of lag, especially with all the dogs! But it was a fun game. Players: gabrielpotumati (1420), Dizaka (1472), bbleft (1651) vs. Franksy (1439), temple (1658), marxod (1124). Map: Caledonian Meadows, Normal, 200 pop, 29:46. Summary: 4/26. This was a memorable game because I didn't make any fields, corrals, fishing boats, or trade! I survived/thrived solely on mountain berries and mountain goats. Players: temple (1667) vs. borg- (2153). Map: Mainland, Small, 300 pop, 22:11. Summary: 5/5. I've played several 1v1's against borg; this was probably my best performance. Players: temple (1664), PTK.SirTopito (1414) vs. Please (JC), Emperior (1356). Map: Ardennes Forest, Medium, 250 pop, 57:32. Summary: 5/5. A Rome war. This was an intense game considering all the cavalry micromanagement. Players: gabrielpotumati (1436), Gizum (1449), bbleft (1614), Liberty (1747) vs. temple (1680), chrstgtr (1517), lapatiencos, PTK.SirTopito (1414). Map: Mainland, Tiny, 150 pop, 1:06:39. Summary: 5/25. I don't think anyone's posted pizza yet? This one was pretty epic. Unfortunately the replay's missing the end of the game, but maybe it's better that I don't relive that. temple.rar

Players: bbgotscrewed (bbleft), xPlague, Emperior, PrincipalityOfZeon vs. nigel87, juanito, bigcam87, borg-. Map: Unknown Land, Normal size, 150 pop. Summary: 1:15:21, but took twice as long with jungle lag. A tough game with lots of decisions about where to defend and where to attack, and another demonstration of the power of massed cavalry. 2017-05-31_0004.rar

That was an amazing game, thanks for sharing.

I wasn't able to get D13 working in Phabricator, but I did get something working through wraitii's GitHub. Units attacked differently, but there was some weird behavior (units walking back to a previous point for no reason, before continuing again), so it's certainly possible that I grabbed the wrong branch or the code isn't up to date or something like that. But I thought I'd share what I found anyway. Currently, units line up in either a circle or a rectangle, when they should line up in the circle/rectangle hybrid. In all three of these cases units take a lot of time walking around to their correct spot at the end of the line. In wraitii's rewrite what happened was that some units (even when selected individually) walked too close to the building, which left room for more units behind them. The effect of this was that the units attacked sooner, since they didn't spend so much time walking. For example, here is a group of archers attacking a civic center. Normally they'd stand shoulder to shoulder in a big arc, but here they stand three or four deep. I like this effect, but normally you'd do it manually: attack the cc, walk forward a bit (or try to capture the cc) for just a second, then attack the cc again. In that time the units in front all walk forward a step which leaves room for the units behind them to get into range. Here's skirmishers attacking a civic center. You can see the outline of the correct circle/rectangle hybrid shape on the outer edge, but again many units walk too close to the building. The main problem with this is that when your units outrange the enemy's buildings or units (they walk too close when attacking units, too), if they walk closer than they need to they risk getting hit. Hopefully I'm just looking at obsolete code. But otherwise I'm not a fan of this change. Also, the "nearest dropsite" bug was still in effect, so I made a ticket for that.

I was wrong with the first line, units are within range if they're within range of the object's obstruction, which is a rectangle for buildings. The problem is in the pathfinding, where it approximates the correct circle/square hybrid shape by either a circle (which works fine) or a square (which doesn't work so well) depending on how the unit's range compares to the size of the obstruction. However, that section of the code's being removed in the D13 rewrite, so for now I think this bug can be ignored.

In the pictures below I ctrl-clicked to attack the cc. The top row shows what usually happens, with units in the back walking around to their place on the ends. But in the bottom row I ctrl-clicked a second time while they were walking around, and the units stopped in place and started attacked the cc. So the units in the top row are definitely in range, but for some reason they walk around to their position on the square before they start to attack. I think currently units are within range if the center of the object is within range? And maybe that's going to be changed in D13 to if any part of the object is in range? That's my understanding. In the first case, the units should line up in a circle around the cc, and in the second case they should line up in a circle/square hybrid, in straight lines parallel to each side, but then 1/4 circles around each of the corners. Playing around in the editor: Archers actually line up in a circle like they should! Slingers and skirmishers line up in a square like in the picture. Hmm, playing around more: Attacking a barracks instead of a civic center, slingers will line up in a circle too! But skirmishers still stand in a square. Skirmishers will stand in a circle to attack a house though. So maybe it has something to do with how the building's "radius" (whatever that means) compares to the unit's range? I'll make a ticket for this. For the other things on my list, I searched for old tickets that mentioned them, and only made a new ticket if I didn't find anything.

