Welcome Guest! To enable all features please Login or Register.

Notification

Icon
Error
84 Pages«<3132333435>»
Sepefeets Silverfish update (OG, Kara, MSG)
Options
Previous Topic Next Topic
sadpwner
#649 Posted : Wednesday, July 20, 2016 3:24:14 AM(UTC)

Rank: Advanced Member

Groups: Registered
Posts: 262
Location: somewhere

Thanks: 3 times
Was thanked: 64 time(s) in 54 post(s)
parakeet;39792 wrote:
Sepefeets has better simulation of your own deck on his to do list. Enemy class based behavior will likely be in the next update. Enemy deck based behavior is a bigger job and requires continual updating. If you demonstrate the value added it might be more likely to happen.


Yes, deck based will require continual updates to keep up with the meta. Though, it should be something anyone can do. Just copy and paste deck lists off hearthpwn or tempostorm.
Back to top
WWW
1 user thanked sadpwner for this useful post.
sepefeets on 7/20/2016(UTC)
Badi
#651 Posted : Wednesday, July 20, 2016 11:10:57 AM(UTC)

Rank: Advanced Member

Groups: Registered
Posts: 46

Thanks: 29 times
Was thanked: 10 time(s) in 9 post(s)
Where can i download the previous versions of silverfish?
Back to top
WWW
1 user thanked Badi for this useful post.
sepefeets on 7/20/2016(UTC)
Dre
#652 Posted : Wednesday, July 20, 2016 11:21:33 AM(UTC)

Rank: Advanced Member

Groups: HSModerator, Registered
Posts: 888
Man
Location: The Netherlands

Thanks: 216 times
Was thanked: 480 time(s) in 315 post(s)
Badi;39801 wrote:
Where can i download the previous versions of silverfish?


Check the message in the first post of this tread
Back to top
WWW
1 user thanked Dre for this useful post.
sepefeets on 7/20/2016(UTC)
Badi
#653 Posted : Wednesday, July 20, 2016 11:30:31 AM(UTC)

Rank: Advanced Member

Groups: Registered
Posts: 46

Thanks: 29 times
Was thanked: 10 time(s) in 9 post(s)
I am sorry for that, overlooked :(
Back to top
WWW
1 user thanked Badi for this useful post.
sepefeets on 7/20/2016(UTC)
Elekid
#654 Posted : Wednesday, July 20, 2016 12:03:09 PM(UTC)
Elekid


Rank: Advanced Member

Groups: Registered
Posts: 101
Location: Nijmegen, The Netherlands

Thanks: 29 times
Was thanked: 65 time(s) in 58 post(s)
It should prioritize more on killing the Flamewakers. Enemy gets so much value out of them in my last couple of turns:

And finally it casts mortal coil on it, but before attacking with minion. So it didn't kill it, missed free card draw.

link
Back to top
WWW
1 user thanked Elekid for this useful post.
sepefeets on 7/20/2016(UTC)
Dre
#655 Posted : Wednesday, July 20, 2016 7:08:44 PM(UTC)

Rank: Advanced Member

Groups: HSModerator, Registered
Posts: 888
Man
Location: The Netherlands

Thanks: 216 times
Was thanked: 480 time(s) in 315 post(s)
When i play my Midrange Shaman (deck is called Midrange Shaman) i Get the error in HR scrolling text:

"Deck not found: Shaman"

Did the same when i played midrange hunter before that.



I dont use the data folders yet. Its just the files in the silverfolder thats it.
Back to top
WWW
1 user thanked Dre for this useful post.
sepefeets on 7/20/2016(UTC)
Elekid
#656 Posted : Wednesday, July 20, 2016 7:42:43 PM(UTC)
Elekid


Rank: Advanced Member

Groups: Registered
Posts: 101
Location: Nijmegen, The Netherlands

Thanks: 29 times
Was thanked: 65 time(s) in 58 post(s)
Not using the battlecry of Corrupted Seer:

432. playcard: Corrupted Seer
after that, same turn:
491. playcard: Murloc Tidehunter
Back to top
WWW
1 user thanked Elekid for this useful post.
sepefeets on 7/20/2016(UTC)
sepefeets
#657 Posted : Wednesday, July 20, 2016 10:35:43 PM(UTC)

Rank: Advanced Member

Groups: Registered
Posts: 899

Thanks: 1028 times
Was thanked: 584 time(s) in 238 post(s)
@Dre - That is HR looking for a deck named Shaman and not finding it. Has nothing to do with SF.

