Which version of the viewer are you using (About in the top menu gives that information)?
How big are the saves in comparison to RAM on your computer?

Version 1.9.2

Trees are between 100 and 170 MB each. I have 16GB of ram

I have not been able to use multiple opponents on the free version. It's either in position, or out of position, correct? Is this solver limited in scope to heads-up pots, or can you use multiway pots.

The viewer should have 4 digits in the version name.
Anyway, make sure you get the 1.9.2.5, you can download it from here:
https://www.dropbox.com/s/6p13pzi7j3...iewer.exe?dl=1

Yes, solving for equilibrium multiway is not viable.
There are two main reasons for it:

1)Even three way with full postflop play is so complex it's not possible to solve to good accuracy. It's possible to approximate using heavy abstractions but those solutions would be very far from real equilibrium

2)Equilibrium in 3+ way is shaky. You can follow it and still lose if someone else deviates. This means there is no "optimal" play in 3+way as there is in heads-up situation. This makes solutions for 3+way not very useful in practice.

Essentially at equilibrium you maximize your EV and minimize your opponents EV, while he is doing the same to you.

The issue is with 3 way pots you can only minimize one opponents EV and maximize your own. So the 3rd player might benefit from your actions which means, as far as we know, there is no equilibrium in multiway games (or at least not necessarily).

There exists equilibrium for 0 sum 3+ way games. The problem is that the equilibrium is "shaky". What that means is that no player can improve their EV by alterating their actions alone (that's what the definition of the equilibrium is) but two other things can happen:

1)one player alters their strategy and they hurt both you and themselves (to the benefit of 3rd player)
2)two other players alter their strategies and they net the benefits while you lose (they can either split the benefits or one of them can gain and other lose but overall they take EV from you).

This is just how the math is. While we may hope to find (or approximate) the equilibrium for multiway games it's not stable nor "optimal" in a way HU equilibrium is.

I cant seem to update it. It still says 1.9.2.0 and it says it is up to date when I use the updater.

Just substitute the file I linked in your Pio folder.

I sent this to them in an email and no response yet...does anyone know any of these answers?


1) If I later feel that I want to upgrade to Pro or Edge, will the purchase price of the lower version be applied to my upgrade, or do I need to pay full price for upgrade?

2) If I want to put Pio on 2 computers, my laptop and desktop (both owned by me), do I need to purchase Pro as it says it includes 2 activations? Or can I purchase one Basic license and still put it on 2 computers?

Thanks!

Your email was sent 10 hours ago during the Saturday night. I advice a bit of patience. Emails are answered within 24 hours, usually much faster than that. We are located in Europe (Central European time) and we sleep during the night (usually), especially during weekend

Answering your questions:

The upgrades are exactly the difference in prices during the first month and then the difference +24$.

You need the pro (or edge) version. That's what "activations" mean.
You can also move them to another computer in the future (within reasonable limits) for purposes like new computers, long travelling etc. With the edge version you can move it around as much as you want without any limits.

I generated a script for 49 flops, flops only and when I try and run a mutiple file aggregate report I get this error.index was outside the bounds of array.

Sent from my QTAQZ3 using Tapatalk

Are you using PioViewer 1.9.2.5? (The link is a few posts above).
Are you sure there are only trees from one script in the folder?

I think it i might have another tree in the folder, but can't figure out which one it is. Thanks for the response.

hey, I just bought pio basic. Everything works fine started playing with it after 5 minutes or so and i am verry happy i finally bought it!
I recorded about 6 hands/spots i struggle with for study and solved them in piosolver at the end of my session, 3bet spots with narrow ranges went reasonably fast.
Now it seems that when i run a SB rfi range vs BB call range my computer is taking quite some time untill it has reached 0.35% accuracy, with 50,100 bet sizes and 3x raise size parameters
(i think -correct me if i am wrong- 50% & 100% size option gives me a decent idea of a ideal betsize for cbets )
I am a microstakes player that is new to GTO (been playing a "exploitative" monkey style of poker for a while now), my expectation of the calculation is to learn with what hands in my range i should cbet with and what size is most optimal.
Having this said i want to keep it simple for now, learn a bit everyday and be able to apply what i have studied on the tables.
So my question is: for trees that are larger to solve, are there any parameters/accuracy settings that i can change to get my desired result considerably faster?
Apologies for my english and this wall of text to explain my situation.

