Open Side Menu Go to the Top
Register
PioSOLVER - postflop equilibrium solver for Holdem PioSOLVER - postflop equilibrium solver for Holdem

09-05-2018 , 11:58 AM
First time running a PIO script overnight. Basic Button v BB SRP spot running for 74 flop subset.

Before running estimated the tree size (around 5000mb had 5800ish mb ram available) Ran the script and went to bed. (Checked once and it had finished a couple).

When i woke up this morning, it stopped after flop 17 because of insuffiecient ram. What could change overnight in my ram usage to not allow it to continue?

IM running a 2017 IMAC on bootcamp.

P.S. planning on buying more ram. Or is it better to run on a server?
PioSOLVER - postflop equilibrium solver for Holdem Quote
09-05-2018 , 10:32 PM
Hi,
I am thinking about updating to edge to do preflop calculations for 3-4 handed ~25bb and under icm spots with limps, minraises, openshoves, reshoves. How big would these trees get with different flop subset numbers? As i understand, solving with low flop subset numbers can give unreliable results, but i am no expert.
Thanks!
PioSOLVER - postflop equilibrium solver for Holdem Quote
09-06-2018 , 04:31 AM
Quote:
I was hoping to get really low benchmarks, but when I run the bench command, I'm getting about 7 seconds. All cores are working and power settings are on high performance. What specs above do I need to improve upon to get those 3 and 4 second benches I've seen in this thread in the past? Any other tips? Thanks!
Are you sure all the cores are working? With the pro version it should be 16 cores without hyperthreading so task manager should show 50%. Maybe for some reason it's only 8+8 (8 with hyperthreading). You can try disabling hyperthreading if your server provider allows that and see if that helps.
One way or another the benchmark time should be better. I get around 9.5 seconds on my i7 quad @ 3.4Ghz (and that's a CPU from 4 years ago).

Quote:
When i woke up this morning, it stopped after flop 17 because of insuffiecient ram. What could change overnight in my ram usage to not allow it to continue?
Windows is bad at seeing freed memory as it sometimes tries to be smart about it. Please use this checkbox:

https://gyazo.com/e2162cad51ad4d6065a1a52befc2d3bd

when generating a script.

Quote:
I am thinking about updating to edge to do preflop calculations for 3-4 handed ~25bb and under icm spots with limps, minraises, openshoves, reshoves. How big would these trees get with different flop subset numbers?
Please notice that Pio is a HU solver that means you can solve spots which are already HU (SB vs BB, steal from CO/MP vs BB etc.).
You should be fine with 64GB of RAM for those trees.
PioSOLVER - postflop equilibrium solver for Holdem Quote
09-06-2018 , 04:50 PM
Quote:
Originally Posted by punter11235
It's true EV of checking and betting is the same for many hands in OOP range.
87 is not one of them though:
https://gyazo.com/59e2794f59e619f90530dfee5bce7441

I don't think OOP is "run over". IP has stronger range. OOP tries to get some value at least by betting while giving up on most hands when checking. If OOP was always checking for example their overall EV would be 10 chips lower. OOP would still be forced to give up a lot after an overbet (try it, OOP is only calling ~17.8% after checking 100%). IP's range is just too strong for OPP to hope to make IP's bluffs break-even.

This is btw a great example of how "1-alpha" thinking fails. A strategy of check-calling enough so IP is indifferent between bluffing and checking costs tons of EV. This only works in one street toy games with symmetric ranges and doesn't apply to real poker at all.

Thanks punter.


Quick question, I know that if I want to compare the different EVs of 33% cbet size vs 60% cbet size for oop player, I can just create 2 trees (one with each size) and compare the OOP EV at ROOT.

What if I wanted to compare the different EV of removing say allowing overbet vs not allowing overbet on the river? Would I still compare EV of these 2 different trees at ROOT, or compare their EV at river?

(I'm guessing for overbetting, comparing at root accounts for multiple runouts, so the ev difference will be smaller, whereas comparing at river will have a larger difference if the runout favors overbetting?)

The reason I ask is because I would like to simplify studying the game tree by reducing the amount of bet sizes available without having too large an impact on EV.

Last edited by HappyLuckBox; 09-06-2018 at 04:56 PM.
PioSOLVER - postflop equilibrium solver for Holdem Quote
09-07-2018 , 02:10 AM
Quote:
What if I wanted to compare the different EV of removing say allowing overbet vs not allowing overbet on the river? Would I still compare EV of these 2 different trees at ROOT, or compare their EV at river?