@Elekid - I have no idea what your problem is with that play. Corrupted Seer aoe's non murlocs and you only had 6 mana which is just enough to cast it. On the next turn it played the tidehunter. I think you might be confusing it with coldlight seer or something else that buffs murlocs?
Back to top
WWW
1 user thanked sepefeets for this useful post.
Elekid on 7/21/2016(UTC)
sadpwner
#661 Posted : Thursday, July 21, 2016 12:37:40 AM(UTC)

Rank: Advanced Member

Groups: Registered
Posts: 262
Location: somewhere

Thanks: 3 times
Was thanked: 64 time(s) in 54 post(s)
Though, how does next nextTurnMaxWide differ from twotsamount?

http://pastebin.com/fwLWJyqH

Turn 6, attacked the peddler instead of the councilman with power overwhelming

http://pastebin.com/HRZeqkhU

Turn 3 or 4. Power overwhelmings the peddler twice when only 1 was necessary.

http://pastebin.com/HnXDdAHq

Turn 3 should have coined out an abusive to clear the cleric

turn 5 should tap first. It decided not to tap at all.

It's strange. When I increase the maxwide calculations, the bot seems to play worse. One of the reasons why I kept it low. I don't mean it ropes more but it actually makes more misplays.

__
http://pastebin.com/E8bAkUai
Turn 3
Idk if anything can be done but it should place the voidwalkers around the dire wolf.
Back to top
WWW
 |  Edit by user
1 user thanked sadpwner for this useful post.
sepefeets on 7/21/2016(UTC)
sepefeets
#662 Posted : Thursday, July 21, 2016 2:21:14 AM(UTC)

Rank: Advanced Member

Groups: Registered
Posts: 899

Thanks: 1028 times
Was thanked: 584 time(s) in 238 post(s)
sadpwner;39822 wrote:
Though, how does next nextTurnMaxWide differ from twotsamount?

http://pastebin.com/fwLWJyqH

Turn 6, attacked the peddler instead of the councilman with power overwhelming

http://pastebin.com/HRZeqkhU

Turn 3 or 4. Power overwhelmings the peddler twice when only 1 was necessary.

http://pastebin.com/HnXDdAHq

Turn 3 should have coined out an abusive to clear the cleric

turn 5 should tap first. It decided not to tap at all.

It's strange. When I increase the maxwide calculations, the bot seems to play worse. One of the reasons why I kept it low. I don't mean it ropes more but it actually makes more misplays.

twotsamount = how many of the "best" boards from the 1st turn should be sim'd further for a 2nd turn, crappy 1st turn boards don't need to all go through a second turn sim.
nextTurnMaxWide = how many boards are done for each "step" of the 2nd turn. Each "step" is how many moves to do in the turn. ie. 0 steps = 0 moves, 1 step = 1 move.

log 1 - data sync.

log 2 - obvious data sync, I think you should know by now that double buffing when 1 is enough is blatant data sync.

log 3 - turn 3 maybe, there is no turn 5.
Back to top
WWW
1 user thanked sepefeets for this useful post.
sadpwner on 7/21/2016(UTC)
sadpwner
#663 Posted : Thursday, July 21, 2016 2:30:32 AM(UTC)

Rank: Advanced Member

Groups: Registered
Posts: 262
Location: somewhere

Thanks: 3 times
Was thanked: 64 time(s) in 54 post(s)
sepefeets;39823 wrote:
twotsamount = how many of the "best" boards from the 1st turn should be sim'd further for a 2nd turn, crappy 1st turn boards don't need to all go through a second turn sim.
nextTurnMaxWide = how many boards are done for each "step" of the 2nd turn. Each "step" is how many moves to do in the turn. ie. 0 steps = 0 moves, 1 step = 1 move.

log 1 - data sync.

log 2 - obvious data sync, I think you should know by now that double buffing when 1 is enough is blatant data sync.

log 3 - turn 3 maybe, there is no turn 5.



