I'm kind of a newb to piosolver. I'm trying to learn how to do node locking right now. I'm node locking the IP player as never raising flop in one example and raising more hands in the other example. I'm noticing that the OOP player's cb frequency with every part of his range is not changing between the two examples. This leads me to think that I'm not node locking correctly. Because you would think if there is a drastic shift in raising strategy by the IP player, the OOP solution would change frequencies right? Anyone know what I'm doing wrong?

Make sure to run the solver again after node-locking is done. Node-locking itself just changes the strategy at your chosen node but doesn't in the rest of the tree (as that's not that easy to calculate). Once you run the solver again (click "go") it will start slowly adjusting to the new conditions (your node-lock).

hi, im trying to simplify my betsize strategy, but my question is, by comparing the overall EV from 2 sims enough to see this? also OOP and IP MES do i have to take this ev into consideration aswell?

Save I have a directory of solves (full saves) across many different flop. Is there any way to run a script or something to go through each and re-save them as small saves?

Thanks. It helped. Any tips for properly feeding the output back into my Python script?

Well, I would need to know exactly what you are doing to form a judgement. In general though yes - you can compare EVs at root (overall EV) to see if adding a particular option is important strategically or not.
You don't need to take MESes into account. Once the accuracy is decent those will be close to EVs anyway.

There is no tool in Pio to do that. You could generate a script like:
and then run it but unless you have a tool to list all the file names in a folder it will mean a lot of typing/editing in notepad to cover all the files.

Not out of the top of my head but I will check if we happen to have some Python code to connect to the solver. Can you please send me a PM so I don't forget to send it to you in case we have it?

Look at this sample code:

https://piofiles.com/other/piosolver...example_python

Oh that's great. Thanks a ton.

Hey guys. So I’m new to Pio and purchased Pro recently. I attempted to run a script for a 95 flop sub-set using BU open and BB 3 bet ranges, 3 bet sizes per street and one donk size for OOP on the turn and river. I originally ran this sim on a dell laptop with an [intel Core i5 8th gen quad core processor] 64 bit with 8GB of memory. The script took about 10 mins per flop and worked my computer pretty hard.

I then tried running the same script on a more powerful laptop with i7-9750H CPU and 6 cores with 16GB

The script still take close to almost 10 mins per flop.

I am running them exploitable to 0.10 of the pot.

So my question is what specs do I need On a PC/laptop to run these scripts faster and how fast can they be run? How long should they take, and what computer could handle these sims easily?

Appreciate any advice guys!

Easy peasy. Thanks!

In general speed is proportional to the number of physical cores multiplied by the based frequency they run on (+around 10-15% for hyperthreading). There might be many system specific things influencing the performance though, most common are:

1)Energy saving mode (this completely kills the performance and is a very common default on laptops)
2)3rd party antivirus (or other junkware)
3)other CPU intensive programs running at the same time as Pio
4)anti-malware scans running at the same time as Pio
5)Intel security mitigations being installed slowly chipping in on the performance (Intel CPUs have tons of security issues which are coming to light by the day and mitigations to those usually makes them noticeably slower)

Please consider that laptops in general use CPUs at lower frequencies as well. If you are looking for a performant computer it's the best to avoid laptops.

Hi, I got a new machine and wanted to install the Pio Solver over there. How can I move my one license key?

Is there any advice on which server set up to use for pio? Im sure there was a thread somewhere on the forum? Id like to know like requirements to run HU SRP solves on 1755 flop subset at a decent speed etc.

I understand this is rather vague, I just dont want to waste money going overboard

Is it possible to run a script over multiple flops with different stack sizes in each sim?

Please follow the instructions here:
https://licences.piofiles.com/

Not really. You need enough RAM to build the trees and the faster CPU you get the faster it will be. There isn't really an optimal setup. The more money you are willing to spend the faster it will be. Providers with the best deals are in general Contabo and Online.net

Right now it's not possible to do automatically (you would need to join the scripts manually and that's not very convenient). It will be possible in the hopefully near future.

Hello.

Just one question, whether it worth for me to use Piosolver against unknown opponents.

I see that many professionals when use PIOSOLVER against unkown opponent...
Consider that opponent has a good range, close to GTO (similar or equal to what HERO itself uses when in place of oponnent)

My fear is that if I do that, I could lose money against unknown players who play a lot of hands or a few hands.

So my question is: if I put in PIOSOLVER that my opponent has a decent range, but he uses a not very good range ... Could I lose money? Or the opponent's preflop errors would make me profit?

Thanks.

The way GTO works is that your opponent losses more by not playing the optimal range than they gain by "exploiting" it later. Yes you will make mistakes against an opponent with a tighter range (for example) on the river if you follow GTO as you will be calling too much but those mistakes will be smaller (on average) than their mistakes of folding too much on earlier streets.

The definition of equilibrium is that no player can improve anymore. If it was possible to have a different range on turn/river and profit from that then it would mean it's not equilibirum.

It's a bit more complicated when you assume preflop ranges. Here you only guaranteed to do well if your assumption is close to GTO. In multiplayer games there is additional concern of collusion (even not intentional) where one of the players deviate from equilibrium and others lose money. While those concerns seem to not matter much in practice they are a theoretical problem in multi-player games.

Hello, can you answer my questions?

I've answered them here:
https://forumserver.twoplustwo.com/s...postcount=5658

(bottom of the post)

I think I've answered your question via email already but let me repeat it here:
No, it's not true that we should bluff the lowest equity hand. In general the best hands for bluffs are hands which block calling range and don't block folding range.
"Bluff with lowest equity, check with higher to somehow win at showdown" is a rule that only applies to toy games with numbers and not to real poker.

It seems to me that your expectations about how the solutions should look like are based on simplified rules. Unfortunately those rules don't really work (along with other things like 1-a defense frequency for example).

email? I haven't seen it?

Sent you another email.

OK thank you!

Just to see if I understand ...
Since Piosolver POSFLOP asks me to put an opponent's range...
If I insert the opponent's range come from the preflop was a decent range (not too far from the GTO)...
I won't lose money to him, even if the opponent's range that come from the preflop was actually be NIT (tight) or maniac (too loose)?

In other words...
My opponent's mistake in choosing his preflop cards ...
They make me profit even when I put the wrong range of the opponent in PIOSOLVER?

Thanks!

Hi,
- I would like to buy PIO EDGE
- Run PREFLOP sims for 100bb deep Cash Games, HU/3Max/6Max

1. Is a 64GB Ram enough to correctly solve these preflop trees (What's the minimum RAM requirement actually, is 16GB enough?) What minimum AMD CPU would you recommand?

2. If it's not too much to ask, could you recommand a PC config? (2K$ budget or more)

3. Would you rather pay for a dedicated server to run preflop trees or is a new 2K$ish PC enough to do the job?

when running aggregated reports I always notice there is 1 empty flop space in excel files with filled in numbers ... seems to always happen ... can ask what causes this to happen? And if you are looking to get accurate averages in excel it should just be thrown out right?

...it doesnt appear to be the averages.