(I'm guessing for overbetting, comparing at root accounts for multiple runouts, so the ev difference will be smaller, whereas comparing at river will have a larger difference if the runout favors overbetting?)

The reason I ask is because I would like to simplify studying the game tree by reducing the amount of bet sizes available without having too large an impact on EV.
If you have 2 different trees it only makes sense to compare at root. The reason is that even if you change an action on the river the optimal strategy across the whole tree might be different and therefore different ranges make it to the river at the point you changed strategic options. This means comparing overall EV there is pointless because to compare overall EVs you need to have the same ranges.

If you want to simplify the process the way to go is to solve one tree (with more options preferably), then create a new tree starting on the river with ranges from that solution (there is a feature called "create subtree configuration" which does that for you) and experiment on that. You still need to compare EVs in root but now the root is on the river and the starting ranges there are the same.
PioSOLVER - postflop equilibrium solver for Holdem Quote
09-07-2018 , 09:09 AM
I have a starting pot 20 and set to finish calculation at 0.5% of the pot (which is ~0.1).
When I run normal sim, it finishes when exploitable for ~0.5 but when I run script, it finishes when "exploitable for 0.1". Why it's not the same?
PioSOLVER - postflop equilibrium solver for Holdem Quote
09-07-2018 , 09:48 AM
HI,

I started to compare the EV of using 3 sizing and 2 sizing On the turn with previously 2 sizing on the flop.

When I got 2% pot EV difference on a certain spot does the result make sense ? Since the strategy on the flop would not be the same

Ex : Board XXX
using 1,5/0,8/0,25 psb (1) VS 1,5/0,8 psb (2) with previously 0,25/0,8 PSB on the flop

(1) we Cbet each sizing on a 25% frequency

(2) we Cbet 40% for 0,8 PSB and 9% for 0,25 PSB

Then after betting the flop for 0,25 PSB when I compare the EV% of the pot using 3 or 2 sizing on a certain turn, does that really make sense Since we do not Cbet at the same % ?

Thanks
PioSOLVER - postflop equilibrium solver for Holdem Quote
09-07-2018 , 05:35 PM
Windows Pro required to run on cloud?

Do I need to upgrade to Windows Pro so that I can use the remote desktop connection to connect to Amazon EC2 cloud?
PioSOLVER - postflop equilibrium solver for Holdem Quote
09-08-2018 , 03:08 AM
Quote:
I have a starting pot 20 and set to finish calculation at 0.5% of the pot (which is ~0.1).
When I run normal sim, it finishes when exploitable for ~0.5 but when I run script, it finishes when "exploitable for 0.1". Why it's not the same?
It's the same. The solver internally reports exploitability in chips (not % of the pot). This is translated by PioViewer to percentage of the pot. If you look at the solver's log on the right side in treebuilding and calculation tab when solving you will exploitability in chips as well.

0.1 out of 20 is 0.5% so it's all adds-up.

Quote:
I started to compare the EV of using 3 sizing and 2 sizing On the turn with previously 2 sizing on the flop.

When I got 2% pot EV difference on a certain spot does the result make sense ? Since the strategy on the flop would not be the same
Yes, that's why comparing EVs only makes sense at the beginning of the tree (root node). I've commented on it several times during the last 200-300 posts. If you have 2p2 set to display 100 posts per page you can find those by searching for "root node" or "comparing".

Quote:
Then after betting the flop for 0,25 PSB when I compare the EV% of the pot using 3 or 2 sizing on a certain turn, does that really make sense Since we do not Cbet at the same % ?
It really doesn't

Quote:
Windows Pro required to run on cloud?
No, just 64 bit Windows.

Quote:
Do I need to upgrade to Windows Pro so that I can use the remote desktop connection to connect to Amazon EC2 cloud?
Maybe there are some versions of Windows where remote desktop is not included. I think it's included in Windows 10 no matter the version although you would need to check. You should be able to install it anyway even if it's not included by default but it's really not the best place to ask as it's purely Windows related issue and it's hard for us to check it on every Windows version out there.
PioSOLVER - postflop equilibrium solver for Holdem Quote
09-09-2018 , 06:35 AM
Hey guys,

For the basic PIO version, I always have to give in the betsizes manually, thus meaning PIO doesn't calculate the best bet sizes automaticly.

Is there any feature that PIO just suggests the best betsizing per street ?

