|
|
metadata
How does the yfpool and efpool settings work? Is it like a Dom- for fortresses?
I tried using it pure, like and it crashes with an invalid argument exception "What: stoi"
|
|
|
|
|
metadata
yf "Phobos Station, The Spire, Tyrolian Outpost, Ashrock Redoubt, Colonial Relay" yfpool 3
Will pick 3 random forts. Same for efpool.
I will make a detailed update post soon.
|
|
|
|
|
metadata
When I run sims in V2.60.6, I get a stream of warnings about The Last Stand missions. It's failing to find 2-3 of the cards. Replacing the xml for that card range hasn't fixed the problem. Any ideas how to fix it?
|
|
|
|
|
metadata
> *Originally posted by **[JangleSam](/forums/2468/topics/920187?page=10#12842098)**:*
> When I run sims in V2.60.6, I get a stream of warnings about The Last Stand missions. It's failing to find 2-3 of the cards. Replacing the xml for that card range hasn't fixed the problem. Any ideas how to fix it?
Devs have partially added some new missions to the xmls, just ignore the messages for now.
|
|
|
|
|
metadata
> *Originally posted by **[DR_F3LL](/forums/2468/topics/920187?page=10#12841648)**:*
> yf "Phobos Station, The Spire, Tyrolian Outpost, Ashrock Redoubt, Colonial Relay" yfpool 3
> Will pick 3 random forts. Same for efpool.
> I will make a detailed update post soon.
This is pretty cool. Can we also use this with Guild Fortresses, like Lightning Cannon and so on?
And by the way of course a big thanks for the awesome job there!
|
|
|
|
|
metadata
#### Another Update:
I'll explain the biggest changes since **v2.59.9** to current **v2.60.7**:
###### Major changes:
1. New flag 'skill' or 'strategy'.
* This works similar to 'mono' or 'factions' flag. TUO will only use cards with specified skill. Eg: "skill jam" will only use jam cards and "skill jam skill strike" will use any card that has jam **or** strike. (only limits assaults and structures, **not** commanders and dominions, unlike the mono-flag)
2. New flag 'no-skill' or 'no-strategy'.
* Inverse to **[1.]**. "no-skill sabotage" will avoid all cards with sabotage. Doesn't work with skill/strategy-flag set.
3. New flag 'no-factions' o 'no-mono'.
* Inverse to 'mono' or 'factions'. "no-factions progenitor" will avoid all progenitor cards. Doesn't work with mono/factions-flag set.
4. New flags 'yfpool' and 'efpool'.
* 'yf "Phobos Station, The Spire, Tyrolian Outpost, Ashrock Redoubt, Colonial Relay" yfpool 3', will pick 3 random forts (similar to the raid.xml pools). Works with CQ or GW forts.
###### Minor changes:
1. Flexible sims are faster now.
2. Support for Crackdown BGE
3. tuo-debug will now always print the runtime (real and cpu) in the last line. (fyi: tuo-debug is slower than normal tuo)
4. Makefile for intel compiler.
5. New flag 'prefered-skill/strategy' has the same effect as 'climb-opts:rb' from previous 'short update' on cards with given skill.
6. 'strategy recent' will only use latest 5% of cards. 'recent-percent 10' sets it to 10%.
7. New flag 'prefered-factor' sets the weigh for how prefered the cards are. eg. 'prefered-factor 5', default is 3
|
|
|
|
|
metadata
> *Originally posted by **[DR_F3LL](/forums/2468/topics/920187?page=10#12844008)**:*
> #### Another Update:
> I'll explain the biggest changes since **v2.59.9** to current **v2.60.7**:
Wow. Thanks for nice updates!)
small misstype in your description above: `recent-precent` -> `recent-percent`
How much faster is `flexible` now? I did not get how properly use it, it seemed to last for ages with no visible progress on deck. Which limitations should be used to get it finished?
|
|
|
|
|
metadata
@DR_F3LL, is receny based or card id? Or does it somehow factor in buffs of older cards?
And what cards does preferred-factor apply to? Does it mean how much to prefer the cards that are in the seed deck?
|
|
|
|
|
metadata
> *Originally posted by **[Demonurg](/forums/2468/topics/920187?page=10#12844050)**:*
>
> small misstype in your description above: `recent-precent` -> `recent-percent`
>
> How much faster is `flexible` now? I did not get how properly use it, it seemed to last for ages with no visible progress on deck. Which limitations should be used to get it finished?
