Alpha 12

mumpsimus
Algorithm
Algorithm
Posts: 63

Re: Alpha 12

Post#81 » 04 Feb 2018, 18:52

RC5 looks good mostly, I have one thing that I'm not sure is a bug. In this screenshot:

farthest_from_enemy.PNG
farthest_from_enemy.PNG (858.9 KiB) Viewed 1836 times


It looks to me like the bot isn't choosing the resource farthest from enemies, but maybe I'm not understanding the calculation? Is it the biggest distance to the closest enemy bot, or the biggest distance to all enemies? If it's the former, the filter should probably be named 'farthest from enemy' not 'farthest from enemies'. In either case, to me it *looks* like the wrong resource is being chosen.

Edit: I checked, and the upper resource is within medium range of the enemy SG, whereas the lower resource is at long range from the sniper.

User avatar
GFX47
Dev
Dev
Posts: 2914

Re: Alpha 12

Post#82 » 04 Feb 2018, 22:36

mumpsimus wrote:how does the new matchmaking work, and why does it obsolete autoqueueing?


The new matchmaking has the same filtering and sorting rules (elo/xp ranges, cooldowns, etc) but instead of just scanning pending matches, it also "reuses" past ones.
It's kind of an hidden auto redeploy but even on draws / defeats.

In ranked matches, the elo gain / loss is only applied to the player who "really" deployed.
Elo gain / loss is also applied to the reused deployments (but not to their original player).

Here's an example, let's assume all players are in the same elo range and that no match was ever deployed:

At T1: Player 1 (P1) pushes deployment 1 (D1) > no past deployment = pending

T2: P2 pushes D2 > he gets matched to D1,
Elo delta is applied to P1, D1, P2 and D2

T3: P3 pushes D3 > he gets matched to D2
Elo delta is applied to P3, D3 and D2 (not P2 because it didn't volontarily engage in this match)

T4: P3 pushes D3b > he gets matched to D1
Elo delta applied to P3, D3b and D1

T5: P1 pushes D1b > he gets matched to D3b
Elo delta applied to P1, D1b and D3b

T6: P1 pushes D1c > he gets matched to D2
Elo delta applied to P1, D1c and D2

T7: P2 pushes D2b > he gets matched to D3b
Elo delta applied to P2, D2b and D3b

Etc

Right now, you don't get too much matches because you have a high elo and only the few alpha 12 players use the new matchmaking.
The more players with similar elo will play, the more you'll have matches, independently of the player activity during your session.

mumpsimus
Algorithm
Algorithm
Posts: 63

Re: Alpha 12

Post#83 » 04 Feb 2018, 22:44

Thanks for explaining this!

User avatar
GFX47
Dev
Dev
Posts: 2914

Re: Alpha 12

Post#84 » 04 Feb 2018, 22:45

mumpsimus wrote:Thanks for explaining this!


I hope it's clear enough?

User avatar
GFX47
Dev
Dev
Posts: 2914

Re: Alpha 12

Post#85 » 04 Feb 2018, 23:16

mumpsimus wrote:RC5 looks good mostly, I have one thing that I'm not sure is a bug. In this screenshot:

farthest_from_enemy.PNG

It looks to me like the bot isn't choosing the resource farthest from enemies, but maybe I'm not understanding the calculation? Is it the biggest distance to the closest enemy bot, or the biggest distance to all enemies? If it's the former, the filter should probably be named 'farthest from enemy' not 'farthest from enemies'. In either case, to me it *looks* like the wrong resource is being chosen.

Edit: I checked, and the upper resource is within medium range of the enemy SG, whereas the lower resource is at long range from the sniper.