Thx in advance guys
PioSOLVER - postflop equilibrium solver for Holdem Quote
09-09-2018 , 02:57 PM
How can I tell if the directory with .cfr files contains nested files or not?
PioSOLVER - postflop equilibrium solver for Holdem Quote
09-09-2018 , 05:09 PM
Just downloaded free version to experiment. I don't see anything in the tree building section like 6max 3b buvsco etc. Do I have to do these manually or something? I'm watching the intro video and he says these are included in the free version - 2:30 into the quick start video

Thanks
PioSOLVER - postflop equilibrium solver for Holdem Quote
09-10-2018 , 04:54 AM
Quote:
For the basic PIO version, I always have to give in the betsizes manually, thus meaning PIO doesn't calculate the best bet sizes automaticly.

Is there any feature that PIO just suggests the best betsizing per street ?
There is not a known theoretical ways of doing it. Most likely you just need to try a lot of sizes and see what's the best.

Quote:
How can I tell if the directory with .cfr files contains nested files or not?
I am not sure what nested files are. Can you describe what you mean in some detail?

Quote:
Just downloaded free version to experiment. I don't see anything in the tree building section like 6max 3b buvsco etc. Do I have to do these manually or something? I'm watching the intro video and he says these are included in the free version - 2:30 into the quick start vide
There are defaults configs here:
https://gyazo.com/94b0424b157232aba56ded8ce8fd6e94

Make sure you unzip the free version to an empty folder and that all the files are there. The common reason for those problems is unzipping to desktop and then removing Ranges and Treebuilding folder.
PioSOLVER - postflop equilibrium solver for Holdem Quote
09-10-2018 , 08:09 AM
i know this have probably been asked a ton of times but i really can't find the answer... , whats the minimum accuracy i should configure pio? i have been solving to 0.40% of the pot but im not sure if thats the minimum realiable solving accuracy
PioSOLVER - postflop equilibrium solver for Holdem Quote
09-10-2018 , 06:17 PM
Quote:
i know this have probably been asked a ton of times but i really can't find the answer... , whats the minimum accuracy i should configure pio? i have been solving to 0.40% of the pot but im not sure if thats the minimum realiable solving accuracy
This is a judgement call. 0.35% should be good for most purposes. You may go for 0.5% if you just need a good overview or for 0.25% if you want more precise solution.
PioSOLVER - postflop equilibrium solver for Holdem Quote
09-10-2018 , 06:34 PM
Quote:
Originally Posted by punter11235
If you have 2 different trees it only makes sense to compare at root. The reason is that even if you change an action on the river the optimal strategy across the whole tree might be different and therefore different ranges make it to the river at the point you changed strategic options. This means comparing overall EV there is pointless because to compare overall EVs you need to have the same ranges.

If you want to simplify the process the way to go is to solve one tree (with more options preferably), then create a new tree starting on the river with ranges from that solution (there is a feature called "create subtree configuration" which does that for you) and experiment on that. You still need to compare EVs in root but now the root is on the river and the starting ranges there are the same.


Thanks punter!
I would expect to get greater difference in EV on the river using this method on certain river cards as opposed to measuring EV between 2 different trees at root.

On another topic, would you happen to know why this subtree config (ott for oop) has only 101.5 combos betting + 53.9 combos checking (155.4 total), when on the prior street there was a total of 166.9 combos betting? There seems to be 11.5 combos missing.