It would be useful to see the whole config ("copy to clipboard" button, paste to pastebin.com and link) but in general using two bet sizes at every point is already quite a big tree so it's expected that it takes a while to solve. You may think about eliminating donk betting on turn/river for example.

I encourage you to start with only one bet sizing and then add them slowly later. It's a good idea to use only one on the flop for example and 2 on the turn/river.

Don't use "add-allin" checkbox lightly as it adds another bet size and makes the tree considerably more difficlt to solve.
Use only one bet size on the flop in most cases.
Eliminate donk-bets on the turn/river.

It's always a trade-off but those are relatively safe things to do.

okay everything you said makes sense, thank you.

how about changing the desired accuracy reached from .35% to 1%-3%?
Since I am not capable in a zoom NL5 game to pinpoint accurate bb v sb call villain ranges to run my range against.

not sure if im understanding u however

solving to →100% accuracy will always give u a better insight, no matter how rough ur premises are (otherwise ur just adding extra wrongness, and wrongnesses compound to become a larger wrongness)

Ok so now its up to date but I still get the same exception: System.outofmemoryexception.

I started my first preflop calculation today.. I built a tree and pressed the "go"-button. on the right side it says "SOLVER started."

1) Is there a way to watch the progress of the calculation? when do i know its finished?

2) Is there a need to use accuracy settings? if yes, which settings are recommended?

Hello,
I recently purchased PIO solver, and am getting a "TurboActivate.dll" file missing error message when trying to open PIO solver. I do see the file in the folder. I'm running Windows 10.
Any help would be greatly appreciated

1% is likely too much unless you only really need ballpark frequencies. I think 0.5% is a good outline if you don't care about details on the river being very close.

We measure accuracy as exploitability per hand (by perfectly adjusting opponent who knows your strategy exactly) so 0% is perfect one.
It's about trade-offs, getting good accuracy takes more time. At some point it doesn't make sense to waste time and electricity when the solution is already almost perfect (at least from practical point of view)

It happens when trying to run reports on full saves sometimes.
I am PM'ing you an unoffical fix for this issue.

It's the same as in postflop one, the info will be printed on the right side in treebuilding and calculation tab. It will just take a longer while on big preflop trees for it to appear.

Aiming at 2bb-3bb/100 should be good enough.

Both are explained in the introductory video for the preflop solver

Have you used the updater and followed the instructions? If no then please do. If yes, please send the screenshot from C:\PioSOLVER to support@piosolver.com and we will take a look what's wrong.

Hi,

as far as I know, the last major update (with improvments/new tools) came out in September 2016. Are you guys still working on the project Piosolver "1" or do you have a new project? like a PLO-Solver? (what I would really like to hear )
Or is a Piosolver 2 coming soon?

Thanks for your time

kind regards

Is there an update in the making where the "set_mes" command is fixed?

That's nature of project when there is one time payment. At some point the devs need to mainly focus on new things.

There will be at least one free update (1.10) version.

New things are ready by now but the release is getting postponed. It should happen soon enough. There are also new things we have in dev for the main Pio product but they have to wait.

set_mes works in dev and will work in 1.10. It's not the first thing to be released though.

Excellent. The reason I ask is that set_mes + the option to cut branches from a solved equilibrium (an option I hope you will keep) will be very useful for simplifying complex strategies in a systematic way against a strong opponent who isn't aware of every little simplification that we make:

Calculate an equilibrium with a complete set of options, snip some branches (say, choosing not to have a check-raise range on a certain board texture) and use set_mes to find our new maximally exploitative strategy with our reduced set of options, while Villain still plays the original equilibrium strategy.

This would model playing against an opponent who plays a strong strategy, but is not able to perfectly counter every simplification we make (in other words, a useful model of a strong-but-not-immediately-countering human opponent).

Give me set_mes and don't touch the branch cutting functionality, and I'll be happy as a clam. ;-)