real_tabasco_sauce

yep, that makes sense to me. One thing to be sure of is to try and not make the system too complicated. Since the upgrade progress from level 1 to level 3, the costs should progress with some logical flow as well. perhaps a flat rate for one non-conflicting resource and some progression for the conflicting resource: wood upgrade: 300f 0w, 300f 300w, 300f 600w stone upgrade: 300m 0s, 300m 300s, 300m 600s metal upgrade: 300s 0m, 300s 300m, 300s 600m the food upgrade could be left as is. I will echo others here and say I also have no issues with the current model.

Yes, but maybe not garrison. Not 100% sure on that. yes, but I'm afraid the distinction and their roles are quite a core structure to the game, and in order for the nomads to fit in they must have some adherence to the standards of the other civs. Right now it feels like a different game almost. The same goes for territory dependence. if they are quick to build, it could work out nicely. I could make a proof of concept sometime. This already sounds better. One risk could be an expansive women boom turning into a flood of archers with an upgrade, so maybe it would be better to promote to this unit? I envision a successful defense from a raid turning into a counterattack using the headhunters. This also adds some nice risk and reward elements.

My understanding is that the upgrades cost 'their own' resource because it forces a decision between investing in long term eco vs short term spending. However, there are other more important examples of opportunity cost (ie training soldiers early vs women), so I'd be open minded to changing the costs. My main concern would be setting the costs so that the upgrades are always a 'no-brainer'. I think its too late to change this for a27, but what I would suggest is submitting this as a merge request for the community mod when it gets launched for a27: https://gitlab.com/0ad/0ad-community-mod-a26 this will be a place to try out gameplay ideas without the risks of releasing potentially negative balance decisions in an alpha. as for beating the AI, don't just put workers on all the resources. for the first 8 or so minutes, you want food and wood because you are training units and building houses and/or barracks. If you have a slinger civ, it helps to start on food and wood and add a few to stone later. This should give you the numbers you need to beat the AI as long as you are constantly spending resources. If you find yourself 'saving up' to try and get an upgrade or reach town phase, that is a problem. A good economy is about rates not about the total resources you have at one time.

Some feedback (one or 2 of these i have mentioned before): allow the CC to train while moving. (both civs) don't allow cavalry to build. buildings can only be constructed in your territory (thus carts can only be unpacked in own territory) Allow Ovoos in p1, exchange their stone trickle for a defense/damage bonus within 60 m. Ovoos instead also provide a territory root, the idea is that you can place one and then move the cc elsewhere. This allows a lot of flexibility while maintaining the core feature of territory influence. I would also move the stone tech back to the storehouse. split the stable and corral, and give the stable a small structure (perhaps for keeping supplies dry). I think the market/food deposit hybrid is fine for xiongnu. I don't really like the maiden archers and cav archers. Being free means a risk of some spammy gameplay. Also it might be confused with normal women. I'd say just let the starting women be fairly strong, which makes sense because of the lack of towers and how weak the buildings are.

yes It is reproducible (big blue tear): When starting a single player game setup. when leaving back to the menu. not when entering lobby. when clicking host and when confirming to host when arriving in multiplayer gamesetup. when clicking start game in gamesetup when leaving game setup to go back to the lobby. when clicking the chatbar to type a message. when clicking away from the chatbar to not type a message. not when viewing player info or the leaderboard when leaving the lobby to the main menu. in game: when opening and closing the chat (either with enter or by selection with dropdown) when pressing backtick to open the console not when viewing the tech tree not when viewing settings/options/hotkeys I can sometimes see fragments of the upper left side of the screen within the tears (for example, lobby participants) my OS is ventura 13.3.1

It only happens in vulkan + fullscreen. Setting borderless fullscreen false does not fix it, turning on windowed mode does fix it. Starting a screen recording also fixes it, which is why I recorded my screen externally.