Code:
#TreeBuilding#V2
#Range0#AdAc:0.676874518,AhAc:0.629342318,AsAc:0.864474475,AhAd:0.676857948,AsAd:0.859189451,AsAh:0.864474237,KdKc:0.883281767,KhKc:0.85762167,KsKc:0.895483375,KhKd:0.883284807,KsKd:0.889685571,KsKh:0.895484447,QdQc:0.886408031,QhQc:0.891928434,QsQc:0.887530029,QhQd:0.886410654,QsQd:0.892690241,QsQh:0.887541354,JdJc:0.879175365,JhJc:0.820463717,JsJc:0.898628116,JhJd:0.879172802,JsJd:0.925167263,JsJh:0.898630798,TdTc:0.901831925,ThTc:0.870325744,TsTc:0.923404515,ThTd:0.901832759,TsTd:0.910465181,TsTh:0.923401535,9d9c:0.939810455,9h9c:0.937686086,9h9d:0.939810038,8d8c:0.785627306,8h8c:0.641194344,8s8c:0.816056013,8h8d:0.785631955,8s8d:0.754764974,8s8h:0.816053808,7d7c:0.767370284,7h7c:0.689374089,7s7c:0.794898331,7h7d:0.767372727,7s7d:0.789862633,7s7h:0.794904768,6d6c:0.839466989,6h6c:0.779500067,6s6c:0.824446261,6h6d:0.839466572,6s6d:0.794280052,6s6h:0.824444115,5d5c:0.789051294,5h5c:0.804077566,5s5c:0.854373753,5h5d:0.789049506,5s5d:0.81599462,5s5h:0.854374588,4d4c:0.654567182,4h4c:0.68121314,4s4c:0.748557389,4h4d:0.654567003,4s4d:0.706419647,4s4h:0.748557329,3d3c:0.545888662,3h3c:0.550690353,3h3d:0.545891404,AcKc:0.747719109,AdKd:0.89325583,AhKh:0.747717202,AsKs:0.916517198,AcKd:0.860280454,AcKh:0.768152475,AcKs:0.780479074,AdKc:0.866752267,AdKh:0.866751969,AdKs:0.822834015,AhKc:0.768154204,AhKd:0.860283136,AhKs:0.780478299,AsKc:0.916551292,AsKd:0.889365017,AsKh:0.916551948,AcQc:0.821076989,AdQd:0.78288424,AhQh:0.821076334,AsQs:0.868858457,AcQd:0.908868551,AcQh:0.81903255,AcQs:0.754443228,AdQc:0.880654216,AdQh:0.880653739,AdQs:0.802652359,AhQc:0.81903255,AhQd:0.908867836,AhQs:0.754458785,AsQc:0.89237982,AsQd:0.86562258,AsQh:0.892379522,AcJc:0.584820032,AdJd:0.603666306,AhJh:0.584816158,AsJs:0.756949425,AcJd:0.79994005,AcJh:0.578066349,AcJs:0.740752935,AdJc:0.600689709,AdJh:0.600686967,AdJs:0.859247506,AhJc:0.578065753,AhJd:0.799940348,AhJs:0.740781367,AsJc:0.77074796,AsJd:0.8333143,AsJh:0.770743728,AcTc:0.750562429,AdTd:0.62978524,AhTh:0.750559986,AsTs:0.777161062,Ac9c:0.892405033,Ad9d:0.920232058,Ah9h:0.892400861,Ac8c:0.0785234794,Ad8d:0.925553381,Ah8h:0.0785232484,As8s:0.781942248,Ac7c:0.0446921699,Ad7d:0.940920949,Ah7h:0.0446918942,As7s:0.838269174,Ac6c:0.0269370154,Ad6d:0.336823672,Ah6h:0.0269371178,As6s:0.320430994,Ac5c:0.840880275,Ad5d:0.859687924,Ah5h:0.840869904,As5s:0.920424938,Ac4c:0.57771337,Ad4d:0.623789549,Ah4h:0.57770896,As4s:0.675729036,Ac3c:0.530831993,Ad3d:0.564031303,Ah3h:0.530813932,KcQc:0.58761555,KdQd:0.74093926,KhQh:0.58761555,KsQs:0.717132151,KcQd:0.863264561,KcQh:0.584834337,KcQs:0.892801762,KdQc:0.655079961,KdQh:0.655079126,KdQs:0.927186012,KhQc:0.584837019,KhQd:0.863264918,KhQs:0.892803252,KsQc:0.894852698,KsQd:0.886640251,KsQh:0.894859672,KcJc:0.806807101,KdJd:0.882113576,KhJh:0.806807756,KsJs:0.833144367,KcJd:0.260895401,KcJh:0.235789523,KcJs:0.278947234,KdJc:0.221104532,KdJh:0.221104771,KdJs:0.281393975,KhJc:0.235789731,KhJd:0.26089561,KhJs:0.278948367,KsJc:0.271371305,KsJd:0.275553763,KsJh:0.271370322,KcTc:0.80757165,KdTd:0.924514651,KhTh:0.807571828,KsTs:0.86731571,Kc9c:0.863259017,Kd9d:0.899754465,Kh9h:0.863262355,Kc8c:0.0153277153,Kd8d:0.483972579,Kh8h:0.0153277507,Ks8s:0.445781678,Kc7c:0.00541189685,Kd7d:0.264011979,Kh7h:0.00541190151,Ks7s:0.256735653,QcJc:0.933168948,QdJd:0.933066487,QhJh:0.93316859,QsJs:0.890171349,QcTc:0.883787334,QdTd:0.958448768,QhTh:0.883787096,QsTs:0.916306257,Qc9c:0.868501246,Qd9d:0.914479017,Qh9h:0.86850822,JcTc:0.957032919,JdTd:0.963019729,JhTh:0.9570328,JsTs:0.912018538,Jc9c:0.865843356,Jd9d:0.912378788,Jh9h:0.865842879,Jc8c:0.174877137,Jd8d:0.237740129,Jh8h:0.174876258,Js8s:0.224678636,Tc9c:0.782523513,Td9d:0.831758201,Th9h:0.782522917,Tc8c:0.650407374,Td8d:0.772501051,Th8h:0.650405645,Ts8s:0.699584067,9c8c:0.736029387,9d8d:0.69014293,9h8h:0.736030579,8c7c:0.575469792,8d7d:0.738988757,8h7h:0.575467527,8s7s:0.724868059,7c6c:0.429018706,7d6d:0.749622464,7h6h:0.429017812,7s6s:0.698889017,6c5c:0.356386393,6d5d:0.378368556,6h5h:0.356384993,6s5s:0.392478555,5c4c:0.122438438,5d4d:0.124609843,5h4h:0.122438356,5s4s:0.126739964
