Boudica

Thanks for uploading. I'm adding my commentary as a spoiler:

Please locate the replay file files (subdirectory with commands.txt and metadata.json files) and attach them because a proof is usually required. @user1 usually handles these issues, so he'll check if your request is legit and ask Dumbledor to give you your points.

If nothing has changed, it's usually easier for moderators to delete your account, so that you can re-create it. I wouldn't recommend creating another account because that could get you banned. Let me tag @user1, but perhaps there is someone else specialized in these issues. By the way, if you find your user.cfg file, there should be a lobby.password entry, which probably contains your password hash and makes your password pre-entered when available.

Hi, have you tried enabling STUN while opening the host from the lobby? If your ports are correctly forwarded, this shouldn't be needed, but it could help otherwise. I had problems with port forwarding because my internet provider doesn't forward ports unless you pay some extra. On the other hand, some people have problems connecting to STUN-enabled hosts due to some security restrictions. If you are on the same network, connecting directly to the host's local IP address should work even more reliably. There could perhaps be problems with firewall settings, but the problem typically is that people enter the wrong address. Maybe try sharing more details about what you're entering. I hope we can get it to work.

Today we played an epic winter battle with several base swaps and a lot of sieging. Let's try uploading more of the good games. RamboXXX + Boudica + Cesar vs. bbgotbanned + Unknown_Player + Stockfish 2018-12-11_Winter-Battle.zip

This could mean it's already fixed in the SVN version, if you don't want to wait. Can you please confirm, @wowgetoffyourcellphone?

The problem is not you being unclear, but you being wrong. The original formula has only one scenario when it is undefined, that is when no game has taken place. Your formula, on the other hand, has an infinite number of cases when the result is undefined (all cases when the player has never won or never lost). It's not really clear what you mean by symmetrical shape of the formula or why your formula should be rated higher in respect to this property. This is where you should be clear. Except from the definability problem detailed above, I don't think your formula is an improvement due to these additional reasons: Most players' rating is gonna be a negative number, generally unbounded, but typically in an interval too narrow to consider rounding to an integer. The original produces ratings between 0 and 1 and can be naturally thought of as a percentage. With the exception of zero, your formula always yields a transcendental number. I'd rather say that my eae rating (i.e. win ratio against the reference player) is 0.1 (i.e. 10%) instead of having to use -2.197 as an approximation. Your formula doesn't give an intuitive idea of how good a player is. You can mostly only tell if they beat the reference player more often than they didn't. When you get a simple 5% instead, you just see they have won one in every 20 games so far. Your formula doesn't provide any more information than that anyway. The original formula is simple enough for anyone to compute in their head; yours isn't.

This formula isn't better in any way, and it makes the infinity problem worse.

Any chance you could be accidentally making a scroll-up gesture on the touch pad, or touching the mouse wheel or some function buttons that have to do with scrolling? Can you perhaps describe more precisely what exactly is happening? When you zoom out, is it instantly going back?

The problem still persists for me on a recently unupdated SVN.

I don't understand anything you are saying now. You did say you win most of our games. I checked that and saw that it's wrong. I didn't really accuse you of lying, but I pointed out that your estimates were wrong. Anyway, I'm glad that you corrected yourself. Top 10 out of the currently active players sounds much more realistic to me than the top 3 you used to rate yourself. Next time you can just avoid making false claims when you know it leads to a predictable chain of responses that prove you wrong.

You could have noticed that I like to encourage people to post interesting replays, so maybe my like had little to do with you being on the losing side. However, in order to understand that, realizing that the world doesn't revolve around you might be needed, so I don't think we can do much here. You might have promoted yourself to a gold player in your terminology, but we can't just unsee all the many games you one-sidedly lost to players you'd like to view as inferior. I only have the full data of replays between us two, but let me help you calibrate your views using this example. Recently you claimed to win 80 % of games against me. The truth? You only really got to 50 % by winning the last one (which took place after your claim). You didn't resign in a single of those games before the host was closed, but that doesn't make your defeat disappear. Your numbers are way off and I think you should be more respectful towards other players. A top player would probably beat a hobby player like me in at least 90 % of the games.