Okay, will do.

I think it might be a simple fix, so I can give it a shot. I haven't done any patches/tickets/etc. before though, so it might take me a few days to figure everything out.

Here's some suggestions I have for the game. I've tried to avoid obvious ones (like improved pathing), but probably many of them have been discussed before. BUGS (1) When constructing buildings, workers should add a certain number of hitpoints to the building each second, but instead they add the floor of that number. This makes buildings take longer to complete in a lot of situations. For example, one unit building a 1000hp defense tower should finish it in 150s. But instead of adding 1000/150 = 6.667 hp to the building each second, they only add floor(6.667) = 6 hp each second. So it takes them 1000/6 = 166.7 seconds to finish instead of 150s. If you've ever just barely failed to finish a tower or fortress or something, this might've been the reason why. The worst example of this is in building a civil center. It has 3000hp and takes 500s to build, which is 3000/500 = 6 hp/s. With ten women, they each build at a rate of floor(6 * 10**0.7 / 10) = floor(3.007) = 3 hp/s. All together that's 10 * 3 = 30 hp/s, so it takes them 3000/30 = 100s to finish. However, with eleven women they each build at floor(6 * 11**0.7 / 11) = floor(2.922) = 2 hp/s, combined 11 * 2 = 22 hp/s, so it takes them 3000/22 = 136.4 seconds to finish. So in this case, adding an extra woman to help "speed up" the building process actually slows it down by over 30s! I think this occurs in the following line in "public/simulation/components/Foundation.js". Is there a reason to use Math.floor? var deltaHP = Math.max(work, Math.min(maxHealth, Math.floor(work * this.GetBuildRate() * this.buildMultiplier))); (2) Workers drop off resources at the nearest drop site, but it seems that "nearest" is calculated using the distance to the center of the drop site instead of the edge. Can this be changed, or is the calculation too complicated? (3) Units stand in a square formation to attack buildings, why? They can attack from a circular range but instead walk closer. IMPROVEMENTS (4) Can we have a hotkey to cycle through buildings where we have units garrisoned? We currently have one to cycle through idle units, but garrisoned units don't count as "idle". More than a few times I've put women in houses to protect them from raiders, only to forget about them for the rest of the game! You can have units garrisoned inside allied structures, so the hotkey should cycle through them too. (Also, houses should get a flag when units are garrisoned inside.) (5) Can we have a hotkey for the town bell? (6) Can we control which units towers, etc., attack? I'd like my tower to focus on the opponent's tower, rather than also attacking their houses and fields. (7) Can we let women and worker elephants place the foundations for every building? Right now it's annoying to select a man chopping wood, have him place the foundation for an outpost or house on the other side of my base, click back to wood so that he doesn't move, and then select the woman or elephant to go actually build the building. It's a lot of unnecessary micromanagement. Either let women and elephants build the foundations themselves (my preferred solution), or if you don't want women building barracks or whatever, then don't allow them to build them period! MORE ANNOYING THINGS (8) In order look at unit stats (to check if your opponent has blacksmith upgrades, for example), you have to stop your cursor over the shield icon and then wait half a second for the tooltip to pop up. I'd like it if that just popped up immediately while you're hovering over the icon. The half second delay is annoying enough, but because I use a pen rather than a mouse it's actually hard to stop exactly on the icon. (9) There's only 13 civs if we count random, do we really need a scroll bar? Also in replays, can we make the player drop-down menu wider (when there's room) and longer, to again remove the scroll bar? (10) In the lobby, can we select and copy chat? There's conversations or links I want to save, but I have to settle on taking a screenshot. EXTRA THINGS (11) I don't know how complicated it would be, but I'd like to see corrals treated different in the summary page. Currently, if you make one sheep you're credited for one unit created, 100f gathered and 50f spent. This is technically correct, but in the game you're at the same point as someone who's simply gathered 100-50 = 50f from berries, and their stats are 50f gathered, 0f spent, with no units created. So instead I'd like sheep to be ignored in terms of "units created", and for them to count as -50f gathered rather than +50f spent. So after one sheep they'll have zero units created, 100-50 = 50f gathered, 0f spent, the same as the berry player. In this way, the unit and resource statistics will be comparable between those who use corrals and those who don't. (12) I'd like "random" civs to be a 1/12 chance of each one, rather than the current 1/9 chance with briton/gaul and athens/mace/sparta grouped together. I think this is it for now. I'm looking forward to Alpha22!