#Range1#QdQc:0.00067006523,QhQc:0.000452663808,QsQc:0.00074933283,QhQd:0.000669583911,QsQd:0.0108797615,QsQh:0.000748965715,JdJc:0.32689029,JhJc:0.156113073,JsJc:0.357390195,JhJd:0.326866955,JsJd:0.400045693,JsJh:0.357371479,TdTc:0.653951883,ThTc:0.360923469,TsTc:0.621383607,ThTd:0.653952539,TsTd:0.702370524,TsTh:0.621389151,9d9c:0.521766782,9h9c:0.462607354,9h9d:0.521774888,8d8c:0.89656502,8h8c:0.792640328,8s8c:0.763873637,8h8d:0.89656502,8s8d:0.89654088,8s8h:0.763856053,7d7c:0.947154641,7h7c:0.797295868,7s7c:0.865231991,7h7d:0.947154403,7s7d:0.951084495,7s7h:0.865223289,6d6c:0.83527559,6h6c:0.367179692,6s6c:0.835107863,6h6d:0.835278869,6s6d:0.8865152,6s6h:0.835093677,5d5c:0.77025646,5h5c:0.625332355,5s5c:0.903715968,5h5d:0.770270109,5s5d:0.942052245,5s5h:0.903723657,4d4c:0.71117723,4h4c:0.760164261,4s4c:0.980995059,4h4d:0.711167693,4s4d:0.981265068,4s4h:0.980995059,3d3c:0.247712597,3h3c:0.140230551,3h3d:0.247817948,2h2c:0.0049877651,2s2c:0.00414513424,2s2h:0.00414388161,AcKd:0.441057384,AcKh:0.441038758,AcKs:0.44112581,AdKc:0.441038072,AdKh:0.441038072,AdKs:0.441127092,AhKc:0.441038758,AhKd:0.441057354,AhKs:0.44112578,AsKc:0.437785506,AsKd:0.441140026,AsKh:0.437785,AcQc:0.306839943,AdQd:0.307676762,AhQh:0.306839943,AsQs:0.305879503,AcQd:0.983776808,AcQh:0.972123623,AcQs:0.976322532,AdQc:0.983283818,AdQh:0.983283758,AdQs:0.940351725,AhQc:0.972090185,AhQd:0.983776867,AhQs:0.976306319,AsQc:0.852843463,AsQd:0.820890248,AsQh:0.852896214,AcJc:0.774354756,AdJd:0.811903059,AhJh:0.774367452,AsJs:0.707346439,AcJd:0.754811943,AcJh:0.782048225,AcJs:0.629811525,AdJc:0.663760245,AdJh:0.663756669,AdJs:0.7142272,AhJc:0.782059908,AhJd:0.754828215,AhJs:0.629831195,AsJc:0.450955182,AsJd:0.515891552,AsJh:0.450958312,AcTc:0.0432398804,AdTd:0.832727432,AhTh:0.0432396568,AsTs:0.38905862,AcTd:0.0317355655,AcTh:0.022571288,AcTs:0.42183274,AdTc:0.0165664367,AdTh:0.016566433,AdTs:0.225600496,AhTc:0.0225712582,AhTd:0.0317354277,AhTs:0.421866775,AsTc:0.197211042,AsTd:0.187032208,AsTh:0.197210953,Ac9c:0.534647524,Ad9d:0.463737577,Ah9h:0.53465867,Ac9d:0.00755624752,Ac9h:0.000835007464,Ad9c:0.0113637075,Ad9h:0.0113634476,Ah9c:0.000835177547,Ah9d:0.00755627174,As9c:0.0116019119,As9d:0.011637046,As9h:0.0116019417,Ac8c:0.00127658714,Ad8d:0.826066554,Ah8h:0.00127658702,As8s:0.336719573,Ac7c:0.000845471513,Ad7d:0.774956405,Ah7h:0.000845469942,As7s:0.381761938,Ac6c:0.000761093514,Ad6d:0.553585291,Ah6h:0.000761092757,As6s:0.388516575,Ac5c:0.632415235,Ad5d:0.657921851,Ah5h:0.632408917,As5s:0.236952767,Ac4c:0.658651173,Ad4d:0.743494272,Ah4h:0.658675313,As4s:0.299745649,Ac3c:0.776864409,Ad3d:0.850118041,Ah3h:0.776929438,Ac2c:0.81610316,Ah2h:0.81610316,As2s:0.815160573,KcQc:0.536858618,KdQd:0.820969462,KhQh:0.53686595,KsQs:0.725366175,KcQd:0.613495231,KcQh:0.517439663,KcQs:0.541640997,KdQc:0.283208221,KdQh:0.28320688,KdQs:0.410732657,KhQc:0.517437935,KhQd:0.613496184,KhQs:0.541643739,KsQc:0.678541243,KsQd:0.495505184,KsQh:0.678525448,KcJc:0.0108299246,KdJd:0.577744901,KhJh:0.0108299134,KsJs:0.426970989,KcJd:0.0106596621,KcJh:0.00895633548,KcJs:0.591433525,KdJc:0.00574227097,KdJh:0.00574227422,KdJs:0.527228057,KhJc:0.008956315,KhJd:0.0106596462,KhJs:0.591443062,KsJc:0.395096064,KsJd:0.328502566,KsJh:0.395099789,KcTc:0.00742454035,KdTd:0.797895908,KhTh:0.00742453057,KsTs:0.594503403,KcTd:0.00214525801,KcTh:0.00136887678,KcTs:0.226597309,KdTc:0.0010301302,KdTh:0.00103012996,KdTs:0.207709149,KhTc:0.0013688762,KhTd:0.00214525708,KhTs:0.22659725,KsTc:0.134425715,KsTd:0.159432903,KsTh:0.134426042,Kc9c:0.862722516,Kd9d:0.846732616,Kh9h:0.862735987,Kc8c:0.000881678599,Kd8d:0.245501935,Kh8h:0.000881678308,Ks8s:0.343196809,Kc7c:0.000716227922,Kd7d:0.0405815132,Kh7h:0.000716228446,Ks7s:0.442755103,Kc6c:0.000975628558,Kd6d:0.586961687,Kh6h:0.000975628966,Ks6s:0.329810202,Kc5c:0.0015517897,Kd5d:0.692625225,Kh5h:0.001551789,Ks5s:0.5763129,Kc4c:0.000322338077,Kd4d:0.0876197889,Kh4h:0.00032233799,Ks4s:0.0989911184,Kc3c:0.0049474258,Kd3d:0.00494785188,Kh3h:0.00494