>
thx, fixed the typo. It runs in ~2/3 of previous time. Very low iterations with flexible still produce good decks, about 20 to 50. I use 20.
> *Originally posted by **[Monkey_k](/forums/2468/topics/920187?page=10#12844111)**:*
> @DR_F3LL, is receny based or card id? Or does it somehow factor in buffs of older cards?
> And what cards does preferred-factor apply to? Does it mean how much to prefer the cards that are in the seed deck?
Yes, it is id-based and won't use buffed cards. prefered-factor applys to 'prefered-skill/strategy', 'prefered-mono/factions' and 'prefered-recent'. These cards get used more often when trying to improve a deck.
|
|
|
|
|
metadata
#### More Updates:
With version **v2.61.0** guild sims are available.
This means you can pass a gauntlet or ';'-separated decks as your deck.
#### Operations:
1. 'sim' works with a deck-group. Every deck from your group will be simmed against every deck from the enemy deck, therefore high iterations result in much to process.
2. 'climb_forts' will check **every** possible fortress combination. The number of fortresses is defined by 'yfpool'.
3. other operations (climb,anneal...) still only work with a single deck.
#### Examples:
1. 'tuo "gauntlet_1" "gauntlet_2" sim 10'
* Will give a score how well gauntlet_1 performs against gauntlet_2
2. 'tuo "guild_1" "Mechmaster Yurich Raid-25" climb_forts 10 yf "LC#2, TC#2, IB#2, DF#2" yfpool 2' raid
* Will calculate best forts against this raid.
3. 'tuo "guild_1" "guild_2" climb_forts 10 yf "LC#2, TC#2, IB#2, DF#2" yfpool 2 ef "LC#2, TC#2, IB#2, DF#2" efpool 2'
* Will calculate the best forts for guild_1 against guild_2 where guild_2 uses two random forts from the pool.
#### Notes:
1. yfpool should not be higher than the amount of passed fortresses.
2. As there will be quite much output from tuo, I added a flag '+so' for simplified output
|
|
|
|
|
metadata
> *Originally posted by **[DR_F3LL](/forums/2468/topics/920187?page=10#12845109)**:*
> #### More Updates:
> With version **v2.61.0** guild sims are available.
Wow. This is really sweet updates! Our sim-master will be glad to use native functions instead of script-based workarounds. Thanks for you effort again!)
|
|
|
|
|
metadata
Here you can find an overview on current parameters and flags: https://github.com/APN-Pucky/tyrant_optimize/wiki/Flags
|
|
|
|
|
metadata
> *Originally posted by **[DR_F3LL](/forums/2468/topics/920187?page=10#12847464)**:*
> Here you can find an overview on current parameters and flags: https://github.com/APN-Pucky/tyrant_optimize/wiki/Flags
Finally someone did it! Much appreciated)
|
|
|
|
|
metadata
@DR_F3LL, is it possible to use `skill` flag somehow to sim in `enhance skill` as well?
|
|
|
|
|
metadata
@Demonurg Do you mean specifying the second skill? For example 'enhance **sunder** 5' or 'evolve weaken to **sunder**'. No there is no way to specify these currently.
|
|
|
|
|
metadata
I have this "bug":
Simming brawl, startdeck has 6 cards, using -L 9 10 flag, anneal.
Result is still a 6-card deck - i guess cause anneal can't find a better deck?
Is there any way to force it or change sth on the anneal iteration (400 100 0.001) to get a bigger deck?
|
|
|
|
|
metadata
I tested the -L flag and anneal was able to change the start deck fitting the 9-10 limit.
The flag works for both anneal and climb in the same way: It denies decks of the same or lower size (if it's below the limit ofc).