Woops, here's the log with turn 5. http://pastebin.com/LrVpGk8N

It registered as a 10/10 so I thought it knew it buffed it twice o.o
Back to top
WWW
1 user thanked sadpwner for this useful post.
sepefeets on 7/21/2016(UTC)
parakeet
#664 Posted : Thursday, July 21, 2016 4:37:30 AM(UTC)

Rank: Advanced Member

Groups: Registered
Posts: 128

Thanks: 18 times
Was thanked: 83 time(s) in 69 post(s)
sadpwner;39824 wrote:
It registered as a 10/10 so I thought it knew it buffed it twice o.o

The data sync is on the first power overwhelming cast, it looks like it still has the peddler as a 2/2.
Back to top
WWW
1 user thanked parakeet for this useful post.
sepefeets on 7/21/2016(UTC)
Elekid
#658 Posted : Thursday, July 21, 2016 4:42:23 AM(UTC)
Elekid


Rank: Advanced Member

Groups: Registered
Posts: 101
Location: Nijmegen, The Netherlands

Thanks: 29 times
Was thanked: 65 time(s) in 58 post(s)
sepefeets;39821 wrote:
@Elekid - I have no idea what your problem is with that play. Corrupted Seer aoe's non murlocs and you only had 6 mana which is just enough to cast it. On the next turn it played the tidehunter. I think you might be confusing it with coldlight seer or something else that buffs murlocs?


My bad, my bad. I indeed mixed up the Coldlight Seer with the Corrupted Seer. Hence I pointed you to the wrong minion and turn. The problem is this:

288. playcard: Coldlight Seer
same turn:
333. playcard: Grimscale Oracle

link
Back to top
WWW
1 user thanked Elekid for this useful post.
sepefeets on 7/21/2016(UTC)
Elekid
#665 Posted : Thursday, July 21, 2016 4:53:19 AM(UTC)
Elekid


Rank: Advanced Member

Groups: Registered
Posts: 101
Location: Nijmegen, The Netherlands

Thanks: 29 times
Was thanked: 65 time(s) in 58 post(s)
I returned to my computer and noticed the bot wasn't doing anything. Just roping out. Apparently it had been doing so for a couple of ours. screenshot

Also the bot stopped making logs @6:29 local time

final logs from before that time.

Any ideas on what caused it?
Back to top
WWW
2 users thanked Elekid for this useful post.
parakeet on 7/21/2016(UTC), sepefeets on 7/21/2016(UTC)
parakeet
#659 Posted : Thursday, July 21, 2016 4:53:21 AM(UTC)

Rank: Advanced Member

Groups: Registered
Posts: 128

Thanks: 18 times
Was thanked: 83 time(s) in 69 post(s)
sepefeets;39821 wrote:
@Dre - That is HR looking for a deck named Shaman and not finding it. Has nothing to do with SF.

Do we need to be concerned about this? If HR feeds the wrong deck name, I might imagine the correct deck files would not be used.
Back to top
WWW
1 user thanked parakeet for this useful post.
sepefeets on 7/21/2016(UTC)
parakeet
#666 Posted : Thursday, July 21, 2016 5:01:00 AM(UTC)

Rank: Advanced Member

Groups: Registered
Posts: 128

Thanks: 18 times
Was thanked: 83 time(s) in 69 post(s)
Elekid;39831 wrote:
I returned to my computer and noticed the bot wasn't doing anything. Just roping out. Apparently it had been doing so for a couple of ours. screenshot

Also the bot stopped making logs @6:29 local time

final logs from before that time.

Any ideas on what caused it?