742533,Kc2c:0.166724473,Kh2h:0.166724488,Ks2s:0.166431323,QcJc:0.150782004,QdJd:0.601207256,QhJh:0.15077956,QsJs:0.200682282,QcJd:0.013644713,QcJh:0.0144798607,QcJs:0.346761018,QdJc:0.0116746211,QdJh:0.0116746388,QdJs:0.301224798,QhJc:0.0144797685,QhJd:0.0136446832,QhJs:0.346776605,QsJc:0.306211501,QsJd:0.250610352,QsJh:0.306212008,QcTc:0.277031422,QdTd:0.785348177,QhTh:0.277027428,QsTs:0.588105857,QcTd:0.0138896406,QcTh:0.0175928585,QcTs:0.104568072,QdTc:0.00629100716,QdTh:0.00629100855,QdTs:0.10138201,QhTc:0.0175932087,QhTd:0.0138896918,QhTs:0.10457354,QsTc:0.108370818,QsTd:0.0462163389,QsTh:0.108370587,Qc9c:0.733235478,Qd9d:0.727152228,Qh9h:0.733245671,Qc8c:0.000623562606,Qd8d:0.0799097046,Qh8h:0.000623562548,Qs8s:0.302903861,Qc7c:9.60039833E-05,Qd7d:0.000905111781,Qh7h:9.60038524E-05,Qs7s:0.0238682069,Qc6c:0.000128438143,Qd6d:0.00526957773,Qh6h:0.000128438041,Qs6s:0.0432019718,Qc5c:0.000159841584,Qd5d:0.000835880637,Qh5h:0.000159841424,Qs5s:0.0429462828,Qc4c:5.84706977E-05,Qd4d:0.000189870785,Qh4h:5.84706286E-05,Qs4s:0.0020278513,JcTc:0.482266337,JdTd:0.635904729,JhTh:0.48226589,JsTs:0.207474947,JcTd:0.391048789,JcTh:0.401658833,JcTs:0.184013993,JdTc:0.38049081,JdTh:0.380490839,JdTs:0.173138142,JhTc:0.401658773,JhTd:0.39104861,JhTs:0.184014514,JsTc:0.261911005,JsTd:0.279997289,JsTh:0.261909902,Jc9c:0.880965471,Jd9d:0.878846526,Jh9h:0.880965829,Jc8c:0.0014129742,Jd8d:0.579990923,Jh8h:0.00141297269,Js8s:0.589393973,Jc7c:0.000128070955,Jd7d:0.00266036228,Jh7h:0.000128070984,Js7s:0.10890805,Tc9c:0.763733029,Td9d:0.731077194,Th9h:0.763732016,Tc9d:0.181234136,Tc9h:0.180648103,Td9c:0.181396008,Td9h:0.181396127,Th9c:0.180648088,Th9d:0.181234181,Ts9c:0.167338103,Ts9d:0.181085318,Ts9h:0.16734387,Tc8c:0.548005283,Td8d:0.868070662,Th8h:0.548000038,Ts8s:0.861813903,Tc7c:0.00103453279,Td7d:0.631756783,Th7h:0.00103453314,Ts7s:0.653528929,9c8c:0.769036829,9d8d:0.656455874,9h8h:0.769024134,9c7c:0.874775827,9d7d:0.784055889,9h7h:0.874764323,9c6c:0.592827201,9d6d:0.557691216,9h6h:0.592846096,8c7c:0.00276471698,8d7d:0.602264464,8h7h:0.00276471791,8s7s:0.796588957,8c6c:0.00170674897,8d6d:0.332177669,8h6h:0.00170674943,8s6s:0.73931092,8c5c:0.000190815612,8d5d:0.00226282608,8h5h:0.000190815816,8s5s:0.13630797,7c6c:0.00125395751,7d6d:0.168372333,7h6h:0.00125395774,7s6s:0.765278161,7c5c:0.000879521074,7d5d:0.0148073556,7h5h:0.000879520667,7s5s:0.74906528,7c4c:0.000156938157,7d4d:0.000679176475,7h4h:0.000156938273,7s4s:0.224468023,6c5c:0.639345944,6d5d:0.524790108,6h5h:0.639321387,6s5s:0.57872504,6c4c:0.728300631,6d4d:0.746344984,6h4h:0.728308022,6s4s:0.632834196,6c3c:0.024677299,6d3d:0.0246159062,6h3h:0.024677299,5c4c:0.710851252,5d4d:0.667516768,5h4h:0.71085602,5s4s:0.10351333,5c3c:0.841328442,5d3d:0.681983829,5h3h:0.841377676,4c3c:0.886044145,4d3d:0.79606092,4h3h:0.886044443,4c2c:0.0288719572,4h2h:0.0288719572,4s2s:0.0283083525,3c2c:0.000189676721,3h2h:0.000189831742
#Board#9s 3s 2d Qc
#Pot#166
#EffectiveStacks#967
#AllinThreshold#80
#AddAllinOnlyIfLessThanThisTimesThePot#250
#CapEnabled#True
#Cap#4
#CapMode#NoLimit
#FlopConfig.BetSize#33.33
#FlopConfig.RaiseSize#3.2x
#TurnConfig.BetSize#66.66
#TurnConfig.RaiseSize#3.2x
#RiverConfig.BetSize#40 80
#RiverConfig.RaiseSize#3.2x
#RiverConfig.AddAllin#True
#FlopConfigIP.BetSize#33.33 50
#FlopConfigIP.RaiseSize#3.2x
#TurnConfigIP.BetSize#66.66
#TurnConfigIP.RaiseSize#3.2x
#RiverConfigIP.BetSize#40 80
#RiverConfigIP.RaiseSize#3.2x
#RiverConfigIP.AddAllin#True
PioSOLVER - postflop equilibrium solver for Holdem Quote
09-10-2018 , 08:35 PM
Nvm, disregard the above lol. Figured it out, some turns block our range more frequently that cbet the flop
PioSOLVER - postflop equilibrium solver for Holdem Quote
09-11-2018 , 03:15 AM
Quote:
I would expect to get greater difference in EV on the river using this method on certain river cards as opposed to measuring EV between 2 different trees at root.
Yes, disadvantage though is that if you make new river trees you won't be sure if introducing new river options doesn't change turn/flop play.