The packable CCs are already fine for this. Also, I would rather nomads have mobility in their offense, not defense.

Are the thracians ruled out? I think the Suebians would a bit more familiar civ structure and thus better received. I totally support the nomadic civs, its just that I anticipate more disagreement on their units and mechanics. Plus, I think work remains on making the two nomad civs look different.

A 2v2 on a27. Pretty good performance all around, but @Palaiologos wouldn't let @BreakfastBurrito_007 and luchomat to full pop metadata.jsoncommands.txt

LuchoMat did this yesterday with no problems: download bundle from the forum install. A warning will appear warning that the previous version will be uninstalled. click 'OK' The installer will suggest a directory (folder) for the installation. This will most likely be the current location of a26. Delete the last part of the 'path' and replace it with something like '0ad_A27'. I think the shortcuts folder should also be separate, but I am not sure.

a little ele balancing: https://code.wildfiregames.com/D5008

Turns out, thats pretty much how it is currently playing, so no need to add explicit counters. The gameplay effect is neat, all that remains is to ensure it is balanced. I am not sure if https://code.wildfiregames.com/D5008 is enough to nerf them, so I would probably support either further reducing splash damage or decreasing health or pierce armor.

Yes, I wasn't trying to make it so explicit as an equation. Such equations aren't really helpful when balancing things like this as there is a lot more influence of context in real battles. Thats why I am calling for playtesting.

yes, I was just talking about the merit of eles vs spearmen in that case.

ok, i just tested the nerfed eles with a mod I made: eleNerf.zip This mod only applies to a27. Eles are still strong even when both sides are fully upgraded (blacksmith upgrades don't apply to eles). The test I did was 50 archers + 10 eles vs 50 spearmen and 50 skirms. (80 pop vs 100 pop) the archers won with 42 remaining and 0 eles remaining. Possibly this is still OP, but 10 eles (big eles) costs 5500 res (2200 metal, 3300 food) while the spearmen only costed 2500 food and wood. Possibly this is still too OP, but we have to remember that these are champions, and also that micro could massively decrease their effectiveness (make the enemy eles chase units). They do go down pretty fast to massed ranged units, so perhaps the balance is fine and they could just be considered a known counter to melee units going forward.

Its pretty exciting for gameplay and all civs have rams. Eles still have pretty decent capability vs buildings, but don't expect to take down a fortress with just 1.

Its true, but I had a huge upgrade disadvantage, slightly outnumbered, and somewhat poorly placed my pikemen. @Philip the Swaggerless probably wasn't aware just how powerful they were, but I think the eles did too well in that fight. You can do unit tests if you want. You will find that 10 eles can kill 100 spearmen and only lose 1 elephant. So either their damage should be adjusted: https://code.wildfiregames.com/D5008 or maybe their HP reduced, now that they are not an alternative ram.

There was a time you could capture catapults and bolt shooters.

Windowed mode seems the only thing that fixes it. The big blue screen tears still happen with vulkan+borderless.fullscreen = 'false'

yeah I'd support making it either a lot darker or a lot lighter, like a silvery grey.

splash attack -10 crush + direct attack + 10 crush One could even move all crush to the direct attack to streamline a bit. Something like direct: 30h 60c, indirect: 20h

metadata.jsoncommands.txt Alright the first posted multiplayer game of a27 rc. Not very balanced, and I had a rough eco. The elephants I trained as @Philip the Swaggerless attacked did unreasonably well, so I think a nerf is warranted. I am thinking 20 hack 20 crush for the splash, and then add back the 10 crush to the direct attack.

is it actually supposed to be false as stated here or "false" like the other options?

ok going out of fullscreen worked perfectly well. Screen tears only occur in fullscreen for me, and borderless fullscreen = "false" did not solve the issue in fullscreen.

borderless.fullscreen = "false" did not fix it. I do have a26 and a27 rc concurrently installed, but not running at the same time. I will try later with a26 uninstalled.