Fixed in RC6! > viewtopic.php?f=4&t=1294#p14896

Ed-209
Algorithm
Algorithm
Posts: 53

Re: Alpha 12

Post#86 » 04 Feb 2018, 23:16

Both links are now windows, no android

User avatar
GFX47
Dev
Dev
Posts: 2914

Re: Alpha 12

Post#87 » 04 Feb 2018, 23:32

Ed-209 wrote:Both links are now windows, no android


Fuuuxed

mcompany
Autonomous Entity
Autonomous Entity
Posts: 872

Re: Alpha 12

Post#88 » 05 Feb 2018, 00:06

GFX47 wrote:matchmaking

I think that's an interesting way of doing it where there's always a match, however, I feel like this setup has a few problems
1. We always join on the right side. Considering the fact that LR was bugged to where the left side always won the middle resource, I think it is a valid worry that there might be some imbalances with always joining the same side
2. No matches that I didn't volunteer for are shown to me. However, I think it is very important that I be able to see the latest loses I have, no matter if I won or loss
3. Furthermore, if I deploy an AI to test a trick against specifically those in my league, this forces me to either used unranked or private matches to do so, lest I want to affect the elo of everyone else
4. Speaking of the elo, I feel like such a system is way more prone to hyperinflation or hyperdeflation in the scores for any active player which constantly wind or constantly loses. At least in the auto deploy method, if you lost, you didn't give your opponent a million opportunities to get to 2000 for free
5. For anyone with multiple accounts, this makes it possible to run into each other even if they try to take all precautions to avoid each other

Imo, I feel like this system will result in more games, but otherwise be worse than auto queue

mumpsimus
Algorithm
Algorithm
Posts: 63

Re: Alpha 12

Post#89 » 05 Feb 2018, 02:27

GFX47 wrote:
mumpsimus wrote:Thanks for explaining this!


I hope it's clear enough?


Yeah, I think I got it. :) The only thing is we're not really talking about an ELO rating anymore if the score is only adjusted for one player, for some matches, but I'm not sure that's super important.

I do see now that the number of matches has gone up dramatically, which to me trumps pretty much everything else.

[edit] One question I have: if I queue a deployment, and then another one, will other players automatically only get matched against my newest deploy, or will both deploys be out there? I'm also not entirely clear how the 10 deployments limit and pushing several deploys of the exact same AI work in this brave new world.

User avatar
GFX47
Dev
Dev
Posts: 2914

Re: Alpha 12

Post#90 » 05 Feb 2018, 12:30

Alpha 12 is live.
I fixed the left/right side bug.

User avatar
Athelinde
Algorithm
Algorithm
Posts: 64
Contact:

Re: Alpha 12

Post#91 » 05 Feb 2018, 19:26

I ran into a strange bug. My windows got screwed up in the last feature update. So I just recently reinstalled everything. I installed stitch, installed Gladiabots there, but outside of making a new username (Athelinda) I didn't use it because I had no bots. Then I saw the 12RCs I downloaded them and created bots from scratch for the new options starting building them in the campaign. Then I wanted to try them out in the 'Carreer', it asked again for a new username and wouldn't take the ones I already have in use. So I created another one (Athalinda). Now Alpha 12 is live I am stuck with the empty Athelinda, and the nolonger accepted 12RC6 has the Athalinda with all my bots. So one this is a bug, and two how do we fix this, at some time I would also love to recover my Athelinde username.

Thanks! (Alpha12 is quite promising!)

User avatar
GFX47
Dev
Dev
Posts: 2914

Re: Alpha 12

Post#92 » 05 Feb 2018, 19:45

Sorry for the inconvenience :(
I'm not sure I fully understand what happened to you but you don't need to uninstall the game before upgrading version.
I synced your Athelinda and Athalinda installs to your Athelinde account.

User avatar
Athelinde
Algorithm
Algorithm
Posts: 64
Contact:

Re: Alpha 12

Post#93 » 05 Feb 2018, 20:16

I didn't uninstall. I ran them side by side. Same computer, same operating system, 2 usernames, that has never happened to me before.