Nice catch - The AI seems to have given you a mystery 3rd card, 'placeholdercard'!
Back to top
WWW
2 users thanked parakeet for this useful post.
Elekid on 7/21/2016(UTC), sepefeets on 7/21/2016(UTC)
sepefeets
#660 Posted : Thursday, July 21, 2016 5:50:04 AM(UTC)

Rank: Advanced Member

Groups: Registered
Posts: 899

Thanks: 1028 times
Was thanked: 584 time(s) in 238 post(s)
parakeet;39832 wrote:
Do we need to be concerned about this? If HR feeds the wrong deck name, I might imagine the correct deck files would not be used.

There's no evidence of this yet. What Dre showed is when HR is picking a deck on the selection screen.

parakeet;39833 wrote:
Nice catch - The AI seems to have given you a mystery 3rd card, 'placeholdercard'!

There's more weirdness, it also doubled up on each enemy minion.
Back to top
WWW
sadpwner
#643 Posted : Thursday, July 21, 2016 6:13:48 AM(UTC)

Rank: Advanced Member

Groups: Registered
Posts: 262
Location: somewhere

Thanks: 3 times
Was thanked: 64 time(s) in 54 post(s)
sepefeets;39774 wrote:
V122.5SE released.

small penalty for dream
Mortal Coil penalty - factor in spellpower
fix _combo.txt switching checking for _mulligan.txt existence
563.1 reduce bonus on enemy card burn
cut duplicate enemy playfields too (from HB) + some cleanup (cutting playfields is a noticeable speed gain but don't view it as a reason to raise enemy turn calculations, they are very dumb, use the gains for your own turn)
625.6 (+1 old log) drastically reduce nerubian egg early bonus
(625.2, 625.3) avoid playing reliquary seeker without 6 minions ((625.1) intentionally not fixed, use cardvalue to fully prevent early play)
(pm) play brann before good battlecries
(491) play frothing berserker before attacking
penalize deadly shot for each small enemy minion



I'm regretting not bumping the major version after writing out the log...

Sepes can you explain what the numbers (563.1, 625.6 and ect.) are?
Back to top
WWW
1 user thanked sadpwner for this useful post.
sepefeets on 7/21/2016(UTC)
parakeet
#644 Posted : Thursday, July 21, 2016 7:40:31 AM(UTC)

Rank: Advanced Member

Groups: Registered
Posts: 128

Thanks: 18 times
Was thanked: 83 time(s) in 69 post(s)
sadpwner;39837 wrote:
Sepes can you explain what the numbers (563.1, 625.6 and ect.) are?

Those numbers are the post number of the relevant misplay. If the post had multiple misplay logs, then the decimal corresponds the the specific log within the post. Example: 625.6 means the misplay was the sixth misplay in post #625.
Back to top
WWW
2 users thanked parakeet for this useful post.
Elekid on 7/21/2016(UTC), sepefeets on 7/21/2016(UTC)
Dre
#667 Posted : Thursday, July 21, 2016 9:14:20 AM(UTC)

Rank: Advanced Member

Groups: HSModerator, Registered
Posts: 888
Man
Location: The Netherlands

Thanks: 216 times
Was thanked: 480 time(s) in 315 post(s)
This is probably my own fault due my combo:

Code:


// [NUDGE] [8] Al'Akir the Windlord x1 (NEW1_010) + [1] Rockbiter Weapon (CS2_045)
NEW1_010,0;CS2_045,0;bonus:50;hero:shaman


But look at this play on turn 9...

http://pastebin.com/rcxaMTiF

Note sure what is going on there...casting rockbiter to kill off a miror image. should i lower the bonus on that combo or remove it entierly. And also did i just rope out there on turn 9 or what? Did it try to attack 3 times with al'akir? Is this sync? Iam confused about everything in turn 9 - help xD
Back to top
WWW
 |  Edit by user
1 user thanked Dre for this useful post.
sepefeets on 7/21/2016(UTC)
84 Pages«<3132333435>»
Forum Jump  
You cannot post new topics in this forum.
You cannot reply to topics in this forum.
You cannot delete your posts in this forum.
You cannot edit your posts in this forum.
You cannot create polls in this forum.
You cannot vote in polls in this forum.

Powered by YAF | YAF © 2003-2011, Yet Another Forum.NET
This page was generated in 0.127 seconds.