Quote:
Nvm, disregard the above lol. Figured it out, some turns block our range more frequently that cbet the flop
Yes, it's about blockers.
PioSOLVER - postflop equilibrium solver for Holdem Quote
09-11-2018 , 03:49 AM
hello again , today i runned many sims and after a while i have noticed my C disk space is a lot smaller than before , which is weird because i have saved all the sims in my D disk help please
PioSOLVER - postflop equilibrium solver for Holdem Quote
09-11-2018 , 04:40 AM
btw i don't know if it helps but i have been opening 3 instances of pio to make 3 sims at the same time every time , maybe that caused some kind of weird error ?
PioSOLVER - postflop equilibrium solver for Holdem Quote
09-11-2018 , 07:36 AM
Hi, I've got problem with my aggregation analysis. The report it spits out in form of "VIII.22","XI.33" etc. How to get rid of that? Thank you
PioSOLVER - postflop equilibrium solver for Holdem Quote
09-11-2018 , 11:38 AM
Quote:
hello again , today i runned many sims and after a while i have noticed my C disk space is a lot smaller than before , which is weird because i have saved all the sims in my D disk help please
I am sorry but this doesn't sound like anythign Pio related. If you saved the trees on D disk then they won't take space on C disk.

Quote:
btw i don't know if it helps but i have been opening 3 instances of pio to make 3 sims at the same time every time , maybe that caused some kind of weird error ?
No, it doesn't matter as long as all of those were saving trees to D disk.