Edit (additional): With this sync I got my old bots from Athelinde back, working on my Alpha12 install, but none of my new ones from 12RC6, which strangely enough are still available when I startup the 12RC6. On second thought I think I know what happened, when this time I installed itch, it asked if I wanted to run my games sandboxed and I said yes. I guess that is the explanation of the difference between the Alpha 12 install and the 12RC6 which is run outside of itch.
Now I just wish for an easy way to copy my bots from my 12RC6 install to the itch Alpha 12.

User avatar
GFX47
Dev
Dev
Posts: 2914

Re: Alpha 12

Post#94 » 06 Feb 2018, 18:49

Athelinde wrote:I didn't uninstall. I ran them side by side. Same computer, same operating system, 2 usernames, that has never happened to me before.

Edit (additional): With this sync I got my old bots from Athelinde back, working on my Alpha12 install, but none of my new ones from 12RC6, which strangely enough are still available when I startup the 12RC6. On second thought I think I know what happened, when this time I installed itch, it asked if I wanted to run my games sandboxed and I said yes. I guess that is the explanation of the difference between the Alpha 12 install and the 12RC6 which is run outside of itch.
Now I just wish for an easy way to copy my bots from my 12RC6 install to the itch Alpha 12.


I didn't know about itch's sandbox install, can you tell me why you choosed it?
What is the account you used on your 12RC6 install?

User avatar
Athelinde
Algorithm
Algorithm
Posts: 64
Contact:

Re: Alpha 12

Post#95 » 06 Feb 2018, 21:33

The sandbox option was not a per game option, but for all games installed through itch. As I sometimes just try out games I thought it might be a good idea, which I now think might be more trouble than it is worth. I don't know how it exactly works, the only thing I have found out so far is that it creates a seperate (hidden) itch-user account in your windows.
The 12RC6 used the Athalinda account, I wish I could merge the accounts, not just one overwrite the other, or just setup the Athalinda account somewhere else, though I hate to have to create another login just for that. I know you merged the Athelinda & Athalinda account into the Athelinde account, but the Athalinda never got the sync option (being deprecated/unsupported and all).

Thanks for looking into it!

pier4r
Skynet
Skynet
Posts: 3389

Re: Alpha 12

Post#96 » 06 Feb 2018, 21:56

GFX47 wrote:
The new matchmaking has the same filtering and sorting rules (elo/xp ranges, cooldowns, etc) but instead of just scanning pending matches, it also "reuses" past ones.
It's kind of an hidden auto redeploy but even on draws / defeats.

In ranked matches, the elo gain / loss is only applied to the player who "really" deployed.
Elo gain / loss is also applied to the reused deployments (but not to their original player).

Here's an example, let's assume all players are in the same elo range and that no match was ever deployed:


I still don't have time to play / fix the unavoidable tournaments (what a period, since summer 2017!). Anyway in this way the elo will be creates out of thin air and this will likely cause the veterans to accumulate more and more of it. This may disrupt easily the 400 points of distance for matching other players. Have you given a thought on this?

Furthermore people that "harvest" the others could potentially reach a score where they do get always 0 points from the others because it is too high (without considering the 400 points delta limit).

I really have to build an elo simulator as well, but again, time....
http://www.reddit.com/r/Gladiabots/wiki/players/pier4r_nvidia_shield_k1 -> Gladiabots CHAT, stats, insights and more ;

User avatar
GFX47
Dev
Dev
Posts: 2914

Re: Alpha 12

Post#97 » 07 Feb 2018, 00:50

Athelinde wrote:The sandbox option was not a per game option, but for all games installed through itch. As I sometimes just try out games I thought it might be a good idea, which I now think might be more trouble than it is worth. I don't know how it exactly works, the only thing I have found out so far is that it creates a seperate (hidden) itch-user account in your windows.
The 12RC6 used the Athalinda account, I wish I could merge the accounts, not just one overwrite the other, or just setup the Athalinda account somewhere else, though I hate to have to create another login just for that. I know you merged the Athelinda & Athalinda account into the Athelinde account, but the Athalinda never got the sync option (being deprecated/unsupported and all).