Increasing the initial temp. to ~500 allows anneal to accept worse simming deck in the early sim process, which will lead to a bigger deck. => 400 500 0.001
If there are still rapid changes in the end at a low temp. and it seems unfinished you can lower the cooling rate to eg. 0.0001, which will result in more simulations close to the end.
|
|
|
|
|
metadata
<raid>
<id>58</id>
<name>Arkadios Transcendent Raid</name>
<levels>26</levels>
<commander>26046</commander>
<deck>
<always_include>
<card>61114</card>
<card>61124</card>
<card>61134</card>
<card>61144</card>
<card>61154</card>
</always_include>
<card_pool amount="5">
<card replicates="2">61114</card>
<card replicates="2">61124</card>
<card replicates="2">61134</card>
<card replicates="2">61144</card>
<card replicates="2">61154</card>
</card_pool>
</deck>
</raid>
If you think you saw more or less of a card - just tell me.
|
|
|
|
|
metadata
Thanks a lot for the Raid file!
I was surprised about how easy this raid seems to be as per TUO.
|
|
|
|
|
metadata
Since several people asked me about supporting this project financially, I created a paypal account: apnpucky@gmail.com
Donations are much appreciate but are **not ** necessary to ensure the continuation of this project.
**Thanks**.
|
|
|
|
|
metadata
I am currently simming for the Return of Razogoth Raid. I have added the following in the raids.xml.
```
<raid>
<id>59</id>
<name>Return of Razogoth Raid</name>
<levels>26</levels>
<commander>26080</commander>
<deck>
<always_include>
<card replicates="2">61746</card>
<card replicates="2">61756</card>
<card replicates="2">61766</card>
<card replicates="2">61776</card>
<card replicates="2">61786</card>
</always_include>
</deck>
</raid>
<!-- END of Raids -->
```
But, I have encounter the following error:
**WARNING: biominal successes > trials in Threads**
[](https://i.imgur.com/cWYQjRh.jpg)
The command line I use is:
tuo.exe "Cassius the Centurion, Cassius Nexus, Command Suit(3), Arctis Half-track(2), Ward Bombardier, Myrmidon Remnant, Badlands Recon" "Return of Razogoth Raid-18" raid ordered -t 4 target 99 -e "CriticalReach" fund 0 -v endgame 1 -L 3 10 climb 5000 dom+
Has anyone else seen that error?
tuo version: v2.61.3
Note that, I have restarted, I have checked the exe versions in conjuction with the latest and all is well.
Thank you in advance...
|
|
|
|
|
metadata
This error happens when tuo evaluates a higher score than the maximum possible score.
The forumla for scoring when you loose a fight is:
raid_damage = 15
+ (p[1]->total_cards_destroyed)
- (10 * p[1]->commander.m_hp / p[1]->commander.max_hp());
where p[1] is the enemy. The reason for scores > 100 is probably that the formula should be 'total_nonsummoned_cards_destroyed' instead of 'total_cards_destroyed', since On-Attacked summon could produce infinite scoring. (=>**v2.61.4**)
|
|
|
|
|
metadata
I simmed few minutes ago and it also happened to me. But after this long spam the TUO simmed correctly.
|
|
|
|
|
metadata
> *Originally posted by **[DR_F3LL](/forums/2468/topics/920187?page=10#12933459)**:*
> This error happens when tuo evaluates a higher score than the maximum possible score.
> The forumla for scoring, when you loose a fight is:
>
> raid_damage = 15
> + (p[1]->total_cards_destroyed)
> - (10 * p[1]->commander.m_hp / p[1]->commander.max_hp());
>
> where p[1] is the enemy. The reason for scores > 100 is probably that the formula should be 'summoned_cards_destroyed' instead of 'total_cards_destroyed', since On-Attacked summon could produce infinite scoring.
So, to overcome the problem now is to use pvp mode instead of raid?
|
|
|
|
|
metadata
> *Originally posted by **[madsam2](/forums/2468/topics/920187?page=10#12933469)**:*
> So, to overcome the problem now is to use pvp mode instead of raid?
**v2.61.4** should be ready to download in some minutes. (progress: https://travis-ci.org/APN-Pucky/tyrant_optimize/builds )
|
 "Joe83"){kind=small}
 "DR_F3LL"){kind=small}
 "JangleSam"){kind=small}
 "TBake07"){kind=small}
 "Demonurg"){kind=small}
 "Monkey_k"){kind=small}
 "Isilrond"){kind=small}
 "madsam2"){kind=small}
 "razor15"){kind=small}