Quote:
Hi, I've got problem with my aggregation analysis. The report it spits out in form of "VIII.22","XI.33" etc. How to get rid of that? Thank you
Use a decent program to open reports. Libre Office or Google Docs are recommended (both are free).
It probably happened because of some settings on your computer or the csv reader causing it to interpret the column values as dates or another object. Default settings in Libre Office should handle it without problems.
PioSOLVER - postflop equilibrium solver for Holdem Quote
09-11-2018 , 01:58 PM
In scripts we can use messages in format

Code:
	- add_line 0 35 105 245 525 1000
	- add_line 35 105 245 525 1000
	etc.
And what about commans to set FlopConfig.BetSize or FlopConfig.RaiseSize? Can we make tree sizing by script using Tree Building v2 format? Or current format is only for handly using?


Code:
#TreeBuilding#V2
#FlopConfig.BetSize#33.33
#FlopConfig.RaiseSize#3.2x
#TurnConfig.BetSize#66.66
#TurnConfig.RaiseSize#3.2x
#RiverConfig.BetSize#40 80
#RiverConfig.RaiseSize#3.2x
#RiverConfig.AddAllin#True
#FlopConfigIP.BetSize#33.33 50
#FlopConfigIP.RaiseSize#3.2x
#TurnConfigIP.BetSize#66.66
#TurnConfigIP.RaiseSize#3.2x
#RiverConfigIP.BetSize#40 80
#RiverConfigIP.RaiseSize#3.2x
#RiverConfigIP.AddAllin#True
PioSOLVER - postflop equilibrium solver for Holdem Quote
09-11-2018 , 02:00 PM
Quote:
Originally Posted by AgressiveDog
hello again , today i runned many sims and after a while i have noticed my C disk space is a lot smaller than before , which is weird because i have saved all the sims in my D disk help please
a) reboot your PC
b) delete "pioviewer_log.txt" and turn off logging in solver.

Did it helped?
PioSOLVER - postflop equilibrium solver for Holdem Quote
09-11-2018 , 02:47 PM
Any thoughts on implementing a playable bot that uses our piosolver .cfr solution files? I'm currently using simplegto but it has some bugs/kinks with pio.

Building on this question, I'm curious if there is a way to nodelock an entire tree at root (say force the solver to raise with 50% less frequency with its bluffs on all streets) and then solve?

This would be really useful for practicing against an exploitable bot that uses the nodelocked solutions. Currently I have to nodelock individual cards on individual streets, but this is really cumbersome when practicing against a bot that is using a large .cfr database of 50+ flops.
PioSOLVER - postflop equilibrium solver for Holdem Quote

      
m