Thanks for looking into it!


I added Athalinda's AIs to Athelinde's list.
You should get the sync conflict message, choose server data.

mcompany
Autonomous Entity
Autonomous Entity
Posts: 872

Re: Alpha 12

Post#98 » 07 Feb 2018, 03:43

mumpsimus wrote:Yeah, I think I got it. :) The only thing is we're not really talking about an ELO rating anymore if the score is only adjusted for one player, for some matches, but I'm not sure that's super important.

I do see now that the number of matches has gone up dramatically, which to me trumps pretty much everything else.

[edit] One question I have: if I queue a deployment, and then another one, will other players automatically only get matched against my newest deploy, or will both deploys be out there? I'm also not entirely clear how the 10 deployments limit and pushing several deploys of the exact same AI work in this brave new world.

The thing is, elo is still a major factor in determining who plays against who, so having any sort of massive amount of inflation/deflation may mean that a player might end up playing against someone much higher/lower than them much earlier and not having as enjoyable of a game. Also, I personally don't think I have anything to fight both of tomcat's accounts are meaningful for me when from my perspective both sets of AIs are the same.

Anyways, I still think that regardless if we use this system or auto queue (which is better imo, we just need more players), there still is a problem that I should be able to see all of my wins and losses, regardless if I agreed to publish that AI or not. Those matches are still fundamentally important pieces of data for any AI to improve

User avatar
Athelinde
Algorithm
Algorithm
Posts: 64
Contact:

Re: Alpha 12

Post#99 » 07 Feb 2018, 08:26

GFX47 wrote:
I added Athalinda's AIs to Athelinde's list.
You should get the sync conflict message, choose server data.


Thank you so much! :D

mcompany
Autonomous Entity
Autonomous Entity
Posts: 872

Re: Alpha 12

Post#100 » 07 Feb 2018, 21:30

mumpsimus wrote:[edit] One question I have: if I queue a deployment, and then another one, will other players automatically only get matched against my newest deploy, or will both deploys be out there? I'm also not entirely clear how the 10 deployments limit and pushing several deploys of the exact same AI work in this brave new world.

So I feel like this point just got noticed in the chat, but I'll bring it up here. What was being found was that over the last few days, players was being matched up with my test AI I deployed around the time RC1 was released (this being noteworthy for me as it was a test AI, did not use any class outside of assaults, and was done with my elo being around the high 1600s). Since then, I deployed a different match around the same time with my full AI, as well as enough matches on Sunday to get myself to the high 1800s. So, under the system, assuming that it deploys my more recent AIs, no one should be able to connect to fight 1600 test AIs anymore as I have many more recent examples to choose from. This is especially true for someone like mumpsimus, who currently is in the top 5 of the playbase with a score healthy in the 1800s or 1900s. However, it was noted that mumpsimus got recent matches like 857023 which still features my test AIs.

This leads to me having 3 questions for the system that has never been given a straight answer
1. What team setups are saved to be deployed against other players? Is it possible to change what AIs are setup for any given player, and is it possible to continue fight a heavily outdated AI like in my case?

2. Why is it setup this way? I know in the chat that you say that the system is to allow players to play against AIs in their elo range, however (especially when deciding to match specifically against a losing AI) there is always a chance that there is an accidental deployment, a test AI, or a demotion AI, which has no baring on the elo of the player at the time of deployment. Likewise, if all of the elo ranges changes meaning (like they will start to do under this system), an AI over a week old is almost guaranteed to not match the rest of the elo range by being so old and being tested against so often.

3. I the case that someone does play through all of the matches that someone makes manually deployments, Are the manual deployments prioritized for everyone over auto generated matches as an alternative towards the system to avoid this mess and actually see my losses against a recent AI

Return to “Release candidates”

Who is online

Users browsing this forum: No registered users and 1 guest