There might not be a predefined shortcut to only select healthy units, but do you know about the Ctrl+O drag modifier that deselects wounded units from an active selection? Also, if you use numbered groups of units, you could just assign the wounded units to a different key, which takes them out of their original group. Suppose you Ctrl+1'd your army before a fight. Now when you O-select the wounded units and press Ctrl+2, pressing 1 again only selects healthy units of the original army. This could be exactly what you need if you want to keep using the 1 key to select your army after sending wounded units elsewhere.

I removed the extra mods from the replay, so MacWolf and others can find it in the game dialog. Regarding that game, I can't believe you didn't resign when the organ eating commando completely destroyed your eco. ;-) commands.txt

Angen was faster to post his reply, so I'll just emphasize what might be the probable cause in my opinion. The name you are entering is not the game name, so this field might have to be unique for each participant. Since you mentioned before that both of you have the same IP address, I'd think you were talking about the public one, which is really not what we need here. You probably don't have to be afraid to say what exactly you are entering, so we can find out possible problems easier. If the values are actually correct, there might perhaps even be a firewall / router problem.

I'd make sure you are entering the local IP address when connecting directly. That one would probably start with 192.168... Are you sure you are entering the local IP? I suppose you are on the same network. Correct me if I am wrong though.

@user1

Good choice.

Except for the fact that that army didn't really stop Feldfeld's activity and that it was borg who ran out of wood.

There are the replays of the final match, fixed to be usable without mods. Thank me later. borg-Feldfeld-final.zip

I think that it's usually required to attach the replay files as a proof. Each game would usually generate a commands.txt and metadata.json files inside a new subdirectory, which you'd optionally zip and mandatorily attach with your post. See https://trac.wildfiregames.com/wiki/GameDataPaths for replay file location.

Smiley is right. I agree that defending ram spam can be hard though. Melee units are generally good against them but they also get killed first in a fight because they are in the front line. You would ideally want to have swordsmen ready in advance near important buildings and then keep resupplying during the fight. You can then use the ram trolling trick where you garrison up to 20 swordsmen in a CC or fortress, set the rally point to the incoming ram and then ungarrison all at once to destroy a ram fast. Keeping soldiers garrisoned in a CC or fortress for longer makes it more expensive for the enemy to protect their rams because their support loses health fast. When the attack is too big and it's clear you aren't able to save the building, it might be better to just focus on weakening the support and only getting to rams later when they aren't well protected. Sending women to kill rams when there are skirmishers near is usually a waste. Instead I'd suggest to keep them at the opposite side of the building repairing the building as soon as possible (use the J + Right Click shortcut to request repairing even when the workers carry resources). The same holds for ranged units if there is no support nearby to kill (otherwise you probably want to focus on killing the support units first). Smiley had a good point when he said that losing a CC doesn't equal the end. You probably want to prevent all your buildings from converting to Gaia, which can be achieved not only by having another CC; a military colony, Iberian monument or a Persian palace are cheaper buildings that save you from getting Gaiaed. Keep in mind that rams are premium in price, so if the enemy loses four rams to destroy a CC, it might no longer be worth for him. Don't let losing a CC stop your military plans. Usually a good counter-attack is possible if the enemy sends all soldiers to your base. On a final note, I think we haven't mentioned heroes in the list of ram counter options. Hero cavalry swordsmen are one of the best units to use against rams. I'd even suggest trying to seek and destroy unprotected rams before the enemy sends them into an attack. While you don't let the enemy have too many rams at once, he won't be able to make a decisive push.

That's false.

Except for the fact that spectators are disallowed by the rules. :-) Thanks for checking the time zone.

Maybe we should add a rule regarding scheduling when the matches should take place. What if one party doesn't cooperate? Anyway, I thought that using Doodle might work, so I set up one for our first match. Doodle does not require creating an account, you just write your name and check the dates and times that look good to you. It also seems to take into account your PC's (or Mac's) timezone, so everyone should just work with their local time (I'm not sure about that one though, so please check which time zone it shows). Boudigtr Vs Lesters: https://doodle.com/poll/ymqvd4q82ninxksq Yeah, I only included six different times a day, I hope that's enough.